Breast Cancer Research and Treatment 12:107-149 (1988) © Kluwer Academic Publishers - Printed in the Netherlands

Abstracts ~FICACY OF AD3qBVAk~C~Ev~q}~RAPY IN OI~KRA~LEN00E NEGATIVE ~RKAST CANCER. AN INTeRgROUP STODY EST 1180. E.G. ~ * , R. (~AY, A.H. ~ATILA, C.K. 0~BCR~, D.C. TCRMEY, K.G. G ~ I S ~ , R. COOPER, G. FAIKS(INAh~ pARTICIPATING ] I q V E S T I ~ , CLEVELAh~ ~IRO~OLITAN ~O~PIT~L, CASE WESTHRN RES~VE U ~ I T Y , CLE~, OHIO 44109. Four htmdr~d six female patients who %alderwenteither a modified radical m a s t e c ~ or total m a s ~ with low axillary node dissection for pot~tially curable breast carcinoma, with estrogen receptor negative (~-) ttmDr or ~ + tu~or equal or greater than 3cm, and having no histopathological evidence of axillary node involvement were randomized to receive adjuvant chemotherapy (C~) or no trea~_nt observation. ~ e CT consisted of six four week eyoles2of Cyclophosphamide (C) iO0~c3/M p.o. day I-I~, Methotrexate (S) 40mg/M I.V. day I ~ 8, 5-Fluorouracil (F) 600rag/M- I.V. day 1 and 8 with Prednisone 4Omg/M p.o. day 1-14. Rar~o~izati~l was stratified on type of surgical ~ocedure, menopausal status, t~nor size (TS) and ~R status. T r e a ~ n t s were balanced with respect to all patients characteristics. Three year disease-free survival rates (DPS%) and p-values for log rank tests are given by trealmlentfor a variety of patient groups. GROO-P

(~

OB~VATI~

p-values

Overall ]~+, TS~3an

--~--

69

,0001

87 82 85 83 90 84 83 84

63 71 72 60

.003 .005 .001

71

.02 .02 .05 .09

~I~-, All TS Pr~pausal Pos~enopausal Premenopausal, ~ + ~usal, ~RPoslm~enopausal, ~I~+ Pos~m~%opausal, ~R-

72 55 62

.01

We conclude that adjuvant C9 for node negative, ~i~-or ER+ TS> -- 3~, provides a sic~ificant difference in DFS over observation at 3 years follow-up. ~ner~ is no difference in overall survival at this time.

3

HER-2/neu ONCOGENE PROTEIN AS A PROGNOSTIC FACTOR IN BREAST CANCER. AK Tandon,* GM Clark, A Ullrich, end WL McGuire, The University of Texas Health Science Center, San Antoni% and the Department of Molecular Biology, Genentech, Inc, South San Francisco. Amplification of the HER-2/ne__~uoncogene has been shown to predict poor clinical outcome in node-positive breast cancer patients. It seems reasonable that expression of a gene protein product may be more closely related to aggressiveness of a tumor than gene copy number. We therefore examined expression of the HER-2/neu oncogene protein fer its prognostic potential in both node-positive and node-negative breast cancer patients. Using a rabbit polycional antiserum produced against the carboxy terminal synthetic peptide, the HER-2/ne_uu protein was determined in 72g primary human breast tumor specimens by Western blot analysis. In node-positive patients (n = 3.50) significantly higher expression of this gene was found in tumors without estrogen receptor (P = 0.02) or progesterone receptor (P = 0.0003), and in patients with greater than three positive lymph nodes (P = 0.04). A significant correlation between expression of the HER-2/neu gene and its amplification was also found (n = #8, P < 0.001), though there were also a substantial number of tumors with significant expression of the HER-2/ne__uuprotein without any amplification of the gene. In these node-positive patients univariate as well as multivariate regression analysis showed that expression of the HER2/ne.....~u gene is a significant independent indicator of both the disease-free and overall survival o~ the patient even when other prognostic factors are considered. By contrast, in node-negative patients (n = 378), expression of HER-2/ne._.u_u does not predict disease outcome. Conclusion: Node-positive patients with higher expression of the HER-2/ne__u_ugene have statistically shorter disease-free and overall survival times than patients with lower expression of this gene.

ELEVATED GLUTATHIONE S-TRANSFERASE ACTIVITY IN VINCRISTINESELECTED MULTIDRUG RESISTANT MCF-7 HUMAN BREAST CANCER CELL LINES. Bridget T. H i l l * , Louise K. Hosking and Richard D. H. Whelan, Laboratory of C e l l u l a r Chemotherapy, Imperial Cancer Research Fund, London WEBA 3PX, England. We reported e a r l i e r (PAACR,25,331, 1984) that a f t e r exposure of MCF-7 c e l l s to f r a c t i o n a t e d X - i r r a d i a t i o n i n v i ~ o the r e s u l t a n t cell lines expressed resistance to both v i n c r i s t i n e (VCR) and etoposide (VP16). To establish whether d i f f e r e n t mechanisms operate or predominate depending on the procedure used to select f o r or 'induce' resistance we have selected a series of VCR- or VPI6treated r e s i s t a n t c e l l l i n e s . These drug-selected sublines expressed the c h a r a c t e r i s t i c multidrug r e s i s t a n t (MDR) phenotype in terms of patterns of cross resistance to other classes of antitumour drugs. In contrast the X-ray-pretreated lines showed no cross resistance to the anthracyclines, so providing another example of atypical MDR. Determinations of levels of glutathione and associated enzymes revealed another major difference between these independently-derived r e s i s t a n t sublines: X-ray-pretreated c e l l s showed no modifications from the parental c e l l s , contrasting with s i g n i f i c a n t elevations of glutathione S-transferase (GST) in the drug-selected resistant cells. The extent of GST elevation (4-8 f o l d ) correlated with the level of VCR resistance expressed (5-15 f o l d , comparing ICs0 values). These data appear to provide the f i r s t evidence of s i g n i f i c a n t elevation of GST a c t i v i t y in VCR-seleeted r e s i s t a n t c e l l s , so extending the range of antitumour agents c u r r e n t l y known, d i r e c t l y or i n d i r e c t l y , to modify the GSTs. The s p e c i f i c isozymes i n v o l v e d a r e c u r r e n t l y being investigated. These data are consistent with the hypothesis that d i f f e r e n t mechanisms operate or predominate depending on the s p e c i f i c way in which drug resistance was derived.

PROGNOSTIC VALUE OF HISTOLOGIC GRADING IN N-- BREAST CANCER, with proposed grade modification based on m u l t i v a r i a t e analgsis of 1282 patients. V. 6e Doussal, M. Tubiana-Hulim, S. Friedman, K. Hacene, F. Spgratos, M. Brunet, J. Rou~ss~. Centre Ren4 Huguenin, 92211 St. Cloud, France A m u l t i v a r i a t e analysis o? 1262 patients with eperable, invasive, ductal breast carcinoma was per£or'med to assess the prognostic ?actors and in par'ticular the value o? the three components o? the Scar.t.?-Z:Ioom-Richardsor, histologic

grading (SBR[. Median time on stud~ was 3 gears ( I - I 0 ) . Overall, nodal metastasis and histograde ~ere the tug most important survival tact ors f o r the time ,if" the study (p=ixlO -9 and 5xi0-5). For N- patients (n=654), ?or DFS, stage (p=Jx10-4) and SBR grade (p=O.01) were the importm~t ?actors; and ?or' Metast.aticFS, grade (p=4xlO-4) and stage (p=0.003) at 5 gears. O? the three components of" the SBR gr.ade ever-all,

tissue

dif'gerentiation

was

the least, and

nuclear ?actc,rs the most discriminaterg (with pleomorphism the highest). The combined nuclear ?actors r.eplaced the o r i g i n a l SBR grade in a l l models. BU rearrangement of` the nuclear scores, a new SBR (NFSBR) was ?greed, which had much higher risk values and better risk separation c,f" patient subpopulati,ans

(usuallg over separate groups

50% of' with

than the original SBR. SBR Grade II patients) was r e d i s t r i b u t e d into 3 statisticallLj diCfer'ent risk ratios

(p=3xlO-S), creating 5 s t a t i s t i c a l l g tiff`far ant sub,2roups over.all. Ir, N- patients, the NFSBRreplaced the SBR and became the m o s t importarlt f a c t o r f`or relapse prediction b,ath ?c,r ])FS (p=IxlO -4) NFS (p=2xlO -6) . This olassif`icatien is more accur.ate and predictive than the star,~at-.dBBR grade and would be particularlu use£ut where the nodal status of

the patient is negative or ur,knc,wm. Patients with high grade node negative breast cancer should be considered ?or adjuvant t.her.apg t r ' i a l s ,

108

Abstracts NODE-POSITIVE (N+) RESECTABLE BREAST CANCER: 5-YEAR RESULTS AS A FUNCTION OF SH-THYMIDINE LABELING INDEX.R.Silvestrini*, N.G. Daidone, P. Valagussa and G. Bonadonna. Istituto Naziohale Tumori, 20133 Milan, Italy.

INI~H: A ~ , . M~V I N T ~ O N SlT~ IN P R E C ~ C E R ~ S Mo~Eathlc.Ml'6. , ~ . I ~ . , Jones.R., Rao,T.R., F~zak,C~., L~e-C~IaII,ES., Hither,M6.,

N~O~.

T ~ / l o r , c ~ . , c o r s m 1 1 , ~ . , ~ o r r i s , V 5. IGrane B i o ~colo~ Institute, "l~T,°14ichiga~ cancer Foundati~,

In a changing reality, traditional staging systems are integrated by biologic concepts. The 3H-thymidine labeling index (SH-thy LI), which has proved to be an important prognostic factor in N- breast cancers, has been used to analyze the impact of systemic adjuvant therapy on N+ resectable tumors. In a series of 557 patients treated with surgery and CMF~ Adriamycin, a direct relation was observed between 3H-thy LI values and B-year risk of relapse and death. When the proliferative index was used as a qualitative variable, a high SH-thy LI.predicted for significantly higher relapse (SI% vs 34%, p=O.O006) and death rates (29% vs 17%, p=O.OOS5). These findings held true regardless of the addition of Adriamycin and the numbers of positive lymph nodes. Moreover, a high SH-thy LI was associated with a higher probability of distant metastases (Z9% vs 26%, p=O.OOS) and with a trend to develop local-regional relapses (21% vs 13%, p=O.08). However, modulation of response to therapy as a function of administered drug dose and proliferative rate was observed in the subset of <3 N+ tumors. Patients with high 3N-thy LI tumors had a lower probability of relapsefree survival than patients with low 3H-thy LI tumors (52% va 78%, p=O.O13) when low doses (<85%) were given but not when full planned drug dosage was administered (70% vs 78%).

~ I H , ~niv. of Western Cmtario, ~ Dept. of Biol. Scie~ce~, Oakland ~htlverlsty, RochesteE, MI 48309-4401. ~ ~roi~i~ ~ in ~ib/c mice by c~miceis and ~ z a ~ e s often contain ore or more newly integrated )EV proviral c x ~ i ~ in IINA(l~irol. 148:360,1986).

Three different high tumor incidence hyperplasias(C4,CS,D2) had one MTV provirus integrated at the s a ~ genumic site (designated int-h), suggesting an insertlc~ml m u t a g ~ i s mechanism in chemical carcinogenesis(ibid), i% partial SausA ger~mic libl-ary was prepared frcm D2 hyperplasia 11~ (which contained only one newly inteqrated N~V at int-h)in the rec A-B-C- Charon 35-DBI161 systeB. MTV ~ and endogenous virus flanking sequence probes were used to obtain a clone which conta/ned a c~plete MI~ copy and 6kb of 5,flanking gencmic IE~. HindIiI digestion yielded a 2kb f ~ t of this flanking ~ which hybridized uniquely to int-h. This fragment was sub~oned in pUCI9 for ~ i n g and expression studies.Int-h mapped to ~ 9,confirming restriction mapping data that the r e g i o n is a u n i q u e routine M ~ i n t e g r a . tion site. No allelic polymorphism was observed at int-h in 5 highly divergent iz~bred mo~se strains. The 2kb int-h sequence is highly conserved in several mammalian species inclu~ng man, a ~ is e ~ r e s s ~ in ~oth mouse and human m ~ x ~ r y cell I ~ . The relative cx~tent of int-h RNA in m ~ m ~ r y ntK~l*-~t,a w i t h ~ i n t e g r a t e d at that locus and in non~l ~ y tissues suggests that MI~ integration at int-h disrupts eE~ression of adjacent cellular genes. ~rted by NIH gl-a~ts (1%44901, C~36399 and an NCI

Supported in part by a grant PFO 87.02813.44.

grant.

7

MODEL STUDIES IN NUTRITIONAL- AND CHEMOPREVENTION OF BREAST CANCER. G N Schrauzer*, T Molenaar, K Kuehn and D Waller, University of California at San Diego, La Jolla, CA, 92093. To demonstrate effects of diet on mammary tumorigenesis, relevant to human breast cancer prevention, model studies were conducted in which female C3H mice carrying the murine mammary tumorvirus (MMTV) were fed simulated American, Bulgarian, and Japanese human diets over their entire postweaning life spans. The incidence of mammary tumors was the highest (82%) in groups of mice receiving the simulated, meat- and fat-rich American diet, which also was low in selenium (0.15 ppm). It was significantly lower (26.5%) in the animals on the simulated Bulgarian diet, which is high in complex carbohydrates (cereals) and adequate in Se (0.25 ppm). These results parallel the variations of human breast cancer incidences in the populations of the two countries. The appearance of the mammary tumors in the groups of mice on the simulated Japanese diet was delayed~ but the final tumor incidence of 46.5% was higher than expected from the comparatively high Se content of 1.07 ppm of the feed in which fish meal was the major source of protein. Since the bioavailability of Se is low in fish meal, an additional life-term experiment was performed in which the mice on the simulated Japanese diet received supplemental bioavailable Se. In this group, the final tumor incidence was oniy 10%, with no adverse effects on survival. These model studies appear to reproduce existing variations of human breast cancer mortalities in different countries and support approaches for prevention.

8

SIMULTANEOUS FLOW-CYTOMETRY (FCM) MEASUREMENT OF PROGESTRONE RECEPTORS (PR) AND DNA INDICES IN BREAST CANCER CELL LINES. ML Graham II*, PA Bunn Jr., KB Horwitz, Divisions of Endocrinology and Ontology, University of Colorado Health Sciences Center, Denver, CO 80262 In addition to axillary lymph node status, factors predicting recurrence and survival in breast cancer include PR, estrogen receptors (ER), % S-phase and aneuploidy. As the first phase in the development of a FCM method that will simultaneously assess PR, ER and DNA indices in small tumor samples, we have used human breast tumor cell lines as models to measure PR and DNA indices. DNA is measured with propidium iodide (PI), and PR with our monoclonal antibodies: AB-52, specific for both the A and B proteins of human PR; and B-30 or B-~4, specific for the B protein. PR-rich, hyperdiploid T47D cells (3xi0 V) are fixed, permeabilized, and first exposed to the MAbs sing y or in combination, or to an isotypic control antibody. Next, the cells are treated with 2rid antibodies - - goat anti-mouse IgG or F(ab') 2 fragments conjugated to fluorescein. Finally, cells are exposed to RNAse and PI, and 15,000 nuclei are evaluated by FCM. By this assay, 95% of T47D cells have high levels of specific PR, with extensive cell to cell heterogeneity in receptor levels. PR measurement and DN~ histograms are optimal with ceil numbers ranging from 1.5 to 4x10~o cells per assay tube. PR are present in all stages of the cell cycle, but are higher in S and G2M; and are increased by pretreatment of cells with progestins. AB-52 plus one of the B-specific MAb's give PR levels similar to all 3 MAb's, suggesting that the B-specific Ab's bind adjacent amino-terminal epitopes. No specific PR are measured in PRnegative MDA-231 cells. Using this assay, we have detected a spontaneous DNA duplication in our T47D cells, rendering them hypertetraploid; the variant cells have even higher constitutive levels of PR. These variants have 5 to 10 copies of chromosome l l, which contains the PR gene. We are now modifying the assay to ceil lines with lower PR levels and similar tumor specimens. An assay that can quantitate PR, assess tumor ceil heterogeneity, and measure DNA indices in a small tumor sample should prove valuable in the management of breast cancer.

Abstracts 9

11

PROGNOSTIC SIGNIFICANCE OF NUCLEAR FEATURES IN PATIENTS (PT~ WITH BREAST CARCINOMA (BC) TREATED WITH ADJUVANT FAC-CHEMOTHERAPY. JY Ro, M Santamaria, AU Buzdar, TL Smith, S Kau, AG Ayala. UT M. D. Anderson Cancer Center~ Houston, TX 77030. Because of the complexicity of the nuclear grading system in BC, in an ongoing histo-pathologic prognostic factor evaluation, we analyzed each of the individual nuclear parameters in nuclear grading, i.e., nuclear size, nuclear pleomorphism, nucleoli, mitoses and chromatin distribution, in a group of initial 115 consecutive cases of BC who had survey at our institute and slides and tissue blocks were available for review, who received adjuvant FAC chemotherapy after mastectomy. Median follow-up of t h e 6 7 pts still alive was 7 years and an estimated 8 years survival was 57%. 72 pts had stage II and 43 had stage III disease. Distribution of pts according to lymph node involvement was as follows: <-<-3 nodes, 42 pts; 4-10 nodes, 46 pts; >I0 nodes, 27 pts. 56 tumors were nuclear grade (NG) I, 52 were NG 2 and 7 were NG 3. Survival was somewhat shorter among pts with NG i tumors (5 year estimate 62% compared to 71% for remaining pts) but the difference was not statistically significant (P=0.12, Wilcoxon test). Individual nuclear features associated with significantly shorter survival experience were larger nuclear Size (P=0.01), nuclear pleomorphism (P=0.04), and increasing number of mitoses (P=0.04). Factors showing no evidence of association with survival were 2 other nuclear features (prominance of nucl¢oli and uhromatin contents); extensiveness of nodal involvement; extranodal extension; vascular, lymphatic or perineural invasion. In a multivariate analysis which adjusted for differing distributions of other important covariates (stage, nodes, menopause status), nuclear size contributed significantly to the fit of the survival model~ This prognosti c information for adjuvant-treated BC pts suggest that individual nuclear features may be of more value than the over all nuclear grade.

12

GRADING OF MALIGNANCY IN PRIMARY BREAST

CARCINOMA EVALUATED ON TISSUE IMPRINT CYTOLOGY ~A. Guazzi. R. Nizzoli, L. Savoldi, G. Bisagni, A. Bowini, G. Ceecoul Medical Oncology Service, Health Physics Dept., Ospedale Regionale, 43100 Parma. Italy Grading of malignancyin breast carcinoma is commonly assessed using nuclear or histologiccriteria. Since fine needle asp/ration (FNA) cytology is increasingly used in diagnosis of breast carcinoma, a cytologic grading should be advisable.In order to find suitable criteria for a possible FNA cytologic grading, in this study we assessed the prognostic meaning of several morphologicparameters on tissue imprint cytology. A consecutiveseries of 75 breast carcinomas submittedto primary surgical treatmeat and provided for ER assay in 1981 was examined. Sixteen cases were excluded from the study because of poor cellular preservation and too scanty material. A single numerical system was applied to 14 cytomorphologicparameters. Each parameter was coded as a single score from I to 3. according to increasing severity. The overall final score (OFS) was obtained adding all the different scores attributed to each criterion. Tumors having score below the 33rd percentyle were considered as low grade (G1), score from the 33rd percentyle to the 66th as intermediate grade (G2). score over the 66th pementyle as high grade (G3). The following table shows the five-year survival related to the OFS, to each single parameter showing a statisticalsignificancyor a trend, to the three most significantparameters CrSP). 5- year survival (%) G1

G2

G3

10

P

OFS 79 59 52 0.07 Nuclear shape 78 37 40 0.003 Nuclear pleomorphism 80 53 46 0.03 Cellular shape 77 56 50 0.04 Nuclear size 87 55 50 0.05 Cellular pleomorphism 71 63 45 0.09 TSP 83 63 45 0.01 Our study suggests that this type of cytologic grading could be used as prognostic indicator at the time of primary biopsy in breast carcinoma. Moreover, the most significantcriteria could be considered for an eventual FNA cytologic grading.

109

INTRA- AND INTEROBSERVER VARIATION IN PATHOLOGICAL GRADING OF BREAST CANCER.*E Chouinard, V Chen, T D'Souza, R Riddell 71cMaster University, Hamilton, Ont.,Canada. L8N3Z5 Intra- and interobserver variation in pathological grade(PG) was measured to determine (1)level of agreement (2)usefulness in identifying a subgroup at high risk of relapse. From Jan-May 1988, 3 pathologists on 2 occasions graded 39 cases of breast cancer chosen at random. Slides were coded differently each time. Pathologists did not confer with each other. This small sample was representative of a larger breast cancer population(80% StageI-II,20% ~50 y.8., 70% invasive duct cancer, 34% receptor -ve). Overall % agreement(%AG) and kappa statistic(K) were used to measure level of agreement on both the standard grading scheme(SG) and the revised grading scheme(RG). The SG used a slight modification of the Scarff-Bloom grading system and the RG incorporated only 2 levels in the mitotic index category (43 or ~ 3 mitoses per hpf). Results: SG RG Intraobserver %AG 75% 85% K 0.67 0.74 Interobserver %AG 63% 76% K 0.44 0.6 One pathologist graded high the first time and more consis tently with the other two the 2nd time. This reflects the difficulty of a grading system which does not use strict criteria, lligh grade correlated with receptor negativity. All potential cases of PG II-III misclassification were identified by receptor negativity. Conclusions: PG can be measured with good to excellent level of agreement. The RG may be superior. PG(in conjunction with receptor status) may be useful in identifying a high risk subgroup for adjuvant treatment. This study will be repeated using Contesso's strict criteria for grading and the effect of education will be measured and ~resented.

HER-2/neu

Oncogene as a Prognostic Factor in Node Negative

Breast Cancer

Paterson, A.H.G., Fourney, R.M., Dietrich, K.D., *Danyluk, J., Jamil, N., Lees, A.W., Kraasc, B., MeEwan, A., Lukka, H., Hanson, L, McBlaln, W.H., Willan, B., **Slamon, DJ., Paterson, M.C. Cross Cancer Institute and University of Alberta, Edmonton, Alberta, *Misericordia Hospital, Edmonton, Alberta, **U.C.L.A., Los Angeles, California. New biological prognostic factors for breast cancer are essential for the appropriate use of adjuvant therapy. This is especially true given the currently advocated use of cytotoxic chemotherapy for all node-negative,ER-negadve patients with breast cancer, many of whom may never have recurrence of their disease. HER-21neu a member of the erbB oncogeae family, has recently been identified as an independent prognostic factor in node-positive breast cancer (Slamon et al., Science 235:{77, 1987). Node negativity is often regarded as a "good" prognostic factor, although approximately 25% of these patients will relapse within 5 years (N.S.A.B,P. data). The purpose of tins study was to examine the abnormal expression of HER-21neu in patients with negative nodes. We identified from our Registry 79 patients with node-negative breast cancer who bad recurred within 5 yrs of initial diagnosis (cases) and matched them for age, tumour size, ER status and menopause with 79 patients who had not relapsed (controls). After locating formalin-fixed tissue sections, each tamour was assessed for nuclear grade and other pathologicalparameters and a 3 ram. core sample was taken for inquiry into HER-2/neu status. Total DNA extracted from tamour samples was subjected to conventional slot-blot analysis, and autoradiograms were scanned by laser densitometry to measure HER-2/neu copy number. Blots were stripped and reprobed with human arginase eDNA to normalize for amount of DNA loaded. Using >__5 copies as the cut-off, oncogene amplification was observed in tumour material from 12 of 79 (-15%) who had relapsed within 5 yrs but in only 4 of 79 (-5%) controls (p<0.02); moreover, tumour DNA from 8 cases, compared to 3 controls, were highly amplified (10-25 fold). No statistically significant association was obtained between HER-2/nea oneogene copy number and ER status, size of primary tumour or nuclear grade. The observed amplification in 15% of node negative patients who have relapsed indicates that further exploration of aberrant structure in this and other oncogenes may be informative.

110

Abstracts

13

AMPLIFICATION AND EXPRESSION OF THE neu/erbB-2 PROTOONCOGENE IN HUMANPRIMARY BREAST CARCINOMA. C. Dati*, R. Muraca, O. Tazartes, M. Giai, E. DeFabiani, P. Sismondi, G. Saglio and M. DeBortoli. Depts. of Anim. B i o l . , Exptl. Med. & Oncol., Obstetr. & Gynecol., 8iomed. Sci. & Human Oncol., University of Turin, I t a l y . The neu/erbB-2 oncogene encodes a 185 Kd growth factor r e c e p t o r - l i k e protein, c-erbB-2__ amplification (reportedly found in up to 30% human primary breast cancers) and/or i t s overexpression has been correlated with negative prognostic factors. We screened a group of f u l l y characterized human primary breast carcinomas for c-erbB-2 amplification and expression. Gene amplification was studied by Southern blot analysis using a eDNA probe. Protein expression was investigated by immunoblot using an antiserum specific for the C-terminal region (donated by Dr. W. Gullick). An amplified copy number of c-erbB-2 was found in about 15% of tumors, and m o r e than 80% of them also expressed the protein. In contrast, erbB-2 protein expression (undetectable in normal control tissue) was observed in about I / 3 of samples. Some of them c l e a r l y overexpressed the protein showing no erbB-2 amplification. No variation of the protein size was observed in any sample. Both erbB-2 expression and amplification correlated with absence of progestin receptor. At present, we have not found any significant correlation between erbB-2 expression or amplification and nodal status of the patient. 11o correlation was seen between the presence of erbB-2 protein and the expression of p21ras. The possible prognostic significance of erbB-2 in human breast cancer is under further i n v e s t i g a t i o n .

14

MONOCLONAL ANTIBODY TAI (ANTl-neu) REACTIVITY WITH ARCHIVAL BREAST CARCINOMAS: CORRELATION WITH CLINICOHISTOLOGIC FEATURES AND SURVIVAL. Thor, A., Koerner, F., Schwartz, L., Fingert, H., Stracker, M., Edgerton, S., Wood, W, Massachusetts General Hospital, Boston, MA 02114; McKenzie, S, Panicali, D, Yin, S. Applied bioTechnology, Cambridge, MA 02142 A murine monoclonal antibody TAI a was generated by i~unization of BALB-o mice with NIH3T3 cells expressing the human ne__~uprotein on their cell surface. The antibody recognizes an epitope on the extracellular domain of the ne___uuprotein~ The recognized antigen is a 185 kilodalton phosphoprotein by immunoprecipltation. MAh TAI will also stain live or fixed cells expressing ne_._uu using immunofluorescenee FACS analyses. In similar experiments, the TAI antibody does not recognize the epidermal growth factor receptor (EGFR) protein from A431 cells, a cell line which expresses a high level of EGFR. Analysis of 17 human breast invasive ductal carcinomas has ~ shown good correlation between neu mRNA levels determined by filter hybridization and end-point dilution immunoperoxidase analysis of formalin-fixed tissues. Using avidin-biotin peroxidase techniques, we have analyzed archival samples of 176 primary adenocarcinomas of the Breast for ne___uuprotein expression. Clinical parameters including T-stage, N-stage, nuclear grsde, estrogen receptor status and survival are well documented for these patients. A minimal follow-up of 6 years (median follow-up of 8.5 years) was available for the study group. Correlation between neu expression and these nlinico-histologic parameters will be presented. asupplied by Oncogenetie Partners, an Applied bioTechnology/Dupont joint venture.

15

BIOLOGICAL RESPONSES OF THE ANTI-ESTROGEN ICI 164,384 IN DIFFERENT MAMMARY CANCER CELL LINES. J R Pasqualini*, and C Gelly, C.N.R.S. Steroid Hormone Research Unit, Fndn for Hormone Res., 26 Bd Brune, 75014 Paris, France. Most of the available anti-estrogens (e.g. tamoxifen, nafoxidine) can have antagonistic or agonistic properties, which is a function of the model or the experimental conditions studied. Recently Imperial Chemical Industries (ICI) have synthesized a potent anti-estrogen, the ICI 164,384, which presents full antagonistic properties. This presentation gives the biological responses obtained with this anti-estrogen in the MCF-7 mammary cancer cell and other hormone-dependent (T-47D) and hormone-independent (MDA 2~I, MDA 438) cell lines. ICI 164,384 at a concentration of 10-~ M can block cell proliferation and iDhibit the stimulatory effect of estradiol (E 2) at 10-~M. The [3H]thymidine incorporation agrees with this data. At this concentration (10- M) the anti-estrogen blocks the stimulatory effect provoked by E 2 on the synthesis of progesterone receptor. In another series of studies was explored the effect of ICI 164,384 in the conversion of estrone sulfate (EIS) to E 2. The data indicate that the inhibition of the conversion with this anti-estrogen is more intense than with tamoxifen in both MCF-7 and T-47D cells. The concentration of E 2 (pg/mg DNA) is of 140 in the incubation with 5xi0 -9 EIS alone and of 10 in the presence of ICI 164,384. No effect in this conversion was observed in the two hormone-independent cell lines. It is concluded that in the MCF-7 cell line ICI 164,384 is a much more potent anti-estrogen than tamoxifen itself and also that this anti-estrogen blocks more intensively the conversion of EIS to E 2. As EIS is quantitatively the most important estrogen in breast cancer tissue, the present data open new possibilities in the knowledge of anti-estrogens action in breast cancer.

16

THE PREVENTION OF SPONTANEOUS MOUSE MAMMARY CANCER BY PROLONGED TAMOXIFEN THERAPY. V.C. Jordan,* M.K. Martin, D.M. Mireeki, and S. Langan, Department of Human Oncology, Univ. of Wis. Clinical Cancer Center, Madison, WI 53792. Tamoxifen, a non-steroidal antiestrogen used successfully as an adjuvant therapy in Stage I and II breast cancer, is being considered as a preventative agent for women at high risk for the disease. The aim of this study was to evaluate the efficacy of tamoxifen in a mouse model of spontaneous mammary cancer. Tamoxifen is considered to be an estrogen in the mouse but preliminary studies in ovariectomized and intact female mice (C3H/OUJ) demonstrated antiestrogenic properties following sustained release administration from silastie implants (200~g tamoxifen/day) between 2 and 6 months. Tamoxifen was detected in serum (20-30 ng/ml) using HPLC with post column fluorescent activation of triphenylethylenes to phenanthrenes. Female mice (C3H/OUJ), bred repeatedly between 2 and 10 months of age, all presented with palpable mammary tumors between 6 and 12 months of age. Unfortunately, tamoxifen is a contraceptive agent in the mouse and the model is inappropriate for study. However, a single pregnancy between 2 and 4 months of age facilitated tumor induction by 10 months of age. Female mice were made pregnant at 9 weeks of age and~ following weaning (16 weeks of age), animals (44) were randomized into 4 groups: control, ovarieetomy, ovariectomy plus sustained tamoxifen therapy or intact animals plus tamoxifen. Mammary tumors started to appear in controls at 6 months of age and 80% had tumors by 9 months of age. Ovariectomy with or without tamoxifen therapy resulted in complete suppression of tumorigenesis at 9 months, and only 20% of intact mice treated with tamoxifen had tumors. Tamoxifen prevents the induction of tumors in this laboratory model of mouse mammary cancer. This study was supported by the American Cancer Society Grant SIG-15.

Abstracts 17

19

PHASE I TRIAL OF CGS 16949A - A NEW AROMATASE INHIBITOR. *A Lipton, H A Harvey, L M Demers, J Hanagan, M Mulagha, S Sanders, S Fitzsimmons and R J Santen, The M i l t o n S. Hershey Medical Center, Hershey, Pennsylvania, 17033, and CIBA-GEIGY Corporation, Summit, New Jersey 07901. OGS 16949A [4-(5,6,7,8 - tetrahydroimidazol [ 1 , 5 ~ ]pyridin-5-yl) benzonitrile monohydrochloride) is a new nonsteroidal inhibitor of aromatase. In this Phase I trial, 16 heavily pretreated postmenopausal patients with metastatic breast cancer were treated with CGS 16949A. Each patient received 0.6, 1.2, 2, 4, 8 and 16 mgm, total daily oral dose (1/2 given at 8 AM and 1/2 at 8 PM) w i t h dose escalations every 2 weeks. Serial hematologic, biochemical, endoerinologic, pharmacokinetic and clinical evaluations were performed. No hematologic or biochemical abnormalities were encountered. Maximtun aromatase inhibition was observed at a daily dose of 2 mgm of CGS 16949A. Suppression of plasma and urinary E 1 and E 2 was equivalent to that of 1,000 mgm A m i n o g l u t e t h i m i d e (AG) plus 40 mg hydrocortisone. At this dose, plasma ACTH and urinary Cortisol were unchanged. One patient allergic to AG had no problems with CGS 16949A. No significant side effects of the m e d i c a t i o n were encountered. Postural hypotension was observed in two patients at the highest doses of CGS 16949A administered. Pharmacokinetic data will be presented. Of 15 evaluable patients there were 2 PR (5 8+ months) and 7 patients w i t h stable disease , + + + + (5,5 ,6 ,7,8,8 ,8 months). We conclude that CGS 16949A is a potent, specific, nontoxic, and clinically active new inhibitor of estrogen biosynthesis in postmenopausal breast cancer patients.

A

IMPROVEMENTOF PROGNOSIS IN NODE NEGATIVE (NO) EARLY BREAST CANCER (BC) THROUGH HIGH DOSE MEDROXVPROGESTERONE ACETATE (HD-MPA) ADJUVANT HORMONOTHERAPY. A 3 YEAR INTERIM REPORT ON A RANDOMIZEDSTUDY. C. Focan, M. Beauduin, J.P. Lobelle, E. Longeval, F. Majois, E. Salamon, A. Tagnon, S. Van Belle, B. Vanderlinden, R. Vandervellen, A. Vindevoghel, f o r the Adjuvant Breast Cancer Project (ABCP) - Belgium. In 1983, the ABCP Belgium i n i t i a t e d a randomized t r i a l comparing hormonotherapy with HD-MPA (group B : 500 mg IM x 28 days, than 500 mg twice weekly f o r 5 months) to no hormonotherapy (group A) f o r e a r l y BC. The therapy began between the 2nd and 4th week a f t e r surgery ; radiotherapy was usually applied. At present, 240 patients entered in the t r i a l out of whom 209 are f u l l y evaluable (106 gr. A, 103 gr. B). Patients were well balanced with regards to age, menopausal status, tumor stage, tumor receptors (ER ; PrG). A f t e r a mean relapse free i n t e r v a l (RFS) of 32 months (mo) and a mean survival time (OAS) of 36 mo, we evidenced a better prognosis f o r patients receiving hormonotherapy : 23/106 relapses in gr. A, 7/103 in gr. B (p ~0.05) ; s i g n i f i c a n t improvement of RFS (log rank t e s t : p(O.05) f o r the whole group (RFS at 36 mo : gr. A : 73% ; gr. B : 94%). The same trend in OAS was also observed (log rank t e s t : p = 0.04) (OAS at 36 mo : gr. A : 89% ; gr. B : 99%). The c l i n i c a l t o x i c i t y was mainly : gr. A : some increase of weight ; gr. B : p a r t i a l l y reversible increase of weight and hydro-saline retention (frequently transient) ; spotting, tremor, headache and cramps were observed in 15% of the patients in gr. B ; more over a l l premenopausal in gr. B developed amenorhea ; in t h i s group, one subject died from severe

lung fibrosis after radiotherapy~

18

111

M A S T E C T O M Y OR TAMOXIFEN? A R A N D O M I S E D CROSSOVER STUDY IN ELDERLY PATIENTS W I T H OPERABLE BREAST CANCER. *J.F.R. Robertson, J.H. Todd, I.O. Ellis, R.I. Nicholson, C.W. Elston, R.W. Blamey, N o t t i n g h a m City Hospital, UK One hundred and thirty-five consecutive patients aged 70 or over w i t h p r i m a r y breast cancer have been prospectively r a n d o m i s e d to r e c e i v e e i t h e r w e d g e m a s t e c t o m y (simple m a s t e c t o m y without u n d e r c u t t i n g of the flaps) or T a m o x i f e n 20 mg b.d. as the initial treatment. Patients r a n d o m i s e d to T a m o x i f e n crossed over and had a wedge m a s t e c t o m y if there was any sign of local progression. The c r i t e r i a for entry were: age > 70 years, operable breast cancer (maximum diameter 5 cms) and a s s e s s e d fit for surgery. Patients r a n d o m i s e d to wedge m a s t e c t o m y w e r e treated w i t h further e x c i s i o n a n d / o r r a d i o t h e r a p y for local recurrence and on failure of these treatments, or on diagnosis of metastatic disease by Tamoxifen. C o m p l e t e r e s p o n s e was o b t a i n e d in 44% of the T a m o x i f e n t r e a t e d group and p a r t i a l r e s p o n s e in a further, 19%. In 30% of patients the tumour continued to p r o g r e s s on T a m o x i f e n t h e r a p y . T h e r e was 0% o p e r a t i v e m o r t a l i t y in the m a s t e c t o m y group. There was no significant d i f f e r e n c e in survival between the two groups. However, the disease free interval in the m a s t e c t o m y group was s i g n i f i c a n t l y b e t t e r than the time to p r o g r e s s i o n in the T a m o x i f e n group. In addition, w e d g e m a s t e c t o m y Gave c o n t r o l of p r i m a r y disease in more patients than Tamoxifen. F r o m our results we b e l i e v e that the o p t i m u m t r e a t m e n t in e l d e r l y p a t i e n t s w i t h o p e r a b l e breast cancer includes wedge mastectomy.

20

POTENTIAL PROGNOSTIC FACTORS RELATED TO PRIMARY HORMONAL RESPONSE IN POST-MENOPAUSAL WOMEN W I T H BREAST CANCER.P. Johnson,P.Bonomi,M.Gale,J.VonRoenn,K.Anderson,J.Wolter,S. Economou;Carle Clinic,Urbana I1 61801,Rush University, Chicago Ii 60621,Northwestern University,Chicago Ii 60611. During the 1970's estrogen(ER) and progesterone(PR) receptors were reported to have predictive value with respect to response to hormonal therapy in advanced breast cancer. The objective of the current retrospective study was to evaluate the potential relationship between response (UICC criteria) to hormonal therapy and traditional predictors of hormonal responsiveness and ER/PR levels. Eligibility requirements w e r e : a d v a n c e d breast cancer,known quantitative ER&PR levels,post-menopausal status,measurable or evaluable disease,primary hormonal treatment with megestrol acetate or tamoxifen.The characteristics of the 105 patients who fulfilled these criteria w e r e : m e d i a n age62 years;median disease-free interval-523 days;soft tissue vs bone vs visceral dominant disease 35%,34%,and 31%;one site of disease-58% vs 2 sites of disease-42%,previous treatment with c h e m o t h e r a p y - 5 1 % ; E R + P R + - 54%;ER+PR- - 12%; ER-PR+ - 16%;and ER-PR- 18%;median ER level - 44fm/mg; median PR level - 52fm/mg.These patient characteristics were included in stepwise logistic analyses in which response was the dependent v a r i a b l e . D e f i n i n g complete and partial remissions as a positive response resulted in a final model (p=.001,R=.57) where log ER (p=.007) and treatment w i t h megestrol acetate (p=.006) were positively related to response,where there was a negative association between response and prior chemotherapy (p=.003).Including stable disease as a positive response resulted in a final model (p=.001,R=.49) where both log ER (p=.001) and log PR (p=.007) were positively related to response.These results suggest that quantitative ER levels supersede the traditional predictors of hormonal responsiveness.

112 21

23

Abstracts THE NEUROENDOCRINEBASIS FOR THE MANAGEMENTOF TAMOXIFEN FLARE, *RF Pommier, MD; EA Woltering, MD; WS Fletcher, MD; Oregon Health Sciences University, Portland, Oregon 97201. A tamoxifen f l a r e is an exacerbation of metastatic breast cancer occurring a f t e r i n i t i a t i o n of tamoxifen therapy. The mechanism behind a f l a r e is unknown, but a f l a r e is considered to be strong evidence o f a hormone sensitive tumor. Accordingly, most c l i n i c i a n s continue tamoxifen therapy a f t e r a f l a r e in a n t i c i p a t i o n of a subsequent response. However, the mean duration of response is only 3 months compared with 9.5 months f o r patients who do not f l a r e . We have demonstrated that tamoxifen markedly stimulates production of dehydroepiandrosterone (DHEA) from the adrenal gland. DHEA is r a p i d l y converted to estrone and estradiol by the enzyme aromatase. Because estrogens have a f f i n i t i e s for the estrogen receptor t h a t are I000 times that of tamoxifen, the beneficial effects of tamoxifen could r a p i d l y be replaced by the deleterious effects of increased estrogens produced under the influence of tamoxifen. This could account f o r the exacerbation observed during a f l a r e . Substantial improvement is only achieved by discontinuing tamoxifen, performing an oophorectomy, and suppressing or removing the adrenal glLands. An a l t e r n a t i v e is a n t e r i o r hypophysectomy. This reduces DHEA and estrogen levels to minimum detectable levels and results in a mean response duration of 21 months. We conclude that DHEA, estrone, and estadiol levels should be monitored in every patient taking tamoxifen. I f the patient experiences a f l a r e , continuing tamoxifen may r e s u l t in a survival disadvantage and patients should undergo t o t a l endocrine ablation or suppression.

22

A POPULATION STUDY OF THE EFFECT OF EDUCATION FOR BREAST SELF-EXAMINATION (BSE). RW Blamey, City Hospital, Nottingham, U.K.

24

89,000 women in the Nottingham Health DistrictS, aged 4065, were invited for education; a self-referral clinic was available. The study is from 1979-1986. Attendance was 55%. 751 cancers arose in the population. Controls are the 751 cancers immediately prior to 1979.

Population

in situ and .< 2 cms

Control Study

281

356

> 2 cms and advanced 358 202

Similarly lymph node stage and histological grade are ameliorated. A case control study shows a risk reduction of death from breast cancer in women who attended for education (relative risk = 0.66, p < 0.025).

IDENTIFICATION OF POTENTIAL STRATEGIES TO INCREASE PHYSICIAN ORDERING OF SCREENING MAMMOGRAPHY. R. Yanagihara. Tne University of Texas Health Center at Tyler, Tyler, IX 75710. We sought to identify methods by which to increase the documented underprescription of screening mammography (sm) by physicians. We contacted by mail all 98 primary care physicians in Smith County TX. Anonymously completed questionnaires were returned from 52 (53%).While 24 subjects (subj) agreed completely with American Cancer Society smguidelines, 23 a~reed only pe~tially 4 Of the latter, 19 (82% of partial substrihers ahd 36% of all subj) objected to the annual rate of sm as too frequent. Over 96% of subj asserted that they ordered sm on > 10% of their eligible patients (pts). Yet a review of sm oYdering by a subset of 6 randomly selected internists at the regional academic center disclosed that, over a 6-month period, only 151 sm were ordered on 3,283 eligible women (4.6%). When subj were asked why they failed to follow sm guidelines, 63% cited high cost; ~3%, simple forgetting; and 35%, excessive frequency of annual studies. When restricted to a single dominant reason, subj most often cited simple forgetting (28%). We queried subj on specific antecedents to and consequences of their sporadic order of a sm. Although 74% cited an educational session as the primary stimulus, their demonstrated underprescription indicated that the receipt of knowledge was incomplete or unconvincing and that it did not translate into practical implementation. For 17%, sm ordering was primarily prompted by a pt's own request. Consequences of sm ordering consisted primarily of internal reinforcement (80% felt intrinsically rewarded by a sense of a task well done) or private recognition (52% received an expression of gratitude from the pt). Only 9% of subj perceived social recognition or public praise as sequelae of their sm ordering. The findings indicate that (a) sm ordering has not yet integrated itself into a physician's customary diagnostic approach, (b) ~hysiciansh~¢e~ inflated perceptions both of their sm orderlng practices and of the effect of education on their performance, (c) certain antecedents to and consequences of sm ordering are amenable to structured modification: Sug$ested strategies include providing to ~hysicians a convincmng rationale for q 1 yr screening, facilitating pts' own requests for sm, and a program of positive reinforcement and feedback on sm ordering performance delivered to primary care physicians by a dedicated advocate.

PRACTICES OF BREAST SELF EXAMINATION (BSE) IN W O M E N W I T H A NON-BREAflT NEOPLASM. Mary Ann Doyle*, William Simonich, American International Hospital, Zion, I l l i n o i s , 60099 C u r r e n t i n f o r m a t i o n on BSE f o c u s e s on p r e v e n t i o n and early detection of b r e a s t c a n c e r in the general p o p u l a t i o n . A n e x t e n s i v e s e a r c h of t h e l i t e r a t u r e shows that the issue of a second p r i m a r y of b r e a s t o r i g i n in p a t i e n t s w i t h a k n o w n m a l i g n a n c y and the n e c e s s i t y of BSE have not b e e n e x p l o r e d . T h i s study addresses the practices of B S E a n d i n t e r v e n i n g factors influencing BSE. Twenty women who are p r e s e n t l y r e c e i v i n g c a n c e r t h e r a p y w e r e given a questionnaire to assess t h e i r b a s e l i n e knowledge of BSE and their past/current practices of BSE. O n l y 5 of 20 (95% C.I.: 6.0% t o 44.0%) p a t i e n t s r e p o r t e d t h a t t h e y r o u t i n e l y p e r f o r m e d BSE a t least once a m o n t h p r i o r to b e i n g d i a g n o s e d w i t h a cancer. T h i s is c o n s i s t e n t w i t h o t h e r s t u d i e s in the general population. Post diagnosis 7 (35%) of 20 (95% C.I.: 14.1% to 55.9%) of t h e p a t i e n t s p r a c t i c e BSE. T h e r e w a s a c o n s e n s u s t h a t B S E is i m p o r t a n t w i t h 17 (85%) r e s p o n d i n g v e r y i m p o r t a n t and 2 (10%) a n s w e r i n g f a i r l y i m p o r t a n t . F o u r t e e n (70%) p a t i e n t s felt t h a t r e s p o n s i b i l i t y for d e t e c t i n g b r e a s t a b n o r m a l i t i e s w a s t h e i r s a n d ii (57.9%) f e l t it w a s a s h a r e d r e s p o n s i b i l i t y with their physician. The role of mamalography w a s e x p l o r e d . M a m m o g r a m s w e r e v i e w e d as i m p o r t a n t b y 80% (16) of the p a t i e n t s and 15 (31%)(95% C.I.: 56.0% t o 94.0%) have had a m a m m o g r a m in the past. Six of t h e p a t i e n t s p e r c e i v e d they had a shared r e s p o n s i b i l i t y w i t h t h e i r p h y s i c i a n in r e c o m m e n d i n g a m a m m o g r a m . The results of this study w i l l b e u s e d to p l a n an in-house educational program.

Abstracts 25

COMPUTERIZED SYSTEM OF CALCULATION OF MAMMOGRAPHICDOUBLING TIME. A. Laffranchi, A. Cerrotta, G. Coopmans de Yoldi, S. Bergonzi, G. Viganotti and E. Galante*. I s t i t u t o Nazionale Tumori - 20133 Milano The prognostic significance of the growth characteristics of the breast cancer have been object of numerous s c i e n t i f i c reports. The mammographic doubling time is considered a val i d parameter of evaluation of growth. In order to reduce the human error of measurement a computerized system of def i n i t i o n of the radiological tumour image and of calculation of tumour volumes was studied. This semi-authomatic system is able to define the contour line of the tumour image by mean of an appropriate algorhythm and to calculate the t h i r d dimension in term of density of the radiological imagecomp~ red to a precodified scale of greys. The interval time between the two successive mammograms was reduced and minimal tumour growth were pointed out in a sample of ten cases ava~ lable f o r a correct experience: the minimal i n t e r v a l time was 17 days and the range of the doubling times, calculated according to the p r i n c i p l e of the exponential growth,was 271070 days. The study is s t i l l ongoing, in order to improve the procedures of elaboration of the not well defined tumour images.

26

27

ULTRASOUND MAMMf)GRAPHY IN THE F O ~ U P OF THE POST IRRADIATED BRFAST. * T.G.Frazier, A.P. Furnary, D. Rose, and J.T. Murphy. Frcm: The Departments of Surgery and ~adiology, The Bryn Mawr Hospital, Bryn Mawr, PA and Thcmas Jefferson University Hospital, Philadelphia, PA 19107

28

Primary radiation therapy has continued to gain rapid acceptance as an alternative in the treatment of primary carcinoma of the breast. Changes in the post-irradiated breast however, do provide diagnostic difficulties for both x-ray~aphy (XRM) and physical examination (PE). Since 7 to 10% of patients will recur in the irradiated breast, the timely diagnosis of such recurrence is essential for salvagemastectomy to be appropriate and effective therapy. To evaluate the use of diagnostic ultrasound mammography (USM) in these patients, 66 consecutive patients treated by segmental resection, axillary dissection and primary radiation therapy were followed with PE, X R M a n d USM (range 11-104 months, mean 36 months). 9 patients (13.6%) had a positive finding on one of the three modes of examination, and underwent either needle or excisional biopsy. 4 patients (6.1%) had recurrent carcincma. 2 of these were diagnosed by USM alone, end one had a questionable XRM, for which 6 month X R M w a s rec(mmlended. 5 patients (7.6%) had negative biopsies. 4 of these had suspicious XRM and 1 had suspicio3s PE. None had suspicious USM. We conclude that U S M w a s superior to XRM in the diagnosis of patients~hDrecurred, and was also superior in assessing four patients with suspicious XRM, who had benign surgical scar or fat necrosis. This improved selectivity and specificity makes USM a valuable adjunct in the follow up of patients undergoing primary radiation therapy for carcinoma of the breast.

113

INTERPRETATION AND FOLLOW-UP OF ABNORMALMAMMOGRAMS. J M Hirsch * , S Eisenberg, P I Tartter. Mount Sinai School of Medicine, New York, NY 10029. Mammography is a useful screening procedure f o r reducing breast cancer m o r t a l i t y . The adequacy of physician follow-up f o r patients not in prospective studies is not known. Charts of patients with mammograms interpreted as suspicious during one year were reviewed to i d e n t i f y f a c t o ' rs influencing follow-up and p o t e n t i a l l y detracting from the e f f i c a c y of screening mammography. We reviewed 1251 mammograms from the Mount Sinai Medical Center performed during 1987. Twenty (1.6%) were highly suspicious and 55 (4.5%) were moderately suspicious f o r malignancy. Twentysix patients underwent excisional biopsy and 13 (50%) were malignant (65% of highly suspicious mammograms and 33% of moderately suspicious mammograms). Eight patients (11%) did not receive follow-up care. Patients referred f o r mammography from primary care physicians were s i g n i f i c a n t l y more l i k e l y to have follow-up than patients of s p e c i a l t y physicians ( p < 0 . 0 5 ) . Black and hispanic patients, and those l i v i n g f u r t h e r away from the i n s t i t u t i o n were less l i k e l y to be followed. Twenty-five percent of the patients scheduled did not keep t h e i r appointments: patientsgiven appointments within 8 weeks were s i g n i f i c a n t l y more l i k e l y to keep t h e i r appointments. Marked differences in i n t e r p r e t a tion of mammograms was noted between the r a d i o l o g i s t s . Radiologist A interpreted 35 mammograms as suspicious but, only 3 (13%) were malignant on biopsy, whereas r a d i o l o g i s t B interpreted 18 as suspicious and I0 (56%) were malignant on biopsy (p
NEEDLE LOCALIZATION AND EXCISIONAL BIOPSY OF MAMMOGRAPHICALLY DETECTED BREAST LESIONS. P Davis *l, G Cederbom 2, and D M o r r i s I, LSUMC, Departments of Surgery 1 and Radiology 2, Shreveport, Louisiana, 71130. We reviewed the records of all patients undergoing needle localization and breast biopsy for occult lesions between January 1984 and December 1987. Patients underwent needle lohalization by our radiologist using the hooked guided wire technique, under local anesthesia, using a film screen ma~mnography unit. All specimens were submitted for specimen mammograph~. The radiologist marked the lesion for the pathologist with a straight needle. Only permanent sections were made. One hundred and six women had ~ii needle loealizations and biopsies. Twenty breast cancers were found in 20 patients. Fifteen percent of these patients were found to have positive nodes after axillary dissection. All primary lesions in these patients were invasive. The majority of the suspicious mammographic lesions that were malignant were noted to have calcifications either alone or associated with a mass or asymetric density. The radiologist's suspicion of malignancy correlated closely with the diagnosis of cancer. Biopsies positive for cancer were called correctly 70% of the time. Lesions negative for cancer on biopsy were called correctly 90% of the time. The most common benign histology was fibrocystic disease. Many patients were found to have atypical intraduetal hyperplasia and thus were marked as being at increased risk for breast cancer. Needle localization and excisional breast biopsy is a safe and accurate technique. Close cooperation between the surgeon, radiologist and pathologist is necessary to find and remove these small lesions and hopefully improve the outcome for patients with breast cancer.

114

Abs~ac~

29

R I S K STATUS OF W O M E N U S I N G A C O M M U N I T Y M A M M O G R A P H Y CENTER. D.C. I v e r s o n , * C. Chrvala, G. W e i n g a s t , and M. M c C l a t c h y . The p u r p o s e of the s t u d y w a s to d e t e r m i n e t h e r i s k s t a t u s of w o m e n who c o m e to a c o m m u n i ty based c l i n i c for a m a m m o g r a m . Risk status w a s d e t e r m i n e d via a m o d i f i c a t i o n of the s y s t e m used by G r o u p H e a l t h of P u g e t Sound. Variables u s e d to d e t e r m i n e risk s t a t u s included: family history, age at m e n a r c h e , age at m e n o p a u s e , rep r o d u c t i v e history, prior b r e a s t biopsies, a l c o hol use, e x e r c i s e s t a t u s , d i e t , SES, B M I a n d t h e p r e s e n c e of b r e a s t s y m p t o m s . R i s k status was d e t e r m i n e d for all p a r t i c i p a n t s ( n = 2 6 0 0 ) as well as f o r s u b g r o u p s based on a g e ( 4 0 - 4 9 years, 50+ years) and m a m m o g r a m history (prior or first mammogram). The results indicated that p a r t i c i p a n t s t e n d e d to be in t h e e l e v a t e d risk categories. F u r t h e r a n a l y s e s were c o n d u c t e d to e x a m i n e the r e l a t i o n s h i p between the mammogram results and reported concerns about the p r o c e s s of having, and results of, t h e m a m mogram. Finally, s t a t e d r e a s o n s for h a v i n g a m a m m o g r a m w e r e e x a m i n e d in light of p a r t i c i p a n t risk status.

30

R E P O R T I N G M A M M O G R A P H Y RESULTS: EXPERIENCE WITH AN O B J E C T I V E R E P O R T I N G FORM. G W e i n g a s t , * D. Iverson, C. Chrvala, and M. M c C l a t c h y P r i m a r y care p h y s i c i a n s c o n t i n u e to be c o n cerned about mammography-related m o r b i d i t y resulting from non-definitive radiological reports. To a d d r e s s t h i s concern, an extensive, objective radiologic report form was developed which i n c l u d e d c a t e g o r i e s on g e n e r a l i m p r e s s i o n , l e v e l of c e r t a i n t y o f i m p r e s s i o n , and recommendations for repeat mammography. When an a b n o r m a l i t y was d e t e c t e d , an o b j e c t i v e f o r m a t r e q u i r e d i n f o r m a t i o n on the type of a b n o r m a l i t y , l o c a t i o n and a s p e c i f i c f o l l o w - u p recommendation. T h e m e t h o d u s e d to d e v e l o p the form a n d exp e r i e n c e in u s i n g t h e f o r m w i l l be r e v i e w e d ( f r o m t h e r a d i o l o g i s t ' s and p r i m a r y care p h y s i cian's p e r s p e c t i v e s ) . To date, the e x p e r i e n c e s h a v e b e e n v e r y p o s i t i v e a l t h o u g h some m o d i f i c a tions to the form seem a p p r o p r i a t e . The c o n t r i b u t i o n of an o b j e c t i v e r a d i o l o g i c r e p o r t form to AMC's c a n c e r c o n t r o l r e s e a r c h e f f o r t is r e viewed.

31

THE IBIPACT OF A N E W C L A S S I F I C A T I O N S Y S T E M FOR ~IAt%.~OGRAPHIC R E P O R T I N G A T A LARGE ~ J L T I S P E C I A L T Y C L I N I C . *GL J a c k s o n , Kt Kemp, PM C o n o l e y , M Htain, P Raphael, Kelsey-Seybeld Clinic and Foundation, H o u s t o n , TX, 7 7 0 3 0 . I n J a n u a r y o f 1984 a new m a n ~ o g r a m classification s y s t e m was i m p l e m e n t e d a t a l a r g e multispeeialty clinic. It facilitated patient triage and c o m p u t e r i z a t i o n for tracking. A c o m p s c i s o n was made b e t w e e n 1 9 8 1 - 3 and t h e f i r s t three yeats o f t h e new c l a s s i f i c a t i o n system 1984-6. The c l a s s i f i c a t i o n s y s t e m was a p p l i e d to 1 4 , 1 9 1 p a t i e n t s f r o m 1983 to 1 9 8 6 : C l a s s 1, normal findings, 4 7 . 1 % ; C l a s s 2 , m a s s or m a s s e s present, probably benign, 4 7 . 7 % ; C l a s s 3, indeterminate lesion found, suggestive of possible malignancy, 4 . 9 % ; C l a s s 4, p r o b a b l e malignancy, 0.3%. Breast biopsies w e r e d o n e on 1,2% o f C l a s s 1 and 2; 3 3 . 5 % o f C l a s s 3, and 95.6% of C l a s s 4 p a t i e n t s . The a v e r a g e f e m a l e population o v e r a g e 35 d o u b l e d . Man~nograms increased from 2,410 to 14,191. The s u r g i c a l consultation for a breast problem increased from 435 to 1 , 3 8 5 , b i o p s i e s increased f r o m 273 to 431. Close communication between the radiologist and t h e s u r g e o n a l l o w e d f o r m a i n t a i n i n g an acceptable, non-mallgnant to m a l i g n a n t biopsy ratio of 431:143 (3.01) which compares quite favorably to t h e BCDDP 5 . 4 .

32

INVASIVE BREAST CARCINOMA WITHOUT PALPABLE MASS WITH NIPPLE DISCHARGE AS SINGLE SYMPTOM. M. Taffurelli, A. Grassigli, D. Santini °, F. Giosa =°, U. Rossati °°°, G. Golematis, G. Urso, D. Marrano. Ist. I^ Clinica Chirurgica, °Anatomia Patelogica, °°Oncologia, °°°Radiologia - Universit~ di Bologna-Policlinico S.O~ sola - Italy Invasive breast carcinoma shows a nipple discharge as single symptom without palpable mass very rarely. Amoung 1768 invasive breast carcinoma surgically treated from 1975 to 1987, 22 cases (1,2%) without palpable mass and with a nipple discharge as single symptom were observed. A spontaneus and clinically caused nipple discharge was pr! sent in 18 (81,8%) and 2 (9%) patients, respectively; no anamnestic data were documented in two cases. The discharge was prevalently hematic (50%) or serum-hematic (22,7%) while less frequently yellow (18,2%), milk-like or serous (4,5%). According to Papanicolau classification, the cytologic examination showed: i0 (45,4%) class V, 7 (31,8%) class IV, 3 (13,6%) class Ill, finally erythrocytes without epithelial elements were detected in 2 cases (9,1%). All patients underwent galactography and mammography; Rx-ma lignant features were only demonstrated in 9 cases (40,9%). In conclusion, according to our e~perience a clinically preoperative diagnosis of breast invasive carcinoma with nipple discharge as single symptom may be very difficult, In fact the cytologic examiflation and Rx-investigations showed neoplastic or suspicious findings only in 77% and 40,9% of cases, respectively. Even if the preoperative exams combined, in 4 cases (18,2%) a correct diagnosis was only performed by histological exam of surgical specimen. Supported by grant Regione Emilia-Romagna

Abstracts

115

33

BREAST CANCER DETECTION : IS WATER-SUPPRESSED PROTON NUCLEAR MAGNETIC RESONANCE SPECTROSCOPY OF PLASMA LIPOPROTEINS USEFUL ? J.M. Nabholtz*, A. Rossignol, N. FARNIBR, S. Friedman, P. Fargeot, J. Cuisenier, P. Gamhert, d. Gerrin. Centre G.F. Leclerc, UASS CNRS, Faculte des Sciences, Laboratoire de Biochimie Medioale, H@pital du Bocage, 21000 Dijon, France. Recently, there has been excitement over detection of malignant tumors by Water-Suppressed Proton Nuclear Magnetic Resonance Spectroscopy (IHNMR) of plasma lipoproteins (N. gng. J.Med, 1986;SI5,19B9-Tg). We performed a similar prospective and blinded study comprising i00 subjects; 40 control (C), 41 untreated breast cancer patients (BC) and 19 patients with benign breast tumors (BT). We measured the parameter WI/g, a mean of the full width at half height of the resonances of the methyl and methylene groups of the lipids of plasma lipoproteins which is inversely related to the spin-spin apparent relaxation time (Tg*). WI/2 values were determined at a fixed baseline of 310 Hz. The WI/2 mean values were 37.6 ~ 9.7 for C, 38.6 ~ 4.0 for BC and 38.1 for BT, essentially identical for all three groups. Furthermore, no differences were seen between any group on scatter pattern that could serve as a basis for a useful detection test. The major difficulty in the determination of WI/2 was due to intemference of metabolite protons (lactate) within the lipoprotein resonance signal. Triglyceride level was seen to co~relate inversely with Wl/2(r= -0.246, p-O.Ol) and no correlation was found between WI/2 and total cholesterol, HDL-cholesterol and (LDL+VLDL)-cholesterol levels. These discrepant results may be related to differing triglyceride-rich VLDL levels in the patient population of each study. We conclude that Water-Suppressed INNMR Spectroscopy of plasma lipoproteins as performed, is not a valid method for detection of breast cancer.

34

DOPPLER SIGNAL IN BLOOD FLOW COMPUTERIZEDANALYSIS OF BREAST LUMPS: EXPERIENCE ON 64 CASES. E. Galante*, M. Stanzani~ E. Gallo~ A. Craveri~ G. Landi~ A. Cerrotta, G. Martelli,G. De Palo. Osp. S.Paolo °, Universit~ degli Studi - Milano - I s t i tuto Nazionale Tumori - 20133 Milano Aim of the study was to ascertain the r e l i a b i l i t y of Doppler Signal in the diagnosis of breast cancer. The bloodflow computerized analysis of 64 c l i n i c a l l y suspicious breast lumps was performed before mammographic and histological examinations, and the values of systolic peak and d i a s t o l i c frequency were compared to f i n a l diagnosis. The values of the s y s t o l i c peak of 42 carcinomas ranged from 2000 to 6500 Hz, with a mean value of 3354 Hz, and the d i a s t o l i c frequency ranged from I000 to 2500 Hz, with a mean value of 1473Hz. The s y s t o l i c peak of three non HD lymphomasshowedthe values of 3000, 3500 and 5000 Hz with a d i s t o l i c frequence of 1700 2000 and 3000 Hz. No typical signals were found in three breast cancers (false negative). The range of values of s y s t o l i c peak of 16 benign diseases (5 fibroadenomas, 9 benign displasias and two local inflammations) was 2000 6000 Hz with a mean value of 2468 Hz: and the range of the d i a s t o l i c frequencies was 300-3000 Hz, with a mean value of 1130 Hz. This f i r s t stage of our experience pointed out that the two parameters, s y s t o l i c peak and d i a s t o l i c frequency, are not able to c l e a r l y distinguish malignant from benign breast diseases because of the wide range of figures found in the last ones. ;n order to catch the c h a r a c t e r i s t i c turbulence of phatological blood flow of breast cancer must be ~dded more specific parameteres ( s y s t o l e / d i a s t o l e radio, Goosling index, window percentage, and Pourcelot index,etc.)

35

FINE NEEDLE ASPIRATION CYTOLOGY IN THE DIAGNOSIS OF BREAST CANCER. G Nicastri,* W P Reed, and B Dziura. Baystate Medical Center, Springfield, MA 01199.

36

FlEW C Y T O M E T R Y A N D CYTOLOGY FROM FINE NEEDLE ASPIRATES OF BREASTS FROM WOMEN AT HIGH RISK FOR BREAST CANCER. J Ensley, T A n , S Martino, D Weaver, P Benitez, F Valeriote* and L Heilhrun, Harper-Grace Hospitals and Wayne State University, Detroit, Mi 48201. As part of a breast cancer prevention program, we have initiated studies on fine needle aspirates from the breast of women who are at high risk for breast cancer. All samples are taken within a week after menses. The aspirates are processed for both cytology and flow cytometric analysis to examine whether there is any correlation with the development of breast cancer; and, whether any manipulation (such as diet modification) can result in a change in any parameter, The majority of women (85%) are at risk because of one or more first degree relatives with breast cancer. The flow cytometric studies demollstrated that 50% of the samples had an aneuploid population although those positive did not always demonstrate such aneuploidy in both breasts. Aneuploidy was defined as samples with a DNA index greater than 1.05. The aneuploid cells were found to be keratin positive a s w e l l . Cytological analysis assessed: a) the degree of epithelial cellularity; b) presence of individually dispersed duet cells; e) presence of apocrlne cells; and d) nuclear atypia of epithelial cells, Of the first 46 patients studied, epithelial cells were not observed in either breast speelmen for 9 patients or in one of the breasts for ii other patients. For those patients lacking epithelial cells by cytology, 3 demonstrated diploidy, 3 demonstrated aneuploidy in one breast and 2 demonstrated aneuploidy in both breasts. For those with increased cellular samples, 7/9 were also aneuploid, for the usual cellular content, 5/14 were aneuploid, for the scantly cellular samples, 12/33 were aneuploid. Supported hy the Marilyn J. Smith Breast Cancer Research Fund.

Fins needle aspiration cytology (RNA)is being used with increasing frequency to diagnose breast cancer. This technique is currently under consideration for use as the sole means of diagnosis in a protocol to evaluate preoperative chemotherapy by a large multicenter cooperative group. To see if a diagnosis of cancer established by FNA is sufficiently accurate to proceed to treatment without tissue confirmation, we have looked at our results in a series of 65 patients undergoing FNA followed by open biopsy between January 1985 and June 1987. There were 4 categories of cytologic reading for specimens evaluated by FNA. These were (i) positive, 15; (2) suspicious, 17; (3) inadequate specimen (too few cells), 8; and (4) negative, 25, All 32 of the specimens with positive or suspicious readings by FNA came from lesions proven to be malignant on subsequent biopsy. Three of the eight specimens that were inadequate, and nine of the 25 negative specimens came from lesions that were later shown to be malignant. We conclude that a positive or suspicious reading on FNA is a highly accurate (100%) indication of malignant breast disease at our institution. Inadequate or negative specimens need further biopsy to exclude malignancy.

116

Abstracts

37

NEEDLE LOCALIZATION BREAST BIOPSY: MORBIDITY AND RESULTS HR Alexander~ DD Dershaw, DW Kinne. Memorial SloanKettering Cancer Center, New York, New York 10021 Five hundred thirty-one consecutive needle localization (NL) breast biopsies were performed during two periods (Jan 1981-June 1984, n=311; and Jan-August 1987, n=220) at MSKCC. Despite the increase in procedures performed (1981, n=89; 1987, n=260), the rate at which carcinoma was identified remained constant (28.6% vs 29.8%) and is compatible with other series. There were 9 syncopal episodes related to the dye NL. Dye injection was either in or adjacent to the suspicious lesion in all but 9 cases, in which the distance was 2 cm or greater; however, excision of these lesions was performed without difficulty. There were 8 wound hematomas and 7 wound infections for a minor surgical complication rate of 2.8%. There were 8 surgically missed lesions in 7 pts (1.5%) requiring a subsequent NL breast biopsy or intervsl mammograms to monitor the lesion. Six of the 8 missed lesions in our series were identified intraoperatively by specimen radiograph that failed to show microcalcifications. Confirmation that the lesion has been biopsied is currently done by specimen radiograph, palpation, or a frozen section diagnosis of carcinoma or calcifications in the specimen. Of 60 invasive cancers found, 39 were less than I em (65%) and 54 were less than 2 cm (90%). In the earlier series 4 of 32 patients (12.5%) were treated with breast conservation while 12 of 26 patients (46.1%) in the later series were so treated, reflecting the current policy of offering breast conservation to suitable patients. NL breast biopsy is an increasingly common radiologic/ surgical procedure associated with few complications. Specimen radiograph should be performed routinely to avoid the consequences of an undetected missed lesion. Carcinoma identified by NL biopsy may frequently be treated by breast conservation therapy.

38

39

RISK OF LOCAL FAILURE IN CONSERVATIVE TREATMENT OF BREAST CANCER OM.Greco,V.Sacchini,M,Merson,V.Oalimberti . NATIONAL CANCER INSTITUTE,MILAN,ITALY

40

The increased risk of local recurrences following limited surgerv is one of the greatest concerns in conservative treatment of breast cancer. The experience of the National Cancer Institute of Milan is fuonded upon 1232 valuable cases of small breast cancer (pT1 tumors),treated since 1970 to 1983 with quadrantectomy,axi]lary dissection integrated by radiotherapy. This kind of treatment has been named QUART,and its outcome, compared with radical mastectomy in a randomized clinical trial has been already published. All premenopausal node positive patients received a standard timed CMF regimen as adjuvant chemotherapy. In postmenopausal was used tamoxifene. The average age of the patients was 48 years,6g% of them were premenopausa] and 32%had positive axillary nodes. All tumors were less than 2.0 centimeters at pathological examination(pT1). The primary tumors were mostly(75%) ductal infiltrating carcinomas. The 49 observed relapses were divided into 32 local recurrences and 17 second ipsilateral tumors, with a calculated risk of 4.1% and 2,7% respectively at 10 years. Same times is not so easy to distinguish between a real local recurrence and a second primary cancer in the same breast. A little increased risk of relapses with increasing the sizes of the tumor was observed and an increased rate of recurrence in negative node patients was observed as w e l l In most cases the treatment of local relapses was a total mastectomy. The overall survival of the whole series at 10 years is about 80%. The disease free survival is about 75%. The comparison of the overall survival from the treatment of the primary tumor with the overall survival from f i r s t relapse shows that survival doesn't seem to be impaired. It is out of doubt that conservative treatment bears an higher risk o f local recurrence than radical mastectomv,but the risk of dying from local relapse is quite heavier a f t e r radical mastectomy than a f t e r QUART,meanwhile the risk of death from distant metastases i~ about the same.

pHASE I-II CLINICAL STUDY OF MONOCLONAL ANTIBODY (MONAB B72. 3-GYK-DTPA-IIIIN) IMAGING IN PATIENTS (PTS) WITH BREAST CANCER (BC). A Buzdar, L Lamki, D Podoloff, L Esparza, R Theriault, F Holmes, S Singletary, J Re, A Zukiwski, G Hortobagyi. The University of Texas M. D. Anderson Cancer Center, Houston, TX 77030. This antibody is a murine monoelonal antibody of IgGl subclass which detects a 200,000-400,000 molecular weight tumor-associated glyeoprotein (TAG-72) expressed in some human BC and colon cancers. Pts with primary BC with or without regional metastatic disease were eligible for this study. Pts with locally advanced BC had chemotherapy (CT) prior to surgery (S). Thirty min after a test dose, pts received a l-hr infusion of 111IN labelled Monab 72-96 hrs before S. Total body nuclear imaging and spot viewing were performed 4,24,48 and 72 hrs later, and findings were correlated with clinical and pathological findings at S. ii pts (i with bilateral tumor), have been registered to date. Initial 6 pts received .2 mg, next 5 pts 2 mg of Monab. Preliminary data are shown in the Table. Ima$in$ Finding to Overall Positive Patholo$icalCorrelation Correlation of Im. True+ TrueFalseBreast (N=I1) I0 1 I 91.6% Nodes (N=6) ~ -? 6 0% The primary BC was detected in all but i pt by imaging studies; in i pt who had no residual disease after CT, imaging studies were also negative. 6 of ii pts had lymph node metastases but none were detected on imaging studies, i pt subsequently found to have (+) HIV test, had palpable bil. axillary uptake but the nodes were histologically negative. The data suggest that at low doses of Monab primary BC, but net lymph node metastases, were imaged in 91% of pts. There were no toxic effects observed. The next 6 pts will be evaluated at 20 mg dose of Monab.

SURGICAL TREATMENT OFBR.EASTCANCER IN THE ELDERLY: THE WIDE LUMPECTOMY IN SAME-DAY SURGERY. EXPERIENCE ON 149 CASES. E. Galante*, A. C e r r o t t a , D. Moglia and G. De Pale. U n i v e r s i t A d e g l i S t u d i - M i l a n o , I s t i t u t o Nazionale Tumori 20133 Milano. T i l l now there i s no well d e f i n e d t h e r a p e u t i c a l approach to brest cancer in the e l d e r l y . We r e p o r t our experience about the s u r g i c a l treatment in day h o s p i t a l regimen. From 1976 to may 1988, 149 p a t i e n t s underwent a wide lumpectomy i n l o c a l anesthesia: 56 pts (37,5%) were operated from 1976 to 1985 and since 1986 the remaining 95 p t s . T h e i r age ranged from 70 to 93 years o l d : the highest percentage (41,6%) was located in the age group 75-80 years. The p a t h o l o g i c volume was ~ 2 cm. in I01 cases (68%), from 2, 1 to 3 cm. in 34 pts (22,7%) and ~ 3 cm. in 9 pts (8%). There was n o - o p e r a t i v e death, and no p a t i e n t needed h o s p i t a l i z z a t i o n . The e i g h t y e i g h t percent o f cases obtained the s u r g i c a l recovery in 2-3 weeks, but only 1 case needed more than 1 month t o heal.Pos~ o p e r a t i v e c o m p l i c a t i o n s were 14 wound seroma and 1 wound dehiscence (about 10% o f the case m a t e r i a l ) . No post o p e r a t i ve t h e r a p e u t i c a l t r i a l s was applied t i l l 1987: one hundred pts had p o s t o p e r a t i v e l y Tamoxifen,25 r a d i o t h e r a p y and 17 radiotherapy+Tamoxifen. The f o l l o w - u p o f 67 pts wasevaluable from 1 toTyears: relapses developed in 17 pts (7 cutaneus recurrences, 3 omolateral a x i l l a r y lymph nodal metastases, 6 bone and 1 l i v e r metastases) but only 6 o f them died. Ten pts died o f diseases d i f f e r e n t from cancer. Five pts developed a second tumor (3 c o n t r o l a t e r a l breast cancer, 1 lung tumor and I large bowel tumor). This analysis shows t h a t wide lumpectomy i n d a y - h o s p i t a l regimen is useful to obtain the local c o n t r o l o f t h e d i s e a s e i n t h e e l d e r ] y pts.

Abstracts 41

FACTORS INFLUENCING LOCAL RECURRENCE A F T E R EXCISION A N D R A D I O T H E R A P Y FOR PRIMARY BREAST CANC E R . ~ R W Blamey, A P Locke~, I OElIis, D A L Morgan, C W Elston, C i t y Hospital, N o t t i n g h a m , U.K.

42

Quadrantectomy, axillary dissection and radiotherapy is by now a consolidated treatment in small breast carcinoma. A statistical analysison 1232 patients treated with QUART from i970 to 1983 at the National Cancer Institute of Milan shows a probability in ten years of local breast failure of 4,1% (2,3-5,9 confidence interval 95%). The probability for second ipsilateral tumor resulted 2,7% .(confidence interval 95%).In about 20% of the patients a deformity worth of sup= gieal correction can be observed. Reconstruction can be done immediately, or it can be delayed after radiotherapy. Radiodermitis and fibrosis subsequent to radiotherapy inter= fere with delayed reconstruction procedures. Breast deformi= ties observed after quadrantectomies (dislocation and/or distortion of areola and nipple, localized deficit of skin and glandular tissue) ape best corrected immediately. Deformities such as asimmetry due to fibrosis and retraction of the gland can be corrected after a few months, most of the times with a contralateral mastoplasty. Reconstructions chances include reshaping of the gland, volume augmentation with custom pro= sthesis, centralization of the areola and local flaps. New techniques for reconstruction of defects after central and inferior quadrantectomies are outlined. Post-op cosmetic re= sults after immediate and delayed reconstructions are shown.

Vascular invasion in turnout P a t i e n t a g e (< 50 years) Tumour size (> 3 cms) On the basis ol this we have a l t e r e d our selection policy.

INTRAOPERATIVE ELECTRON BEAM THERAPY FOR BREAST CANCER R.R. DOBELBOWER, JR., M.D.~ Ph.D.7 A. EiTaki, M.D., H.W. Merrick, M.D.S* D.G. Bronn 9 M.D., Ph.D., Medical College of Ohio, Department of Radiation Therapy and Department of Surgery~* C.S. #10008, Toledo, Ohio 43699 Between 1984 and 19881 seven women with breast cancer were treated with intraoperative electron beam therapy (IOEBT). All patients were Caucasian. Patient age ranged from 37 to 62 (median 53) years. Mistopathology was infiltrating duct carcinoma (6/7) or medullary carcinoma (1/7). Four tumors were staged TI NO MO, one TI NI MO, and T2 NI MO, and one lesion was recurrent. At the time of primary lumpeetomy or axillary node dissection, a dose of i0 or 15Gy IOEBT was administered to the tumor bed with 6 or 9MeV electrons through the lumpectomy wound. All patients received 45 or 50Gy over five to six weeks with 6MeVp photons to the breast and, and in four patients, to the regional nodes. Two patients received adjuvant chemotherapy and Tamoxifen. No post-operative complications were ohserved. All patients are alive as of April, 1988, with no evidence of recurrent disease. Excellent cosmesis is the rule. Clinical and technical aspects of treatment will he presented in detail along with the rationale for the use of IOEBT in the primary definitive radiotherapeutlc management of breast cancer.

RECONSTRUCTION AFTER CONSERVATIVE TREATMENT FOR BREAST CAN= CER. A.Grisotti, A.Luini, M.CallegaPi, M.Nava, V.Sacchini, V.Zanini. National Cancer Institute - Milan.

263 p a t i e n t s with p r i m a r y b r e a s t c a n c e r have been t r e a t e d by t u m o u r e c t o m y and whole b r e a s t irradiation. 56 local r e c u r r e n c e s have occurred in the t r e a t e d b r e a s t (giving an a c t u a r i a l local r e c u r r e n c e of 23% at 8 years), l g of t h e s e have been m a s s i v e uncontrollable local r e c u r r e n c e . We have analyzed a number of f a c t o r s to d e t e r m i n e which influence local r e c u r r e n c e and then carried out a Cox multiv a r i a t e analysis Ior the independence of I a c t o r s identified. The following proved independently s i g n i f i c a n t :

43

117

44

COSNESIS A F T E R IRRADIATION OF SUBCUTANEOUS BREAST PROSTHETIC IMPLANTS: LONG-TEPJ4 FOLLOW-UP IN SIX PATIENTS. *J Halpern, M MeNeese, S Kroll, Clinical Radiotherapy and General Surgery, UT M. D. Anderson Cancer Center, Houston, Texas 77030 The cosmetic effects of irradiation of breasts after prosthetic implants are not well documented. At MDAH, 6 patients7 aged 34-53, were followed for 3 to 8 years after irradiation of their implanted breasts. Five of the six patients had augmentation by subcutaneous implants without a previously known diagnosis of breast cancer. Four patients had bilateral breast augmentations 7, 5, 5 and 3 years prior to their radiotherapy. Two patients had reconstruction: in one, microscopic cancer was found after prophylactic mastectomy for fibrocystic disease, and the second one developed recurrence 5 years after reconstructive surgery for a Stage I breast cancer. Follow-up photographs show that only two patients enjoyed good cosmetic results. Both these patients had good cosmesis prior to radiotherapy, but in the 4 patients with poor cosmetic results, cosmesls was poor before initiation of the radiotherapy, with capsular contractures in both the irradiated and the intact breasts. The contraetures, however were more prominent on the irradiated side. Both patients who enjoyed good eosmesis received only 45 Gy over 5 weeks, whereas the other 4 patients had slightly higher doses or were treated in a shorter time period. It may be concluded that even though irradiation of breasts with subcutaneous prostheses is not recommended, if the aesthetic quality of the implant is good prior to treatment, the radiotherapy may produce acceptable cosmetic results. Whenever feasible; 45 Gy/5 weeks is preferred over higher doses and fractions. (Supported in part by NIB-CA-06294).

118 45

47

Abstracts

PAGET'S DISEASE OF THE BREAST: A CONTRAINDICATION TO BREAST PRESERVATION ? BE Krause,* BD Willan, RN Ludwig, Cross Cancer Institute, Edmonton, Alberta, Canada Between 1963 and 1984, 123 cases of histologically verified Paget's disease of the breast (PD) were registered at our institution. The records of 67 patients (54.5%) presenting with PD without a clinically palpable mass (TO), were reviewed with regards to their pathology, prognosis and suitability for breast preservation. Pathologic examination of mastectomy specimens in 65 cases revealed an associated invasive carcinoma in 24, intraduct carcinoma in 33 and 8 patients with PD alone. Sixty seven percent had non invasive disease.The ten year disease free survival in all TO patients was 87%. Mammograms were performed in 39 patients, allowing correlation with mastectomy findings. In 20, the radiographs were available for review, in the remainder, original reports were utilized. When the radiographs where negative or abnormalities were limited to the subareolar region, tumor was usually confined to the nipple areolar complex. In conclusion, patients presenting with TO PD frequently have small tumors at the base of the nipple whose presence can be predicted mammographieally. In selected patients, removal of gross disease by excision of the nipple areolar complex should result in acceptable local control after radiation therapy. Preservation of the breast mound, even with loss of the nipple may be an alternative to mastectomy in some patients.

46

AXILLARY LYMPH-NODE DISSECTION IN BREAST CANCER: ANALYSIS OF 1446 CASES. V.Galimberti, *V.Sacchini, A.Luini, G.Farante, S.Marchini, Istituto Nazionale Ttm~ri 20133 Milano Italy 1446 patients affected by operable breast carcincma, treated with complete axillary dissection since 1983 to 1986, were evaluated for the distribution of the axillary nodes by levels. The median age of the patients was 52 years (range 23-86 Y.). The average ntmJ3er of nodes remuved was 20.3 per patient: 13.5 at the first level, 4.4 at the second and 2.3 at the third level. 839 patients (41.9%) had nodal involvement. No difference between Nand N+ patients was noted in terms of nodes removed. The average nt~nber of lymph-nodes was 20.7 in 167 Halsted mastectomy, 20.4 in 732 Patey mastectomy, 20 in 340 quadrantecton~ with axillary dissection in continuity, and 20.3 in 207 quadrantectomy with axillary dissection in discontinuity. The n~nher of nodes involved resulted correlated to tuner size. 59.2% of the T1 group resulted N-, while in the T3 group the patients N- were 22%. 19.2% of the TI group had just one node involved, and 8.2% more than i0 nodes involved. In the T3 group just one involved node was found in 10% and more than i0 nodes in 38%. In the T1 group with axillary metastases only the first level was involved in 64%, while all levels in 14.5%. In the T3 group the first level was positive in 25.6%, and the three levels in 38.4%. 96.4% of the N+ patients had a regular distribution of the metastases through the levels, while 3.6% had skipping distribution. The first level was skiped in 1.3% (i0 cases with only the second level positive and in 1 case only the third). The second level was skiped in 2.4% of the cases (19 with positivity of the first and the third level together, and in 1 case only the third.)

48

'A CORRECT SURGICAL APPROACH FOR AXILLARY DISSECTION INTENDE~ AS STAGING PROCEDURE FOR RESECTABLE BREAST CANCER Muscolino G., Marolda R., Luini A. Istituto Nazionale Tumori - Milano The role of axillary dissection in breast cancer is being debated. Lymph node metastases are now considered "indicators but not governors" of prognosis. Thus, the removal of metastatic nodes does not affect "per se" disease evolution. Never theless, the assessment of nodal status might theoretically help in designing adjunctive treatments. Based on these premisses we began performing axillary dissection saving pectoralis muscles and nerves. From 1984 to 1985, at Istituto Nazionale Tumori of Milan,lOS consecutive patients were treated by a surgical team using this procedure, whereas 108 patients consecutively underwent classical surgery by another team of surgeons in the same Institute. Regarding the number of lymph nodes removed, they averaged 19.0 in the former and 18.9 in the latter. Disease free survival and overall survival analysis did not show significant differences between the two groups of patients. Although the knowledge of lymph nodal status has not determined dramatical impact on clinical course of the disease, through the use of adjunctive treatments, it might be useful to select a group with high risk of relapse, in which interesting investigations can be carried out on biological behaviour, to design more rational treatments.Accordingly, as long as axillary dissection is considered as a "dogma" in resectable breast cancer, this technique should be taken into consideration.

TAMOXIFEN AS AN ALTERNATIVE TO SURGICAL RESECTION FOR SELECTED GERIATRIC PATIENTS WITH PRIMARY BREAST CANCER. *RS Foster Jr, R Margolese, University of Vermont, Burlington, Vermont 05405 and Jewish General Hospital, Montreal, Quebec H3T IE2. Elderly patients tend to have breast cancers that are well differentiated and are commonly positive for estrogen and progesterone receptors. We are reporting on 26 women (median age 82) with primary-operable invasive breast cancer who, because of infirmity or refusal of surgery, were treated with the antiestrogen tamoxifen I0 mg. twice daily, after needle biopsy only. Seventeen had reduction of tumor volume (5 with complete regression and 12 with partial regression [>50%]). Six patients had minimal change. Three patients had progressive disease. The latter 3 and 5 others, who after an initial response had recrudescence of disease (mean response of 33 months), were treated locall/ with surgery or radiotherapy. No patient has developed uncontrollable local-regional disease. Twenty-two patients are still alive at a mean interval of 43 months since initiation of treatment (range 6-90 months). One of the 8 patients requiring localregional treatment has died (mean follow-up 52 months). Three of the 18 not requiring local-regional therapy have died (mean follow-up 36 months). For most older primary-operable breast cancer patients, the principles of local-regional management with surgery +/- radiotherapy are similar to those for younger women. In selected geriatric patients where avoidance of hospitalization and/or a surgical procedure is an important part of quality of life considerations, treatment with tamoxifen permits a delay of surgery, which for some exceeds their life expectancy.

Abstracts 49

SUBSTITUTED BENZOPHENONE DINITROPHENYLHYDRAZONES WITH ANTITUMOR ACTIVITIES IN H O R M O N E D E P E N D E N T BREAST TUMORS. L R Morgan, L E Gillen, and C L Hooper, Dekk-Tec, Inc., a n d T h e K i n g F o u n d a t i o n f o r C a n c e r R e s e a r c h , N e w O r e l e a n s , LA. 7 0 1 1 9 Dinitrophenylhydrazones of s u b s t i t u t e d benzephenones are described which are devoid of estrogenic (uterotrophic) activity and have greater antiuterotrophic activity than has tamoxifen (TAM). Previous studies with the ZR-75-1 estrogen dependent human breast cancer cell clonogenic assay documented t h e i n h i b i t o r y a c t i v i t y of 4 , 4 ' dihydroxybenzophenone-2,4-dinitrophenyl-hydrazone ( A - 0 0 7 ) in t h e p r e s e n c e or a b s e n c e of e s t r a d i o l ( A A C R 29, A b s t 965, 1 9 8 8 ) , O p t i m u m a n f i t u m o r activities were seen with A-007. In t h e D M B A - r a t b r e a s t c a n c e r m o d e l , A - 0 0 7 in d o s e s of 0 . 3 7 , 0 . 7 4 , a n d i . i i m g / k g d a i l y s.c. f o r 28 d a y s p r o d u c e d a 2 5 - 5 0 % r e d u c t i o n in i n i t i a l t u m o r v o l u m e as c o m p a r e d to a 1 2 5 % i n c r e a s e p r o d u c e d b y 0 . 1 8 m g / k g T A M a n d an 8 5 0 % i n c r e a s e s e e n in c o n t r o l s . O v e r a l l , a 3 8 % C R w a s s e e n in t h e r a t s t r e a t e d w i t h A - 0 0 7 . A - 0 0 7 h a d no i n f l u e n c e on T A M ' s p o t e n t i a t i o n of t u m o r g r o w t h . Additive estradiol blocked both TAM and A-007 activities in t h e D M B A m o d e l . In 50 d a y o l d f e m a l e r a t s w h o h a d r e c e i v e d 20 m g D M B A 30 d a y s p r e v i o u s l y , d o s e s of 200 u g of A-007 given daily s.c., prevented the developm e n t of E R / P R p o s i t i v e t u m o r s . Long term continuous u s e of 300 m g of A - 0 0 7 d a i l y to D M B A t r e a t e d r a t s w i l l b e d i s c u s s e d . Acute and long term toxicities of A - 0 0 7 in r a t s and mice will be discussed.

50

51

EFFECTS CF A LONG-ACTING IH-RH A N A L O G , ~ OR IN COF~INATIC~ WITH TAM3)~F~N, ON []~BA-I~DUCSDRAT MAMMARY TtMOF~. F.Pessa, ~S.Zenardi,R.DeMenesh, M.Landucci,G.Cerfuti,F.Alesssndri,F.Boccardo~titutoNszionaleper la Ricerca sul Cencro, 16132 Geneva, Italy. Both te~noxifen(TMX)andanalogs of LH-RH(LH-P~-A),which act through dif ferent mechanisms,sppear to be effective in bresst cancer ther~0y.In the present study the effects of a icr~-acting LH-RH-A(single injsotien of S.6 n~ D-TRP-6 LH-~ in a depot formulaticn),aloneor in ccmbination with TMX (50 u~/d~y), were investi~ted in the DMBA-indu~ed rat memmary tt~or model. Vehicle treatsd intact or c~ariectomized animals served as controls. After a ~-week trea~nent period t[~Dr s~face end cytosol sex steroid receptors were as follows: ~SOP IUMOR SU~ACE" STEROID RECEPTORS° Lnitial Finsl ER PgR CONTROLS (n=ll) 349 + 45 635+102 11.5 9.1 OVARIECTd~ (n: 9) 322 + 46 172 + 57** 17.0 0 TMX (n=lO) 294 + 29 289 + 77* 5.5 31.0 LH-RH-A (n=12) 329 + 43 185 + ~ 11.7 0 LH-RH-A + TMX (n:lS) 338 + 25 263 + 75** 2.7 i0.3

52

'Mesn(_+S.E.M.)ttnorarea per rat in sq.~m; °medien values in fmol/mg cytosol protein; ~p<0.05 end ~p
119

STRUCTURE-ACTIVITY RELATIONSHIP OF METABOLITES OF TAMOXIFEN TO CONTROL BREAST CANCER CELL GROWTH. CS Murphy,* R MeCague and VC Jordan, Department of Human Oncology, University of Wisconsin Clinical Cancer Center, Madison, WI and Cancer Research Campaign Laboratory, Institute of Cancer Research, Sutton, Surrey, England. An estradiol-responsive subline of T47D human breast cancer cells has been developed to evaluate structureactivity relationships of metabolites of the nonsteroidal antiestrogen tamoxifen. Fixed-ring derivatives of tamoxifen metabolites were synthesized to test the activity of hydroxylation at various positions without the effects of isomerization. Estradiol (I0 -I0 M) increased cell proliferation 5-fold in a 7-day growth assay. The nonfixed-ring t r a n a form of metabolite E (tamoxifen without its side chain) isomerized to the more estrogenic cis form and was a full agonist. However, the fixed-ring form of t z a n s metabolite E was only a weak partial agonist. Both the fixed-ring and nonfixed-ring form of c i s metabolite E were full agonists (EC50 = 2 x 10-it M). The addition of an aminoethoxy side chain to the estrogenic c i s fixed-ring form of metabolite E (forming t r a n s monohydroxytamoxifen) resulted in the loss of estrogenic activity. This compound is a potent antiestrogen (ICs0 = 10-s M). Addition of the side chain to the t r a n s fixed-ring form of metabolite g to produce cia m0nohydroxytamoxifen produced a weak antiestrogen (IC50 : 10 -7 M). The results demonstrate that metabolite E can isomerize from its partial agonist t r a n s form to a full agonist cis form in tissue culture. The positioning of the hydroxyl group appears to be important in determining the potency of the agonist activity of these compounds. The addition of an aminoethoxy side chain abolishes estrogenic activity. Supported by NIH grant CA-32713. CSM was supported by NIH traning grant 0A-09471.

THE EFFECT OF LHRH AGONIST, ZOLADEX, ON O V A R I A N HISTOLOGY. *J.F.R. Hobertson, K. Williamson, I.O. Ellis, R.I. Nicholson, R.W. Blarney, Nottingham City Hospital, UK In premenopausal women with advanced breast cancer the luteinising hormone releasing hormone (LHRH) Zoladex produces response rates similar to surgical oophorectomy. There have been proposals both to use Eoladex as adjuvant therapy and in benign disease: irreversible ovarian suppression would be an unwanted side effect in these patients. We have looked at the histological changes in the ovaries of patients with advanced breast cancer treated by Zoladsx. In a phase 1 clinical trial of Zoladex as first line hormonal therapy in premenopausal women with advanced breast cancer, patients o n progression of their disease underwent surgical oophorecotmy. The histology of 39 ovaries from 23 women treated with Zoladex (Zx) has been compared to 68 ovaries from 34 patients who underwent surgical oophorectomy as primary, therapy (Ox). Histologically both the Zx and Ox groups show similar follicular phase development. Follicular cysts were seen more often in the ovaries of Zx treated patients than in the Ox patients (p < 0.05). Corpus lutea were seen more often in the ovaries of Ox patients than in the Zx patients (p < 0.01). Zoladex appears to arrest development of follicles with formation of follicular Cysts. Despite low FSH levels folliculogenesis was not inhibited.

120 3

Abstracts T H E L H R H AGONIST, ZOLADEX, A N D A N T I O E S T R O O E N , TANOXIFEN - A R A T I O N A L C O M B I N A T I O N IN P E E M E N O P A U S A L W O M E N W I T H B R E A S T C A N C E R ? *J.F.R. Robertson, K e r r y Walker, R.I. Nicholson, R.W. Blamey, N o t t i n g h a m C i t y H o s p i t a l a n d T e n o v u s Insfitute, Cardiff, U.K.

4

In the p r e m e n o p a u s a l w o m a n the u s e of Z o l a d e x r e s u l t s in t h e p r o d u c t i o n of c a s t r a t e levels of oestradiol. The c o m b i n a t i o n of Z o l a d e x a n d T a m o x i f e n has been proposed in p r e - m e n o p a u s a l women. This requires endocrinological assessment to s h o w t h a t t h e i r c o m b i n e d e f f e c t s are not a n t a g o n i s t i c (because T a m o x i f e n a l o n e s t i m u l a t e s h i g h o e s t r a d i o l peaks). W e h a v e s t u d i e d 34 p a t i e n t s w i t h o e s t r o g e n receptor positive and negative tumours treated with Z o l a d e x a n d T a m o x i f e n as i n i t i a l s y s t e m i c t h e r a p y . Z o l a d e x w a s g i v e n as a m o n t h l y s u b c u t a n e o u s i n j e c t i o n (3.6 mg) a n d T a m o x i f e n 20 m g b.d. E n d o c r i n e d a t a was a v a i l a b l e in 23 p a t i e n t s . As w i t h Z o l a d e x a l o n e patients on Zoladex and Tamoxife~ show transient s t i m u l a t i o n of F S H a n d LH l e v e l s o v e r the first 7 - 10 days (peak m e a n values 7 mU/ml and 14.9 mU/ml respectdvely) w i t h s u b s e q u e n t l o w g o n a d o t r o p h i n levels (mean < 2 m U / m l a n d < 1 m U / m l r e s p e c t i v e l y ) . Serum oestradiol and progesterone levels were reduced to c a s t r a t e levels. ConClusion T h e r e is no e n d o c r i n o l o g i c a l c o n t r a i n d i c a t i o n to the u s e of Z o l a d e x a n d T a m o x i f e n t o g e t h e r in prem e n o p a u s a l w o m e n w i t h b r e a s t c a n c e r e i t h e r as adjuvant therapy o r in treating advanced disease. One a d v a n t a g e of Z o l a d e x o v e r s u r g i c a l oophorectomy, particularly as a n a d j u v a n t t h e r a p y , is t h a t t h e castration effects are reversible on stopping Zoladex therapy.

55

The Frequency of Flare Reactions in Post-Menopousal Women with Advanced Breast Cancer Treated with Megestrol Acetate (MA) or Tamoxifen (Tam). P. ,.9onomi~ J. Von Roenn, P. Johnson, J. Wolter, !<. Anderson, S. Economou, Rush University & Medical Center and Northwestern University & Medical Center, Chicago, 111. 60612 Flare reactions may occur in patients receiving additive hormonal therapy f o r advanced breast cancer. The purpose of this retrospective study was fo evaluate the frequency of flare reactions in relation to steroid receptor levels in post-menopousol women with advanced breast cancer treated with MA or Tam as primary or secondary hormonal therapy. Flare reactions were defined as the appearance of hypercalcemia (3 pts)t increase in size or number of soft tissue lesions (3 pts), appearance of o malignant pleural effusion ¢I pt)occurring within 14 days of starting MA or Tam. The patient sample consisted of 114 post-menopausal women with advanced breast cancer. Primary hormonal therapy was: MA 65 pts and Tam - 49 pts; and secondary hormonal therapy was MA - 39 pts and Tam - 38 pts. The following patient characteristics were observed: median age 60 years~ disease free interval < 2 years 72 pts~ soft tissue dominant disease 40 pts, bone dominant disease 39 pts, and visceral dominant disease 35 pts. Defining_< 10 fm/mg protein as "positive", the distribution of estrogen !ER) end progesterone receptors (PgR) was: ER+PgR+ 61 pt% ER+PgR- I4 pts~ ER-PgR+ 20 pts, and ER-P#R- 19 pts. Flare reactions were observed in /4 (/+.6%) Tam treated pts and 3 (3 q'~) MA treated pts. Two pts. who experienced a flare reactlon on MA were ER-PR+ and one was KR+PR+. Tam induced flares occurred in 2 ER+PR+ pts and 2 ER+PR- patients. No MA flares occurred in PgR-pts. and no Tam flares occurred in KR- pts. In conclusion, flare as defined in this study is uncommon during treatment with MA or Tam. These observations suggest that PgR plays a role in MA flare and that ER plays a role in Tam flare.

THE E F F E C T OF T A M O X I F E N O N T H E E X P R B S S I O N OF T U M O U R A N T I G E N S B Y I M M U N O C Y T O C H E M I S T R Y IN B R E A S T CANCER, *J.F.R. Eobertson, I.O. Ellis, J a n e Bell, R.W. B l a m e y N o t t i n g h a m C i t y Hospital, U N A n u m b e r of t u m o u r a n t i g e n s a r e k n o w n to be expressed by breast cancers. W e h a v e s t u d i e d the effect of hormone therapy on tissue antigen expression. We h a v e e x a m i n e d 3 t u m o u r a n t i g e n s (CEA, I 1 5 D 8 a n d N C R C - I I ) b y ir0rnunocytochemistry of t u m o u r specimens before and during Tamoxifen therapy. I m m u n o c y t o c h e m i c a l s t a i n i n g w a s s c o r e d 0 - 4. NCRC-II I15D8 CEA Staining (n = 28) (n = 27) (n = 25) score Pre Post Pre Post Pre Post 0 i 2

0 5 2

0 6 0

0 4 0

0 2 2

18 5 1

20 3 2

2 22

2

i

0

3

7

3

4

14

19

23 0 0 N.S, N.S. N.S. There was no significant change in the e x p r e s s i o n of t u m o u r a n t i g e n s in t u m o u r s p e c i m e n s before and during Tamoxifen treatment. The I15D8 and N C R C - I I m o n o c l o n a l a n t i b o d i e s w e r e r a i s e d to b r e a s t cancer cells and show similar staining patterns. T h e r e is a s i g n i f i c a n t d i f f e r e n c e in t h e s t a i n i n g p a t t e r n s b e t w e e n the two e p i t h e l i a l m e m b r a n e antigens, I15D8 a n d NCEC-II,

56

and CEA

(p < 0.001, X 2 test).

ACTIVITY OF G L U C O S E - 6 - P H O S P H A T E DEHYDROGENASE IN RELATION TO THE R E C E P T O R STATUS IN BREAST C A N C E R . H-3 L~ck and :~ Hilfrich, Medical School H a n n o v e r , Dept, of O b s t e t . and Gynaecology ( D i r e c t o r : P r o f Dr 3 Schneider) D-3000 Hannover 51, Podbielskistr. 380, G e r m a n y , M e a s u r e m e n t of e s t r o g e n r e c e p t o r in t u m o r tissue of patients with b r e a s t c a n c e r proves to be useful in s e l e c t i n g p a t i e n t s likely to respond to endocrine therapy, H o w e v e r , a positive ER does not g u a r a n t e e a response to endocrine t r e a t m e n t since about #0% of these p a t i e n t s fail to respond. The additional m e a s u r e m e n t of the p r o g e s t e r o n e r e c e p t o r i n c r e a s e s the s u c c e s s of an endocrine treatment, The aim of this study was to took for further information about the hormone responsiveness of breast cancer by the measurement of estradiol-inducible glucose-B-phosphate dehydrogenase (G-B-PDH). In 145 tumor cytosols (39 pre- and 106 postmenopausal) ER and PR were determined by DCC assay (pos.>20fmol/mg protein), G-6-PDH was measured photometrically at 365 nm in nmol/min/ mg protein at 25°C. As a function of receptor status we found the following results: (pre/postmenopausa]) G-6-PDH

ER+PR+ 19/22

ER+PRlg/i0

ER-PR+ 2#•32

ER-PR10/5

The results indicated that G - 6 - P D H could be an additional indicator for hormone responsiveness in breast cancer, The G-6PDH assay is not inlluenced by steroids in tissue and blood. The loss of enzyme activity in tumor tissue [rom postmenopausal women with the receptor status ER+PR- may be a loss of the protein inducing function of the estrogen receptor,

Abstracts 57

MONOCLONAL ANTIBODIES MBrl AND MBr8 AS PREDICTORS OF RESPONSE TO OOPHORECTOMY IN A D V A N C E D BREAST CANCERS.

58

M.Greco,iR.Agresti, E.ieo, N.Cascinelli.

121

F A C T O R S INFLUENCING ADVANCED BREAST C A N C E R . *R W Blarney, M R Williams, C W Elston, R I Nicholson, C i t y Hospital, N o t t i n g h a m , U.K.

NATIONAL CANCER INSTITUTE, MILAN, ITALY O n e hundred and t w e n t y - n i n e patients have r e c e i v e d endocrine therapy as initial t r e a t m e n t for distant m e t a s t a s e s . In all patients both the histological g r a d e and ER s t a t u s of p r i m a r y t u m o u r tissue~ the lymph node s t a g e and menopausal s t a t u s a t mastectomy~ and the d i s e a s e - f r e e interval and sites of initial m e t a s t a s e s are available for analysis. Four of those f a c t o r s h a v e been found to c o n t r i b u t e independently to survival a f t e r the initiation of t r e a t m e n t : t u m o u r grade, ER s t a t u s , d i s e a s e - f r e e i n t e r v a l and sites of m e t a s t a s e s . Employing a m u l t i v a r i a t e analysis, t h r e e groups of patienlcs have been identified w i t h survivals of 67, 37 and O per c e n t at 18 months on therapy.

The r e s p o n s e to o o p h o r e c t o m y in 40 p a t i e n t s with a d v a n c e d b r e a s t c a n c e r was e v a l u a t e d as a function of the p r e s e n c e or a b s e n c e of the a n t i g e n s recognized by the m o n o c l o n a l a n t i b o d i e s MBrI,MBrS, MOv2 on cell surface of the p r i m a r y tumor. Two groups of p a t i e n t s were evaluated. T h e s e groups had: a) the same d i s t r i b u t i o n of lymph node status, d i a m e t e r of primary, and p r e s e n c e of e s t r o g e n receptors at time of first tre~ment; b) a c o m p a r a b l e e x t e n t of the d i s e a s e at the time of oophorectomy. The 5 years survival r a t e s a f t e r o o p h o r e e t o m y was: 100% in one g~oup, and no s u r v i v o r s in the second. The f r e q u e n c y of c o m p l e t e r e s p o n s e s after oophor e c t o m y was s i g n i f i c a n t l y h i g h e r in p a t i e n t s whose tumors did not e x p r e s s the a n t i g e n s r e c o g n i z e d by the m o n o c l o n a l a n t i b o d i e s MBrl and MBr8. F o u r t e e n out of the 22 p a t i e n t s with tumors not r e a c t i n g with MBrl and ii out of the 18 n e g a t i v e to MBr8 had c o m p l e t e response. No c o r r e l a t i o n with r e s p o n s e was o b s e r v e d a s s o c i a t e d to the m o n o c l o n a l a n t i b o d y MOv2. MBrl and MBr8 w e r e both p r e s e n t in ii p a t i e n t s and both a b s e n t in other ii. The sim u l t a n e o u s a b s e n c e of the two a n t i g e n s was r e l a t e d with the h i g h e s t c o m p l e t e r e s p o n s e rate (9 out of ii patients).

59

POST-CHEMOTHERAPY AMENORRHOEA AS A PROGNOSTIC FACTOR IN BREAST CANCER PATIENTS TREATED WITH ADJUVANT CHEMOTHERAPY. S. Toma*,T. Coialbu, L. Repetto, A. Giacchero, M. Costantini and R. Rosso. *Ist. di Oncologia, Univ. di Genova-lstituto Nazionale per la Ricerca sul Cancro, Genova - Italy. Prognostic importance: of ovarian suppression by adjuvant chemotherapy in breast cancer patients is still object of debate. Variables such as age and hormonal receptors status may interfere in patient's outcome. We examined 288 consec~ tire patients who underwent adjuvant chemotherapy with a CMF based regimen for 10-12 courses. Minimum follow-up was 3 years. Patients were classified in 3 age classes:<43 years 55 patients, 45-54 years ii0 and >54 years 123. At the time of diagnosis 186 patients were postmenopausal and 102 were premenopausal. Ovarian suppression was induced in 59 of 74 premenopausal patients, in which this data was known(79.7%), during or immediately after the end of chemotherapy program. 3 years disease free survival was 72% in the group of patients with chemotherapy induced amenorrhoea (R.R.=0.86) and 48% in the premenopausal group in which amenorrhoea was not obtained (R.R=l.63)(p=n.s.); 3 years survival was 93% (R.R= 0.83) and 83% (R.R.=l. S8)(p=n.s.) respectively. At multivariate analysis young age (<45years) and increasing number of positive nodes were associated with a worse prognosis. Induced emenorrhoea seems associated with a reduction in the risk of relapse: however this observation does not reach statistical significance.

60

LONG TERM EFFECT OF ABJUVANT CHEMOTHERAPY ON BREAST CANCE~. G.H.Rong, Y.Z.C,,, W.~.Miao, YoC.Mao,and D.M.Yang. Department of Sur~.~cal Oncolo~y, People's Hospital, BeiJin~ Medical University, BeiJin~, CHINA. Clinical trial initiated in 1972 at People's Hospital included 312 female patients with invasive breast cancer ( Stages I, K285 and ~27 cases). After either radical or modified radical mastectomy, the patients were randomized into two group,- adjuvant chemotherapy administered ~n 144 patients and 168 not treated.lu the chemotherapy ~roup, sixty two pa+,ients (43%) received TSPA(2.4 courses, 150200mR each course) and eighty two (g?%) CMF(cytoxan,metho. trexate, ~-fluorouracil, 6-12 courses) postoperatively for about one year. All of the patients wer~ fol~owed up for at least over f i w y~ars and the longeet fifteen years. The median duration i~ nine years. No significant difference was observed in five year survival between adjuvant chemotherapy ~rcup and control in natients without axillary lymph node metastases. It ~s likely to increase 5 year survival of the the chemotherapy group (83=) with axilla~y lymph nodes metastases 1-3 in number versas control group (63~)o When the patients with axillary lymph nodes ~ 4 in number~ the 5 year survival o~ chemotherapy ~roup is significantly higher (6~) than the controls ( 2 5 ~ (p~o.o3). There is no apparent difference of the survival between TSFAand CMFcbemotherapeutic regimens. In both re~:mens no severe toxicity nccured as well.

122 61

Abstracts IMPACT OF ADMINISTRATION- RELATED FACTORS ON OUTCOME OF ADJUVANT CHEMOTHERAPY FOR PRIMARY BREAST CANCER. E. Campora W, P. Pronzato, D. Amoroso, G. Bertelli, F. Botto, P.F. Conte, M.R. Sertoli, R. Rosso. Istituto Nazionale per la Rieerea sul Canero, V.le Benedetto XV, 10, 16132 Genova, Italy.

62

The success of adjuvant chemotherapy for primary breast cancer depends on various factors, the majority of which are intrinsic to the patient or the tumor. However, factors zela ted to the modality of chemotherapy administration such as total number of courses received, time lapse between surgery and the start of chemotherapy and dose intensity may also s! gnificantly effect treatment outcome. Survival of 229 consecutive patients treated with adjuvant CMF (all drugs given i.v. on day 1 q 21 days for 12 courses) following surgery for primary breast cancer was analyzed to evaluate the impact of the above mentioned administration- related factors. All parameters were found to be significantly associated with survival of patients in a unlvariate analysis. Multivariate analysis confirmed the independent importance of the time from surgery to the start of CMF and of dose intensity. Although prospective studies are needed to confirm these results, clinicians should be aware that unnecessary delays in comeneing adjuvant therapy or dose reductions during therapy could be detrimental to patient survival.

63

HIGH-DOSE MITOXANTRONE/ETOPOSIDE/THIOTEPA WITH AUTOLOGOUS MARROW SUPPORT FOR RELAPSED BREAST CANCER. F Dunphy,* G Hortobagyi, A Buzdar, K Dicke, S Jagannath, J Yau, J Spinolo, J Ellis, B Radcliffe, and G Spitzer. U T M. D. Anderson Cancer Center, Houston, TX 77030. We employed high-dose mitoxantrone/etoposide/thiotepa (MVT) with autologous bone marrow support in 20 patients (pts) with metastatic breast cancer. 7 pts are too early to evaluate and are not reported. Pt characteristics at entry: median age 40 (1657), 8 premenopausal, 5 peri- or post-menopausal, 8 ER status <10 fmol/mg, 2 ER unknown, 3 ER +; 11 pts had >_2 sites of involvement. Disease sites: lung 54%, soft tissue/breast 85%, nodes 38%, bone 15%, brain 8%, liver 15%. Median disease-free interval from diagnosis was 61 weeks (range 0-205). 7 had firstrelapse breast cancer while 6 suffered multiple relapses (see table). Median number of chemotherapy cycles prior to MVT were 7 (2-31). 9 pts received 1 cycle and 4 pts received 2 cycles of MVT (M 30mg/m 2, V 1200mg/m 2, T 750rag/m2). There have been no toxic deaths. Response rate for first-relapse pts with progressive disease (PD) was 80%; overall response rate for multiple-relapse pts is 100%. 6 pts (46%) relapsed; 1 died 11 weeks post-MVT. Median follow-up from MVT is 13.5 weeks (535). Conclusions: high-dose MVT is an active combination for poor-prognosis breast cancer, including a) first-relapse breast cancer with PD to standard therapy and b) responsive or PD multiple-relapse breast cancer. Toxicity is acceptable. Follow-up is short; however, 46% of our pts have shown disease progression. Further follow-up and pt accrual is needed to more accurately evaluate this combination. Tumor characteristic Response Progression-free immediatelyprior to MVT* @ors to MVT survival(weeksl 1st relapse ist relapse S D t 1st relapse P D

i 1 S

Multiplerelapse MultiplerelapsePD

PR SD 4 PR, 1 P D

5+ iO+ 5,5+,8+,24+

$

S PR

14,22+,S0

8

1 CR, 2 PR

17,8,9

**

* remponsetostandardchemotherapy;~ SD = stable;** I pt NED postmurgery

64

HIGH DOSE CHEMOTHERAPYWITH AUTOLOGOUS HEMATOPOEITIC STEM CELL SUPPORT IN THE TREATMENT OF REFRACTORY STAGE IV BREAST CARCINOMA. Lynne S. Kaminer*, Stephanie F. Williams: Jan Beschorner, Janet &olick, Sheila O'Brien and Jacob D. Bitran University of Chicago Medical Center and Michael Reese Medical Center, Chicago, IL Nineteen patients with r e f r a c t o r y metastatic breast cancer were treated with high dose chemotherapy and autologous hematopoietic stem cell rescue in one of three successive phase I / I I studies. All patients received cyclophos~hamide (7.5 gm/m over 3 days) and thiotepa (150-225 mg/m over 3 days), 3 patients in addition received melphafan (4.5 mg/kg): and 12 patients received carmustine (150750 mg/m~). T o x i c i t i e s included pancytopenia, i n f e c t i o n , hemorrhagic c y s t i t i s , skin rash, nausea, vomiting, d i a r rhea, and mucositis. There were three toxic deaths (16%). One patient died secondary to sepsis and v e n t r i c u l a r tachycardia, one from sepsis and r e s p i r a t o r y f a i l u r e , and one from hepatic and pulmonary f a i l u r e . Granulocyte counts reached 500/ul a median of 19 days and p l a t e l e t counts reached 50,O00/ul a median of 49 days a f t e r stem cell r e i n fusion. The overall response rate was 75% including a 12% complete response rate. The overall median survival f o r a l l patients was 6.2 months (range 2-22 months). The median survival f o r nonresponders was 3.5 months. The median duration of response was 89 days (range 40-262). In our experience high dose chemotherapy with autologous stem cell reinfusion produces a high response rate in r e f r a c t o r y breast c a n c e r . However, because of the short duration of response and overall s u r v i v a l , we feel t h i s type of therapy should be u t i l i z e d e a r l i e r in the course of disease.

IMPROVED SURVIVAL IN LOCALLY ADVANCED AND INFLAMMATORY CANCER OF THE BREAST. M J Lopez,* W G Kraybill, A Khojasteh, Departments of Surgery and Medicine, Ellis Fischel State Cancer Center, Columbia, Missouri 65203 and Department of Surgery, Washington University School of Medicine, St. Louis, Missouri 63110. A/,ong 908 patients with breast cancer treated between 1975 and 1984, 34 (3.7%) presented with untreated locally advanced disease without demonstrable distant metastases at the time of diagnosis. Only patients with stage IIIB (T-4abc, NX-2,MO) or inflaramatory carcinoma were included in this review to provide some uniformity in a notoriously heterogeneous group of cancers. During the first 5 years(1975-1979) one half or 17 patients were primarily treated with sequential radiotherapy and chemotherapy (Group A). From 1980 to 1984 the management consisted of 3 to 4 courses of neo-adjuvant multidrug chemotherapy followed primarily by mastectomy (Group B). The mean followup for the most recent Group (B) is 48 months and no patient was lost to follewup. While the local control rate was the same for both groups (76.4%), the survival was remarkably different. Group A patients experienced a median survival of 15 months, over 50% were dead within one year and only one survived 5 years. By contrast, in Group B the median survival was 46 months with 9 (53%) patients alive between 30 and 66 months, six of whom are already 5 year survivors. While the overall mortality of patients with i n f l a ~ a t o r y breast cancer was greater in both groups, the survival of patients in Group B was better than Group A for both inflam/aatory and noninflarmmatory cancers ( P ~ 0 . 0 1 ) . Estrogen receptor and menopausal status did not influence survival. This data suggest a significant improvement in survival and delay in the establishment and progression of distant disease with the use of neo-adjuvant chemotherapy.

Abstracts

65

EFFECTOF POST RELAPSE SYSTEMIC THERAPY ON SURVIVAL OF

66

PHASE II TRIAL OF MITOXANTRONE, VINDEBINE AND METHOTREXATE IN ADVANCED METASTATIC BREAST CANCER (AMBC), J.M. Nabholtz*, P. Fargeot, A. Vanoli, D. Fric, D. Dramais S. Friedman, F. Mayer, C. de Gislain, J. Guerrin, Centre G.F. Leelere, 21000 Dijon; Hotel-Dieu, 7120B Le Cmeusot, France. A prospective phase II trial was designed to assess the efficacy and toxicity of a polyehemotherapy containing Mitoxantrone (MIT) i.v. (iSmg/m2), Vindesine (VDS) i.v. (2 mg/m2) and Methotrexate (MTX) i.m. (8 mg/m2) administered every 21 days in AMBC. Thirty-two evaluable patients (pts) received 127 courses (mean 3.97). Their characteristics were as follows : ages 39-77 (median B8.7) ; ECOG performance status 0 (9 pts), i (ii pts), 2 (12 pts) ; pmemenopausal (4 pts, 12.5 %), postmenopausal (28 pts, BT.B %]; metastatic sites were lung (20 cases), bone (17 eases), liver (7 cases), lymph nodes (5 cases), skin (5 cases) and CNS (4 cases) with 20 pts having 2 or more metastatic involved sites. All pts had been previously treated, 22 (88.7 %) with hormonotherapy and/or 25 with chemotherapy (78.1%), of which 14 pts had received Adriamycin-based regimen (mean dosage : BlO mg/m2, 200-800). Objective responses included 0 Complete Remission, 17 Partial Remissions ( 5 3 . 1 % ma~or responses), 9 No Change (28.1%) and 6 Progressive Diseases (18.8 %). Toxicity, assessed according to the WHO grades (O), was as follows : Granulopenia GI-2 (Ii pts), GS (9 pts), G4 (4 pts) ; Anemia gl-2 (15 pts), G3 (3 pts) ; Thrombopenia GI-2 (4 pts), GS (3 pts), G4 (i pt) ; Alopeeia G1 (3 DEs), G3 (3 pts) ; Emesis GI-2 (12 pts) ; Cardiac Rhythm GI (i pt), G2 (2 pts), G4 (I pt). Our results suggest that the combination of MIT, VDS and MTX is an effective regimen with acceptable toxicity in AMBC.

68

OUTPATIENT AFb[ (ADRIAMYCIN, FLLOR(X]RACIL, bIET}'I]IREXATE ~ R A P Y IN ADVANO~D BREAST CANCER. WP Vaughan and J Garvey. U. of Nebraska M e d i c a l C e n t e r , Omaha, NE 68105. Combination chemotherapy for breast cancer without a l k y l a t i n g a g e n t s may induce c o l l a t e r a l s e n s i t i v i t y or a t l e a s t not induce r e s i s t a n c e . A v o i d i n g the use of a l k y l a t i n g a g e n t s i n a d j u v a n t t h e r a p y may reduce l a t e c o m p l i c a t i o n s . AFM r e g i m e n s a r e b e i n g investigated as "remission i n d u c t i o n " p r i o r to c o n s o l i d a t i o n w i t h h i g h dose a l k y l a t i n g a g e n t s and a u t o l o g o u s marrow s u p p o r t . Most use h i g h dose methotrexate, require hospitalization and a r e q u i t e e x p e n s i v e . We have used a Q 3 wk regimen of A 20 mg/M~ and F 200mg/M 2 d a i l y f o r 3d by IV b o l u s f o l l o w e d by M 20 mg/M2 IV on d l 0 t o t r e a t 16 p a t i e n t s ( p t s ) w i t h s t a g e IV b r e a s t cancer and 5 p t s w i t h s t a g e I I I b r e a s t cancer f o l l o w i n g m a s t e c t o m y . A f t e r 5 c y c l e s , ABel was f o l l o w e d by r a d i a t i o n t h e r a p y i n t h e s t a g e I I I p t s and by c r o s s o v e r to nonA d r i a m y c i n c o n t a i n i n g t h e r a p y i n the s t a g e IV p t s . Ten of 21 p t s r e q u i r e d dose r e d u c t i o n f o r g r a n u l o c y t e count n a d i r <1000 and 9 / Z l p t s r e q u i r e d d e l a y of one or more c y c l e s because of f a i l u r e to r e c o v e r g r a n u l o c y t e count to >2000. One e p i s o d e of n e u t r o p e n i c s e p s i s o c c u r r e d , b ~ u c o s i t i s was seen i n a l l 21 p t s but was s e v e r e i n o n l y 6. V o m i t i n g was seen i n 8 p t s and d i a r r h e a in 2. All pts suffered alopecia. Other toxicities included fatigue (3 p t s ) , d e p r e s s i o n (2 p t s ) , and p h l e b i t i s , vaginal mucositis, s e v e r e c o n s t i p a t i o n and s e v e r e anemia (1 p t e a c h ) . Seven of 8 p t s who had s t a g e IV d i s e a s e and no p r i o r chemotherapy a c h i e v e d a complete or p a r t i a l r e s p o n s e . Only 3/7 p t s who had r e c u r r e n c e a f t e r p r i o r a d j u v a n t chemotherapy r e s p o n d e d . Median d u r a t i o n of r e s p o n s e w i l l be >6 mo. A l l s t a g e I I I p t s remain d i s e a s e - f r e e w i t h f o l l o w - u p from 3 t o 21 mo (median 8 mo). This outpatient AFM r e g i m e n was w e l l t o l e r a t e d and produced a h i g h r e s p o n s e r a t e w i t h a c c e p t a b l e t o x i c i t y i n t h i s pt p o p u l a t i o n .

PATIENTS DEVELOPING METASTATIC BREAST CARCINOMA J. Rou~ss~, S. Friedman, I. Gnash-Jordan, K. Hacene, M. Brunet. Centre Ren~-Huguenin~ St. Cloud, France A retrospective analgsis bg Cox' survival model was performed in 760 patients uith relapsed breast c a n c e r who uere initiallg without detectable metastases (MO), for the effect of post relapse therapg. Initial anal~sis included the factors of menopause, stage, clinical tumor aggressiveness (PEV), (ar,~ adverse factors associated with) pre relapse chemotherapw and time to relapse. Both time to relapse (p<.OOO0011~ and factors associated with giving prelapse chemc,therapg (p(.O00001) negativelg a£fected s u r v i v a l . Post relapse therapy (theme p<.O001, and hormone ther.ap~ <,00001, replacing chemotherapy in the model) positivelg affected o v e r a l l s u r v i v a l (measured from diagnosis) ~or both operable and inoperable subgroups. In the inoperable group~ PEV was an additional ris~ factor (p<.O005). However, ~or the operable group, when histologic grade and nodal status were introduced into the analgsis, no post relapse thera~9 showed effect on overall survival in ang subgroup o~ relapsing patients (initially NO, NI-3, NS+) measured From relapse, nor. was an e f f e c t of adjuvant chemotherapg on survivat seen measured from diseasqe onset. The addition of site of relapse, did not change these figures. Ristograde (p
67

ESCALATING HIGH DOSE INTENSITY CYCLOPHOSPHAMIDE, METHOTREXATE AND 5-FLUOROURACIL (CMF) G I V E N O V E R 24 H O U R S P L U S V I N C R I S T I N E (V) A N D P R E D I S O L O N E (P): PROSPECTS FOR I N C R E A S E D C U R E R A T E S IN POOR R I S K BREAST CANCER. L A P r i c e * and B r i d g e t T Hill ** • 111 H a r l e y S t r e e t p L o n d o n W I N IDG and ** Imperial Cancer Research Fund Labs., L o n d o n W C 2 A 3PX, U.K. Increased dose intensity adjuvant chemotherapy may i n c r e a s e the c u r e rate in h i g h risk b r e a s t c a n c e r ( H r y n i u k and Bush, J C l i n O n c o l . 1984, 2 : 1 2 8 1 ) . T h i s c a n be a c h i e v e d by g i v i n g d r u g s o v e r 24 h o u r s ( P r i c e and Hill, Proc. ASCO, 1986, 5:62). 15 p a t i e n t s w e r e g i v e n 81 c o u r s e s of o u r 24 hour CMFVP protocol. Myelosuppression was moderate with maximum f a l l s of 70% and 65% in b a s e l i n e neutrophil and platelet counts. Treatments were g i v e n e v e r y 21 d a y s in 12 p a t i e n t s . The o v e r a l l r e c e i v e d ,!rse i n t e n s i t y ( R D I ) c o m p a r e d w i t h standard Cooper projected dose intensity were C 0.45:1, M 3.1:1 and F 3.48:1. The o v e r a l l RDI w a s 2.3:1. T h i s r a t i o c a n be s a f e l y i n c r e a s e d to 5.4:1 by q u a d r u p l i n g the d o s e of M . The a v e r a g e total d o s e of P a n d V w e r e 450 mg and 10 mg. W i d e r use of this m e t h o d m i g h t i n c r e a s e the c u r e rate in p o o r rlsk b r e a s t c a n c e r .

123

124

Abstracts

69

FOLINIC ACID MODULATION OF LOW-DOSE FLUOROURACIL INFUSION IN REFRACTORY BREAST CANCER. A PRELIMINARY REPORT. K. Jabboury, F. Holmes, G. Hortobagyi. UT M. D. Anderson Cancer Center, Houston, Texas 77030. Folinic acid (200 mg/m2/day I . V . bolus) modulation of low-dose d a i l y f l u o r o u r a c i l continuous infusion (200 mg/m2/ day) was explored aiming at therapeutic enhancement of the fluoropyrimidine. Ten heavily pretreated(median, 5 regimens) patients with r e f r a c t o r y breast cancer and p r i o r f l u o r o u r a c i l exposure in 9 were treated in this dose-searching study. The median age was 40 years (range, 26-59). Median performance status (Zubrod) was I (4-0). Estrogen receptor was positive in 2 patients, negative in 7 and unknown in 1. A median of 2 p r i o r hormonal therapies (O-3)was administered. The dominant metastatic s i t e was visceral in 5, bone in I and soft tissue i n 4 for a median of 2 sites (1-4). A t o t a l of 32 infusion courses were given for a duration of 5-10 days. Eight-day i fusion courses appeared to be optimally tolerated with a median rest period of i0 days between courses. Oral mucosal t o x i c i t y ( g r a d e I and I I ) was noted in 31%,and hand-footsym drome(grade l ) i n 13%, of courses, however reversible within i week. Other episodic observation of possible relevance were nausea (3 p a t i e n t s ) , lacrimation (1 course), nasel congestion with blood-tinged discharge (2 p a t i e n t s ) , fatigue (4 patients) and soft stools in h a b i t u a l l y constipated patients (4 courses). Myelosuppression was not observed. Responses were seen in 7 patients including I complete and 6 p a r t i a l responses for 1+-3.5 months. Two patients responded despite p r i o r disease progression on protracted low-dose fluorourac i l infusion ( I primary, and I secondary resistance). This regimen is active and well tolerated in r e f r a c t o r y breast cancer. Shorter rest periods may be achievable without augmenting the cumulative nature of mucocutaneous t o x i c i t y of prolonged f l u o r o u r a c i l infusion. Further evaluation i n a pat i e n t population with more favorable prognostic factors i$ warrante~L.A s~h~dule-optimization study is currentl~ underwa~

70

71

IMPACT OF SECOND-LINE THERAPIES ON SURVIVAL IN METASTATIC BREAST CANCER. D. Frye, G.N. Hortobagyi, A.U. Buzdar. The University of Texas M. D. Anderson Cancer Center, Houston, TX 77030. Combination chemotherapy (CT) in metastatic breast cancer can palliate symptoms in a large proportion of patients and may result in some improvement in survival. The median time to progression on Doxorubicin-containing regimens (DCR) has remained unchanged between 1973 and 1982. Secondary chemotherapy and hormonotherapy regimens after initial treatment with DCR induce objective remissions in 15-30% of patients (pts) The effect of these treatments on survival remains unclear. At our institution, in the past 14 years, a progressively increasing fraction of pts have been treated with secondary regimens after relapse following front-line CT. A consecutive series of 1424 pts with metastatic breast cancer treated between 1973 and 1982 in Doxorubicin-cont a i n i n g t r i a l s was evaluated. 1332 of these pts (93.5%) have developed progressive disease. Pts were divided into 3 groups according to the date of progression after Doxorubicin: 217 pts between 1973 and 1975, 588 pts 1976 to 1978, and 527 pts 1979 to 1982. Performance status, dominant disease and the extent of metastases at the start of front-line UT were comparable in each time period. Survival was calculated from the date of failure to Doxorubicin therapy. These data are shown in the following table: Survival in Months Time Intervals 75% 50% 25% 1 9 7 3 - 1 9 7 5 3 6 13 1976-1978 3 7 15 p=<0.001 1979-1982 4 9 20 These data illustrate that the increasing utilization of second-line therapies in recent years is having a modest hut favorable impact on survival.

72

REUTILIZATION OF DOXORUBICIN (DX) IN BREAST CANCER ~BC) PATIENTS (PTS) WHO FAILED ADJUVANT CHEMOTHERAPY. S. Kau, G. Fraschini, A. Buzdar, G. Hortobagyi. UT M. 0. Anderson Cancer Center, Houston, Texas 77030. Between 1975 and 1982, 736 pts with Stage I I - I I I BC were entered in consecutive t r i a l s of DX-containin 9 adjuvant chemotherapy and 317 eventually relapsed. A retrospective review of these records i d e n t i f i e d 53 pts who had s u f f i c i e n t follow-up documentation and were rechallenged with DX alone or in combination a f t e r relapse. With exception of two who had received inadequate doses, the pts were retreated only i f relapse had occurred >6 months a f t e r the l a s t dose of DX. All had received adjuvan~ DX by i . v . bolus a d m i n i s t r a t i o n , a t a median cumulative dose of 290 mg/m2 (range, 40-420). Median time from l a s t dose to relapse was 25 months (range, 178), median Zubrod performance was ~ (range, 0-4). Dominant site of disease was visceral in 42 pts, bone in 8, and soft tissue in 3. Prior to DX rechallenge, 26 pts received hormonotherapy and 12 received other chemotherapy. Eighteen pts received additional DX by bolus, while continuous infusion over 48 to 96 hours was employed in 35. Dx was combined to new agents (mitomycin-C or v i n b l a s t i n e ) w i t h a c t i v i t y against BC in 9 pts while 44 received some or all of the same drugs used in the previous adjuvant regimen. Median additional DX dose was 200 mg/m2 (range, 50-650). Objective responses (4 complete, 21 p a r t i a l ) were documented in 47% of the pts, independently of the combination with new, active agents. Median time to progression was 13 months (range, 3-82+) for responders and 4 months o v e r a l l . The response rate did not c l e a r l y correlate with the time from last adjuvant DX dose to relapse. Three pts developed congestive heart f a i l u r e , medic a l l y compensated,at t o t a l DX doses of 405 and 520 mg/m2 by bolus, and 650 mg/m2 (320 by bolus), respectively. In conclusion, DX-based chemotherapy may be e f f e c t i v e l y r e u t i l i z e d in pts who relapse a f t e r similar adjuvant chemotherapy.

CLINICAL TRIAL OF CONTINUOUS INTRAVENOUS INFUSION (CTI) OF 5-FU OR ADRIAMYCIN (ADRIA) IN METASTATIC BREAST CARCINOMA: AN EASTERN COOPERATIVE ONCOLOGY GROUP STUDY (ECOG). AK Hatfield*, DS Neuberg, RF gerris, DC Tormey, MA Abeloff, and Participating Investigators. Carle Cancer Center CCOP, Urbana, IL 61801, and University of Wisconsin, Madison, WI 53706 Eighty-six patients (pts.) with hormonally unresponsive metastatic breast carcinoma were treated with either CTI 5-FU (250 to 300 mg/mZ/day) or CTI Adria (2.5 to 3 mg/m2/ day). Pts. with less than 300 mg/m 2 prior exposure to Adria were randomized between treatments, while pts. with greater than 300 mg/m 2 prior Adria exposure were assigned directly to 5-FU. Eligible pts. had received at least one, but not more than 3 prior chemotherapy regimens. The purpose of the study was to estimate response rates to each drug. A two stage design was used to stop accrual early if no responses were observed. Seventeen pts. received Adria and 69 pts. received 5-FU. Dominant disease sites by treatment were: 5-FU: visceral - 44 pts., bone - 7 pts., soft tissue 5 pts. Adria: visceral - 8 pts., bone - 5 pts., soft tissue 4 pts. Fifty pts. are available for analysis of response at this time (33 - 5-FU, 17 - Adrda). No life threatening or lethal toxieities have occurred and treatment was generally well tolerated. Mucositis, diarrhea, mild leukopenia, and the hand-foot syndrome have been observed. There have been six partial responses to 5-FU (18%). Thirty of the 33 5-FU treated pts. had received prior exposure to 5-FU bolus therapy. Four of the responding pts. had received prior 5-FU and two had not. No responses to Adria have been documented. CTI 5-FU has definite activity in metastatic breast cancer and the ECOG is integrating this therapy into other treatment programs.

Abstracts 73

A FAIR TRIAL FOR TRIMETREXATE (TM~X) IN ADVANCED BREAST CANCER. ~E Costanza, AR Eorzun, L Norton, MARice, WC Wood, Cancer and Leukemia Group B (CALGB) Brookline, MA 02146 A continuing problem in Phase i i breast cancer studies has been the attempt to evaluate new agents by using them in heavily pretreated patients (pts). Low or absent response rates may r e f l e c t unfairly on potential drug activity as was the case with adriamycin when used as a Phase II agent in heavily pretreated breast cancer pts. However, studying Phase II agents "upfront" may affect response rates to subsequent standard chemotherapy and/or ultimate length of survival. To evaluate the riskbenefit ratio of using Phase II agents first, CALGB has launched a major trial in the treatment of metastatic breast cancer pts. 350-400 pts will be randomized to either standard therapy (CAF) or 4 cycles of a Phase II agent followed by CAr. Eligibility re-quirements include: no prior chemotherapy for metastatic disease; performance status 0-I, survival expectation of > 4 mo, prior adjuvant therapy is permitted. In a projected series of 6-8 Phase II agents, the first drug to he evaluated is ~MTX a 2,4 diaminoquinazoline derivative of methotrexate and an effective folic acid antagonist. Pts randomized to receive TMTX received 12mg/m 2 per day, IV qd l-5,q 21 days. Unless progression occurred, 4 cycles were given before beginning CAr. 20 of 22 pts on study have follow-up. Toxicities were: Infection: 5% lethal. Allergy: 15Z grade 3-4. Platelet, WBC; ]5% grade 3-4. 16/22 pts had no prior chemotherapy; 6 had had adjuvant therapy completed Z I yr previously; none had had chemotherapy for their metastatic disease. No responses were seen. Further details regarding TMTX will be presented. We conclude: TMTX received a fair trial; TMTX is inactive in breast

74

125

CEI~BINEDTP45~IT£NTFOR LOCALLYADVAt,ICEDBREASTCANCER(LABC) M.Colozz~S.Gori ,V.Bel santi ,A.Mosconi,R.Rossetti ,C.Basurto,P.Anastasi ,P. Latini,E.Maranzano,F.Checcaglini,U.I'4ercati,M.Tonato. Division of ;4edical Oncology,Radiotherapy and Surgery, Policlinico,06100 Perugia,ltaly Frc~ August 1985 to May 1988 a co~ined ,~dality therapy was given to 21 patients (pts) with LABC (2 stage ILIA,19 I I I B including 11 IBC).A fine needle biopsy was ~erfonmed in all pts. All pts received an induction chemotherapy (CT) with Cisplatinum 2Q~J/m2 x3 days,Doxorubicin 4(~4*ffm2 on day I, Cyclophosphanide200~/m 2 x3 days (PAC),frml 4 to 6 courses. Pts achieving CR or PR unde~ent a radical n~stectc,~i (~gfollowed by external RT(45 Gy over 5 weeks on the chest wall) and CT(CWi.v.) for 6 courses.The overall objective response rate to induction CT in 17 evaluable pts (3 pts too early and 1 lost to foll(~-up) was 71% with 1 CR,II PR.Twelve responsive pts and t~e pts out of 4 with NC considered technically operable underwent a ZM(I PR pt had a MRafter RT and consolidation CT).In the pt who achieved CR to CT,a pathological CRwas documented;one pt ( I I I A) with PR to CT appeared to have residual disease only in the nodes and fibrosis in the breast;the other 12 pts were rendered disease free by~.l]qe toxicity of the induction CT was G.I. grade 2-3 in 17 pts, alopecia grade 3 in all pts, hematologic grade I-2 in 7"pts, neurotoxicity grade 1 and ototoxicity in 1 pt.All pts ~ith objective responses and one pt NC to induction CT have c ~ l e t e d the full schedule of treat~it.Lneo-regional control has been obtained in I0(7/~of responsive pts.With a ,ledian folls~ up of 66 ~eeks, 7 responsive pts and one pt v~th NC are disease free.Five of 12 responsive pts have relapsed in distant sites and 4 are died.i,t~dian survival and median time to progression have not been reached.The results obtained v~th the induction CT are similar to those ~e observed in a phase I I study of r,~etastatic disease.lhe conbined modality treatn~nt se~ns feasible with quite acceptable toxicity.

cancer.

75

CHEMOTHERAPY WITH ESTROGENIC RECRUITMENT VS CONVENTIONAL CHE MOTHERAPY IN METASTATIC BREAST CANCER (MBC). G. Gardin*, P. Pronzato, D. Amadori, F. Botto, T. Guido, P. Miccoli, L. Miglietta, C. Monzeglio, M. Pace, L. Repetto, A. Rubagotti, R. Rosso, P. Sismondi, P.F. Conte. Istituto Nazionale per la Ricerca sul Cancro, Genova-ltaly and North West Oncology Group. We report the results of our second randomized cooperative trial comparing conventional chemotherapy (CT) to a cytokin~ tic regimen based on estrogenic recruitment induced by diethylstilbestrol (DES) in MBC. Two-hundred and six patients (pts) were randomized to receive CEF (CTX 600mg/sm; Epi-Dx 60mg/sm; 5-FU 600mg/sm d 1 q 3 weeks)or DES-CEF(DES img/die per os d 1-3, CEF d 4, q 3 weeks). Response: CEF= complete remission(CR) 10%;partial remission(PR) 36%; stable (S) 32%; progression (P) 22%; DES-CEF= CR 11%; PR 42%; S 41%; P 6%; pattern of response was significantly in favour (chi square 5.08) of DES-CEF for pts not previously treated with adjuvant CT: CEF= CR 11%; PR 33%; S 30%; P 26%; DES-CEF= CR 14%; PR 42%; S 40%; P 4%. Overall survival(S) and progressionfree survival(PFS) were similar in the two arms: median S and PFS were 18.5 mos and 9.5 mos for CEF pts and 21 mos and 10.5 mos for DES-CEF pts. Among pts not previously treated with adjuvant CT an advantage in PFS was seen in DES-CEF arm (p=0,001). Toxicity was superimposable with the exception of myelotoxicity; leukopenia was present in 30% of CEF pts and in 48% of DES-CEF pts (p<0.05).

76

FOLLOW-UP PROCEDURE FOR RETRIVING DATA IN A LO573 T~RM BREAST CANCER STUDY. *V. Sacchini, P. Boracchi, V.Galimt~rti, E.Beretta, S.Marchini, M.Greco, A.Luini, Istituto Nazionale TtmDri, 20133 Milano Italy For the attendibility of any survival analysis the patients lost to follow-up should not be more than 10%. In fact all analysis methods consider the loss to follow-up unrelated to the subsequent risk of dying, and analysis can be incorrect if the lost patients tended to have a better or worse prognosis. Our study regards the survival analysis of 2357 patients with breast cancer operated frcra 1973 to 1984 at the National Cancer Institute in Milan. At the beginning of the follow-up research for the statistical analysis, 13% of the patients were dead and the follow-up procedure were curried out on 2060 patients. 63% of these resulted regularly controlled in the Outpatients Dept. In the other 37% the procedure to retrieve outcome data was started. The first step consisted in thelephoning the patient for arranging an appointrent, or for finding out the usual outside checking clinic. In 10% this first approach permitted the conclusion of the search (to arrange an appointment, or to know of the death of the patient. In 22% of cases information was retrieved frc~n outside physicians who controlled the patient. No difference in the patient "s status was found in a group of these patients checked a second tJ/De by our staff. In the other 5% the registry office of the city of residence was contacted to know whether the patient was alive, the date and the cause of the possible death and the possible re~no~ral. In 35% of the patients just a search was enough to retrieve sure outcome data. In 4% more than 4 searches were necessary, qhe resource cost of this fol low-up research resulted justified for the decrease of number of patients lost to follow-up: 6%.

126 77

Abstracts

MAJOR OBSTACLE TO PROTOCOL ACCRUAL: PATIENT INELIGIBILITY.*N J R o b e r t , M E Reilly, D M a r c h a n t , D. Wazer, T J Smith, Breast Health Center, *Division of Hematology/Oncology, New England Medical Center, T u f t s University School of Medicine, Boston, MA 02111

78

The Breast Health Center is a m u l t i d i s c i p l i n a r y c e n t e r f o r b r e a s t disease. All new patients with primary b r e a s t c a n c e r are evaluated by a team o f surgical, radiation, and m e d i c a l oncologists. All patients are presented at a weekly breast management conference. All patients are e x a m i n e d f o r potential participation in national cooperative group protocols including examination of case record by a designated data manager. We r e v i e w e d t h e c a s e s o f 134 consecutive patients that were seen. We found that only 19% of these patients were eligible. Of these patients 70% participated in a randomized cooperative group study. However, 81% o f p a t i e n t s were found to be ineligible. Reasons for ineligibility included age, inadequate r e c e p t o r data, second c a n c e r , p o s i t i v e m a r g i n s , and time delay between surgical procedures. From this survey it appears that the major obstacle to p a t i e n t a e c r u r a l is not p a t i e n t refusal or lack of physician interest but ineligibilty defined by the available protocols.

79

MDR-1 GENE EXPRESSION IN BONE MARROW CELLS OF BREAST CANCER PATIENTS AFTER COMBINATION CHEMOTHERAPY. *J Yau, D Li, J Tsao, T Kuo, and K Lu. The University of Texas M.D. Anderson Cancer Center, Houston, Texas 77030. Morphologically normal bone marrow cells from breast cancer patients were studied to determine if combination chemotherapy can induce over expression of rndr-1 m R N A in bone marrow cells. Both intrinsic and acquired multidrug resistance (mdr) in tumor cells have been shown to be associated with over expression of m d r - I gene. High expression of m d r - I m R N A has been found in normal adrenal, colon, kidney, and liver. Very low level of expression is found normal bone marrow cells. A human cDNA library was constructed from m R N A prepared from a lympboblast cell line (CEM/VLB). A human mdr gene probe was isolated by screening the library with Chinese hamster mdr DNA probe. The human eDNA clone was identified to be m d r - I by D N A sequencing. We have employed this probe to study ten morphologically normal bone marrow samples from breast cancer patients. All these patients have been treated with a medium of 3 (range 1=5) combination chemotherapy regimens. All but one had received adriamycin. The medium dose was 210 (120-310) m g / M 2. The other chemotherapeutic agents include cisplatinum, fluorouracil, methotrexate, etoposide, thiotepa, and vinblastine. The medium age was 47 (34-60). Time from initial diagnosis was 30 (6-132) months. Time from the last chemotherapy was 30 (23-568) days. Metastatic sites include lymphnodes (9), skin (6), lung (4), liver (1), bone (1), and pericardium (1). The expression of m d r - I m R N A level was studied by northern, dot blot, and in-situ hybridization techniques. No significant expression of m d r - I m R N A was detected in any of these samples. The data suggests that combination chemotherapy in breast cancer patients does not induce over expression of m d r - I mRNA in bone marrow cells.

N S A B P / I N T E R G R O U P NODE N E G A T I V E STUDIES: EVALUATION OF DISEASE FREE SURVIVAL DATA USING A DECISION A N A L Y T I C MODEL. *N J Robert, A J Moskowitz, M H Eekman, S G Pauker, Divisions of H e m a t o l o g y / O n t o l o g y and Decision M a k i n g , N e w England Medical Center, T u f t s School University School of Medicine, Boston, MA 02111 Recent release of results from two randomized studies, comparing chemotherapy versus no t r e a t m e n t in node n e g a t i v e patients with breast cancer (NSABP B-13 and I N T 0011), revealed disease free survival advantage for patients receiving chemotherapy. However, there was no differenee in o v e r a l l s u r v i v a l r e p o r t e d . The significance of this finding is d i f f i c u l t to interpret. We have developed an analytic model for deciding on the benefit of adjuvant chemotherapy for node positive patients ( p r e v i o u s l y p r e s e n t e d at this symposium). The a d v a n t a g e of this model is that u s i n g disease free s u r v i v a l can evaluate t r e a t m e n t intervention f r o m the perspective o f quality of life benefit, a useful clinical endpoint. With this model we found that for the c h e m o t h e r a p y treated patient t h e r e is a d e f i n i t i v e q u a l i t y o f l i f e b e n e f i t a m o n g the d i f f e r e n t p a t i e n t subsets d e s c r i b e d . However, these findings will need to be reviewed as m o r e i n f o r m a t i o n is p r o v i d e d a b o u t t h e s e studies.

80

A TUMOR CELL BIRTH RATE END-POINT THAT IMPROVES THE PREDICTABILITY OF XENOGRAFT ASSAYS FOR DRUGRESPONSEOF BREAST CANCERPATIENTS. D Decker ~, d Grudalen, T Rao, d Pettlnga and B McGrath. G r a c e Bio-Oncology Laboratory, Inc., Oakland Unlversity, Hematology/Oncology Assoclates, Troy, Michigan 48084, Ni11Cam Beaumont Rosplta1-Troy, Troy, MI 480gB. Xenograft assays for druB response based on net volumetric tumor growth have proven too insensitive to predict response of indlvldual cancer patients to chemotherapy or hormonal therapy, Our klnetlc data of primary breast tumors and breast tumor fragments grafted subcutaneously in athymic mice indlcate that the i n s e n s l t l v i t y is due to tumor cell loss, which approaches 90% per daM, even though v i r t u a l l y a l l fragments of every tumor, independent of hlstologlcal type, successfully vascularlze. Ne have validated an end-polnt based on drug-lnduced i n h i b i t i o n of tumor cell DNA synthesis to circumvent the problem of cell loss in primary tumor xenografts. The assay is sensitive enough to test v i r t u a l l y every operable primary or rcurrent tumor in a I0 day assay format. The major parental tumor ceil lineages, defined by thymidlne labaelllng index, ploidy, and estrogen response were preserved in the I0 day xenografts. Inhlbltion of tumor DNA synthesis by tamoxifen adminlstered for 9 days as a single 25 umg s.c. pellet dose correlated with tumor estrogen receptor levels. Xenografted tumors demonstrated a wide range of s e n s i t i v i t y (0-80% inhibition of DNA synthesis) to a slngle Maximum Tolerated Oose og six chemotherapeutic agents assayed 4 days a f t e r i.p. drug injection. CI~nica] response to both hormonal therapy and chemotherapy correlated with xeno~raft test results in 4 patients Bested, with >50~ i n h i b i t i o n of DNA synthesis indicating early c l i n i c a l response.

Abstracts 81

THERAPEUTIC EFFECT OF N-(4-HYDROXYPHENYL) RETINAMIDE (4-HPR) ON METHYLNITROSOUREA (MNU) INDUCED RAT MAMMARY CANCER. *K Dowlatshahi, R Mehta, C F Thomas, N M Dinger and R Moon. University of Chicago, Chicago, IL 60637 and I I T Research Institute, Chicago, IL 60616. The chemopreventative role of 4-HPR alone, or combined with hormonal modification, on MNU-induced mammary carcinoma is well established. This study was designed to investigate the therapeutic influence of 4-}IPR on established rat mammary carcinoma. Virgin Sprague Dawley rats received a single intravenous injection with 40 mg/MNU/kg body weight at 40 days of age. After the first palpable mammary tumor reached 5 mm in size, the animals were sequentially allocated to one of four groups: a) placebo diet; b) i0 mg tamoxifen subcutaneously x 3 weekly; c) 3mmol 4-HPR/kg diet; or d) both (b) and (c). Weekly measurements of the initial and subsequent tumors were made throughout the study. Results indicate that:

82

TOXICOLC~IC STUDIES ON THE ORGANOMETALLIC ANTICANCER AGENT, VANADOCENE DICHLORIDE (VDC). L.N.Rao, R.A.Coldschmidt, J.C. Dohnal, L.Y.Kuo, K.Sriram, M.S.Murthy, T.J.Marks. Evanston Hospital & Northwestern University, Evanston, IL 60201. A n t i t u m o r e f f e c t s o f v a n a d o e e n e d i c h l o r i d e (CP2VCI2) , a representative metallocene dihalide, against TA3Hamouse m , ~ a r y carcinoiL~ (~ILS, 90) are comparable to those of cisplatin (~ILS, 83) (Inorganiea Chimica Aeta 152: 117, 1988). In contrast to eisplatln, VDC exerts no nephro- or gastrointestinal toxicity. Furthermore, the mechanism of a n t i t ~ o r action of VDC appears to differ markedly from that of eisplatin (J. Am. Chem. Soe. 108: 7263, 1986). The present study focuses on the toxicologic effects of VDC against murine hepatic and pancreatic tissues. Strain A mice injected i.p. with 80 m~/kg VDC were sacrificed 8 hr to 30 days after VDC treatment. Blood, liver and pancreas were collected from these and untreated c o n t r o l group of mice. Eight hr after VDC treatment, average serum levels (U/ml) of blkaline phosphatase, LDE, AST, ALT, and amylase were (control values in parenthesis): 223 (156), 9609 (84~, 5603 ( 9 9 . 8 ) , 6372 ( 5 5 . 2 ) , and 2406 ( 2 0 0 6 ) , r e s p e c t i v e l y . The l i v e r enzyme l e v e l s r e t u r n e d t o n o r m a l i n 3 t o 15 d a y s . One day a f t e r VDC t r e a t m e n t , h e p a t o c y t e s showed k a r y o l y s i s , k a r y o r r h e x i s , and s t e a t o s i s . By day 3, h e p a t i c r e g e n e r a t i o n was e v i d e n t and by day 15, t h e l i v e r assumed a n o r m a l appearance. The p a n c r e a s showed v a c u o l a r d e g e n e r a t i o n o f t h e a c l n a r c e l l s , w h i c h p e r s i s t e d f o r a t l e a s t up t o 30 days. Serum a m y l a s e l e v e l s a l s o w e r e e l e v a t e d d u r i n g t h i s period. In conclusion: the pattern of organ toxicity of VDC differs from that of cisplatin, and the liver toxicity but not the pancreatic toxicity is reversible. The mechanism of VDC toxicity and the potential use of VDC i n c o m b i n a t i o n c h e m o t h e r a p y w i t h c i s p l a t i n will be d i s c u s s e d . We are grateful t o the Peter Garard Memorlal and the Dee & Moody Funds for the support.

84

WHOLE BODY HYPERTHERMIA (WBH) I N METASTATIC BREAST CANCER, R o b e r t D. Levin*, R. M i c h a e l W i l l i a m s , R a n u l f o Sanchez, Y o u n g Doo Kim, A l f o n s o M e l l i j o r , M a r y A n n Doyle, a n d W i l l i a m Simonich. A m e r i c a n International Hospital, Zion, Illinois, 60099, a n d Northwestern University Medical School(RMW),Chicago

4-HPR administration resulted in a complete regression (non-palpable state) of the initial 5 mm tumor in six animals (22%) and partial regression or non-progression of the first palpable tumor in five animals (19%). Tamoxifen alone induced a partial response in nine animals (33%). In animals receiving both tamoxifen and 4-HPR, 19% of the tumors regressed completely and 26% exhibited a partial response. Based on these data, it is apparent that 4-HPR has therapeutic value in the treatment of MNU-induced mammary carcinoma. With its known minimal side effects in humans, it is reasonable to suggest a clinical trial, either as an adjuvant or as a new agent, for recurrent breast cancer. Retinoid receptor evaluation in these tumors are in progress.

83

P R O G N O S T I C F A C T O R S / S U R V I V A L IN M E T A S T A T I C BREAST CA P A T I E N T S E L I G I B L E F O R WHOLE BODY H Y P E R T H E R M I A (WBH). R,Michael Williams*,Robert D.Levin,Ranulfo Sancbez,Young Doo Kim, Alfonso Mellijor, Mary Ann Doyle, and William Simonieh. A m e r i c a n International Hospital, Zion, Illinois, 60099 and Northwestern U n i v e r s i t y Medical School (RMW), Chicago, II. 106 metastatic breast cancer patients f r o m 1982-1986 who were medically eligible to receive WBH plus chemotherapy were retrospectively analyzed for prognostic factors related to survival. Variables included age, p e r f o r m a n c e status (PS), ER and PR status, # prior chemotherapy regimens for metastatic disease (#PCs), # of metastatic organ sites, specific metastatic sites, time f r o m diagnosis of metastases to first A I H therapy, WBH, and interactions of WBH with the other factors. Using a proportional hazards model, only ER- status (p=.0001), #PCs (p=.0001), and # of metastatic sites (p=.0493) were independent predictors of decreased survival. O f the 56 patients who received WBH, only poor prognosis patients had improved survival. WBH was associated with increased s u r v i v a l in patients with liver mets (p=.0040), i f ER- without liver mets (p>.0001), i f ER- without lung mets (p=.0016), and i f ER+ with PS >2 (p=.0030). Proportional hazards analysis of the 45 ER+ patients showed the expected association of PS>2 and #PCs with decreased survival (p>.0001 in both). WBH was associated with decreased survival only in patients with lung mets but without liver mets (p=.0006). The median s u r v i v a l time (MST) of all unmatched patients was not significantly a f f e c t e d by whether or not WBH was utilized (MST=57 vs 60 weeks, p=.68). Analysis of the 37 patients with liver mets also showed that #PCs was associated with decreased survival (p=.0171). When 14 pairs of these patients were matched for #PCs, WBH was associated with increased s u r v i v a l (MST=58.7 vs 18.5 weeks, p=.0101). Among 69 patients without liver mets, ER- (p<.0001) and #PCs (0=.0029) were also predictive of decreased survival. A f t e r m a t c h i n g for ER, lung mets, and #PCs, WBH did not appear to influence s u r v i v a l in 20 pairs of liver negative patients (MST 89 vs 88 weeks,p=.58). Prospective randomized trials are u n d e r w a y of WBH +/-chemotherapy.

127

o f 139 p a t i e n t s ( P t ) with metastatic breast cancer s e e n 1 / 8 2 - 1 / 8 7 , 28 e v a l u a b l e p a t i e n t s r e c e i v e d W B H a n d d o x o r u b i c i n . T h e s e 28 w e r e r e t r o s p e c t i v e l y m a t c h e d w i t h 28 w h o r e c e i v e d d o x o r u b i c i n w i t h o u t WBH, u s i n g Z u b r o d status, # of p r i o r c h e m o t h e r a p y (C) r e g i m e n s , h o r m o n e r e c e p t o r (HR) status, a n d + / v i s c e r a l mets. 3 Pt h a d 1 W B M therapy, 12 h a d 2, 3 h a d 3, 4 h a d 4, 2 h a d 5, and 4 h a d >5 WBH. In P t w i t h o u t p r i o r C, 14 c o n t r o l Pt (CON) h a d 1 CR, 6 PR, a n d 2 M R w h i l e 11 W B H Pt h a d 1 CR, 5 PR, & 2 MR. In Pt r e l a p s i n g from p r i o r C, 14 C O N h a d 4 PR, & 2 MR, w h i l e 17 W B H Pt h a d 1 CR, 3 PR, & 3 MR. M e d i a n s u r v i v a l in W B H & CON w a s 443 & 423 days, respectively (p=ns). In H R n e g a t i v e Pt, m e d i a n s u r v i v a l in W B H a n d C O N w a s 488 & 298 days, resp. (p<0.04). M e d i a n t i m e t o d i s e a s e p r o g r e s s i o n (TTP) in H R p o s i t i v e Pt w a s 279 & 314 days, r e s p . , w h i l e in M R n e g a t i v e Pt, 261 & 117 days, r e s p . ( p < o . 0 4 ) . D u r i n g 71 W B H , t o x i c i t i e s > ECOG grade 2 included f a t i g u e a f t e r 17, f l u i d r e t e n t i o n a f t e r 14, w e a k ness a f t e r 6, 2nd d e g r e e skin b u r n s a f t e r 2, a n d A R D S a f t e r v e n t r i c u l a r t a c h y c a r d i a in 1 WBH. In 6 p a t i e n t s w i t h p r i o r s p i n a l i r r a d i a t i o n of 4000 r a d s ( 0.2 -16 y e a r s p r e v i o u s l y ) , no p a r a l y s i s developed. In p a t i e n t s r e c e i v i n g d o x o r u b i c i n f o l l o w i n g WBH, t h e r a p y is w e l l t o l e r a t e d a n d a s s o c i a t e d w i t h a 54% i n c r e a s e in TTP, a n d in H R n e g a t i v e p a t i e n t s , a 63% i n c r e a s e in survival.

128 85

Abstracts LOCALLY RECURRENT BREAST DISEASE TREATED WITH COMBINATION RADIATION AND HYPERTHERMIA. S Jampolis*,A DiStefano,J Kelly G Blumenschein,C Kahlig,B Firstenberg,J Kong,JE GomezYeyille. Arlington Cancer Center, Arlington, TX 76012.

86

DIETARY FAT QUALITY AND METASTASIS FROM THE 13762 bDAMMARY T L ~ R (MT) in F ~ RETIRED BREEDER (RB) RATS. E B Katz and E S Boylan*, Queens College and The Graduate School, The City University of New York, Flushing, NY 11367. Previously, we reported that a high fat (23%) corn oil diet (HFCO) stJ/nulates metastasis to the lung from the 13672 MT as c c ~ e d to that in rats fed a low fat (5%) corn oil diet (LFCO) . This HFCO effect is seen in intact or ovariectomized RBs and in 10+ me old virgins, but not in young virgin rats. Primary tumor growth was ccr~oarable in both the high and low fat diet groups. To assess this phenomenon further, we investigated whether the type of fat might affect metastasis from the 13762 MT. RB (Fischer 344 strain) were fed 1 of 5 diets representing mainly [x]lyunsatttrated fat (HFCO, LFCO) , monunsaturated fat (20% and 5% olive oil, HFO0 and I/DO), and saturated fat (29% beef tallow, HFBT). After 4 weeks on the diets, a 2 m m piece of the t%m%or was i~la-nted sc into each rat. Primary tumor growth was monitored weekly. Forty days after implant, necropsies were performed, metastatic lung nodules counted, and total metastatic tumor vol%a~e estirm~ted. The average voltnne of pulmonary metastases in the HFCO animals (n=30) was significantly greater than in the other 4 groups. Among the 4 groups which did not differ significantly from each other, the rank order in extent of pulmonary metastasis was: HFBT (n=26) > I/OO (n=25) ~ HFOO (n=25) > LFCO (n=lS) . Growth of the primary tumor did not vary appreciably among the 5 groups despite the significant difference in p u ~ n a r y metastasis volLm~e. These results suggest that the c ~ l i t y of dietary fat ~ be an important determinant of pulmonary metastasis from the 13762 M T in RB rats. Supported by the American Institute for Cancer Research.

88

CHEEOPREVENTION BY POWDERED SOYBEAN CHIPS (PSC) OF HAHMARY TUMORS IN RATS. S.Barnes*, C. Grubbs & K.D.R. Setchall. Depts. Pharmacol., Biochem. & Nutr. Sci.~ and Comp. Cancer Center, UAB, Birmingham, AL 35294, and Dept. Pad. Gastrosnt., Children's Bosp., Cincinnati~ OH 45229. The chemopreventive properties of PSC were tested in a N-methyl-N-nitrosourea (~NU} rat model of breast cancer. Female Sprague Dawley rats were fed (throughout the study from 25 days of age) isocaloric, isonitrogenous AIN-76A diets in which casein was partially or totally replaced by PSC. In another group of experiments PSC was autoclaved for 60 min to denature protease inhibitors (PSC-A). MNU was administered (45 r~g/kg BW) i.v. at 50 days of age. At necropsy (140 days after MNU), all tumors were removed for histological classification. Cytoplasmic estrogen (ER)and progesterone (PqR) receptors in tumors were measured by a multidose radioassay. Results: Rats on diets containing PSC had significantly lower numbers of mammary tumors/rat than those on unmodified AIN-76A. Tumor cytosol ER levels, but not PgR levels, were positively correlated with the average numbers of tumors per rat. Dietary PSC Tumors/rat Receptors (fmol/mg protein) protein (%) PSC PSC-A ER P~R 0 7.7 7.7 57.6 • 4.9 334 ± 83 25 3.1 5.2 19.1 • 4.9* 336 & 66 ns 50 5.5 5.4 40.0 ± 6.0* 271 • 49 ns i00 4.3 4.3 32.4 £ 4.9* 288 • 43 ns • P < 0.05 versus control (0%) group, ns = not significant. Conclusions: Since both PSC and PSC-A are chemopreventive in MNU-treated rats, protease inhibitors do not cause this effect. Tumor steroid receptor data suggest that antiestrogens in PSC may be responsible for the inhibition of mammary tumors in this model. Supported by ACE, AICR and Archer Daniels Midland Co.

From March ]982 to May 1988, 67 patients(pts) with recurrent and locally extensive breast cancer were treated with low to moderate-dose radiation and superficial hyperthermia(HT). All patients had previously received surgery, radiation therapy, chemotherapy, or hormonal therapy. Superficial and interstitial HT was given to 26 of the 67 pts while the remaining 41 pts received superficial HT only. Follow-up observation ranged from ] to 48 months. 42 patients are still alive and evaluable. Local control was as follows: 52 of the 67 pts(78%) remained locally free of disease in thetreated area at their last follow-up or a t t i m e of death. 7 of the 67 pts(]0.5%) subsequently developed recurrence in the treated area. At the completion of therapy | of the 67 pts (1.5%) had persistent disease, however, there had been a greater than 50% reduction in tumor volume. Mean disease free survival from start of HT to last follow-up was 10.9 months. Evaluation of response was further broken down with relation to tumor size. ]8 pts had tumors <3cm of which 17 had local control of the tumors at the completion of treatment. 49 pts had tumors >3cm of which 43 had control of the tumors at completion of treatment. 7 of the 67 pts were subsequently lost to follow-up and could not be evaluated. Significant complications were present in 5 pts who suffered 2 ° to 3 ° burns that subsequently required skin grafting over the treated region.

87

EFFECT OF DIETARY GLUCARATE ON ESTROGEN RECEPTORS AND GROWTH OF 7,12-DIMETHYLBENZ[a]ANTHRACENE-INDUCED RAT MAMMARY C A R C I N O M A S . *Z W a l a s z e k , E F l o r e s , a n d A K A d a m s , T h e U n i v e r s i t y of T e x a s M.D. A n d e r s o n C a n c e r C e n t e r , Science P a r k - R e s e a r c h Division, S m i t h v i l l e , T e x a s 78597. O u r previous studies h a v e shown t h a t d i e t a r y g l u c a r a t e i n h i b i t e d c h e m i c a l c a r c i n o g e n i n d u c t i o n of r a t m a m m a r y tumors, in part, by reducing the levels of estradiol a n d prolactin. The purpose of this s t u d y w a s to d e m o n s t r a t e t h a t g l u c a r a t e affects the level of estrogen receptor (ER) a n d g r o w t h of r a t m a m m a r y t u m o r s i n d u c e d by 7 , 1 2 - d i m e t h y l b e n z [ a ] a n t h r a c e n e (DMBA). Female S p r a g u e Dawley r a t s received a single dose of DMBA (15 mg in 1 ml of sesame oil), by gavage, at 50 days of age a n d were kept on a s t a n d a r d rodent chow. Fifteen weeks after DMBA, all r a t s with a t least one t u m o r over 1 em in d i a m e t e r were randomized into 3 groups o f l 8 animals. The r a t s were then fed the AIN-76A diet s u p p l e m e n t e d with 2% (w/w) of calcium (+)glucarate (group 1 for 8 wk a n d g r o u p 2 for 4 wk) or an equivalent a m o u n t of calcium (+)tartrate (group 3, control, for 8 wk a n d group 2 for 4 wk following g l u c a r a t e diet). T u m o r sizes were determined weekly by caliper measurements. After 8 weeks on the diets all r a t s were sacrificed by decapitation. ER a n d progesterone receptor (PgR) were m e a s u r e d in t u m o r cytosols by use of dextran-coated charcoal assays. Continuous feeding with g l u c a r a t e p r o d u c e d a m a r k e d i n h i b i t i o n of t u m o r growth. Cessation of glucarate t r e a t m e n t (group 2) resulted in resumption of growth. The level of ER in m a m m a r y carcinomas excised from glucarate-fed r a t s w a s reduced by 48% (p<0.01) in comparison with control tumors. A positive correlation between ER a n d PgR levels was observed in a majority of tumors. In conclusion, g l u c a r a t e , a n o n - t o x i c m i c r o n u t r i e n t , reduces ER a n d inhibits the g r o w t h of h o r m o n e - d e p e n d e n t r a t m a m m a r y carcinomas induced by DMBA.

Abstracts

129

89

STEREOMICROSCOPY OF INTRACYTOPLASMIC LDMINA IN BREAST CARCINOMA. M.J. Song,* M. Hussain, J. Jed, M. Marko, and D.F. Parsons, Wadsworth Ctr. for Labs and Res., NYS Dept. of Health, P.O. Box 509, Albany, NY 12201-0509, and Dept. of Pathology, Albany Medical College, Albany, NY 12208. Intracytoplasmic lumina (ICL) are seen occasionally in hyperplastic or normal mammary duct epithelium, but mainly in metastatic breast cancer. Their role in the cell phyalology or pathology remains unknown. In seeking their role in invasive breast cancer, we have investigated the development of ICL, correlation with blood-vessel invasion, and whether they communlcate with the exterior of the cell. We h a v e s t u d i e d seven biopsies of breast carcinoma. Of these, light microscopy revealed infiltrating ductal carcinoma in three of the cases and infiltrating lobular carcinoma in the remaining four, two of which were characterized by mucicarmlne-positive cells with slgnet-rlng patterns. Two slgnet-ring cells were serial-sectioned at 0.25 and 0.5 micrometers respectively. The sections were photographed in stereo in the highvoltage electron miurograph and computer 3-D reconstructions were made. The reconstruction revealed the ICL to be entirely inside the cells, without conuumlcation with the exterior, with possible origin from the Golgi apparatus. Correlation with blood-vessel invasion is high. This work was supported by NIH grant RR01219.

90

SOLITARY AND MULTIPLE INTRADUCTAL PAPILLOMASOF THE BREAST. P Lepera, I Del Prato, A Crippa, G F a r i s e l l i , E Galante*, G De Palo, I s t i t u t o Nazionale Tumori, 20133 Milano, I t a l y . A series of 153 s o l i t a r y and 230 multiple intraductal papil= lomas was found, from January 1976 to December 1985, in 383 patients treated by excisional biopsy of a lump (88 cases) or by resection of the subareolar duct system (295 cases). Clinical records were revue to evaluate c l i n i c a l features of the lesions and cancer risk of the patients.~The mean age of the patients treated f o r s o l i t a r y papilloma was 47.8 years (range 20-83) and 46.6 years(range 16-80) f o r multiple pa = pillomas. The predominant c l i n i c a l feature of multiple pa= pillomas was a spontaneous nipple discharge. I t was present in the 79% of cases, was hematic in 68, serous-hematic in 58 serous in 56 and associated with a subareolar lump in 31. A lump was the only symptom in 48 cases and was located in the subareolar region in 11. A palpable lump occurred in the 37% of s o l i t a r y intraduGtal papilloma, was located in the suba= reolar region in 32 and was the only symptom in 39 cases. Nipple discharge was hematic in 50 cases, serous-hematic in 44 and serous in 20. During the follow-up, ranging from 3 to 126 month, breast cancer occurred in the same breast of 4 patients treated f o r multiple papillomas and of 6 treated f o r s o l i t a r y papilloma. Divergences with others report are correlated, of course, with our wide surgical approach of nipple discharge.

91

CANCER OF TEE BREAST WITH NIPPLE INVOLVEMENT: OF MULTICENTRIC DISEASE

92

CA1MONOCLONAL ANTIBODY IDENTIFIES DUCT0-LOBULAR ANTIGEN ASSOCIATED WITH 'HIGH' RISK HISTOLOGY. *S P Courtney, Susan Williams, R E Mansel. Departments of Surgery and Pathol~gy, University of Wales College of Medicine, Cardiff CF4 4XN, U.K.

AN EXPRESSION

R.S. Menon* and A.N. van Geel, Departments of Pathology and Surgery, The Dr. Daniel den Hoed Cancer Center, P.O.Box 52019 3008 AE Rotterdam, The Netherlands

In this study of 33 consecutively submitted mastectomy specimens, instead of the customary single sagittal section, multiple transverse sections of the nipples were examined. In 19 nipples (58%) malignancy was seen in one or more levels, with some shoving skip areas. Of these, 17 showed nonlnvaslve carcinoma; either ductal or lobular type, with one of them having local invasion. Contrary to the generally held view9 direct extension from an underlying tumor situated in the breast proper, was found only in a single case. The remaining case shoved tumor in the Intramamillary lymphatics in association with massive axillary lymph node metastases. In 4 cases, the type of tumor in the nipple was dissimilar to that found in the breast. The most significant finding was the eccentric location of tumor in 14 nipples (74Z). A single central sagittal section would have detected only 5 (26%) cases involving the centrally situated duct. There was no significant correlation between nipple involvement, and the size, quadrant location, multicentricity, type of tumor in the breast and metastases in axillary lymph nodes. It is evident from this study that malignant changes in the nipple occur more frequently than realized, and that it is also one of the sites of multicentrlc origin of the tumor. These factors as well as the close proximity of the subareolar lymphatic plexus should be taken into account in planning various conservative therapeutic programs.

It has been suggested that CAI and 323A3 monoclonal antibodies (MAb) identify 'high' risk breast histology. To establish which aspect of staining was responsible for this we have compared the staining of benign breast biopsies with the staining of similar histological components associated with in-situ carcinoma of the breast. Sixteen cases of lobular or ductal carcinoma in-situ were studied: 50 prospectively studied patients with benign breast biopsies and 77 breast biopsies obtained from other departmental studies. Apical staining in ducto-lobular tissue with CAI MAb showed that 29 out of 50 stained strongly in in-situ carcinoma group. This difference was statistically significant (0.i > p >0.05) No differences in staining for cytoplasmic staining in dueto-lobular tissue with CAI or 323A3 were noted and no differences in staining were noted in apocrine tissue between control groups and the in-situ carcinomas for either 323A3 or CAI MAbs. Strong ~pical staining with CAI MAb in ducto-lobular tissue may be a useful marker of cancer risk in women undergoing breast biopsy.

130 93

Abstracts ERB B-2/NEU GENE AMPLIFICATION - A POSSIBLE NEW PROGNOSTIC FACTOR IN NODE-NEGATIVE BREAST CANCER (TI/T2, N0). J.S. Ro, M. Bliek~ J.Y. Ro, D. Frye, A. Buzdar, H. Fritsche, and G. Hortobagyi. UT M. D. Anderson Cancer Center, Houston, TX 77030. Erb B-2/neu gene analysis by DNA dot blot and Southern blot methods was done in consecutive samples (N=56) obtained at surgery and kept in liquid nitrogen since 1978. All patients (pts) were node-negative. No pts received adjuvant chemo~ or hormonal treatment, and 19 pts received postoperative XHT. The median age was 58 years (range, 36-83). Twenty-six tumors were <~2 em, 28 were 2-5 em, and 2 were >5 cm. Histologically all tumors were invasive ductal or lobular carcinoma. Twenty-two tumors were nuclear grade (NG) I. 23 were NG II, and Ii were NG III. Twenty-six tumors had estrogen receptor (ER) contents (8S) ~5 fm/mg, and 30 had >5 fm/mg. Erb B-2/neu gene was amplified in 13 tumors (23~. The median follow-up was 85+ me. Nineteen pts (34%) developed local or distal recurrence, and 9 pts (16%) died of metastatic d i s e a s e . Seven of 13 pts with erb B-2/neu amplification developed recurrence, while 12 of 43 pts without erb B=2/neu amplification developed recurrence (P=0.09). The results of disease-free-survival at 5 yrs and 7 yrs are shown in the table by erb B-2/neu gene amplification. erbB~2/neu No. of % Disease-Free at amplification pts 5 yrs 7 yrs

94

STRESS RESPONSE PROTEIN srp27 PREDICTS RECURRENCE IN NODE-NEGATIVE BREAST CANCER. GC Chamness, A Ruiz, L Fulcher, GM Clark, and WL McGuire. The University of Texas Health Science Center, San Antonio, TX 78285, USA. The stress response protein srp27 was discovered by our laboratory in MCF-7 human breast cancer cells as an estrogen-induced protein originally called 2t~K. Purification of this protein led to production of a rnonoclonal antibody and then a eDNA, whose sequencing showed identity of the protein with a human heat shock protein described in HeLa cells and called hsp27 by Hicks et al. Because srp27 was found by immunohistochemistry in benign breast disease with atypia and hyperplasia, a small subset of benign disease with an ominous prognosis, and because it appeared to vary widely in malignant breast tumors, we performed a pilot study of its possible significance for clinical outcome in breast cancer. In particular, we measured the protein by a quantitative Western blot procedure with densitometry in cytosols from 188 node-negative breast cancer specimens, with a median clinical follow up of 51 months. The median value of srp27 in these cytosols was t~ (in arbitrary units), with a wide log-normal distribution. A wide range of c u t o f f values showed significant differences in disease-free survival between high and low srp27. Of these, the best cutoff value was l i units - - the 22% of patients with tumor levels of srp27 above this value had a relative risk of recurrence of 3.5 (p=.017). In spite of this relationship, tumors with high srp27 levels were very likely to be positive for estrogen and progesterone receptors (usually found to be correlated with a better prognosis); there was no correlation of srp27 with tumor size or patient age. In a multivariate analysis using all of these factors, high srp27 and negative progesterone receptor were independent predictive factors for a short diseasefree survival. These results, based on a relatively small sample of 188 patients, are clearly preliminary, but it seems possible that srp27 will be a useful prognostic factor for defining node-negative breast cancer patients at high risk of recurrence.

96

THE ROLE OF DENSITOI~TRIC VflLUES OF ERR-B ONCOUENE IN HU~N BREflST C~CER. *S S dung, C Schultz, T Geoghgan, T J Wieman, I C Kim Catholic ~edical College, Seoul, Korea and University of Louisville, Keatuck~ 40292. The mechanism of growth regulation of human breast cancer is not fully understood. Previous model of the steroid hormone and their receptors could not explain completely the molecular machinery of how the breast cancer induces and progresses to more malignant form. Recently there is increasing evidence of involvement of nncogene-initiated pathways in the biology of cancer as well as invasiveness of metastasis. We evaluated the r01e of erb-B nnc0gen messenger RIqfi in cancer and metastatic lymph node of the same patient by densitometric comparisiou of dot blot hybridization in 4 patients, flverage r e l a t i v e deasitometric values of erh-B messenger ~fl was increased 156~ in metastatic lymph nodes group, compared to cancer group i t s e l f . Paired T-rest revealed s t a t i s t i c a l significance (R > @.@5). This data indicates that metastatic tumor c0nsistently express higher levels of messangerRNfl for E~F receptor /erh-B oncogene than primary tumor and suggested that the metastatic growth potential of tumor cells is related to the expression of EGF receptor/ erb-B oncogene. ~easit0mctric values of erb-B messenger I~fl may be useful as prognostic indices of human breast cancer. We are going to get more analyses of this information and ~i[1 report later.

yes 13 54% 54% no 43 80% 71% There was no correlation between erb B-2/neu amplification and other clinical parameters mentioned above. These data suggest that erb B-2/neu gene amplification nmy be a new prognostic factor in early breast cancer.

95

EXPRESSION OF 323/A3 PROTEIN PREDICTS RELAPSE AND SURVIVAL 1N BREAST CANCER. AK Tandon,+ GM Clark, and WL McGuire, The University of Texas Health Science Center, San Antonio, Texas 7828¢. We have earlier described a monoclonal antibody (323/A3) against a surface protein of breast cancer MCF-7 ceils (Cancer Research t~6:1306-1317, 1986). While in these cells the 323/A3 antigen is a 43 kDa protein, in a total protein extract of breast tumor tissues it is identified by Western blot analysis primarily as a ~ 2 0 0 kDa protein with much smaller amounts of the 43 kDa form. Here we examine the.-~200 kDa 323/A3 protein as a breast cancer prognostic factor~ alone and in combination with other factors. Using our 323/A3 monoclonal antibody, tissues from 381 breast cancer patients were evaluated by Western blot analysis. The level of the~200 kDa 323/A3 protein in individual tumors was determined by densitometrlc scanning of the autoradiogram and calculated in arbitrary units. The median 323/A3 protein values were 26 units for nodenegative patients and 19 for node-positive. However, when various cutoff values were used to separate high and low 323/A3 protein levels according to their ability to predict good and bad disease-free survival, the best cutoff value for the node-negatives was 10 and for the node-positives was t~0. The node-negative cutoff value identifies 6t~]lpt¢- node-negative patients with low 323/A3 protein levels and a relative risk of 1.7 (P = .0t~) for recurrence and 3.0 (P = .001) for death. The node-positive cutoff value identifies 126/187 node-positive patients with low 323]A3 protein levels and a relative risk of 1.9 (P = 0.003) for recurrence and 2.9 (P = 0.0002) for death; this is similar to the predictive effect of the number of involved nodes~ and higher than that of other factors (PgR, tumor size) which also enter a multivariate model. CONCLUSION: Low expression of t h e ~ 2 0 0 kDa 323/A3 protein predicts short disease-free and overall survival in both nodenegative and node-positive breast cancer patients.

Abstracts 97

PROGNOSTIC IRDICATORS IN GERIATRIC CAIICER. *a Pother, AE Paoat~tas, G Lemick. Department of ~ g e r y N Y Io0~

BREAST From the

98

The Mount Sinai Medical Center, Hew York,

The ad.~vant treatment of breast eanoer in the geriatric population (->?0 years old) is not always based on data collected in this age group. Althou~ treatment recommendations are the same as for younger postmenopausal women, most major study ~otocols that guide treatment policy have excluded geriatric patients. In a retrospective study we examined the records of 99 women ->70 years with breast cancer operated upon from 19?7-i987, to determine the association between actual prognosis and prognostic factors that may affect further treatment options. Comparison was made to 214 women 50 - 69 years of age. Hormone receptors (liR) were graded as negative if both estrogen and progesterone receptors were less than 10 fnVmg cytosol protein. Lymph nodes (LI,T) were graded as 0, 1-3 or /& or more and tumor ~fferentiation (TD) was graded well to moderate or poor. I~ultiveriate analysis using SAStBI~DP statistical software revealed that for woman ->70, HR were the most significant predictor of disease free survival (DFSD while LN were not significant. LN were the most significant indicator among womem50-69. VARIABLE UNIVARIATE p. STEPP¥ISE @ COEFFISD ~?0__ 5O-69_ >_?0__ 50-69 _~70 50-69 HR 0.000/~ 0.1168 0.0002 0.0189 -3.97 -2.42 TD 0.0924 0.0004 0.0~420 0.0012 -2.16 -3.31 LN o.41ot o.oooo ...... o.oooo ~.?i AGE 0.9202 0.0823 ...... 0.0168 2.3?__ Among women with negative nodes, those _>70with negative H R had a 5 year DFS of 137.vs ?4x for ages 50-69 (p=0.003) while in those with positive H R the respective figures were 88"g vs 697. (p=o.o6). These results clearly show that I-IR in this elderly population are more important prognostic indicators than Li,I and suggest that among women with negative nodes and nogative HR, systemic treatment should be considered as their prognosis appears to be unexpectedly poor.

99

PLASMA HORMONE LEVELS IN BREAST CANCER PATIENTS. *R K Mehta, and T K Des Gupta, Division of Surgical Ontology, University of Illinois at Chicago College of Medicine, Chicago, Illinois 60612. Plasma levels of estrone (El) , estradiol (E2) , androstenedione (A2) , prolactin (PRL) and lutelnizing hormone (I/{) were determined in 70 premenopausal women with node positive breast carcinoma receiving chemotherapy. All women received standard therapeutic dose of cyclophosphamlde, methotrexate and 5 fluorouracil. Blood samples were obtained on day 1 and 8 of each chemoeyele. Hormone levels were determined by standard radioi~unoassays. Fifty four of seventy (77%) women became amenorrheic within 0-I1 months of initiating chemotherapy (Group I). Mean time to amenorrhea was 2.83±2.48 (S.D.) months. The incidence and time to develop amenorrhea were inversely correlated with patients age. Sixteen women (23%) continued to have regular cycles through the course of chemotherapy (Group II), Plasma E 1 and E 2 levels in Group I patients declined to their baseline (E 1 56.45±52.05; E 2 34.24±27.35, Pg/ml) from initial levels (E 1 90.14~65.26, E 2 67.75~61.93, Pg/ml) on onset of amenorrhea. Plasma LH levels were maintained at >30 mIU at this time. Plasma estrogen levels were higher in Group II as compared to in Group I patients. Plasma A 2 and PRL levels did not change during chemotherapy and were not significantly different in these groups. Twenty two of fifty four (41%) women in Group I and 5/16 (31%) women in Group II recurred within 10-83 months. Mean time (months) to recurrence was 23.81±16.68 (Group I) and 35.0±16.62 (Group II). Detailed analysis wlth respect to plasma estrogen levels, and other prognostic factors such as obesity, steroid receptor status and number of positive lymph nodes, suggested that patients with high estrogen levels, negative steroid receptor status and >3 positive lymph nodes have poor prognosis of the disease. Supported by NIH, POI CA31827.

100

131

THE UTILIZATION OF A MODIFIED CUMULATIVE ILLNESS RATING SCALE TO ACCOUNT FOR COMORBIDITY IN BREAST CANCER. H. Ownby*, R. McCune, W, Satariano, E. McKnight, J. Russo, S. Brooks and the Breast Cancer Prognostic Study Associates. Michigan Cancer Foundation, i i 0 E.Warren, D e t r o i t , MI 48201 The Cumulative lllness Rating Scale (CIRS) of L i n n e t . e l . was modified and applied t o the medical history data collected f o r 1123 patients entered in the Breast Cancer Prognostic Study. The mean follow-up time f o r those s t i l l a l i v e is 83 months. The CIRS sum f o r t h i s group ranged from zero to 15. For purposes of analysis, three strata were developed: O-2(n=233),3-7(n=694), and 8-15(n=196). Three hundred t h i r t y breast cancer deaths, 98 deaths due to other known causes and eleven undetermined deaths have been recorded t o date. Frequency analysis shows no association between the CIRS and stage at diagnosis, lymph node status, maximum tumor diameter, or estrogen receptor status. Based on l i f e t a b l e analysis, the CIRS is highly predictive of a l l deaths with those individuals having a ClRS of 0-2 having the best survival (x2:12.92, p<.O02, Breslow's Generalized Kruskal-Wallis). The CIRS does not predict breast cancer deaths,(x2=5.76, p=.0563). Based on the Cox's Proportional Hazards Model, when maximum tumor diameter, lymph node status, nuclear grade, estrogen receptor, and stage at diagnosis are considered, the ClRS is p r e d i c t i v e of both a l l deaths and breast cancer deaths with the highest ClRS score group(8-15) showing the greatest risk of death in the a l l causes model(2 df Score t e s t p:O.OO05) and the middle CIRS score group(3-7) showing the highest r i s k of death in the breast cancer death model (2 df Score t e s t p:0.045). The CIRS should prove useful in c l a r i f y i n g some of the real effects of various prognostic indicators by explaining the comorbidity factors t h a t impinge on the patients' status.

URINARY ANDROGEN EXCRETION IN OPERABLE BREAST CANCERUP-DATING AT 30 MONTHS OF POST-SURGICAL FOLLOW-UP. S.Oriana? G.Secreto, G.L.Riboldi, C.Reechione, P~Ballerini, S.BShm. Istituto Nazionale Tumori - 20133 Milano - Italy Previous works on therapeutic ovsriectomy in patients with advanced breast cancer reported that 24-hours urinary levels of testosterone (T) and 5-alpha-andrcstanediol (Adiol) were useful markers of response to hormonal therapies, in addition to steroid receptors determination. In this work, T and Adiol were measured in 107 consecutive women with operable breast cancer. Measurements were taken before and after surgery, and every six months afterwards for five years, in order to find if differences might occur. Although follow-up period is too short (at least 30 months for each patient), no variations were found among pre-surgical and post-surgical or follow-up mean values of T and Adiol. In premenopausal patients with axillary metastatic nodes (N+), an increased rate of recurrences was observed among those with high androgen levels. In fact, 85.7% of relapsed patients had basal levels of T and/or Adiol ~ 75th percentile, while only 33.3% of clinically cured patients had basal androgenic values in the upper quartile. Our data suggest a relationship between high androgen levels and a worse prognosis: this finding, if confirmed on a longer follow-up, might lead to more appropriate adjuvant endocrine therapy.

132 101

103

Abstracts

FLOW CYTOMETRY IDENTIFIES A GROUP OF NODENEGATIVE BREAST CANCER PATIENTS WITH LOW RISK OF RECURRENCE. GM Clark,* LG Dressier, MA Owens, and WL McGuire, The University of Texas Health Science Center, San Antonio, Texas 75285. It has been suggested that ali patients with clinical and pathological stage I and stage n breast cancer, regardless of age, menopausal, nodal, or receptor status, should be considered for systemic adjuvant therapy. But is it appropriate to t r e a t all nodenegative patients when 65-70% are cured of their disease by surgery and/or radiotherapy? Perhaps other factors, alone or in combination, might identify groups of patients either at high or low risk of recurrence. We have performed flow cytometry to determine ploidy status on tumor specimens from 3z;6 node-negative breast cancer patients. S-phase fraction was also measured on 255 of these tumors. A total of 233 specimens (67%) had aneuploid DNA content. The median followup for all patients was 55 months and the actuarial 5-year recurrence rate was 22%. Patients with aneuploid tumors had significantly worse disease-free survival than patients with diploid tumors (p = 0.02). S-phase was not a significant additional predictor of disease-free survival among patients with aneuploid tumors, but was highly significant (p = 0.009) among patients with diploid tumors. The actuarial 5-year recurrence rate for the 97 patients whose tumors were diploid with low S-phase fraction (28% of the total study sample) was only 9% compared to 27% for the remainder. Conclusion: The vast majority of breast cancer patients with node-negati.ve disease whose tumors have diploid DNA content with low S-phase fraction have long disease-free survival. Therefore, few patients in this group will benefit from the addition of systemic adjuvant therapy.

KI 67 IN PRIMARY BREAST CANCER, *J.F.H. Robertson, Kerry Walker, R.I. Nicholson, A.P. Locker, I.O. Ellis, R.W. Blamey, N o t t i n g h a m City H o s p i t a l and Tenovus Institute, Cardiff, UK

102

A MODEL FOR PREDICTING DISEASE-FREE SURVIVAL IN NODE-NEGATIVE BREAST CANCER PATIENTS. RE Miles,* GM Clark, and WL McGulre, The University of Texas Health Science Center, San Antonio, Texas 78285. A mathematical model has been developed that provides estimates of the likelihood of recurrence at any point in time for patients with node-negative breast cancer. The model was derived using demographic and tumor characteristics from 1855 patients with node-negative disease who did not receive systemic adjuvant therapy. The median followup for these patients was 52 months. The actuarial 5-year recurrence rate was 25%. Characteristics from a random sample of 1552 of these patients were used to develop the model, and the remaining 293 patients were used to validate the model. The model was constructed in a stepwise manner using the logarithms of estrogen receptor (ER) and progesterone receptor (PgR) concentrations, the logarithm of tumor size, the age of the patient at diagnosis, quadratic terms for each of these factors, and all possible 2-, 3-, and 5-way interactions as potential prognostic factors. The final model contains terms for the following factors, in order of importance: ER squared, tumor size, ER, and age. None of the interaction terms entered the model, while each of the included factors was significant with p < 0.02.

The model predicts that older patients with small, ERpositive tumors are at low risk for recurrence. The combination of quadratic and linear functions of ER predicts that the risk of recurrence is high for ER-negative tumors, decreases as ER concentration increases, but increases again for tumors with very high ER levels. Application of the model to the validation set of 293 untreated patients with node-negative breast cancer resulted in a strong correlatinn (p < 0.0001) with actual clinical outcomes. Conclusion: This model can be used to predict the likelihood of recurrence for patients with node-negative breast cancer, and can be used to identify patients that might benefit from systemic adjuvant therapy.

104

m ~E ~

c

(~)

~

oF o e m ~

~

~

(~c)

E~HESSIGN (IF ITS H~i~rEPATIVE ACTIVITY? *A. Parediso, A. Ym~gia, A. ~ s a ,

S. Tcnnmsi, A. Racanelli, M. De Lena. 0ncolo~y

Institute,

70]24 BARI, ITALY Ki 67 is a p r o l i f e r a t i o n antigen a s s o c i a t e d with a c t i v e l y c y c l i n g cells. 104 p a t i e n t s w i t h p r i m a r y o p e r a b l e b r e a s t c a n c e r had Ki 67 m e a s u r e d on tumour sections by i m m u n o c y t o c h e m i s t r y . Forty-five patients w e r e p o s i t i v e for Ki 67 b i n d i n g (> 5% t u m o u r nuclei stained). T h e r e was an e x c e l l e n t c o r r e l a t i o n w i t h grade: No. of tumours p o s i t i v e for Ki 67 Grade

I

(n = 23)

6

The TH gradient

(~TH)

is generally ccnsidered related to the

vasculerity and metabolism of bregst cancer. In order to verify if the proliferative activity of OBC can be predicted by a A T H evaluation, in the last t~o years, we have performed a contact thenmegraphy just before mastectomy, cn a continuous series of i01 women with OBC. Tumors were classified in 5 A T H categories, A,B, C-,C+,D fras the icwest to the highest ATH. After surgery, proliferative activity was evaluated cn fresh

(%)

(26%)

tumor frs~nente by autoradiographic method (incubation for I hr at ~7"C with ~-Thymidine)

and expressed as % of labeled tomc~ cells (LI). At the

moment, 89 pte are evaluable (median age 57 yre; 32 (36%) in pre-pe~i Grade

II

Grade

III

(n = 33)

12

(38%)

33

(69%)

menopausal staOas). Primary breast tunc~ lesions resulted TI,T2,T3,T~ respectively in 21~o;58%;~£;19% of cases; N O in 43°/o. No ATH category A has

(n = 48)

X2=3O.7;2d.f.;p<0.001 T h e r e w a s no c o r r e l a t i o n between Ki 67 and tumour size nor lymph node stage. Patients with Ki 67 positive tumours had an increased number of recurrences.

been observed. Median LI value was ~go. No significant correlation was fcl~d between LI value orATH category and clinical-biological feat/res of t~aors. Category

Kb

ATH

cases

Mean LI (%) Total series

pre~eri

post

B

8

1.5

3

1.2

C-

8

2.9

0,9

3

C+ D

26 47

2.3 1.8

2.15 2.8

2.35 1.75

Proliferative activity shows no relationship with the TH aspect of OBC verified by contact thermo~mphy.

Abstracts 105

pBOLIF~PATr~ ACTIVITY

~D

H~%~N~

F~0~[~

CON~

IN ~ K T I O N

TO

106

MA~4DGBAPHIU P ~ OF Qp~ABLK ~ A S T ~ (08C)*A. Paradias, V. Ventrella, S. Tommasi, A. Yer4~ia, G. Simone, A. Racsnelli, M. De Lena. Oncology Institute, 70124 BARI,Italy To verify the relationships of me~mographic (mx) pattern with hormone receptor ccntent (ER,P~R) and proliferative activity of 0BC, we studied iii wc~en (6~% post-menopausal; mediem 8ge 54 yrs) in the l~t two year~. Before mastectc~y the patient% were submitted to mmrmo~rsphy and the tumors classified in 5 ~x categories ascording to Broberg: i) spiculated appearance with or without calcifications; 2) struclx~l va~iaticn with c~ without calcifications (density); 3) clusters of calcifications; 4) eircun~cribed; 5) no visible c ~ as ccnpar~d with nornel breast. The enzlysis of ER end PgR content (cut-off 10f~ol/mg cyTosol protein) was performed by DCC method on 811 pts. Autoradiogvaphic method with incubation of fresh t%mor fre~nent% for 1 hr at 37 °C (3H-Thymldine) was utilized to evaluate proliferative activity (LI-~/~. labeled ceils) on 47 tumors. Mean ER and P~R content resulted no significant differer~ amor~ the 5 n~ categories, whereas the ~R asd/or PgR sta~as, has been found percentuslly different in the follcwir~ c~tegories: I vs IV (ER:p~0.Ol;RPg:p~O.Ol;ER+~g+:p~O.006; ~R-RP£-:p~O.001); II vs V (~9-RPg-:p~O.00£); Ill vs IV (RPg: ~0.05). LI value resulted significantiy different only in III vs IV category (p=O.04). We suggest to ccnsider mx I cat%&<~ry predictive of high probability of Mx No ~+ pgR+ No hormzne recepto~ positivity 8rid categ cases (%) (%) C~ LI IV mx cat%gory of a low probabiI 31 84 71 12 2.29 lity. An3,w~y, the little rmmber II ~ 62 50 14 3.41 of cases does not permit any III i0 60 90 4 6.32 kind of conclusion ~aout n~ and IV 33 48 36 17 2.U7 LI relaticnshJps. V 5 80 80 -

107

CA15-3 LEVELS IN BREAST CYSTIC DISEASE AND IN EARLY STAGE BREAST CANCER: A PROSPECTIVE STUDY. MK Gupta,* R Arciaga, J Snoga, and S Grundfest-Broniatowski, The Cleveland Clinic Foundation, Cleveland, Ohio 44195 The u t i l i t y of CA15-3 as a tumor marker in monitoring advanced breast cancer is well documented. However, i t s diagnostic s e n s i t i v i t y and i t s prognostic value in early stage breast cancer and in benign breast diseases has not yet been established. We measured CA15-3 levels in 40 women with benign breast lesions. Twenty-nine of these had breast cystic disease and CA15-3 was measured in both serum and cyst aspirates of these patients. Also, sera from 78 patients with newly diagnosed breast cancer (48 lymph node negative (LN-), 20 lymph node p o s i t i v e (LN+), and I0 with metastases) were analyzed f o r CA15-3 and CEA p r i o r to surgical resection of cancer. The levels of CA15-3 in benign diseases (mean S.D., 17 ± 5.6 U/ml; range, 5.8 to 28 U/ml) were identical to levels obtained in 48 normal women (mean S.D., 17.3 ± 5.3 U/ml; range, 6.6 to 27,9 U/ml). However, 34% of patients showed a 2 to 12 fold increase in CA15-3 levels in cyst f l u i d s as compared to serum levels. This high cyst/plasma r a t i o suggests increased local synthesis of this antigen by these lesions. CA15-3 levels > 28 U/ml were also observed in 16.6% (8 out of 48) of patients with LN-, 30% (6 out of 20) of patients with LN+, and 80% (8 out of i0) of patients with metastatic breast cancer. CEA in these patients showed 15%, 2 5 % , and 20% s e n s i t i v i t y , respectively. The prognostic value of elevated CA15-3 levels in cyst f l u i d or in serum from patients with early stage cancer is c u r r e n t l y being assessed.

108

133

COMPARISON O F CEA, L A S A - P A N D C a 1 5 - 3 IN B R E A S T C A N C E R . I N T E R I M R E P O R T . * F r a n c i s 9, A r e n a , MD, Hematologic Research Foundation, Roslyn, NY 11576. T h i s is a f o l l o w - u p r e p o r t c o n c e r n i n g the comparison of CEA, L A S A - P , a n d C a 1 5 - 3 a n t i g e n s in p a t i e n t s w i t h b r e a s t c a n c e r , 50 p a t i e n t s h a v e b e e n e n r o l l e d in t h i s s t u d y . 36 p a t i e n t s h a v e p r e s e n t e d w i t h no e v i d e n c e of d i s e a s e w i t h a m e a n f o l l o w - u p t i m e of 16 m o n t h s (range 1-32 months). 14 p a t i e n t s w i t h d o c u m e n t e d metastatic d i s e a s e h a v e b e e n f o l l o w e d f o r a m e a n t i m e of 7 months (i-28 m o n t h s ) . Of t h o s e i n i t i a l N E D patients, 30 h a v e r e m a i n e d w i t h o u t r e c u r r e n t disease. Of t h e s e 30 p a t i e n t s , 2 3 / 3 0 (76%) h a v e all 3 m a r k e r s in n o r m a l r a n g e . 6 patients have a n e l e v a t e d L A S A - P a n d 1 p a t i e n t h a s all 3 markers elevated. Of t h e 6 o r i g i n a l l y NED that have recurrent breast cancer, 3/6 patients had an e l e v a t e d C a 1 5 - 3 a n t i g e n p r i o r to or at recurrence, 1 p a t i e n t h a d an e l e v a t e d L A S A - P , a n d 1 p a t i e n t h a d a n e l e v a t e d CEA. i patient showed no elevations of a n y t u m o r m a r k e r s , despite recurrence. In 4/6 p a t i e n t s w i t h r e c u r r e n t d i s e a s e , t h e r e w a s a m e a n l e a d t i m e of 6.2 m o n t h s b e t w e e n a p o s i t i v e t u m o r m a r k e r elevation and documentation of r e c u r r e n t disease. In t h o s e 14 p a t i e n t s w i t h k n o w n metastatic d i s e a s e , 1 2 / 1 4 h a d an e l e v a t e d C a 1 5 - 3 a n t i g e n , 7 / 1 4 h a d an e l e v a t e d C E A a n d 1 0 / 1 4 h a d an e l e v a t e d L A S A - P . O n l y o n e p a t i e n t h a d no e l e v a t i o n of a n y m a r k e r . Using a combination of a l l 3 a n t i g e n s , 1 3 / 1 4 (92%) w e r e p o s i t i v e , The c l i n i c a l c o u r s e of p a t i e n t s w a s p a r a l l e l e d by t h e l e v e l s of t h e s e e l e v a t i o n s , especially the Ca15-3 antigen.

DETERMINATION OF CA 15-3 LEVELS IN NORMAL, BENIGN AND MALIGNANT HUMAN BREAST TISSUES. Kouyoumdjian S.C., Beaune L, Rymer LC. and Feuilhade F. Laboratoire d'Oncologie Clinique et Fondamentale de l'Universit6 Paris-Val de Marne. H6pital Henri Mondor 94000 Cr~teil FRANCE The CA 1543 is a carcinoma associated antigen which is detected at higly levels in sera of breast cancer patients. The ORIS solid phase radioimmnnoassay utilizes two monoclonal antibodies (115D8, DF3). One of these antibodies, 115D8, was raised against antigens of human milk.fat globules membranes. The DF3 antibody was prepared against a membrane enriched fraction of a human breast carcinoma. Recent date suggest that this marker is also expressed sometines in benign diseases. However, the exact significance of the local production into the cells is not yet clearly demonstrated. In our study we have determined, prior the treatment, the levels of CA 15-3 into the tissues of four groups of women.- GROUP 1 : normal breast tissues - GROUP 2 : benign mastopathies,adenofibrosis and fibrocystic mastosis - GROUP 3 : adenocareinomas - GROUP 4 : metastatic cells mostly from pleural effusions. Simultaneously the level of ER, PR and aromatase activity are measured in same samples. Different clinical parameters are also determined such as SBR histopronostic grading. Our data show, a great variation of CA 15-3 expression levels from samples to samples in each group and also between different groups. In normal cells, the concentration of CA 15-3 is very low. In benign tumors the levels of CA 15-3 is higler and more frequent However the concentration of the marker in the sera is frequently very low. In adenocarcinomas and mestastic tissues most samples exhibited a high level of CA 15-3 expression which is correlated with the concentration present in the sera of the same patients. In this groups no correlations was observed with receptors status or SBR grading. A local identification of CA 15-3 expression using an immuno-histochemical assay with DF3 as specific antibody is presently under investigation.

134 109

Abstracts 110

MULTIPLE TLTMOUR MARKERS IN BREAST CANCER, *J.F.R. Robertson, M. Price, C. Selby, D. Pearson, R.W. Slamey, Nottingham City Hospital, UK The future management of systemic treatment in advanced breast cancer depends upon the o b j e c t i v e assessment of tumour mass. We have studied in 386 patients the pattern of serum concentration using 3 potential breast cancer serum markers carclnoembryonic antigen (CEA), CA 15-3 and NCNC-II. Mean serum values (n = 386) Group NCRC-II CA15-3 CE___AA (U/L) (U/L) (ng/ml) Normal controls (NC) 49.9 16.8 1.7 Benign breast disease 33.8 14.9 2.1 Primary breast cancer 79.9 17.5 2.8 Advanced primary cancer 79.9 42.9 4.4 Advanced systemic cancer 116.6 131.1 17.3 (ABC) (Analysis of variance) p<0.0004 p<0.00001 p<0.00001 The means + 2 SD for NCRC-II and CA 15-3 in the NC group were 182 and 33.3 respectively. The upper limit for serum CEA was taken as 6 ng/ml. The percentages of patients with elevated markers above these levels were: Advanced systemic N o r m a l Controls cancer + benign NCRC-II CA 15-3 CEA

20.5% 67.1% 36.8%

3,2% 2.6% 3.88

S E E M THYMIDINE KINASE A TUMOUR MARKER IN ADVANCED BREAST CANCER, *J.F.R. Robertson, K. O'Neill, G. McKenna, R.W. Blamey, City Hospital, Nottingham and University of Ulster, Coleraine, UK. The enzyme thymidine kinase (TK) is associated with DNA synthesis. i.) TK serum levels were studied in normal controls, primary breast cancer (PBC) and advanced breast cancer -

(ABC).

All values are given as p moles ml "I hr -I . Group NO. of patients Mean value SD Normal 20 4.22 ± 1.08 PBC 60 6.22 ± 1.24 ABC 20 9.79 ± 7.56 Analysis of variance p < 0.005 2.) Twenty PBO patients were followed up a f t e r m a s t e c t o m y by serial 3 m o n t h l y TK levels in the disease free interval. Four patients have developed systemic recurrence with a mean serum value of 14.3. 3.) Ten ABC patients had serial TK levels 2 monthly during the first 6 months of primary hormone therapy. External assessment of response to treatment by UICC criteria has been obtained. The serum values fell in all 5 responders (mean 9.12 to 4.78) and rose in all 5 progressors (mean 8.62 to 38.56). Serum TK reflects stage of disease b e i n g significantly elevated in advanced breast cancer both on serial estimations and in comparison with controls. Serial TK levels in ABC patients measures response to systemic therapy. TK appears to be a useful biochemical marker in breast cancer.

77.1% of the AHC patients had one or more markers elevated, The 3 antibodies in combination show great promise for the measurement of tumour mass in patients with advanced disease.

111

PATIENTS112

LONGITUDINAL STUDIES OF CA 15-3 AND CEA IN WITH METASTATIC BREAST CANCER. AM DNISTRIAN*, EJ GREENBERG, D SCHWARTZ, C SMITH AND MK SCHWARTZ. MEMORIAL SLOAN-KETTERING CANCER CENTER, NEW YORK, NEW YORK 10021. Thirty nine patients with metastatic breast cancer were followed for 2 to 6 years. All patients had recurrent disease or were at high risk for recurrence. At the time of entry,ll/39 (28%) h a d a b n o r m a l C A 1 5 - 3 () 30 U / m l ) , 4 / 3 9 (10%) h a d a b n o r m a l C E A (> 5 n g / m l ) , 3 / 3 9 (8%) h a d b o t h m a r k e r s e l e v a t e d , a n d 2 1 / 3 9 (54%) d i d n o t e x h i b i t e l e v a tions in either marker. In patients with initial tumor marker elevations,increases in CA 15-3 and CEA paralleled disease progression. During followu p o f t h e 21 p a t i e n t s w i t h i n i t i a l l y n o r m a l m a r k e r levels, CA 15-3 and CEA first became abnormal in 12 a n d 5 p a t i e n t s , r e s p e c t i v e l y . Both markers r e m a i n e d in t h e n o r m a l r a n g e in t h e 3 p a t i e n t s w h o were disease free throughout serial monitoring. These results demonstrate the clinical utility of the combined measurement of CA 15-3 and CEA in assessing disease progression in b r e a s t c a n c e r patients at high risk for recurrence. Ultimately, breast cancer markers may provide sufficient lead time for early intervention resulting in increased disease free interval and survival. (Funds were provided by the Joseph Alexander Foundation through the Breast and Prostate Cancer Research Foundation. CA 15-3 reagents were provided by Centocor.)

IMx: BREASI CANCERMARKER J.G. Konrath*, G.L, Manderino L.W, Przywara. (Abbott Laborator es, Abbott Park, IL 60064)

An automated [IMx) enz~#ne immunoassay for deslalylated breast cancer associated mucin has been developed. The f u l l y automated system performs a l l assay pipettin~ steps fluorescence detection and data reduction in less than 45 min. Using a two-step competitive inhibition format, breast cancer associated mucin-like glycoproteins in sera are detected. Monoclonal antibody M85C34 has been shown to react strongly with deslalylated breast cancer associated mucins, During the assay, a serum sample is desialylated and allowed to react with alkaline phosphatase ( A P ) labeled M85C34, M u c i n coated microparticles are then added. Remaining free M85C34Ap binds to these microparticles, The microparticles are washed and a fluorogenic substrate added. A fluorescent product is produced at a rate inversely proportional to the arcount of the mucin present in the serum, A clinical comparison with the commercially available CA 15-3 bead RIA assay was performed using cutoff values (R * 2SD) for both the IMx (23 unitsJ and CA 15-3 (30 units) assays derived from the same panel of documented breast disease-free normals (n=22). The penel results were as Follows. Controls Benign Breast Stage I Stage I I Stage IV

IMx -'2-/30 2/29 4/15 5/15 20/29

CA 15-3 3/30 2/29 l/IS 4/15 15/29

In this study of a limited population the performance of the IMx breast cancer assay suggests assay s e n s i t i v i t y superior to CA 15-3.

Abstracts 113

OB3ECTIVE MEASUREMENT OF RESPONSE IN ADVANCED BREAST CANCER. * R W Blarney, M R Williams, A Turkes, City Hospital, Nottingham, U.K.

114

The ideal method to measure response would be with circulating tumour markers. Individual markers have not proved sufficiently sensitive. We have investigated 20 possible markers in 179 patients receiving endocrine therapy. Levels were measured before and a f t e r 2, 4 and 6 months treatments. Five markers proved sensitive if used at very high levels: CEA, CRP, Orosomucoid, Ferritin and ESR. These have been combined to produce an index: UICC Criteria at 6 Months Response Static Progression

Serum Index Response Progression

35

16

13

2

/4

47

The index is simpler to apply and should replace the UICC criteria: it may also be used to direct therapy.

135

THE INFLUENCE OF E X O G E N O U S H O R M O N E S ON E S T R O G E N AND P R O G E S T E R O N E R E C E P T O R S IN PRIMARY BREAST CARCINOMA. B Heppefle*, BD Willan, BE Krause, AW Lees, HR Lukka, DW Morrish, HJ Jenkins, M. Grace. Cross Cancer Institute, Edmonton, Alberta, Canada. T6G 1Z2. Exogenous hormone is known to alter estrogen (ER) and progesterone receptor (PgR) levels, thus potentially influencing the choice of a therapeutic regimen. We evaluated the effect of oral contraceptives (OC) and estrogen replacement drugs used prior to and at the time of biopsy in 1334 breast cancer patients. ER levels (mean + S.E.M.) were significantly lower among current users (24 + 8 fmol/mg protein) and previous users of OC (43 + 4 fmol/mg protein) compared with never users (120 + 7 fmol/mg protein; p < 0.02 and p < 0.001 respectively). Current users of estrogen replacement drugs had significantly reduced ER values (58 + 10 vs 94 + 6 fmol/mg protein, p < 0.05); whereas previous users had higher values (126 + 14 vs 94 + 6 fmol/mg protein, p < 0.05). PgR levels were lower only among current users of OC but this difference was not statistically significant. Duration of hormone use (< 1 yr to > 10 yr) was > 10 associated with lower ER but not PgR levels. Any previous OC use (discontinued 1 - 4 yrs prior to biopsy, 5 - 10 yrs or > 10 yrs) resulted in decreased ER values (22 + +6; 43 + 9; 43 + 6; respectively) compared with never users (127 + 8, p < 0.001). Similar recency results were observed with previous use of estrogen replacement drugs. There was no association between duration or recency of exogenous hormone use and PgR levels. We conclude that both current and previous hormone use significantly alters ER concentration and must be considered when making therapeutic decisions in the treatment of metastatic breast cancer.

1 15

COMPARISON OF IMM~/NOHISTOCHEMICAL ESTROGEN RECEPTOR ANALYSIS IN PARAFFIN AND FROZEN SECTIONS WITH STEROIDB~NDING ASSAYS IN HUMAN BREAST CANCER. J Andersen*, H S Poulsen, W J King, I Christensen, B B Rasmussen, S M Thorpe and C Rose. Danish Cancer Society, Dpt. of Experimental Clinical Oncology, N~rrebrogade 44, DK-8000 Aarhus C, Denmark. The Finsen Institute, Strandboulevarden 49, DK 2100 ~¢benhavn ~, Denmark. Abbott Laboratories, North Chicago, Ii 60064, USA. Tumor specimens from postmenopausal women with stage If breast cancer were independently assayed for estrogen receptors by a conventional dextran-coated charcoal steroid binding assay (DCC), by an immunooytoohemical assay utilizing frozen tissue sections and monoclonal ER-antibodies in a peroxidase-sntiperoxidase amplification system (ER-ICA), and by an immunohistochemioal assay in paraffin sections using a monoclonal ER-antibody and an avidin-biotin-peroxidase amplification system (ER-~A). The results are shown below: ER-ICA pos

neg

DCC poa DCC neg

63 1

9 16

total

64

25

ER-PA

total

i

total

pos

neg

72 17

58 2

25 21

83 23

89

60

46

106

The DCC-positive, ER-ICA or ER-PA negative tumors had an ER-content below lOO f~ol. Comparison of ER-status determined by ER-ICA and ER-PA in 84 tumors showed concordance in 73 and discordance in ii, all of whom were ER-ICA positive and ER-PA negative (end with a Dec ER content below i00 fmol). we conclude that imml~nohistochemical ER analysis offers a valuable alternative, but the sensitivity is lower in t~llors with an ER content < I00 fmol, particularly when using peraffin-em~oedded tissue.

116

AN IMPROVED NaSCN EXCHANGE ASSAY OF E S T R O G E N R E C E P T O R USING A M M O N I U M SULFATE PRECIPITATION. K R Gertz*, Y M Chen, C B Vaughn, O n c o l o g y D e p a r t m e n t , P r o v i d e n c e Hospital, 16001 W. Nine Mile Road, Southfleld, M i c h i g a n 48075. NaSCN has been used for s o m e t i m e in e s t r o g e n receptor (ER) assays. It has the a d v a n t a g e over hlgh-temperature e x c h a n g e assays b e c a u s e it is performed at 4°C; thus p r e v e n t i n g the d e n a t u r l z a tlon of the labile receptors. NaSCN d i s s o c i a t e s the e n d o g e n o u s hormones from the h o r m o n e - f l l l e d receptors but may also Inactivate the unfilled h o r m o n e receptors wlth Iong-tlme contact. 58% a m m o n i u m sulfate p r e c i p i t a t i o n of the r e c e p t o r s after just one hour of NaSCN digestion, Is a fast, safe method to separate the free receptors from the NaSCN. Now the unfilled r e c e p t o r s can form hormone receptor complexes without I n t e r f e r e n c e from NaSCN. The assay of ER with AS and NaSCN was 117-199% of the control value. The dlfference obtalned between the specific activities of assays wlth and without NaSCN digestion represents the endogenously filled hormone receptors. Specifically, ammonium sulfate p r e c i p i t a t i o n f o l l o w i n g NaSCN d i g e s t i o n greatly e n h a n c e s the s p e c i f i c activity of the total receptors.

136 117

Abstracts EVALUATION OF NUCLEAR ER AND PgR IN RABBIT UTERI BY NaSCN DIGESTION AND BY ASSAY USING AMMONIUM SULFATE PRECIPITATION. YM Chen*, CC S t r o u d , and CB Vaughn. Oncology Research Laboratory, Providence Hosptlal, Southfleld, Michigan 48075. The p r o p e r u t i l i z a t i o n o f NaSCN d i g e s t i o n o f the nuclear p e l l e t and assay o f ER and PgR by ammonium s u l f a t e p r e c i p i t a t i o n have been p r e c i s e l y studied. The n u c l e a r p e l l e t s e p a r a t e d from t h e c y t o s o l and w e l l - w a s h e d w i t h t h e b u f f e r was f r o z e n (-22°C) overnight or longer. TEDG b u f f e r c o n t a i n Ing 10 mM M o l y b d e t e was used. The f r o z e n n u c l e a r pellet was suspended In t h e b u f f e r c o n t a i n i n g 20 nM o f T r i t i u m - l a b e l l e d E2 o r p r o g e s t i n and I n c u bated a t 4 C f o r 30 m i n u t e s . 1 p a r t o f 6 M NaSON In t h e b u f f e r was then mixed w i t h 9 p a r t s o f t h e Incubation mixture. The i n c u b a t i o n and d i g e s t i o n proceeded t o g e t h e r a t 4°C f o r a hour f o l l o w e d by

118

AS (ammonium sulfate) precipitation at 50% saturat6~n and centrifugation at I05,000 g, 30 min., 4 C. A pellet containing hormone receptor complexes was formed. The pellet f r e e d of N a S C N was then suspended In t h e b u f f e r and c e n t r i f u g e d under the same conditions to obtain the solubilized nuclear fraction in t h e supernatant (S.). A pellet was obtained by AS p r e c i p i t a t i o n ofmS,. The pellet was suspended In 50% s a t u r a t e d AS ~olutlon and centrifuged to obtain washed pellet for the final end product. The total activity in t h e final AS p e l l e t w a s p r e s e r v e d In freezer. The non-speclfic activity was o b t a i n e d by t h e same procedure using 20 nM of t h e hot hormone plus 100 fold cold hormone. The high y i e l d of n u c l e a r ER and P g R was o b t a i n e d by t h i s procedure.

119

ESTROGEN R E C E P T O R GENE METHYLATION IN HUMAN BREAST TUMORS. NS F a l e t t e , SAW Fuqua, GL Greene) SM Hill, and WL McGuire, The U n i v e r s i t y of T e x a s H e a l t h Science C e n t e r ) San Antonio, T e x a s 78285. The presence of e s t r o g e n r e c e p t o r (ER) in b r e a s t tumors i d e n t i f i e s a group of p a t i e n t s m o s t likely to respond to endocrine t h e r a p y . But the control m e c h a n i s m s underlying ER a c t i v a t i o n and expression a r e unknown. E p i g e n e t i c c h a n g e s in DNA m e t h y l ation could provide a m e a n s to control ER gone expression. We h a v e d e t e r m i n e d t h e d e g r e e of m e t h y l a t i o n of t h e ER g e n e in human b r e a s t t u m o r biopsies. Two m e t h y l a t i o n - s e n s i t i v e r e s t r i c t i o n endonucleases, Hpall and MspI, w e r e used in c o m b i n a tion with EcoRl to d i g e s t t u m o r DNAs, and Southern blot analysis was p e r f o r m e d . Using an ER c D N A probe obtained f r o m G e o f f r e y G r e e n e , w e d e t e c t e d d i f f e r e n c e s in m e t h y l a t i o a of s p e c i f i c ER r e s t r i c t i o n f r a g m e n t s . When t h e f r e q u e n c y of m e t h y l a t i o n of a 2.9-3.5 kb band was e x a m i n e d as a function of t u m o r ER s t a t u s as d e t e r m i n e d by ligand-binding assays) s i g n i f i c a n t d i f f e r e n c e s w e r e observed. No d i f f e r e n c e s w e r e seen in the m e t h y l a t i o n of this ER f r a g m e n t in normal human placenta. We have localized this m e t h y l a t e d f r a g m e n t to the region containing the hormonebinding and 3' u n t r a n s l a t e d domains of t h e ER gone. Conclusions: i. W e have observed specific methylation of the ER gene in human breast tumors. 2. These analyses will help establish a relationship between the methylation status of specific sites and ER expression,

120

CLINICAL SIGNIFICANCE OF THE IN SITU STABILITY OF ESTROGEN AND PROGESTERONE RECEPTORS IN HUMAN BREAST TUMORS KEPT AT ROOM TEMPERATURE. J. Tabachnick,~ P. Kim, A. Desai, R.G. gomers, and E. Catalano. Depts. of Surgery and Pathology, Albert Einstein Medical Center, Phila., PA 19141. The in situ stability of estrogen (ER) and progesterone (PR) receptors in breast tumors is tumor dependent. (Breast Can. Res. and Treat. 10:102, 1987.) Most studies with once-thawed tumors report on steroid receptor stability in tumor oytosols, a mixture of proteases and other cell contents extracted from cancer and non-cancerous tumor cells. By keeping the tumors 1 and 2 hrs at room temperature (RT) and then extracting the remaining protease stable receptors we are recording intracellular proteolysis of ER and PR in cancer cells. Partial tumor dehydration (15-25%) at RT would tend to seal in cell fluid. Receptors were classified as ER and PR stable (> 50% remaining), unstable (< 30% remaining), and borderline (31-49%). After 2 hrs at RT, 29/31 (94%) ER and 23/32 (72%) PR retained 10-150 fmoles/mg protein with high affinity (K n < 0.4nM). To determine possible clinical significance-of tumor differences in ER and PR stability after 2 hrs at RT, tumors were grouped as ER/PR Stable 9/32 (28%), ER/PR Unstable 9/32 (28%), ER Stable/PR Unstable 12/32 (38%), and ER Unstable/PR Stable 2/32 (6%). This preliminary data revealed no clear correlation between in situ receptor stability at RT and patients at risk for early recurrence, metastasis or response to therapy. Receptor stability at RT may need narrower redefinition. A longer interval than 2 hrs at RT may be required. One patient with ER Stable/PR Unstable receptors after 2 and 18 hours at RT (135 and 97 fmoles remaining) had early recurrence with metastasis. Receptor stability in cytosols and nuclear receptors are also under investigation.

DETERMINATION OF PROGESTERONE RECEPTORS IN BREAST CANCER SPECIMENS BY BIOCHEMICAL AND IMMUNOCYTOCHEMIOAL METHODS. G. Di Fronzo, C. Patriarca, V. Cappelletti, G. Cattoretti, G. Granata, P. Miodini, D. Coradini. Istituto Nazionale Tumori, 20133 Milan, Italy. Three monoclonal antibodies (mAbs) kindly supplied by D. Edwards, were used for immunocytochemieal detection of progesterone receptors (PgR) on frozen sections of human breast cancer samples. One, called A/B-52 reacted both with the 94KDa A and the 120 gDa B receptors; the other two mAbs, B30 and B-64, recognized B receptors only. The immunocytochemical detection of PgR was performed with an indirect streptoavidin-biotin peroxidase method using carbazole as ehromogen. Immunoperoxidase staining was predominantly nucleus-restricted; there was cell to cell hetoregeneity and patchy positivity. In a series of 74 human breast cancer specimens immunocytoehemical findings were correlated with cytosolic PgR (PgRc) determined by a radioligand assay, and with nuclear PgR (PgRn) determined by an enzyme immunoassay. An overall agreement on positivity and negativity was observed in 67.5% of cases when the immunoeytoehemical results were compared with the PgRc levels, and in 81.0% when the PgRn were compared. 46.9% of PgRc+ (>i0 fmol/mg protein) tumors were immunocytoehemioally defined as PgR- and 4% of PgRc- specimens gave a positive staining. About twenty percent of PgRn+ (7100 fmol/mg DNA) tumors did not stain for PgR. Conversely, 19.2% of PgRn- tumors were characterized by a positive immunocytochemical staining. Practical implications of these findings will be discussed in the presentation. Supported in part by a grant PFO 86.02813.'44.

Abstracts 121

HOP~IONE RECEPTOR DETERMINATION BY BINDING ASSAY (BA) VERSUS ENZYME IMMUNOASSAY (EIA): WHICH PATIENTS DON'T MATCH AND WHY ? F.A. Holmes, H.A. Fritsche, A.M. Geitner, R. Sutton, A.U. Buzdar, G.N. Hortohagyi. U.T.M.D. Anderson Cancer Center, 1515 Holcombe, Houston, Tx. 77030. We retested estrogen (ER) and progesterone receptors (PR) on i00 frozen tumor samples by standard BA (BA88) and EIA methods~ Both measurements were compared to the initial hormone receptor determination performed by BA when the tissue was obtained (BA1). Specimens were collected from 5-82 thru 12-86. Any value >i0 fm/mg protein was considered positive;
SA2fE

72

83

122

77

POS to NEG 3 14 6 NEG to P0S 25 3 17 CORRELATION COEFF 0.96 0.85 0.71 The high number of ER neg values by BAI which became pos by BA88 may reflect 1-addition of molybdate to buffer, 2different portions of tissue analyzed. 14 samples pos By BA88 were neg by EIA. This is the reverse of what is usually seen~ and may be caused by hydrolysis of EH between the steroid-binding and DNA-bdnding sites. The BA is pos if steroid-binding is present; a poe EIA requires the presence of both sites. Characteristics of the 13 pts whose pos ER by BA88 was neg by EIA were consistent with ER neg status clinically (<50 yrs old, short ddsease-free interval and survival). In these 13 pts, 9 pts had neg PR by both BA88 and EIA; 2 had pos PR by both BA88 and EIA. Overall agreement between BA88 and EIA was good. In the small subset of pts whose measurements were discrepant, EREIA may be a better predictor of outcome. Measurement of ER by EIA is rapid, requires less tissue and can be used confidently for pt management.

123

PERCENTILIZATION VS STANDARD CUT-OFF IN DEFINING

PROGNOSTIC VALUE OF ESTROGEN RECEPTOR (ER) IN BREAST CARCINOMA (BC). A STUDY FOR A NEW PROPOSAL. *G.Cocconi, R. Passalacqua, G. Bisagni, C. Bozzetti, A.Guazzi, N. Naldi, M. Bella, A. Borrini. Medical Oncology Service, Health Physics Dept., Ospedale Regionale, 43100 Parma, Italy. ER concentration in BC is expressed as a quantity per weight unit, usually fmol/mg.prot.(fmp). Values are considered as positive or negative when over or under a given concentration, named "cut-off". As time goes on, increasing sensitivity of the method tends to raise the ER contents, even using the same method. Moreover, interlaboratory variability is high in spite of the efforts of the standardization groups and, in the largest published series, the percentages of ER+ are highly different even using the same method and the same cut-off value. In this study, a series of 369 evaluable pts had ER determination by DCC method at time of diagnosis on a 3-year period (1970-1982). They were assessed for the 3-year overall survival (OS). Different values of fmp or different percentiles (pct) of all ER determinations were used as cut-off in separating pts with lower vs higher ER. In the table, chi-square values are reported for comparisons, by log-rank test, between two classes ofpts having ER below or over each cut-off. Cut-off points were calculated as absolute values, anti expressed as fmp, or calculated as percentiles values based on the overall number of patients (o-pct) or on patients distinguished by menopausal status (m-pct) or by year and by menopausal status (ym-pct). chi-square fmp chi-square pct o-pct y-pct m-pct ym-pcl 10 20 30 40

33.i5 14.03 19.80 21.31

10% 20% 30% 40%

19.45 27.38 19.85 19.55

17.10 30.09 22.86 21.09

19.45 21.89 23.42 23.85

26.77 38.50 30.48 29.70

Differences are highly significant using all methods. However ym-pct method mostly showes an higher chi-square value than any o~er, suggesting a better biological fitting. We propose to use percentilization of ER in distinguishing pts with ER-poor or ER-rich tumors. Supported by CNR PF"Oncologia"; co. no. 86.00641.44

137

INFLUENCE OF PRE-FREEZING DELAY ON ESTROGEN (ER) AND PROGESTERON RECEPTOR (PgR) CONCENTRATION IN SPECIMENS OF BREAST CANCER ~VI.Benecchi, G.Cocconi, C.Bozzetti, A.Guazzi, R.Nizzoli Med.Oncol.Service, Ospedale Regionale, 43100 Parma, Italy. Discrepant proportions of ER positive patients (pts) reported by different laboratories, even using the same technical procedure and the same cut-off level, support confusion and disagreement about the ER prognostic value. This discrepancy may be ascribed to differences both in analytical steps and in specimen handling. Only die former differences may be monitored by quality controls on lyofilized standards. In this study we compared receptor concentration, assayed by DCC method, in a consecutive series of tumor samples under different conditions of pre-freezing in liquid nitrogen. ER and PgR concentration was assayed on the same tumor sample, partly immediately frozen (TO) and partly frozen after 1 hour in ice-bath (3"1) (lst experiment, 10 pts); then, the comparison was carried out between samples directly frozen (TO) and samples kept 1 hour at ambient temperature ('I'2) (2nd experiment, 10 pts). The samples were cut in two homogeneus parts, randomly divided in the two arms of each experiment and respectively assayed for hormonal receptors. Only tumors containing more than 3 fmol/mg prot ER and PgR were included in the study. The ER and PgR concentrations were compared by sign test for paired comparison and by regression analysis of T1 or T2 versus TO values. N. AA+ P b a r ER PgR ER PgR

T1 T1 T2 372

vs vs vs vs

TO TO TO TO

9 8 8 9

8 5 7 8

1 3 1 1

0.02 0.36 0.03 0.02

0.79 0.87 0.6i 0.50

16.2 12.4 24.6 26.6

0.97 0.98 0.93 0.93

The decrease of ER after 1 hour in ice-bath was statistically significant, as well as that of ER and PgR after 1 hour at ambient temperature. A decrease of about 20%, 40% and 50% might be respectively estimated. We propose a quick sending of samples to receptor laboratory and a quick freezing. Supported by CNR Finalized Project "Oncology" co. n.86.00689.44.

124

FLOW CYTOMETRIC ANALYSIS OF C - e r b B - 2 AND r a s ONCOGENE PRODUCTS IN HUMAN BREAST CARCINOMA CELLS. K.L. Trimpe 1, D.M. Dombkowski 1 , P.J. Hamer 1, S. McKenzie 2, H. Rabin 1, and W.P. Carney 1. 1Medical Products Dept., E.I. duPont de Nemours & Co.,

N. Billerica, MA 01863; 2Applied BioTechnology, Cambridge, MA 02142. There is increasing evidence that the activation of cellular protooncogenes is involved with the initiation or progression of cancer. Activation of the ras oncogene product may result from specific mutations or from alteration in the regulation of expression of the gene. The c-erbB2 (neu) gene, which is related 1o the epidermal growth factor receptor (EGF-R) gene, has been shown to be amplified in a variety of human carcinomas. We have developed two mAbs to these oncogene products. One, termed RAS-10, is reactive with both the normal and mutated forms of the ras oncogene product, p21. The second, TA1, recognizes the ce r b B - 2 oncogene product, pi85, in immuncprecipitatlon and immunofluorescence techniques. To detect intracellular ras p21, fixed cells were stained with ras-10 and fluoresceinated goat anti-mouse Ig then analyzed by flow cytometry. Specimens from solid human tumors were meehanicatty dissociated to prepare single-cell suspensions for immunofluorescent staining, The level of ras p21 expression was increased in carcinoma cells as much as 7-fold above lymphocyte controls. The results obtained by flow cytometry were consistent with results from immunoblotting and immunohistochemical staining of tissue sections with ras-10. To examine the expression of p21 during the cell cycle, cells were incubated with Hoechst 33342 and pyronin Y for estimation of DNA and RNA content. Initial studies indicated that p21 increased as cells progressed from G la through G 1b to S phase. Cells from human breast carcinoma specimens and cell lines were stained with anti-p185 mAb TA1 for flow cytometric analysis. The level of p185 expression ranged from normal 1o 5x above normal Studies of cerbS-2 expression during the cell cycle show that p185 levets remain constant within all phases. Two-color flow cytometric analysis of ras vs. c - e r b B - 2 expression on breast carcinoma celt lines indicated little correlation between levels of p21 and p185 on most of the lines tested. Studies are ongong to evaluate the expression of ras p21 and c-erbS-2 p185 as diagnostic and prognostic indicators in human cancer.

138

Abstracts

125

TRANSCRIPTIONAL CONTROL OF c-erbB-2 GENE EXPRESSION IN HUMAN MAMMARY EPITHELIAL CELLS. GK Scott,* JM Dodson, and CC Benz, Cancer Research Institute, University of California, San Francisco, CA 94143. Overexpression of the proto-oncogene c-erbB-2 (HER-2/neu) predicts for poor survival in patients with early breast cancer and may occur without accompanying gene amplification in both primary tumor specimens and established breast cancer cell lines. We have shown that normal breast epithelial ceils harvested from 4 different reduction mammoplasty specimens increase their levels of c-erbB-2 mRNA 3- to 25-fold during short-term culture, reaching or exceeding transcript levels found in cultured breast cancer cell lines. To investigate mechanisms accounting for c-erbB-2 overexpression, we have compared 4 human breast cancer cell lines (MCF-7, ZR-75, MDA-231, MDA-453) that exhibit a 30-fold range in baseline c-erbB-2 mRNA levels. Southern and Northern analyses indicate that these differences in transcript levels are not due to gene amplification, expression of estrogen receptor, or responsiveness to estradiol. Nuclear run-on assays indicate that the number of RNA polymerase complexes participating in c-erbB-2 transcription is increased 3- to 10-fold in ZR-75 and MDA-453 ceils as compared to MCF-7 ceils, when expression is normalized to Ih-microglobulin and l~-tubulln. No differences are detected in the hypomethylated status of the multiple HpalI-MspI restriction sites located within a 0.55 kb SmaI-SmaI fragment upstream of the mRNA start (cap) site in each of these cell lines. Using an indirect end-labeling hybridization technique, we have examined almost 2 kb of promoter-containing sequence upstream of exon 1 and have identified a single major DNase I hypersensitivity site in overexpressing MDA-453 cells. These findings indicate that e-erbB-2 ovcrfxpression is determined by transcriptional mechanisms possibly involving a single 5' hypersensitive domain close to the exon 1 cap site. We are currently investigating the possibility that trans-acting factors interacting with 5' c-erbB-2 DNA sequences account for the variable degree of expression of this growth factor receptor in mammaryepithelial cells.

126

127

PARALLEL ENHANCEMENTOF RAS PROTO-ONCOGENE EXPRESSION AND OF ESTRADIOL-16a-HYDROXYLATION IN HUMANMAMMARYTERMINAL DUCT-LOBULAR UNITS (TDLU) BY A CARCINOGEN. N T Telang,* A Basu, M J Modak, H L Bradlow, and M P Osborne, Breast Cancer Research Lab, Memorial Sloan-Kettering Cancer Center, New York, New York 10021, Dept. of Biochemistry, UMD-NJ, Newark, New Jersey 07103, and Biochemical Endocrinology Lab, The Rockefeller U n i v e r s i t y , New York~ New York 10021. We have previously demonstrated t h a t benign (noninvolved) mammary tissue from breast cancer patients exhibits higher c o n s t i t u t i v e levels of ras p21 and higher extent of estradiol (E2) 16~-hydroxylation r e l a t i v e to t h a t in tissues from-patients with no pathological evidence of breast cancer. The purpose of the present study was to examine molecular and metabolic response of benign, 'high' r i s k mammary tissue from mammary cancer patients to chemical carcinogen. Ten-day old explant cultures of benign terminal duct-lobular units from mastectomy specimens (TDLU-M) and of mammary f a t (MF) were exposed to 1 ~g/ml benzo(a)pyrene (BP) f o r 24 hr, and were e i t h e r processed f o r determination of ras p21 l e v e l s , or were incubated with [C-16m-3H E2] to measure the extent of E2 16m-hydroxylat i o n . Treatment of MF to BP did not enhance p21 expression or 16m-hydroxylation. TDLU explants a f t e r exposure to BP exhibited a 2 . 6 - f o l d increase (P = 0.05) in c o n s t i t u t i v e p21 l e v e l , and a 12.7-fold increase (P = 0.001) in 16mhydroxylation, r e l a t i v e to that in MF. The specific enhancement of ras p21 levels and of E2 16a-hydroxylation in TDLU, but not in MF, suggest that the two q u a n t i t a t i v e parameters may constitute useful molecular and metabolic markers in human mammary carcinogenesis. (Supported by NIH grant R2g CA 44741 and ACS grant ACS-IN114-I)

128

TYROSINE KINASE ACTIVATION CONFERS TARGET CELL RESISTANCE TO TNF. J Klostergaard*, TC Suen, RU Rodrigaz X Zhang and MC Hung, The University of Texas M. D. Anderson Cancer Center, Houston, Texas 77030. Previous studies have implicated expression of the E1A oncogene in conferring target cell susceptibility to direct cytotoxicity by macrophages (M0) and NK ceils, as well as to lysis by TNF. Since both M0 and NK cells are thought to play a survelliance role at several steps in the metastastic process, we have examined possible mechanisms of subversion of this surveillance related to transformation. Using the NIH 3T3 expression system, we determined that both the parental and H-ras (E.J.) transfectant were susceptible to rHuTNF (Biogen) in the 103 - 105 units/ml range. Thus, immortalization is irrelevant to the expression of a particular phenotype of TNF-snsceptibility; furthermore, an activated p21 is also excluded from a role in this phenotype. However, the transfectant obtained by expression of the v-abl oncogene demonstrated profound resistance to TNF, or even slight growth enhancement, ata dose as high as 7.5 x 104 units/ml for as long as 96 hr. This pattern was essentially mirrored in the response to cytotoxic activated macrophages. Thus, activation of tyrosine kinase in target ceils during transformation may be one pathway for escape from the surveillance of monocytes/macrophages exerting TNF-dependent cytotoxicity. Supported by NCI CA45265 and ACS CD-360 (MCH) and ACS IM-419 (JK).

PAGET'S DISEASE OF THE EXPRESSION OF c-erbB-2 Rosemary R Millis + and Oneology Units, *Guy's smith Hospital, LONDON,

NIPPLE SHOWS A HIGH INCIDNECE OF PROTEIN. * D i a n a M Barnes ~ G A Lammie + W L Gullick ° , ICRF Clinical Hospital, LONDON, SEI gET and °HammerW I 2 0 H S , DK

Significant amplification of the proto-oncogene c-erbB2 has been d e m o n s t r a t e d in between 10-25% of human breast carcinomas. Gene a m p l i f i c a t i o n correlates well with overexpression of the protein product as detected immunohistochemically. An affinity purified polyclonal antibody, 21N, raised against the predicted c-terminus of the c-erbB-2 protein, has been used in a study of 42 cases Of Paget's disease of the nipple diagnosed over a lO-year period. This r e p r e s e n t e d less than 2% of the mammary carcinomas diagnosed in the unit. Paget's disease was associated with either pure in situ, or infiltrating ductal carcinoma associated with in situ carcinoma, in the u n d e r l y i n g breast in all cases. Positive membrane staining of the Paget cells was seen in 37 cases (88%). The u n d e r l y i n g carcinoma, either in situ or infiltrating, showed comparable staining in each case. This r e l a t i o n s h i p between c-erbB-2 a m p l i f i c a t i o n and the phenomenon of m a m m a r y c a r c i n o m a associated with Paget's disease of the nipple merits further investigation.

Abstracts 129

INMUNOCOMPETENCE OF BREAST CANCER (BC) STAGE II WITH ADYUVANT TREATMENT (A Tr) OF RADIOTHERAPHY (R) AND R PLUS CHE M O T H E R A P Y (CH). Eu~is A.,Levin M., Caballero C., Rumi L.and Batagelj E. Centro Oncol6gico Buenos Aires. (Argentina).

130

BREAST CANCER IN THE C3Hf/He MOUSE-A RELEVANT HUMAN MODEL. T. Mafwijiw and P. Etkind*. The Montefiore Medical Center/ Albert Einstein College of Medicine, Bronx, New York 10467. Although infection by the Mouse Mammary Tumor Virus (MMTV) is not known to play a role in human breast cancer, the presence in the human genome of DNA sequences homologous to MMTV proviral sequences suggests that these human MMTVlike sequences could be involved in human breast cancer. In fact, activation of a specific endogenous MMTV proviral sequence in a unique mouse model system, the C3Hf/He mouse, results in mammary tumors which arise late in the lifetime of these foster-nursed mice which never see exogenous MMTV. With exogenous MMTV infection, mammary tumors arise early in the lifetime of the mouse and are often associated with integration of exogenous MMTV sequences at the int-i and/or int-2 loci and specific RNA transcription at these loci. It has been suggested that human homologues of these mouse loci may be similarly involved in human breast cancer. We have studied these loci in the C3Hf/He mouse which we consider a very relevant model of human breast cancer development since these mice see no exogenous MMTV yet develop mammary tumors late in their lifetime as a result of endogenous proviral activation. We have examined, to date, over 50 C3Hf/He mouse mammary tumors for the presence of int-i and/or int-2 integrations. We have detected only 7 tumors which contained possible int-i integrations and only 2 tumors which showed possible int-2 integrations. RNA expression at the int-i and int-2 locus was seen in only 18 and 3 mammary tumors respectively. If the C3Hf/He mouse is indeed a relevant model for the study of human breast cancer, than our data suggest that a third integration site in these tumors, which we have reported (Etkind and Sarkar,J.Virol.45:l14,1983), may play a more important role than either the int-i or int-2 locus in not only breast tumor development in the C3Hf/He mouse but also in the human.

ACTIVE IMMUNIZATION TO PEPTIDE HORMONES AS AN ANTINEOPLASTIC STRATEGY. P M Ravdin, The University of Texas Health Science Center, San Antonio, Texas 7828#-7884. Several peptide hormones have been implicated as playing important roles in the control of growth of breast cancer cells. Good antagonists for the actions of m a n y peptide hormones are not available. In an attempt to develop general techniques for blocking the actions of peptide hormones, states of autoimmunity with immunoneutralizlng titers of antJbodies to endogenous peptide hormones, have been generated in mice by active immunization. The basic technique is to couple a peptide hormone to a foreign protein and to use the conjugate as an i m m u n o g e n to elicit antibodles to epitopes on the entire conjugate. This technique has been used in mice to generate states of autoimmunity to endogenoas E G F (epidermal growth factor), a peptlde hormone implicated in the growth of some types of murine m a m m a r y cancers. Animals were immunized to K L H (keyhold limpet hemocyanin) alone or K L H - E G F conjugates. All K L H - E G F immunized animals developed antibodies that were capable of immunoprecipitating E G F and which were detectable in ELISA assays. Even 1:100 dilutions of the antisera were capable of bloc~dng the bInding of 1251 E G F to routIne 3T3 ceils in radinreceptor assays. Immunized animals were used as sources of lymphocytes for producing hybridomas which secrete high affinity routine monoclona/ antibodies. Conjugate immunized animals had no obvious systemic toxiclties detectable on histologk examh~ation of major organs~ and had similar weights to control animals. Active immunization to endogenous hormones can be easily achieved, The success of this technique sets the stage for the exploration of the effects of active and passive immunization to E G F in murlne m a m m a r y tumor models. This work is part of an effort to develop a n e w strategy ~ the control of hormone dependent neoplasms.

The use of R and/or CH are two well-known possibilities of post-operative A Tr. The inmune system (IS) plays and impor tant role in tumor-like development, being lymphocytes T and their various subpopulations of vital relevance in defending the host from tumor aggresion. Since R as well as CH can pro duce changes in the IS,different lymphocite subpopulations were evaluated;in 21 patients (pts) bearing BC stage II pre (Pr) and post treatment (Po Tr). 60ml of total heparinized blood were processed by means of a gradient of Ficoll Hypaque 3x106 mononuclear cells were incubed with 20ml of a i/i0 dilution of the following monoclonal antibodies T1,T4,TB,Ia. Them they were incubed with an anti inmunoglobulin antibody of rat marked with fluorescein. 200 cells were reas in a fluorescence microscope. The presence of surface (S) IgG was also studied by means of direct inmunofluorescence techniques. Results:Pts.with Pr Tr R N ° II:TI:60,33~ 4,45, T4:52~ 2,05, T8:19~ 2,52 Ia:12,16~ 3,8 S IgG:3~ 0.91.Po Tr TI:58~ 6,69, T4:45,16~ 6,!0, T8:18,16~ 2,36, Ia:5,16 ~ 1,46 SIgG:3,33,~ 1,28 Pts.with R plus CH Pr Tr N ° I0:TI:44,66~ 8,65,T4:48~ 2,99, T8:16,5~ 2,28, Ia:6,25~ 1,33,SIgG:6,66~ 0,86 Po Tr:TI: 48,33~ 7,35,T4:34,75~ 7,14,T8:13,75~2.99 Ia:i0+2,33 SIgG: 6,33_+ 0,90. Statistically significant results were not obtained in any of the groups (p ~ 0,05) that's why the study is still open.

131

139

132

DISTRIBUTION OF MOUSE MILMMARY TUMOR VIRUS (MMTV) IN ASEAN WILD MICE. *S Imai, M Okumoto, M Iwai, Y Hiasa, N Miyashita, K Moriwaki,Nara Medical University, Nara 634,Japan, Radiation Center of Osaka Prefecture, Osaka, Japan, and National Institute of Genetics, Shizuoka, Japan. Mouse mammary ttaaor virus (MMTV) is thought to be widely distributed among inbred mouse strains as provirus . However, it is known that some wild mice lacked the endogenous MMTV. It is not clear how provirus negative mice had lost their provirus. The present study was initiated to study the distribution of MMTV in various Asean wild mice such as M.m. musculus sub-Jyg, MOM, M.m. cas-Tch or M.m. cas-Bgr. DNAs from various organs were digested by restriction enzymes, such as Eco-RI, Bgl-II. Integretation of MMTV genomes were analyzed by Southern blot hybridization. Hybridization probes were MMTV-LTR, env, gag-pol, int-lc, 2c, 2j and int-3. In M.m. musculus sub-Jyg, two lines of mice which had endogenous MMTV or lacked endogenous MMTV were segragated. MOM showed no MMTV genomes in any organs. M.m. castaneus showed 1-2 copies. In mammary tumors in M. m. musclus sub-Jyg, rearrangement was found in 1/4 in int-lc, 2c and int-3 legions. However, no rearrangement was found in the mammary tumor in MOM. These results provide support for the involvement of the cellular int loci in mammary tumor development. But, there are clearly many other genes which may be involved in these processes.

140 133

Abstracts Z ~ R A L ~ N O N E , ITS ROLE IN BREAST AND S~X ORGANS ~JMORS. R.Schsental,Department of Pathology,Roy~l VeteTi~ary College,University of London,London N ~ 0 ~ , England. Zearalenone, ~he secondary m~tabolite of Fusarium Kraminearum (Gibberella zeae)and of certain ether FUSARIUM spp. of soil miorofungi, has first been found In maize ~ (indian corn), that caused in livestock genital organs abnormalities. Eearalenone has significant estrogenic activity in animals and humans. Rats exposed perinatally to a few,relatively large~ doses of zearalenene (males and females) developed tumors of ~he sex organs (Schoental,Ad~.Cancer R~s.,

134

24K Monoclonal antibody (MAb) identifies an antigen that is oestrogen regulated. We have studied the expression of 24K MAB in benign breast tissue in 2 groups of women to determine whether oestrogen effects this regulation. Firstly, those taking the oral contraceptive pill for greater than one year for family spacing and secondly, those taking the 'Pill' for less than I year. Sixty-one women had taken the 'Pill' one year or less and seventeen for more than one year. Of the sixty-one short term 'Pill' users thirty-seven exhibited apocrine change in their biopsies and thirty-three stained strongly in the cytoplasm. In the long term 'Pill' use group eleven contained apocrine change and only six stained strongly in the cytoplasm which is statistically significant (p > 0.05). No difference was shown in the ducto-lohular tissue between the two groups.

45, 2i7,i~85). Sporadically high levels of z~aralenone hays been found in cereals and ~n v~rious other agricultural produo~s harvested during wet weather, and/or stored inappropriately. Yet,Japanese researcheTs did not find zearalenone-producing strains of Fusarium in th~ Japanese rice and rice paddies,"although ric~ is an appropriate substrata for zearalenone production (lshii et al.,Appl.Microbiol.27,625,1974). Could the absence of zearalenone in Japanese rice be a factor of the low incidence of cancers of the breast and of genital organs in the Japanese men and w a m e n ? This possibility deserves serious consideea~ien. The various lesions and tumors found in rate exposed perinatally to zearalenone ~ l l be presented.

135

LACTOGENIC HORMONES IN BREAST CANCER. P R Maddox, D Jones, R E Mansel, University Department of Surgery, University of Wales College of Medicine, Cardiff CF4 4XN, U.K. The role of prolactin (PRL) and to a lesser extent growth hormone (GH) in the development and growth of mammary tumours in rodents is well established, but much controversy remains about the involvement of lactogenic hormones in human breast cancer. However, the radioimmunoassay (RIA) levels reported may not necessarily reflect the biological activity of the hormones. A sensitive bioassay (BA), (Tanaka et al 1980 J Clin.Endo. Metab: 51; 1058-63), measures the mitogenic activity of lactogenic hormones for the rat Noble (Nb) 2 lymphoma cell line. We have modified the assay replacing the laborious cell counting technique by a spectrophotometric method which uses light scattering to measure cell growth. The total lactogenic hormone activity (TLH) in serum was determined by both BA and RIA in 31 patients with Breast cancer. Basal levels and peak responses to thyrotrophin releasing hormone (TRH) were evaluated. Mean TLH (GH + PRL) by BA was found to be significantly greater than by RIA for both Basal (13.0 ng/ml vrs 5.1 ng/ml p < 0.0001) and TRH-stimulated (74.1 ng/ml vrs 38.2 ng/ml p < 0.0001), samples, indicating that non-immunogenic forms of the hormones (or serum factors) may be present, causing increased biological activity. These early results suggest that previous attempts to assess the importance of lactogenic hormones by RIA in breast cancer may have failed to measure the biologically important lactogenic activity.

~4KMONOCLONAL ANTIBODY EXPRESSION IN APOCRINE METAPLASIA Ig SUI~PRESSED BY LONG TERM ORAL CONTRACEPTIVE USE. *g P Courtney, Susan Williams, R E Mansel. Departments of Surgery and Pathology, University of Wales College of Medicine, Cardiff CF4 4XN, U.K.

It appears that long term pill use has an inhibitory effect on the expression of 24K in the cytoplasm of apocrine tissue and would support the epidemiological evidence that long term pill use has a suppressive effect on breast tissue.

136

ANDROGENS A N D B R E A S T C A N C E R IN P R E M E N O R A U S A L WOMEN. P Toniolo~,G Seoretn,P Pis~ni,C Recchione,A Cavalleri. and F B e r r i n o , Istituto Nazionale T u m o r i , M i l a n o , I t a l y , and N e w Y o r k U n i v e r s i t y M e d i c a l C e n t e r , N e w Y o r k , N.Y. A case-control study was conducted to i n v e s t i g a t e t h e r o l e of a n d r o g e n s in p r o - m e n o p a u s a l breast c a n c e r . C a s e s w e r e 63 b r e a s t a d e n o c a r c i n o m a s and c o n t r o l s 70 h e a l t h y w o m e n o f s i m i l a r age. Laboratory a n a l y s e s of s e r u m t e s t o s t e r o n e iTS), dihydrotestosterone (DHT), a n d r o s t e n e d i o n e (ADIONE), dehydroepiandrosterone sulphate (DHEAS), progesterone (P), s e x - h o r m o n e - b i n d i n g globulin binding capacity ( S H B G ) , and of u r i n a r y testosterone (TU) and a n d r o s t a n e d i o l (ADIOL) were performed in s a m p l e s o b t a i n e d in the m i d - l u t e a l p h a s e of t h e m e n s t r u a l cycle. The most remarkable finding was a strong association of b r e a s t c a n c e r w i t h h i g h l e v e l s of TS and, a l t h o u g h l e s s s t r o n g ly, w i t h TU and DHT. T h e r e l a t i v e r i s k s (RRi w e r e 3 . 4 ( 1 . 6 - ? . 3 ) , 2.1 ( 0 . 9 - 4 . 8 ) , and 2 . 5 ( 1 . 1 - 5 . 9 ) , respectively. No a s s o c i a t i o n w a s a p p a r e n t for other androgens, for P or for S N B G . W h e n t i m e to the n e x t r a t h e r t h e n t i m e s i n c e the p r e v i o u s cycle was considered, t h e r e w e r e no e v i d e n t a s s o ciations in w o m e n w h o s e b l o o d h a d b e e n d r a w n w i t h i n 4 d a y s of m e n s t r u a t i o n . T h e R R s for w o m e n d o n a t i n g 5 to 9 d a y s b e f o r e m e n s t r u a t i o n (midl u t e a ! p h a s e ) w e r e s i m i l a r to t h o s e o b s e r v e d in the t o t a l p o p u l a t i o n . For women donating 10 or m o r e d a y s p r i o r to m e n s t r u a t i o n , the RR for TB i n c r e a s e d to 3 B . l ( 6 . 8 - 1 5 3 . 0 1 , for TU to 1 4 . 5 (2.0-27.0), and f o r D H T to 1 4 . 7 ( 1 . 8 - 1 2 2 . 5 1 . T h e s e d a t a s u g g e s t that i n c r e a s e d a n d r o g e n i c act i v i t y m a y p l a y a r o l e in b r e a s t c a n c e r .

Abstracts 137

PROLACTINAS A PREDICTOR OF BREAST CANCF~RRISK. S Martino*, R Gala, F Valeriote, P Kim, L Heilbrun, C Kresge, A Boomer, Wayne State University/Harper-Grace Hospital, Detroit, Michigan 48201, Wesley Medical Center, Wichita, Kansas 67214. Clinical strategies directed at breast cancer prevention w o u l d b e greatly facilitated if one could identify laboratory parameters that correlate with breast cancer risk and subsequent cancer develo~nent. Sertm~prolactinmaybe such a marker, and is being evaluated serially in a group of women who are at high risk for developing breast cancer. Thus far, Ii0 high risk w a m e n h a v e been enrolled in a breast cancer prevention program. Criteria for entry are I) 18 years of age or older, 2) a first degree relative with breast cancer, 3) a prior breast biopsy demonstrating epithelial hyperplasia or atypia, and 4) P2 or D y W o l f e mammogram pattern. The wcmen are randomized to a standard diet (SD) versus a 15% fat diet (LFD). Blood samples are obtained at times zero, 3,6,9,12,18, and 24 months. Sanloles are assayed for sertm~prolactin, growth hormone and bioactive lactogenic hormones. Levels are determined by both radioinlmlno-assay (R/A) and t h e N b 2 lynlohcraa cell assay (BA). Preliminary data is available on 67 women at time of entry into the diet study. Prolactin levels range from-7-124 ng/ml by R/A, and 5-179 ng/ml by BA. Nine women (5 on the SD and 4 on the LFD) have clearly elevated levels by both assays 0 2 5 ng/ml). All women will be followed to determine subsequent hormone levels, and their correlation with diet and breast cancer risk.

139

COMPARATIVE ANALYSIS OF GROWTH FACTOR-EXTRACTION PROCEDURES IN BREAST CANCER BIOPSIES. V. Cappelletti, P. Miodini, D. Coradini, G. Granata and G. Di Fronzo. Istituto Nazionale Tumori, 20133 Milan, Italy.

Proliferation of breast cancer lines is influenced by growth factors. It is, therefore likely that growth factors should play a role also in breast solid tumors. 0nly a few results have been reported in biopsies from solid tumors and data are still controversial. The main reason for the lack of consistent data on growth factors in human tumors is the difficulty in obtaining high yield-growth factors from neoplastic tissue due to the action of proteolytic activities and to segregation by binding proteins. In the present study we carried-on a careful comparative analysis of different procedures for growth factor extraction from human breast cancer biopsies. Our work was mainly focused on EGF, TGFa and other peptides able to interact with the EGF receptor. The isolation procedure for transforming growth factors based on acid/ethanol extraction originally proposed by G.J. Todaro (PNAS 77, 3494-3498, 1980) was slightly modified and compared to an acetic acid/acetone extraction protocol. Direct assay on cytosol treated by acid/ethanol was also performed. Growth factor preparations, obtained by using the different approaches were submitted to qualitative and quantitative determinations by radioreceptor assay and radioimmunoassay. Detailed results will be illustrated and practical implications will be discussed. Supported in part by a grant PF0 no. 87.02813.44

141

138

Growth factor requir~nents of normal h~nan msmmary epithelial cells in vitro. S.P. Ethier and K.C. Cundiff. The Michigan Cancer Foundo+ion, Detroit, RI 48201. i** previous work, we have characterized the growth factor requirements of normal rat fnammary epithelial cells and subsequently identified altered growth factor requirements of cells f r ~ n p r i m a r y m a n m ~ r y carcinomas. The present studies were aimed at characterizing the growth factor requirements of normal hlmmm mammary epithelial (}~E) cells in serlm~free culture. HME cells were obtained by enzymatic dissociation of reduction mammoplasty specimens. Cells were grown in Hepes-buffered Ham's F-12 medium supplemented with transferrin, ethanolamine, insulin (I), hydrocortisone, epidermal growth factor (EGF), cholera toxin (CT), prolactin, progesterone, triiodothyronine, sodium selenite and bovine serum albumin. Examination of growth factor requirementswas performed by deletion of individual factors free the medium. Deletion of I, EGF or CT did not significantly affect }]ME cell growth in primary culture. Therefore, the effect of growth factor deletion on long-term growth was examined. HME cells grown in the continuous absence of I underwent the same number of population doublings before senescence as cells grown in cempletemedium. Primary culture of H M E e e l I s in complete medium prior to growth in I - f r e e m e d i u m y i e l d e d a subpOpulation of cells that underwent 30 population doublings in the absence of I. By contrast, deletion of DGF bad dramatic effects on }IME cell growth resulting in complete growth cessation within 5 to 6 doublings of factor removal. Deletion of CT had intermediate effects as HME cells underwent 7 to 9 doublings before growth cessation. These results indicate that ~4E cells require ~F,F for long term proliferation. These cells have only a marginal requirement for CT and at least a subpopulation of cells do not require I for long term growth.

140

CHROMOSOMALASSIGNMENTS OF HUMAN 27 kDa HEAT SHOCK PROTEIN (hsp27) RELATED SEQUENCES. *S E McGuire, S A W Fuqua, S L Naylor, D A Helin-Davis, W L McGuire, The University of Texas Health Science Center, San Antonio, Texas 78284. We have previously demonstrated an estrogen-regulated protein, termed 24K, in human breast cancer cell lines and human tumor biopsies. We have cloned a p a r t i a l cDNA to 24K and have determined that 24K is identical to human heat shock protein hsp27. Therefore, hsp27 is dually regulated by both estrogen and heat shock, and represents a valuable m o d e l system to study t r a n s c r i p t i o n a l regulation. We have determined the chromosomal locations of hsp27 sequences by somatic cell hybrid analysis. There are at least three related hsp27 sequences on chromosomes 3, 9, and X. These chromosome assignments are i n t e r e s t i n g due to the reported karyotype abnormalities of chromosomes 3 and 9 in several human breast tumor cell lines. We are c u r r e n t l y examining the genomic organization of hsp27 sequences in human breast tumor biopsies for hsp27 polymorphisms. Conclusion: Human hsp27 related sequences are located on chromosomes 3, 9 and X as determined by somatic cell hybrid analysis.

142 141

143

Abstracts

THE CORRELATION OF KI-67 ANTIBODY DETERMINED GROWTH FRACTIONS OF MAMMARY CARSINOMA WITH ESTROGEN RECEPTOR, TNM AND MENOPAUSAL STATUS. *K v. S m i t t e n , S N o r d l i n g . The Fourth Department of S u r g e r y , H e l s i n k i U n i v e r s i t y Central Hospital and the Department of Pathology, University of Helsinki, SF-00130 Helsinki Finland. Immunohistological staining with the monoc l o n a l a n t i b o d y K i - 6 7 h a s b e e n s u g g e s t e d as a s i m p l e m e t h o d to a s s e s s t h e g r a d e of m a l i g n a n c y of b r e a s t c a n c e r , a n d e a s i e r to p e r f o r m t h a n t h y midine labelling index and flow cytemetry. We therefore evaluated the correlation of K i - 6 7 l a belling with features known to affect the outcome o f b r e a s t c a n c e r , i.e. T N M a n d e s t r o g e n r e c e p t o r status. F r o z e n s e c t i o n s of 30 i n v a s i v e b r e a s t c a r s i nomas were simultaneously immunostained for Ki-67 and strogen receptor (Eri). In 18 p a t i e n t s t h e cytosol estrogen receptor ( Erc ) was also assayed. Values exceeding 10 f m o l / m g c y t o s o l p r o tein were considered positive. Ki-67 positivity was correlated to Eri, Erc, T N M a n d m e n o p a u s a l status. F i f t y % of t h e c a r s i n o m a s were No and Ki-67 positive. Eri was positive in 53 %, w h i l e E r c w a s positive in 72 %. T h e r e w a s n o c o r r e l a t i o n betw e e n K i - 6 q s t a i n i n g a n d TNM, Er, as w e l l as m e n o pausal status. Our somewhat discouraging data express the n e e d of l o n g - t e r m s t u d i e s to e v a l u a t e t h e p r o g n o s t i c v a l u e of t h i s m e t h o d b e f o r e it w i l l b e routinely performed.

PERCUTANEOUS ELECTRICAL TREATMENT OF LARGE MALIGNANT TUMORS IN MICE. *D M Morris and A A Mar±no, Departments of Surgery and Orthopaedic Surgery, LSUMC, Shreveport, La. 71130. Several reports in the literature describe the use of electrical energy to treat tumors. We are trying to develop a useful and practical system for using electrical current to destroy tumor tissue. Lewis lung carcinoma was implanted on the back of C57/BL6 mice 1.5 cm caudal to the base of the skull after induction of anesthesia. Animals were treated when the tumors were at least one cm minimum diameter (12 days post-implant). All animals were treated with lidocaine prior to treatment with Percutaneous Electric Treatment (PET). Lidocaine was necessary to prevent cardiac arythmias in the model. Animals were randomized into four groups. Group 1-one PET treatment, Group 11-2-3 PET treatments, Group Ill-complete excision of the tumor with closure of the skin, Group IV-untreated controls. Data are shown in Table I: Table I Survival Time and Primary Tumor Weight In Mice Treated With PET Survival Tumor Treatment N Time (days) Weight (gms) GI - PETI 18 *23.5±5.0 5.3±2.5 GII- PET 2&3 26 *25.2±7.7 *4.8±2.8 Gill- Surgery 16a *37.5±9.0 *4.5±3.7 GIV - Control 16 15.6±4.3 7.0i3.7 a, 2 survivors *p 0.05 (two-tailed t test) Our data shows that PET produces a therapeutic effect in this animal model and that PET is not as beneficial as surgery. The most direct path to clinical utility for PET is as an adjuvant treatment with standard therapy such as surgery, radiation, or chemotherapy.

142

144

DISTRIBUTION OF EPITHELIAL AND MYOEPITHELIAL CELLS OF HUMAN MAMMARY GLAND IN BREAST CANCER. *S Hiroishi, S Imai, Y Taniguchi, Y Hiasa, Shiraimatsu Shinyaku Co., Ltd., Shiga 528, Japan, Nara Medical university, Nara 634, Japan. We have prepared two monoclonal antibodies which is reactive with epithelial cells (epi-l) and myoepithelial cells (myo-l) in the human mantmary glands. In breast cancer, epi-i reacted with carcinoma cells, while myo-i reacted specifically with stromal cells. These stromal cells including the extracellular matrix may play an important role for tumor development and progression. Therefore, we tried to clarify the precise characteristics of the stromal cells by use of myo-i antibody iramunohistochemically or electronmicroscopically. Breast cancer tissues were obtained from Nara National Hospital. They were fixed with glutalaldehyde and stained by various specific staining (Elastica-Van-Gieson, Azan-Mallory, Victoria-blue stainings). In carcinoma cells, connective tissue and collagen fibers proliferated mainly in the stroma. Elastic fibers or reticulin fibers, however I were not proliferated. Immunohistochemically, myo-i reacted with stromal cells, which is suggestive to be mainly myofibroblasts by electronmicroscopy. We are now investigating to identify the myo-i positive cells precisely in the stroma of breast cancer by colloidal-gold electronmicroscopy.

CHARACTERISTICS OF BREASTCARCINOMA IN THE ELDERLY. A. Cerrotta, B. Corradini, C. Clemente and E. Galante.* I s t i t u t o Nazionale Tumori - 20133 Milano A serie of 149 wide ]umpectomies f o r breast cancer in elderly patients gave us the opportunity to study i t s characteristics. The phatological volume was 2 cm. in 101 cases (68%), from 2.1 to 3 cm. in 34 cases (22.7%) and 3cm. in 12 cases. The i n f i l t r a t i n g duct carcinoma accounted f o r 66.4% of the case material (99 cases), but 30 cases (20.1%° were invasive lobular carcinomas and the other h i s t o l o g i c a l v a r i e t i e s accounted f o r the remaining 13.5% (9 mucinous carcinomas, 8 p a p i l l a r y carcinomas, 3 medullary carcinomas). The Peritumoral lymphatic i n f i l t r a t i o n (PLI) was negative in 86 (91.5%) of 94 studied cases and the necrosis was absent in 65 (95.6%) of 68 examined cases. The oestrogen receptors (ER) were positive in 102 (85%) of 120 studied cases and the progesterone receptors (PgR) were positive in 71 (67%) of 102 studied cases. Duct i n f i l t r a t i n g carcinomas were ER+ in 86,4 and PgR+ in 70% of cases; the i n v a s i v e l o b u l a r carcinomas were ER+ in 84,6% and PgR+ in the 66.6% of cases. We conclude that in our experience breast cancer in e l d e r l y is characterized by a percentage of invasive Iobular carcinomas higher than in the whole population and by a low percentage of PLI and necrosis as in the well d i f f e r e n t i a t e d carcinomas. In contrast the hormone receptor status of breast cancer in e l d e r l y is similar to that of the whole population.

Abstracts 145

PATTERNS OF TREATMENT IN ELDERLY PATIENTS WITH BREAST CARCINOMA D6.~ Crivellari*, E. ~alligioni, U. Tirelli, Foladore, Conte, O Nascimben, M. Amichetti, ~" Spagnolli, E. Recaldin, S. Monfardinl. G OCCNE gruppo Oncologico Clinico Cooperativo del Nord Est, 3 3 0 8 1 A w a n o , Italy.

146

Since little information is available in Europe on treatment of elderly pts with breast carcinoma, a retrospective analysis of pts aged 70 years or more referred to our Institute and to 6 parteeipating Centers in 1985 was performed One hundred and three pts are presently evaluable (i male and 102 females) with median age of 75 years (70-93) who account for the 16.8% breast carcinomas observed in 1985. Sixty-three (61%) pts presented operable cancers [32 stage I and 31 stage If), 35 (34%) locally advanced cancer and 5 patlents (5%) metastatic disease (bone 3, lung 1 and soft tissue 1 . Forty-one pts (40%) presented associated comorbid conditions. Among the 63 operable cases, only 2 were not surgically resected because of respiratory ~ailure. Postsurgical chemotherapy (CMF) was used only in 1 out of 31 stage II pts, while ~he remaining Dts were treated with tamoxifen (24)(80%) or orchiectomy ~i male pt); 5 Dts had n o treatment at all. Among the 35 pts wlth locally advanced disease, first therapv consisted of surgery followed by hormonal therapy in 21 pts (and radiotherapy in 9), hormonal therapy alone (7 pts) or followed 5y surgery (2 pts), radiotherapy alone ]3 pts), chemotherapy (2 pts). 0nly one pt was excluded from surgery for concomitant illness. Almost all metastatic pts were treated with hormonal therapy, only one with FEC regimen. Conclusions: i) In our patients' population, advanced disease seems more frequent than lhat reported in equivalent case series in American literature; 2) In spite of old age, adjuvant therapy with tamoxifen was not differenf from fhat of younger postmenopausal D ts. 3) In locally advanced disease most pts were treated with a combined approach of local modality plus tamoxifen which optimal sequence needs to be befter defined.

147

AUGMENTATION MAMMOPLASTY WITH SILICONE-GEL-FILLED IMPLANTS: ITS EFFECT ON BREAST C A N C E R PROGNOSIS. M 3 Silverstein ~, Neal Handel, P G a m a g a m i , 3 R Waisman, R 3 Rosser, W Colburn. The B r e a s t C e n t e r , 14624 Sherman Way, Van Nuys, California 91405. Breast a u g m e n t a t i o n is now one of the most frequently p e r f o r m e d plastic surgical procedures in the United States. Epidemiologic studies r e v e a l t h a t silicone-gel implants do not cause c a n c e r . They are, however, radio-opaque) obscuring as much as 80% of b r e a s t tissue during m a m m o g r a p h y . In addition, they c a u s e various d e g r e e s of capsular scar as well as com= pression ol surrounding b r e a s t p a r e n c h y m a , adding to the difficulty of m a m m o g r a p h i c i n t e r p r e t a t i o n . F r o m March 1981 thru February 1988, 27 p a t i e n t s with prior a u g m e n t a t i o n rnammoplasty were diagnosed and t r e a t e d for breast c a r c i n o m a . All p a t i e n t s had unilateral i n f i l t r a t i n g c a r c i n o m a s . None of the c a n c e r s were occult (discovered m a m m o g r a p h i c a l l y ) . All p r e s e n t e d as palpable masses. F i f t e e n of 27 (56%) had m e t a s t a s e s to axillary lymph nodes. During the s a m e t i m e period The Breast C e n t e r t r e a t e d 901 nonaugmented b r e a s t c a n c e r p a t i e n t s : 237 (26%) had in situ lesions; 230 c a n c e r s (24%) w e r e occultl 238 (26%) had involved axillary nodes. A u g m e n t a t i o n m a m m o p l a s t y d r a m a t i c a l l y reduces the ability of m a m m o g r a p h y (our best diagnostic tool) to visualize the breast p a r e n c h y m a . In addition) it c a u s e s various d e g r e e s of capsular scar which m a y reduce t h e physicians' ability to palp a t e c a n c e r . When c o m p a r e d with our own b r e a s t c a n c e r population, a u g m e n t a t i o n p a t i e n t s with b r e a s t c a n c e r present with more a d v a n c e d disease. They h a v e a higher p e r c e n t a g e of both invasive lesions and positive nodes, resulting in a worsened prognosis.

148

143

BODY FAT DISTRIBUTION AND BREAST CANCER RISK, D.Sehapira,* N.Kumar, G.Lyman, C.Cox, H.Lee Moffitt Cancer Center and the University of South Florida College of Medicine and Surgery, Tampa, Florida 33612 We wished to determine whether a particular pattern of obesity placed women at an increased risk of developing breast cancer. Obesity in women can involve predominantly abdominal fat deposition (android) or deposition on the buttocks and thighs (gynecoid). Fat cells on the abdomen are larger and differ adrenergic receptors compared to fat cells on £he buttocks and thighs. Women with android obesity have significantly higher levels of serum testosterone than women with gynecoid obesity. Anthropometric parameters were determined in 200 patients with breast cancer and 444 age matched controls. The measurements were made using standard anatomical landmarks and the percentage of body fat was determined by the Durnin method. The measurements for cases and controls are outlined in Table I. Table I Breast Ca. Pts. Controls P-value (Median Values) (M.V,) Abdomen skin fold (mm) 42.6 39.8 0.02 Thigh skin fold (~n~) 39.5 42.0 0.02 Abdomen/thigh ratio 1.14 0.97 ~0.001 Waist circumference (cm) 32.6 30.2 <0.001 Hip circumference (cm) 39.4 39.1 0.05 Waist/hip ratio 0.82 0.77 <0.001 Percent body fat 40.5 38.7 <0.001 Weight (lh) 151.1 143.8 <0.003 Quatelet Index 25.9 24.7 ,0.003 The distribution of covariates in separate groups were compared utilizing the Wilcoxan Rank sum test. The association of covariates in the disease category were studied in a multivariate fashion utilizing a logistic regression model. Coefficient estimates were based on maximum likelihood. Conclusion Patients with breast cancer had significantly more body fat and abdominal obesity, when compared to controls. In this study android obesity represented a significant risk for breast cancer.

AUGMENTATION MAMMOPLASTY WITH SILICONE-GEL-FILLED IMPLANTS: ITS EFFECT ON MAMMOGRAPHY. M 3 Silverstein*, P G a m a g a m i , E D Gierson, 3 R Waisman, Neal Handel. The Breast C e n t e r , 1t~624 Sherman Way~ Van Nuys, California 91405. The m o s t f a v o r a b l e (occult) breast c a r c i n o m a s , those with 90 to 93% 5-year survival r a t e s , are currently found using film screen m a m m o g r a p h y and d i r e c t e d biopsy in a s y m p t o m a t i c women. We must be concerned about any e l e c t i v e surgical procedure t h a t d e c r e a s e s our ability to m a m m o g r a p h i c a l l y visualize breast tissue. Sixty women were studied before and after augmentation mammoplasty with low dose f i l m screen mammography. Postaugmentation mammograms were done in the standard fashion, with implant compression, as well as with a new technique which displaces the implant. Visualized breast tissue was measured using a transparent grid. Following subglandular augmentation (44 pts.), standard compression mammography revealed an average decrease in visualized breast tissue of 45%; when displacement mammography was used, the average decrease was 31%. Sixteen submuscular augmentation patients did better; standard compression mammography yielded a 20% decrease in visualized tissue; displacement mammography showed a slight 4% increase. Intramammary scarring yielded numerous postaugmentation mammograms with parenchymal distortions. Many mammograms revealed areas of compressed bandlike breast tissue in which it would he exceedingly difficult to see subtle architectural distortions or m i c r o c a l c i f i c a t i o n s . A u g m e n t a t i o n m a m m o p l a s t y reduces the ability of m a m m o g raphy~ our best diagnostic tool, to visualize b r e a s t tissue. State of the a r t m a m m o g r a p h y is not possible in most patients augm e n t e d with silicone-gel-filled implants. Should t h e s e p a t i e n t s develop breast c a n c e r it would likely be diagnosed in its palpable r a t h e r than occult f o r m .

144

Abstracts

149

T-CELL CHIMERISM IN LYMPHOTROPHICMAMMARYCANCERCELLS. Untae K1m, N. Kadohama, S. Srivastava and R. Penetrante, Roswell Park Memorial Institute, Buffalo, NY 14263 The most c r i t i c a l attribute of human breast cancer is its capacity to metastasize to regional lymph nodes and beyond, a property peculiar to lymphoid cells. However, most mammarytumors developed rapidly in laboratory animals lack such property, regardless how undifferentiated or fast growing they may be. I t was postulated, therefore, that the lymph node-homing property of human breast and other carcinoma cells might have been acquired by incorporating lymphocyte 9enomes through somatic hybridization with the host cells during immunological encounters over the l i f e time of patients. This supposition was tested by fusing the nonmetastasizing rat mammary tumor cell line NM-081 with syngeneic thymocytes in v i t r o , and evaluating their growth characteristics in vivo. The hybrid mammary tumor cells manifested not only lymph node-homing capacity generation after transplantation generation, but also expressed the thymocyte-specific enzyme, terminal deoxynucleotidy] transferase (TdT), various T-cell associated antigens and antigen-receptors immunocytochemically.,, With antibodies to TdT, the presence of TdT was further verified by immunoblotting analysis of proteins separated by SDS-PAGE. Thus, i t was concluded that the acquisition of lymphotrophicism by the rat tumor cells was achieved by the establishment of T-cell chimeras in them through somatic hybridization with immature T-lymphocytes. A similar T-cell chimerism was observed in the lymphogenously metastasizing rat mammary tumor SMT-2A, the cell lines MCF-7 and BT-20, and also in Stage-ll human breast cancer celts, while in Stage-I cancer cells some had the T-cell markers and others did not. The significance of these findings and possible mechanisms of the lymphotrophicism will be discussed. (Supported in part by American Cancer Society Grant PDT-306).

151

A PJ%K~)OMIZED TRIAL (]OMPARING 12 WEEKS WITH 36 ~ E K S OF ADJ~ CHEMOTHERAPY IN STAGE II BREAST CANCE~. M. Levine*, M. Gent, W. Hryniuk, L. Levin, H. Abu-Zahra, A. Arnold, J. Skillings, V. Bramwell, S. DePauw, for the Ontario Clinical Oncology Group, Hamilton, Ontario L8N 3Z5. Traditionally, adjuvant chemotherapy in wc~en with Stage II breast cancer has been administered for at least 6 months. A randomized trial has been performed in which Stage II patients were allocated to either a 12-week treatment ( C ~ + AT): cyclo~hospba/uide 80m~/m 2 p.o. daily x 8 weeks; methotrexate 35 m~j/m2 and 5 ~J 500 l!%g/m2, both I.V. weekly x 8; vincristine 1 mg/m2/week x 4 and then every 2 weeks x 4; prednisone 50 mg p.o. x i0 days, and then tapered; a d r i a ~ c i n 20 m~/m 2 weekly, weeks 9 through 12; and tamoxifen i0 r0~ p.o, b.i.d. throughout; or 36 weeks of (IMFVP (Cooper, Cancer 1979; 44:793-798). The median follow-up is currently 29 months. The disease-free survival of the 223 women randomized to the 12-week treatment is 58% c o ~ e d to 70% for the 216 women in the 36-week group, p = 0.03; the overall mortality rates are 18% and 13%, respectively. The number of patients with recurrences by subgroup (age, nodes) are: Subqroup 12-week treatment 36-week treatment Age(years) Nodes Patient Recurring Patient Recurring No. NO, < 50 1-3 56 34.9% 53 17% • 50 1-3 82 20.7% 77 23.4% -~ 50 > 4 34 52.9% 34 38.2% > 50 -~ 4 52 55.8% 52 42-3% • nese findings suggest that 12 weeks of adjuvant chemotherapy is insufficient for women with Stage II breast cancer.

150

152

ADJUVANT THERAPY OF POSTMENOPAUSAL WOMEN WITH BREAST CANCER - A N ECOG PHASE Ill STUDY. *H C Falkson, R Gray, W H Wolberg, K W Gilohrist and G Falkson, University of Pretoria, Pretoria, South Africa, Daca-Farber Cancer Institute, Boston, MA 02115 and Wisconsin Clinical Canner Center, Madison, WI 53792 A prospectively controlled randomized trial to compare the adjuvant efficacy of 12 cycles of oyclophosphamide, methotrexate, 5-fluorouranil, prednisone and tamoxifen (CMFPT) followed by observation or a total of 5 yrs of continuous tamoxifen (TAM) vs 4 cycles of CMFPT followed by observation in postmenopausal women with operable node positive breast cancer was started by ECOG in 1982. In 1986 the study was modified to allow patients (pts) on CMFPTz12 + continuous TAM to be randomized after completing 5 yrs of TAM to either continue for life or to stop therapy. Pts were stratified for number of involved nodes and ER status. Of the 962 pts on study, 832 were eligible. There were 283 on CMFPTx12 + continuous TAM; 279 on CMFPTx12; and 270 on CMFPTx4. Toxicity data are available for 823 pta; life-threatening toxicity occurred in 77 and lethal toxicity in 7 pts. Median follow-up is 3.1 yrs; 233 pts had recurrent disease. Time to relapse is significantly longer for pta on CMFPTx12 ÷ continuous TAM than for CMFPTx12 (P=0.02) and CMFPTx4 (P=0.02). Differences between 4 or 12 cycles of CMFPT are not significant. Relapse-free rates at 3 yrs are 76% for CMFPTx12 + continuous TAM; 71% on CMFPTx12 and 65% on CMFPTx~. Treatment differences in overall survival are not significant, pts alive at 3 yrs for CMFPTx12 + continuous TAM, 83%; CMFPTx12, 80%; CMFPTx4, 85%. Association of ER and number of involved nodes with time to relapse and survival are highly significant.

A MONOCLONAL ANTIBODY COCKTAIL FOR DETECTION OF MICROMETASTATIC TUMOR CELLS IN THE BONE MARROW OF STAGE IV BREAST CANCER PATIENTS. M Taha,* NG Ordofiez, SS Kulkarni, M Owen, CL Reading, G Hortobagyi, KA Dieke, and G Spitzer. U T M.D. Anderson Cancer Center, Houston, TX 77030. We investigated if monoclonal antibodies (MoAbs) reactive against both acidic and basic keratins alone were sufficient to use for detection of minimal numbers of contaminating breast tumor cells in bone marrow. Anticytokeratin MoAbs (AEI & AE3) were used for staining 14 cases of breast carcinomas. 9 cases (64.3%) showed high positivity and 5 cases (35.7%) were low to moderately reacting. A group of 9 breast MoAbs which proved to be unreaetive to bone marrow cells (using immunealkaline phosphatase histo-chemistry and light density marrow cells) were screened for reactivity to breast carcinomas showing low and moderate positivity only with keratin MoAb, 3 of 9 MoAbs showed a high percentage of positivity and were selected to supplement the immunohistochemical use of anticytokeratin antibody for detection of minimal marrow disease in breast cancer patients. The MoAb cocktail was further tested for reactivity to another 5 cases of breast carcinomas and compared to cytokeratin staining alone, The cocktail labeled 100% of carcinoma cells in all the examined cases. To determine the sensitivity of this panel to detect minimal numbers of contaminating tumor cells in bone marrow, in vitro mixing experiments were carried out. T47D breast carcinoma cells were mixed with mononuclear cells at a ratio up to 1 tumor cell in 1 x 10° marrow cells, and cytospin preparations were stained with the cocktail by immune-alkaline phosphatase method. This methodology could detect 1 tumor cell in 10s hematopoietic cells. Staining of bone marrow samples from stage IV breast cancer patients could demonstrate the presence of single alkaline phosphatase-positive tumor ceils in 18 of 46 patients tested (39%) which were negative by the routine examination of bilateral bone marrow biopsies and aspirates. The clinical significance in those patients is being evaluated.

Abstracts 153

155

RAPID DETECTIONOF HUMANBREASTCARCINOMAXENOGRAPHSUSING TECHNETI~M-ggM 323/A3 FAB' IN AN ATHYMICNUDEMOUSEMODEL K.Y. Pak , M.A. Nedelman, W.E. Fogler and W.L. McGuire and P.E. Daddona, Centocor, Malvern, PA 19355 and Univ. of Texas Health Science Ctr., San Antonio, TX (78284). The monoclonal antibody 323/A3 shows specificity for primary and metastatic human breast tumors by immunohistochemical techniques (Cancer Res. 46;1306, 1986). We investigated the effectiveness of technetium-ggM-labeled Fab' fragment (Tc-Fab') and indium-ill labeled F(ab')2DTPA coupled fragment (In-F(ab')2) of 323/A3 in imaging progressively growing MCF-7 breast carcinoma in athymic nude mice. In in vitro cell binding as~ays~ Tc-Fab' an~ In:F(ab') 2 exhibited-K a-values of I x 10° M"I and 5 x 10~ M - I - , respectively. Tumor bearing (0.274 gm ± 0.116) animals were injected IV with either the Tc-Fab' or InF(ab') 2 and imaged with a gamma camera equipped with a pinhole collimator. Additional animals were necropsied and evaluated for radiolabel biodistribution. Distinct tumor visualization was possible as early as I hour post IV administration with the Tc-Fab'. In contrast, at least 5 hours were required for unequivocal detection using the In-F(ab')2. Tumor-to-blood and tumor-to-liver ratios for % ID/G were 157% (p=O.O01) and 60% (p=O.1) higher for the Tc-Fab' relative to the In-F(ab')2 at 5 hours. The biological h a l f - l i f e for Ehe Tc-Fab' and In-F(ab') 2 in blood was 4.5 and g hours, respectively. In conclusion, both 323/A3 antibody fragments showed strong positive tumor localization in this breast carcinoma animal model. However, the Tc-Fab' showed earlier detection of tumor, better tumor-to-tissue ratios and a more rapid blood clearance relative to the In-F(ab') 2. These results suggest the Tc-Fab' may have potential clinical u t i l i t y for the rapid diagnostic imaging of breast carcinoma.

TFE PN~I~81S CF AFOO~N~ H~JkST CA~IN~JI~S. N.J. Rrdrr~ , Rosemary %~ll~r, Gilli~ ~'~nite, H~len Stewart, W.R. ~.,Eller. D~i:~rhT~nts of SLmgery enJ Pathology, Lhiv~aity of Leiceste~~.

DepBr~nt

of Clininal S~ry,

lhiversity of Riin~gh.

/~ecrine change in h~sst cancer is of urban ~r~rostic import. An ~ stain for ZincS. Gly~oprote'in,(l~n st~ocrine l~s bern used to mjectivery identify apoorine ohm]go ~lJ deterrr~ne ff t h e expression of a~esries differentiatio~ l~y tutors is of pmgnestic value.

I~P.sst c a l c ~ samples ~ in paraffin from 145 m m n v~3re tested. ~ n i m m fol!av up ~as five years. Sixty s i x (45%) of t m did not stain, 52 (36-°~) eedibited a!:~crine stairfing ~nilst 27 (19%) ~ r o a:laivesal. 31~ ~ of staining w s not affected 17/noJo status, rca~trual status, tunam grade or size, oentrogm r o ~ t ~ r (F~R) or l~ogesterene receptor status. Patients x ~ s o turrets eg~nibited ~!~crino staining had a significantiy shorter disease free interval (I~.04) and st~wival (t~0.05). 31~ ~ of al~crine staining (p=0.006) sdOaJ significantly to tl~ p r o g ~ t i c value of r ~ e status (i~.0002), nm'~truai status (p=0.0006), tumor size (~M).0008) and E2R status (1~0.0004) a C ~ ' s ProNortimate t-~zards ~l~el ~vas caioalated n n l t i ~r~riately. Nc~ of the ether s6wen factors tested ~ r e inclu£ed in ths ~ad~l. /~Fuerine differentiation in ~astcar~r ap~ irde~t predictor of ~oor imx~ec6is ~ .

N.J. Rmdred, 603 - l O ,

W.R. ~iller, R.A. ~al~.

to be an

Histo~athol ~987:

154

145

COMPARATIVE ANALYSIS OF PLOIDY AND CELL KINETICS OF PRIMARY BREAST CANCER AND LYMPH NODAL METASTASIS. G. Del Bino ~, M.G. Daidone, M.R. Zucconi, C. Selvi and R. Silvestrini. Istituto Nazionale Tumori, 20133 Milan, Italy Ploidy and cell kinetics have been defined for a series of 600 primary breast cancers and analyzed as functions of clinioo-pathologic features. DNA content was evaluated by a FACS IV in nuclei released from frozen tissue and stained with propidium iodide and cell kinetics were determined as 3H-thymidine labeling index (3H-thy LI). The overall frequency of aneuploid tumors was 71%. Aneuploidy was not related to presence or number of positive lymph nodes and only slightly related to pathologic tumor size, nor was 3H-thy LI. Diploid and aneuploid tumor subsets had similar median 3H-thy LI (3.2% vs 3.7%). Thus aneuploidy is a largely autonomous variable and could give independent prognostic information, in addition to 3H-thy LI and to the most important pathologic features. In a series of 58 patients, ploidy and 3H-thy LI were defined for both primary and synchronous lymph nodal metastases. Agreement of ploidy profiles and DNA indices of primary and metastatic lesions from the same patient was observed in 90% of eases. Similar median 3H-thy LI values were observed for primary or metastatic lesions (4.8% vs 5.4%) considered as a whole. However, this was true for aneuploid tumors (5.7% vs 4.7%) but not For diploid tumors, for which the 3H-thy LI of metastases was three times that of primary tumors (10.1% vs 3.4%, p=O.Ol). Supported in part by a grant PFO 86.02813~44

156

AUTOCRINE FUNCTION OF BREAST CARCINOMA IN APOCRINE TISSUE IDENTIFIED BY 323A3 MONOCLONAL ANTIBODY. *S P Courtney, Susan Williams, R E Mansel. Departments of Surgery and Pathology, University of Wales College of Medicine, Cardiff CF4 4XN, U.K. It has been shown that 323A3 monoclonal antibody (MAb) identifies 'high' risk breast pathology. We have compared 323A3 MAb staining of benign histological components from patients with benign breast biopsies with similar benign histological components from patients with breast carcinoma. 79 patients with breast carcinoma were compared with two different control groups. (50 patients with prospective benign breast biopsies and 77 patients with benign breast biopsies available to the department from other studies). Staining in the cytoplasm of ductolobular tissue and apocrine metaplasia was assessed as was basal membrane staining in apocrine tissue. 35 patients with breast carcinoma exhibited apocrine metaplasia in their biopsies, 20 in the prospective benign controls and 42 in the departmental benign group. Strong cytoplasmic staining in apocrine tissue occurred in 18 of the 35 benign tissue adjacent to breast carcinoma whereas in the departmental controls this was 2 of 42. This difference was statistically signifant (p < 0.001). Similarly, the prospective group stained strongly in i of 20 cases. This was again statistically significant when compared to the breast carcinoma group (p < 0.005). The increased expression of the 323A3 MAb in the cytoplasm of apocrine metaplasia adjacent to malignant breast tissue suggests the tumour has an autocrine function on apocrine metaplasia. Should these appearances pre-date overt malignancy then this monoclonal antibody may be a useful indicator of pre-clinical cancer.

146 157

159

Abstracts EFFECTS OF FIBRINOLYTIC AGENTS ON EXPERIMENTAL TDMOR METASTASIS. M.S.Murthy, L.J.Summaria, E.F.Scanlon. Evanston Hospital, 2650 Ridge Avenue, Evanston, IL 60201. TA3Ha mammary carcinoma cells metastasize to the lungs and lymph nodes~ but rarely to the liver or kidney of syngeneic Strain A mice (J.Snrg. Oncol. 35: 44, 1987; Cancer Res. 47: 774, 1987). Partial nephrectomy or hepatic wedge resection, by the milliwatt CO 2 laser, followed immediately by i.v. injection o f TA3Ha cells resulted in the development of 28/58 (48%) and 21/53 (40%) tumors at the site of trauma. Injection of tumor cells after 5 or more days of surgery reduced the frequency of tumor formation at surgical site (Breast Cancer Res. Treat. I0: 93, 1987). Thus, the earlier events in wound healing appear to favor tumor implantation better than t h e later ones. Since fibrin deposition is an important early event in wound healing, we have investigated to s e e if tumor formation can be retarded or prevented by fibrinolytic agents. Strain A mice subjected to partial nephrectomy or liver wedge resection by electrocauterywere injected i.v. with 300 units of B-chain-streptokinase complex (B-SK) or saline followed immediately by i.v. injection of TA3Ha cells. Fourteen days later, these mice were sacrificed and autopsied. I n the c o n t r o l g r o u p , 9 / 1 2 (75%) a n d 8/13 (62%) m i c e showed tumors at the s i t e of renal and hepatic injury, respectively. I n t h e B-SK g r o u p , 1/8 (13%) a n d 3/19 (16%) mice showed tumors at these respective sites. When B-SK was administered to non-surgical mice immediately before, or 1,2, o r 3 days after i.v. injection of TA3Ha tumor cells, 3/10, 4/10, 7/10, and 6/10 mice developed tumors in the lungs compared to 8/10 in the controls. Thus, B-SK appears to affect the implantation of tumor cells at both surgical and non-surgical sites. These and the results obtained with other fibrinolytic agents (streptokinase, urokinase, and tissue plasminogen activator) on tumor implantation and growth will be discussed.

HYPERTHERMIC POTENTIATION OF MACROPHAGE/TNF TUMOR CYTOTOXICITY. J. Klostetgaard*, M. Barta, and S. Tomasovic, M. D. Anderson Cancer Center, Houston, TX 77030. The cytotoxic activities of macrophages (MO) and TNF against tumor cells were modulated by h y p e r t h e r m i a and displayed striking treatment sequence dependencies. B COactivated MO showed pronounced augmentation of their ability to secrete TNF, if endotoxin-triggering preceded 1 hr heating at 42 ° or 43°C by 1.5 to 4 hr compared to when the same heating was concomittant with or preceded triggering by a similar interval. Qualitatively similar observations were made when 24 hr heating at 39 ° or 40.5°C preceded or was eoneomittant with t r i g g e r i n g by a s i m i l a r interval. M~3 coeuhured with various tumor cells, when e n d o t o x i n - t r i g g e r e d prior to heating, exhibited retained or enhanced function, whereas heating simultaneously with t r i g g e r i n g significantly i n h i b i t e d effector cell cytotoxicity. Profound treatment sequence dependencies were also seen when addition of TNF to tumor cells was varied about the heat treatment. Heat-induced sensitization to TNF of TNF-responsive L929 and TNF-resistant EMT-6 tumor cells was up to 10,000-fold i f monokinc administration preceded 1 hr heating at 43°C, whereas the reverse sequence markedly reduced sensitization. Both tumor cell types were also sensitized to TNF if treatment preceded 2 4 h r heating at 40.5°C. These results support the h y p o t h e s i s that appropriately constructed sequences for MO p r i m i n g / triggering or TNF treatment of tumor ceils combined with hyperthermia can significantly potentiate the cytotoxic interactions between effector cells or exogenously added TNF and their tumor cell targets. (ACS IM-419 to JK and USPHS CA-32745 m SIT).

158

TUMORIGENIC AND METASTATIC BEHAVIOR OF A HUMAN,BREAST CARCINOMA CELL LINE IN NUDE MICE. J.E. Price , L.M. Daniels and R.D. Zhang. The University of Texas, M.D. Anderson Cancer Center, Department of Cell Biology, Houston, TX 77030. There are few reports in the literature of experimental models of the growth and metastasis of human breast carcinomas in athymic mice. To establish such a model we concentrated i n i t i a l l y on estrogen receptornegative cell lines. Nhether the site of tumor cell injection influenced tumorigenic and metastatic behavior was tested by comparing growth of cells injected into the mammary f a t pad (m.f.p.) or s.c. in 6-to-8-week old female nude mice. For one cell l i n e , MD~-MB-435, we found a 100~ tumor take at a low cell dose (i0 J c e l l s ) at the mfp s i t e , compared with the same dose injected s.c., which produced tumors in only 20% of mice. The trophic e f f e c t of the orthotopic site (vs "ectopic" s.c. site) appeared to be specific f o r the breast carcinoma cell l i n e , as three other human tumor cell lines, from colon carcinoma, renal cell carcinoma and melanoma, did not show enhanced tumor growth when injected in the m.f.p, compared with s.c. injections. The MOA-MB-435 cells implanted in the m.f.p, grew as 1ocally invasive tumors which disseminated to numerous sites, including regional and distant lymph nodes, lungs (in 80-I00~ of experimental animals) and less frequently to skeletal muscle, heart and brain. Cells isolated from lung, brain, and lymph node metastases were established in tissue culture, and reinoculated into nude mice to test whether these new variants represent sub-populations of breast carcinoma cells metastatic to specific organ sites. We propose that this model is suitable for experimental studies of the course of advanced metastatic breast carcinoma. (Supported by a grant from Triton Bioseiences).

160

MOLECULAR DETERMINANTS OF THE ANTIPROLIFERATIVE EFFECT OF INTERFERON-=~ (IFN-~G) AND INTERFERON~ ~ (IFN- ~ ) IN HI~Xq BREAST CARCINOMA. Raj K Tiwari, H Kurihara, N Telang, TB Hakes, MP Osborne. Breast Cancer Research Laboratories, Memorial Sloan-Kettering Cancer Center, New York, NY 10021 The antiproliferative effect of i n t e r f e r o n ~ a n d (500 U/ml) was examined in a human breast carcinoma cell line MCF-7 using the following end points (a) cell number; (b) anchorage independent (AI) growth in soft agar; (c) expression of a panel of interferon inducible genes 2A(HLA), 561, i-8, and 6-16. A five-day exposure to I F N - ~ and IFN- ~ resulted in a 37% and 48% decrease in cell number respectively. The decrease in cell number was dosedependent. I n t e r f e r o n s ~ a n d ~ inhibited AI growth of O MCF-7 in soft agar. A 14-day exposure of I F N - ~ a n d IFN-Y reduced the colony forming efficiency (CFE) by 51% and 64% respectively. MCF-7 was found to be highly responsive with respect to the expression of interferon inducible genes. The expression of these genes were measured by mRNA analyses using specific cDNA probes. All of the genes (2A, i-8, 561, 6-16) were induced by I F N - ~ a l o n e and in the presence of a protein synthesis inhibitor cycloheximide indicating that the expression of these genes is a primary effect of IFN treatment. The kinetics of expression of these genes reveals significant levels of their mRNAs after 48 hours of IFN exposure. The results show that the induction of specific interferon-inducible genes are the initial events which precede the growth inhibitory effect of these cytokines and bring about the non-proliferative state of the cell.

Abstracts 161

REGULATORY AUTOANTIBODIES AND ANTI-IDIOTYPIC IMMUNE RESPONSES IN MAMMARY TUMOR-BEARING RATS. Susan M. North*, Department of Tumor Biology. The University of Texas M. D. Anderson Cancer Center, Houston, TX 77030.

162

The objective of this study was to determine the effect of a s y n g e n e i c monoclonal antibody (MAb) MT10:21 on rat mammary tumor (MTLn3) metastasis. This MAb was generated by the fusion of splenocytes from an MTLn3 tumor-bearing rat with the Y3 Agl.2.3 myeloma. MTLn3 is a metastatic subclone of the rat 13762NF m a m m a r y adenocarcinoma. One i.v. injection of MAb MT10:21 (50 g g ) resulted in the detection of serum antibodies (ELISA) which react with a wide variety of normal tissue components, including cytoskeletaI p r o t e i n s (Immunohistochemistry) and rat immunoglobulins. These antibodies are therefore considered to be "autoantibodies". Concurrently, anti-Idiotypic (Id) antibodies with specificity for MAb MT10:21 were also induced in tumor-bearing rats (Ab2). The presence of autoantibodies in the serum i n f l u e n c e d the p r o d u c t i o n of anti-Id antibodies in vivo. Induction of serum antibodies after injection of M A b is modulated b y antigen, i.e., tumor. I f the rats have small tumors (< 1 cm in diameter), the h u m o r a l immune response is enhanced (increased titers of I g M autoantibodies), but i f the tumors exceeds 2 cm in diameter there is suppression of s e r u m antibody s u g g e s t i n g that injection of M A b influences a n ongoing immune response to the tumor. Using MAb MT10:21 as immunogen we have generated MAbs with auto- and anti-Id reactivity which have s i m i l a r s p e c i f i c i t i e s to the serum antibodies. Supported by NIH Institutional grant RR5511-23.

163

INTENSIVE INDUCTION CHEMOTHERAPY AND INTENSIFICATION WITH AUTOLOGOUS BONE MAI~OW REINFUSION IN PATIENTS WITH STAGE IIIB AND IV BREAST CANCER. P.H.B.Willemse*, E.G.E.de Vries, D.Th.Sleijfer, P.O.M.Mulder, C.Th.Smit Sibinga+, N.H.Mulder, Dept. of Internal Medicine, and +Regional Red Cross Bloodbank Groningen-Drenthe, Universify of Greningen, The Netherlands. Intensive induction chemotherapy in pts with surgically incurable breast cancer should lead to a high percentage of complete resp0nders, and intensification of chemotherapy after obtaining a complete response could lead to cure. This hypothesis is tested in premenopausal previously untreated pts (n=39) with stage IIIB (11=22) or IV (n=lT) breast cancer. Induction therapy consists of minimal 4, maximal 6 courses q 4 wks of MTX 1500 mg/m 2 and 5-FU 1500 mg/mZ day i, Adriamycin 50 mg/m2 and Vineristine i mg/m2 day 14, and prednison orally 60 mg day 2-14. If complete remission is reached, intensification therapy comprising cyclophosphamide 7 g/m2, VP 16-213 1.5 g/m 2 , and autologous bone marrow reinfusion is given. Finally, radiotherapy is applied to T4 lesions and pts are maintained on %amoxifen for 2 years thereafter. In 21 pts evaluable for response on the induction regimen, 7 partial remissions (33%) and 13 complete responses (62%) were seen (response rate 95%). Fifteen pts received intensification so far. No toxic deaths have occurred during induction or intensification. The predicted DFS in al] pis is 49% at 2 years, median duration since entry is now 18 months. This induction regimen yields a high complete response rate and intensification with ABMT is feasible after this regimen. In some pis a long term DFS is obtained.

164

147

~ (~{CWfHFACTC~ IN BREAST CYST FLU]]]S. ~S.Zanardi,G.Valenti,D.Cama ta,G.~ , T.Fassio,P.Bruzzi,°A.Barreca,gP,Del Fmnte,°F.Minute,F.Boccardo. Istituto Nazianale per la Ricerca sul Cancro,and °Cattedra di Fisiopatologia Endecrina UniversitA di Geneva, 16132 Geneva, Italy. GroSS cystiC disesse(GCD)of the breast has been associated with an in_ creased hresst cancer risk.A high risk has been reported especially for wo_ men bearing cysts lined by apocrine epithelium,which ccotain high potassium (K+)emd dehych-cepiendrostercnesulphate(D~AS)levels. In 95 breastcyst fluids aspiratedfrom 86 wc~n with C£D we masstmedthe levels of sodium(Na+),K+,Db~AS snd EpidermalGrowthFactor(E~) .which has been shownto be involvedin the control of normal and neoplastic cell growth.The Na+~+ ratio showed a clear-cut bimodal disthibutian,allowingto classify breast cysts in two distinct subpopulaticrs according to their matico ocntent(Na+/K+<3 or>S).Both D~FAS and Ff~ levelswere found to be inversely correlated with the Na+/K+ ratio, median ccncentratiens in the two cyst classes being as follows: Na+K+ <3 Na+/K+> 3 Statistical median(~) median(rsnge) significance u~dl 3615(178-2278) vs 480(30-7260) p< O.OOl F~ n~ml 103.26(1.90-433.15) vs 57.22(0.25-2~0.85) p< 0.(301 E~F was detectable in all cyst fluid tested.Considering that this 6F is al m o s t t~detectable in plasma,co the basis of the present results the hypothe sis can be made that E ~ found in cyst fluid is locally produced by the epi thelitm lining breast cysts. The Na+/K+< 3 cysts seem to he particulsrly active in this regard, a finding that could provide ~q explasatico for the increased risk of subsequent breast cancer associated with this cyst type. In additico,E~ and D}~AS levels appeared to be directly correlated: this might suggest that E6~ productico,which has been shown to be under testoste r e n e control in the mgtse subssxillary glsnd,might be androzen modulated el s o in breast tissues.

ADJUVANT CHEMOTHERAPy FOR POOR PROGNOSIS ESTROGEN RECEPTOR NEGATIVE STAGE II, III BREAST CANCER ONE YEAR VS. 2 YRS CMFVP. *S. Rivkin, S. Green, B. Match, C.K. Osborne, W.A. Knight, R. McDivitt, A. Cruz, D. Tesh, J. Costanzi, S. Balcerzak, R. Stevens. For the Southwest Oncology Group, Seattle, WA 98104 and San Antonio, TX 78229. Between July 1979 and May 1984, 444 patients with estrogen receptor negative Stage II and III (T N • . l-3a' i' M ) breast cancer were randomized to recelqe one or two 0 2 years of CMFVP: Cy~lophosphamide (60mg/M po daily), Meth~trexate (15mg/M-- IV weekly), 5-~luorouracil (400 mg/M IV weekly),Vincristine (.625mg/M ~ IV weeks i-i0; 53-62 ~iso for 2 year patients), and Prednisone (30,20, 10mg/M~ po daily, weeks 1-2,3-4,5-6; and weeks 53-58 for 2 year patients). Patients were treated within 42 days of surgery. Analysis is on 409 eligible patients (26 of the 35 ineligible patients had therapy start after 42 days). Treatment arms were balanced with respect to number of nodes, age, and menopausal status. Compliance on the 2 year arm was poor; among patients who did not die or relapse before year 2, only 38% completed 2 years of treatment. With a median of 5.4 years follow-up (maximum 8.6 years), there are no significant differences in survival between the two treatment groups either overall (2-sided log rank, p=0.37), or within premenopausal, postmenopausal, or 1-3 positive node subgroups. However, there may be a survival advantage in patients treated on the 2 year arm with >3 axillary nodes involved. The estimated 5 year survival for the 1 year arm is 43% as compared to 57% for the 2 year arm. This treatment advantage within the >3 node subgroup is strongest for premenopausal patients. In conclusion, 2 year duration of chemotherapy may prolong survival in patients with >3 axillary nodes.

148 165

Abstracts NEO-ADJUVANT CHEMOTHERAPY FOR OPERABLE BREAST CANCER. PRELIMINARY RESULTS FROM A RANDOMIZED TRIAL. L. Mauriac*, M. Durand, J. Pigneux, J.M. Dilhuydy, E. Bussi~rcs, A. Avril, D. Mar~e. Fondation Bergoni6,180, rue de Salnt-Gen~s F-33076 Bordeaux.

166

A PHASE II STUDY IN RECURRENTBREAST CANCER PATIENTS (PTS) WITH LIPOSOMAL ENCAPSULATEDDOXORUBICIN (LED). J. Treaf*, A. Greenspan, D. Forst, A. Rahman, Divlslon of Medical Oncology, Lombardl Cancer Research Center, Georgetown U n i v e r s i t y , Washington, D.C. 20007 A p r i o r Phase I study of LED established the maximum tolerated dose of 90 mg/m2, This Phase I I study has been undertaken to f u r t h e r evaluate LED t o x i c i t y and its therapeutic efficacy in pts with recurrent measurable breast carcinoma. LED are prepared by using cardlolipln, phosphatidylcholine, cholesterol and etearylamine. A dosage of 75 mg/m2 or less, depending upon prlor chemotherapy and/or radiation has been administered q3 weeks l.v. A total of 20 pts have entered in this Phase II trial with ages ranging from 32 to 75 years. All pts had received adjuvant chemotherapy including two pts who received CAF. In addition, 17 pts had previ~sly failed first llne non-anthracycllne therapy for recurrent disease. Measurable lesions included skin, lymph nodes, pulmonary and hepatic metastases. Ten of 18 evaluable pts have responded in pulmonary (5), hepatic (2), skin (2), lymph nodes (I) sites. Response durations of I0+, 6+, 5+, 5, 4+, 4, 2+, 2+, l+, I+ have been seen. Myelosuppression with WBC
168

PREDICTION OF RESPONSE (N), TO SECOND LINE CHEMOTHERAPY(CTI IN PATIENTS (PTS) WITH METASTATIC BREAST CANCER (MBC). *G. Fraschini, H.Y. Yap, T. Smith, L. Esparza, G. Hortobagyi. M. D. Anderson Cancer Center, Houston, Texas 77030. In a retrospective analysis of second or subsequent chemotherapeutic t r i a l s given to 436 MBC pts, we have attempted to i d e n t i f y pt, tumor, or treatment related characterist i c s that may predict R. The pts received single-agent or combination regimens inclusive of v i n b l a s t i n e , vindesine, mitoxantrone, bisantrene, etoposide, c i s p l a t i n , and e l l i p tinium acetate that produced comparable R rates, between 16% and 37% (average, 26%). With the exception of 55 pts who had only received one adjuvant regimen, all had received one or more regimens for metastatic disease. A s i g n i f i c a n t l y higher R rate (38%) was observed in the adjuvant-only treated group. Otherwise, number and duration o~ and response to prior chemotherapies did not c l e a r l y correl a t e with R. We have also observed s i g n i f i c a n t correlation of R with performance status (35% in asymptomatic, 28% in symptomatic, and 14% in p a r t l y bed-ridden pts), serum l a c t i c dehydrogenase (32% i f normal, 23% i f intermediate, 15% i f markedly elevated), and extent of disease, as assessed by the coding system reported in Cancer Res 39:1552, 1979 (31% i f l i m i t e d , 26% i f intermediate, 15% i f extensive). Not s i g n i f i c a n t l y higher R rates were associated with normal serum alkaline phosphatase, albumin and calcium concentrations, and absence of weight loss. Reduction of CT dosage, as demanded by pt tolerance, did not a f f e c t R. No correlation with R was seen for age, hormonal receptor and menopausal statu s , CEA, hemocytometric and other biochemical parameters. A m u l t i v a r i a t e regression analysis is being added to f u r t h e r assess the prognostic value of these factors. The i d e n t i f i c a t i o n of prognostic factors of response may assist in comparing and s t r a t i f y i n g pt populations in c l i n i c a l t r i a l s and in selecting pts who may benefit from additional standard chemotherapy.

From 01.01.1985 to 10.31.1987, 187 patients with an infiltrative breast adenocarcinema T 2 > 3 0 a m or T v N Oor N 1, without inflammation have been included in a randomized trial comparing efficacy of two therapeutic options: group I: initial mastectomy + lymphadenectomy + adjuvant chemotherapy in case of node involvement or estrogen and progesterone receptor negativity (EPR') [n=94J ; groupe II: nee-adjuvant chemotherapy followed by adjusted locoregional treatment [n=93]. Adjuvant and nee-adjuvant polychemotherapy included 3 cycles with epirubicine 50 mg/m2+ vincristine 1 mg/rn 2 + methotrexate 20 mg/maand 3 cycles with mitomycine C 10 rag/m2 + thiotepa 20 mghn 2 + vindesine 4 mg/m 2, each course delivered every 3 weeks. Mean age, hormonal status, mean clinical tumor size, histological types, ScarffBloom-Richardson grading, estrogen and progesterone receptor contents were the same in the two groups. In group I, 24 patielits did not receive any adjuvant treatment (EPR + and N'); 70 were treated with adjuvant chemotherapy. In group II, all patients received nee-adjuvant chemotherapy; 34 were in complete clinical remission and were treated with external radiotherapy and curietherapy; 23 were treated.with tumorectomy + axillary node dissection +extemal radiotherapy for a residual tumor smaller than 30 a r e ; 35 were treated with mastectomy (Patey) for a residualtumor bigger than 30 a m . EPR" tumors responded better than EPR*; their mean size reduced from 42 to 18 mm and 42 to 28 mm respectively (p=.002). Breast conservative treatment was performed in 78 % of EPR- and 49 % in EPR÷(p=.014). With a 18 months median follow up, metastatic relapses are the same in two groups.

167

HIGH-DOSE CYTOXAN/VP-16/PLATINUM (CVP) INTENSIFICATION FOR H O R M O N A L L Y UNRESPONSIVE (HU) METASTATIC BREAST CANCER (MBC). F Dunphy,* G Spltzer, J Ellis, B Radcliffe, A Buzdar, G Hortobagyi, L Horwitz, S Jagannath, J Yau, l Spinolo, K Dicke, U T M. D. Anderson Cancer Center, Houston, T X 77030. High-dose CVP intensification with autologous bone marrow rescue has been used in 46 patients (pts) with estrogen receptor-negative (ER-) or H U MBC. O f this group, 7 are too early to evaluate, and 3 have suffered early deaths. This report concerns 36 evaluable pts. Pt characteristics at entry: median age 40 (29-62); 25 pre-menopausal, 4 peri-menopausal, 7 postmenopausal; 21 ER-, 12 ER+, 3 ER unknown; 23 had ~2 sites of disease. Disease sites: soft tissue/breast 41%, lung/ mediastinum 52%, nodes 25%, bone 33%, liver 28%. Median disease-free interval from diagnosis is 58 weeks (0-344). Pts received 2-10 courses standard induction chemotherapy, usually adriamycin 60mg/m s + Cytoxan 6O0mg/m z. Induction was followed by 2 courses CVP in 35 pts and 1 course in 1 pt (Cytoxan 4.5-5.25 g / m s, VP-16 750-1200mg/m 2, platinum 120-18Drag/mS). T~)xiqi*ies; 60% incidence of moderate nausea and vomiting, 76% incidence of F u n or documented infection. Response to induction: CR 31%, PR 50%, overall response 81%. 7 pts (19%) had stable or progressive disease. Rgsoonse oost-CVP: 67% CR, 28% PR, 2 pts had stable disease, 1 rendered CR after surgery; 69% total CR rate. 19 pts progressed and of those, 6 pts died. Median followup intervals from induction and CVP intensification are 58 and 39 weeks, respectively. Median survival from induction is estimated at 111+ weeks, median survival from 1st CVP is estimated at 98+ weeks. From induction, 10 pts are long-term disease-free (7 between 1 to 2 years, and 3 >2 years). Conclusions: a) CVP intensification increases CR rates in MBC; b) toxicity i s acceptable; and c) unmaintalned, continued longterm CR o£ 10 pts from 1-2+ years appears promising; further follow-up is needed.

Abstracts 169

P L A T I N O L A N D CONTINUOUS INFUSION 5-FLUOROURACIL IN REFRACTORY STAGE IV BREAST CANCER. *J D Bitran, R K Desser, and M F Kozloff, Joint Section Hematology-Oncology, University of Chicago, Pritzker School of Medicine, Michael Reese Medical Center, Chicago, Illinois 60616. We entered 24 women with refractory stage IV breast cancer into a phase II trial of Platigol (P) 100mg/m I.V. day i and 5-Fluorouracil (F) 1000mg/m~ continuous infusion days 1-5 every 21 days. The 24 patients (pts) entered on study had a median age of 42 years (range 37-62 years) and a median PS of 1 (range 0-3). All 24 pts received prior chemotherapy including anthracyclines and the median number of prior chemotherapy regimens was two (range 1-4). The dominant site of metastases included skin and soft tissues, i0 pts; parenchymal lung, 6 pts; liver, 4 pts; and bone and bone marrow, 4 pts. Time to treatment failure (T~F) or death was calculated from day 1 of P and F. Results are shown in the table below: Dcminant Site NO. of TTF of Metastases Pts CR PR NR (Months) Skin and soft tissue i0 0 8 2 2+,3+,3,4,4,5, 6,6+,7+,12+ Lung 6 0 6 0 2,3,4+,5,6+,7+ Liver 4 0 0 4 2,3,4,4 Bone 4 0 0 4 2r3,5,5 Overall 24 14 i0 The overall response rate was 57%, and all responses were partial. The median T T F w a s 4 months (range 2-12+ months). Toxicities included mucositis, grade 3 or 4 in "5/22 pts; par~nthesias, 3/22 pts and leucopenia, median 2400 cells/rsa ~ (range 1300-3000) on day 12. Grade 4 neutropenia was seen in 2 pts. Grade 4 thrc~bocytopenia was not encountered. F and P have significant activity in pts with refractory stage IV breast cancer albeit a short duration of response.

149