Dig Dis Sci (2006) 51:2213–2219 DOI 10.1007/s10620-006-9274-2

ORIGINAL PAPER

The Use of Early and Midpoint Adenoma-Carcinoma Sequence Biomarkers in Prediction of Neoplastic Progression in Patients with a History of Colorectal Neoplasia Hassan Albataineh · Violeta Yordanova · Jessica Bowman · WeiLi Zhou · James Hatfield · Michael J. Lawson · Paula Sochacki · Martin Tobi

Received: 21 September 2005 / Accepted: 1 March 2006 / Published online: 1 November 2006 C Springer Science+Business Media, Inc. 2006


Abstract Since significant neoplasia after initial colonoscopy is low, we conducted this pilot study to compare the predictive role for colorectal neoplasia recurrence of anti-DCC with that of Adnab-9 binding to colonic effluent of high-risk patients. DCC and Adnab-9 effluent ELISA were performed at baseline colonoscopies. The results of followup colonoscopies were reviewed. To ensure specificity, immunohistochemistry and Western blot was performed with anti-DCC and for Adnab-9 where optimal fixation times were also evaluated. Mean follow-up was 2.6 years. Of 21 patients, 6 of 10 who progressed to CRN and 2 of 11 who did not had a positive Adnab-9 ELISA result (P = 0.08). Despite an initial good correlation with Adnab-9 ELISA results in a smaller dataset, we were unable to obtain consistent subsequent DCC immunohistochemistry or Western blot data using antibody from two different sources. However, the original dataset of Adnab-9 results was reproducible on repetition of the ELISA with a larger set of samples that included this initial dataset and optimal fixation time was 20 min. We conclude that Adnab-9 appears to be a promising prognostic marker for neoplasia in the high-risk population. Industry standards H. Albataineh · V. Yordanova · J. Bowman · W. Zhou · J. Hatfield · P. Sochacki · M. Tobi Departments of Internal Medicine and Pathology, John D. Dingell VAMC and Wayne State University School of Medicine, Detroit, Michigan, USA M. J. Lawson Kaiser Permanente Medical Group, Sacramento, California, USA M. Tobi (
) John D. Dingell VAMC, Med II/GI, Division of Gastroenterology, 4646 John R, Detroit, Michigan 48201, USA e-mail: [email protected]

need to be developed for DCC monoclonal antibodies that may have similar utility. Keywords Deleted in colon cancer . Adnab-9 . Monoclonal antibodies . Colonic effluent

Introduction Approximately 148,000 new cases of colorectal cancer (CRC) are diagnosed each year in the United States. The lifetime incidence for patients at average risk is 5%. In the United States, CRC ranks second to lung cancer as a cause of cancer death, and it is third in frequency of cancers, in men and women [1]. Most colorectal cancers arise from adenomatous polyps, some of which progress from diminutive (<5 mm) to large polyps ( ≥ 1.0 cm) and then to dysplasia and cancer [2]. There have been few large-scale prospective studies comparing the above methods for surveillance, but where this has been done [3], colonoscopy has been deemed the modality of choice compared to barium enema. However, it has since been shown that most of the benefit of colonoscopy in the reduction of colorectal cancer is from the initial examination, as was initially reported using a Micro-Stimulation Screening Analysis model [4] because the incidence of significant neoplasia on surveillance colonoscopy, particularly cancer [5], is low. Thus, a concerted effort to develop a prognostic marker that could be performed at the time of initial endoscopy to define the population at risk is warranted and could avert fruitless follow-up colonoscopic examinations. Noninvasive or minimally invasive detection of tumorassociated antigens (TAAs) such as Adnab-9 in the stool or the colonic effluent may reflect the presence of a field effect of carcinogenesis [6]. Effluent (colonic washings) testing

Springer

2214

has been recognized as a vehicle of biomarker development [7]. Accordingly, we used effluent collected at the time of the initial colonoscopy to test candidate biomarkers for early and intermediate markers of colorectal neoplasia (CRN) risk. Adnab-9 is a monoclonal antibody raised against constituents of colonic adenomas [8] and a marker for this sequence. The antigen occurs early within the sequence but the exact interposition of the antigen within the sequence is not known [9, 10]. Adnab-9 has been shown to identify an 87-kDa protein antigen found in the cytoplasm of cells in colonic adenomatous but not carcinomatous tissue, resembling Paneth cells, a source of growth factors in the gut [11]. The presence of Adnab-9 staining in nonneoplastic colonic epithelium has been associated with an increased risk for the development of bowel cancer [6, 8–12] and the diagnostic potential may be related to the presence of an Adnab-9-defined field effect [6]. Another recent study showed that positive labeling of severely dysplastic polyps by Adnab-9 is more significantly correlated with the presence of synchronous neoplastic lesions and neoplasia on subsequent colonoscopic follow-up than that of p53, a late-stage marker for the adenoma carcinoma sequence [13]. Thus, Adnab-9 appears to be ideally suited for evaluation of future neoplastic risk determination in conveniently obtained colonic effluent samples. In seeking to characterize the Adnab-9 as a risk marker in the setting of effluent testing, it was necessary to use a marker that has been shown to have a recognized prognostic function. Deleted in colorectal cancer (DCC) gene mutation occurs when this gene is deleted from its location on chromosome 18q. DCC labeling of colorectal neoplastic tissue has been reported to be a prognostic marker in patients with CRC, in that the absence of DCC in tumors is linked to poor survival among these patients [14]. DCC can be regarded as a midpoint marker within the adenoma-carcinoma sequence [15]. Functionally, DCC is identical to the netrin-1 receptor [16] that was thought to direct growth of specific neuronal axons and is expressed in the first neuronal clusters of the developing brain [17]. Subsequently it was found to induce apoptosis dependently when disassociated from its specific ligand. The down-regulation observed in colorectal cancer is due to both the expected loss of heterozygosity associated with mutation of the encoding genes and epigenetic influences, the latter perhaps related to histone deacetylase activity [16]. In addition, since expression of these receptors inhibits cell anchorage-independent growth and deletion restores these tumor attributes, DCC may act as a tumor suppressor [16]. Thus, DCC is also an ideal prognostic candidate for evaluation and has the advantage of having an elucidated mechanism but has not been tested for in colonic effluent. One possible pitfall regarding the mode of bowel preparation leading to effluent collection and its effect on ELISA testing had been encountered in earlier work and we planned to address it in this study. A previous study showed that Springer

Dig Dis Sci (2006) 51:2213–2219

oral colon lavage solutions containing polyethylene glycol interfere with ELISA detection of tumor associated antigens distinct from Adnab-9 or DCC in colonic effluent [18] and we were concerned that this commonly used bowel preparation might affect ELISA results. Patients prepared by this method were included in this study and analyzed for the potential effects of ELISA interference. In addition, since immunohistochemistry [IHC] is a convenient method of testing for both DCC [14] and Adnab-9 [10–13] and a moderate reduction of ELISA binding has been demonstrated when reacting Adnab-9 with formalin fixed-antigen [10], we performed IHC for both DCC and Adnab-9. We first tested for reproducibility, and once this was established, the effect of timing of formalin fixation was further evaluated. Thus, the main objective of this pilot study was to investigate the mechanics leading to demonstrating potential prognostic utility of both Adnab-9 and DCC as markers in effluent samples for the development of CRN. This would be achieved by looking at the interrelationship of these two markers and that of different modalities of bowel preparation and IHC. This was a retrospective pilot study performed on archival samples to investigate the mechanics and potential of two markers and is not designed to provide the power to definitively yield a diagnostic marker.

Materials and methods This study was performed at the John D. Dingell VA Medical Center (VAMC, Detroit, MI) and Kaiser Permanente Medical Center (Sacramento, CA). Patient effluent samples aspirated at the initial colonoscopies had been stored frozen at –70◦ C. The patients underwent a follow-up colonoscopy that was clinically indicated. The findings of these procedures were reviewed, and the presence of lesions manifesting the typical morphology of colorectal neoplasia (CRN), either adenoma or cancer, as determined by routine pathologic examination was considered as neoplastic progression. Mixed polyps, defined as polyps with an adenomatous tissue component less than 20% of the polyps, were considered as nonprogression of CRN [19]. Exclusion criteria were symptoms unexplained by colonoscopic findings, cancers at initial colonoscopies where colonoscopy was incomplete at the time of surgery with the presence of small polyps therefore not documented at baseline, and patients with a follow-up of less than 3 months. A standardized ELISA for native, nonfixed antigen was performed on the effluent samples using DCC (Santa Cruz Biotechnology, Santa Cruz, CA) and Adnab-9 monoclonal antibodies (MAbs). The streptavidin-biotin-peroxidase complex (ABC) method was used for DCC and Adnab-9 staining using the Vectastin kit (Vector Laboratories, Burlingame, CA). Positive ELISA results were defined

Dig Dis Sci (2006) 51:2213–2219

as optical density minus background >0.05. If the initial ELISA was successful, a second ELISA assay to ensure reproducibility was performed on the same samples on a later occasion. Immunohistochemistry (IHC) with the above MAbs was also performed using positive controls as per the manufacturers’ recommendation (human brain for DCC and colonic adenoma for Adnab-9) to demonstrate specificity of labeling of nonnative, fixed antigen. The type of bowel preparation was also noted and Adnab9 ELISA results were compared with different preparations such as polyethylene glycol (PEG)-based or electrolyte purge non-PEG (magnesium citrate or sodium phosphate)-based reagents. The study met the requirements of the respective institutional review boards. Western blotting was performed after electrophoresis on 10% polyacrylamide gels with two membrane-enriched brain extracts (kindly provided by Dr. A. Sloan, Department of Neurosurgery, Wayne State University School of Medicine) and transfer as described previously [8–10, 12]. Similarly, an adenoma extract was similarly run against Adnab-9. All lanes were loaded with 5 µg total extract protein. For the antibody where IHC labeling was shown to be reproducible and specific, we also attempted to define optimal fixation times for labeling. We investigated fixation in formalin at 10, 20, and ≥ 30 min in serial frozen sections from nine adenomatous polyps for Adnab-9 and brain sections for DCC. Sections reacted with nonimmune mouse sera (Biogenex, San Ramon, CA) in place of the primary antibody served as negative controls for background labeling determination. Labeling intensity was graded 1 to 4 + as reported previously [11].

Statistical analysis Data were statistically analyzed for linear correlation by the Pearson correlation test and for significant proportional differences by Fisher’s exact test. Differences in the means of groups of individual parameters were evaluated by a twotailed Student’s t-test and ANOVA using a computerized statistical package (Instat by GraphPad Software Inc,. San Diego, CA). Significance was defined as P < 0.05.

Results Twenty-one patients (all male) were included in this study. Eleven patient samples were excluded: four with insufficient follow-up documentation (n = 1) or with cancers at the baseline colonoscopy (n = 3) where small polyps were not excluded and seven with continuing unexplained symptoms. Ten patients had neoplastic progression, 11 patients did not, and the difference in median duration of follow-up between

2215 Table 1 Characteristics of patient groups at baseline and at end of follow-up P value

Patient parameter

Progression

Nonprogression

Number Age, mean yr ± SD Gender (% male) Median ± SD follow-up (range), yr Baseline cancer/Hgda Baseline adenoma Died Polyethylene glycol purge Other cancers

10 65.6 ± 4.5 100 3.0 ± 3.2 (2–10.1) 3 (30%) 6 (60%) 4 (40%) 38%

11 61.6 ± 11.86 100 2.5 ± 1.9 (0.5–7.2) 3 (27%) 3 (27%) 4 (36%) 75%

N/A 0.3 1 0.3

6/10 (60%)

6/11 (55%)

1

a

1 0.2 1 0.32

High-grade dysplasia.

the progressive and the nonprogressive patients was not statistically significant (mean ± standard deviation, 3.0 ± 3.2 vs 2.5 ± 1.9 years, respectively; P = 0.3). Other demographics are summarized in Table 1, showing no statistical significance in ages (mean, 65.6 vs 61.6 years, respectively), baseline numbers of cancer or high-grade dysplasia (30% versus 27%), adenomas (60% vs 27%), or death as an end point (40% vs 36%), respectively. In addition, other background cancers were similar (60% vs 55%, respectively) and no statistical difference was seen in the type of bowel preparation (38% vs 75%, respectively). Initially there was a good correlation (Fig. 1) between DCC and Adnab-9 on 11 paired samples (r2 = 0.95, P = 0.0001). We were unable to reproduce this initial result and were therefore unable to make a direct comparison of DCC ELISA results to Adnab-9 results on the larger series of samples. Adnab-9 ELISA results were reproducible on the 11 samples performed (Fig. 2) on two different occasions (r2 = 0.9, P < 0.0001). Adnab-9 ELISA results tended to be positive in 6 of 10 patients who progressed to neoplasia (Fig. 3), compared to only 2 of 11 patients who did not (OR, 6.75; CI, 0.93–49.3; P = 0.08). DCC IHC staining on brain tissue was positive but showed high background staining (Fig. 3a), but this result was not reproduced later in a reproducibility experiment (Fig. 3b). In contrast, specific Adnab-9 adenoma labeling was demonstrated with a negative background (data not shown), similar to previously published results [10]. Differing bowel preparations did not appear to significantly interfere with the Adnab-9 ELISA but there was a weak trend to a lower proportion of positive results in patients receiving a PEG-based bowel cleansing preparation (22% vs 60% magnesium citrate-based bowel preparation; OR, 0.19,CI, 0.018–2.06; P = 0.3). Western blotting after gel electrophoresis was performed for two DCC antibodies and Adnab-9. Multiple reactive species were detected when two Springer

2216

Dig Dis Sci (2006) 51:2213–2219

Fig. 1 Initial correlation of DCC and Adnab-9 ELISA results. A significant linear correlation is seen between 11 paired samples (r2 = 0.95, P < 0.0001). X-axis, Adnab-9; Y-axis, DCC and depicts results expressed as OD – background/5 µg protein

different brain extracts were reacted with the DCC antibody supplied by Santa Cruz (the same as used in the comparison ELISA with Adnab-9). In comparison, the same extracts, when reacted with the second DCC antibody obtained from U.S. Bioscience did show single bands, but of a lower molecular weight than would be anticipated. These extracts did not react with Adnab-9 but a specific single band was obtained at 87 kDa when Adnab-9 was reacted with an adenomatous polyp extract (Fig. 4). Since reproducibility experiments were positive only with Adnab-9 IHC, we proceeded to timed fixation experiments only with this antibody. Figure 5 shows the effect of fixation times in representative serial sections fixed at different time intervals. The mean grades of intensity over background from 10 sections evaluated at time 0 were 0.6 ± 0.5; at 10 min, 0.7 ± 0.8; at 20 min, 1.4 ± 0.8; and at ≥ 30 min, 0.8 ± 0.3 (P > 0.05 by one-way ANOVA). It appears that 20 min tended to be the optimal fixation time for Adnab-9 labeling, showing an almost 50% higher intensity compared to the mean of combined intensities at the other fixation times (P < 0.03). Background staining (Fig. 5) was also signifi-

Fig. 2 Correlation of reproducibility of effluent Adnab-9 assays. Results obtained plotted after performing ELISA assays on 22 samples on two different occasions (r2 = 0.9, P = 0.0001). X-axis, first assay; Y-axis, second assay

Springer

cantly higher (r2 = –0.91, P < 0.02) at shorter fixation times when data points for 40 and 60 min are considered separately from the remainder of the ≥ 30 min group.

Discussion Colorectal cancer is a common, lethal but preventable disease. The literature provides evidence that colonoscopic polypectomy testing reduces the disease-specific CRC mortality rate [20] despite the use of historical controls used in that study. The current methods of surveillance are invasive, with a definable mortality rate with a relatively low yield [5], favoring the development of predictive markers to identify patients in whom CRN is likely to recur. Detection of TAAs in effluent samples is here proposed as a new minimally invasive modality to aid in targeted surveillance. In our study, two TAAs were used, DCC and Adnab-9 MAb binding. Like Adnab-9, DCC is a relatively early (midsequence) event occurring in adenomas [21] and may influence carcinogenesis [22]. Although initially there was a good correlation between

2217

Mean Intensity

Dig Dis Sci (2006) 51:2213–2219

1.75 1.5 1.25 1 0.75 0.5 0.25 0 0 10 20 >30 Fixation Time (Minutes)

Fig. 5 Graph showing optimal fixation time for Adnab-9 labeling. The Y-axis depicts the mean grade of intensity of Adnab-9 labeling in serial frozen sections of nine polyps fixed for differing time periods with 10% buffered formalin minus the background staining; the X-axis depicts the time in minutes. At times 0, 10, and ≥ 30 min one can see that the intensity equals ≤ 1, but clearly at 20 min (as depicted by the filled diamonds) there is an almost 50% increase in the intensity of labeling compared to that of the combined other times. Comparing labeling intensities at 20 min with those at other fixation times, the difference is significant at P < 0.05 (Student’s t test). The background is depicted by filled squares and is significantly inversely proportional to the fixation times when data points for times >30 min are analyzed separately (r2 = –0.91, P < 0.02)

Fig. 3 Immunohistochemistry using DCC MAbs on human brain. (A) Initial: Labeling shows neuronal and other cells with dark cytoplasm stain but high background staining. Substrate was 3-amino-9-ethylcarbazole. (Original magnification, × 100.) (B) Reproducibility: Tissue shows the absence of staining in neurons under similar conditions and magnification

Fig. 4 Western blot with DCC and Adnab-9 antibodies. PAGE followed by transfer and binding to specific banding was performed for two DCC antibodies and Adnab-9. The DCC antibody, supplied by Santa Cruz (the same as used in the comparison ELISA with Adnab-9), showed multiple reactive species in lanes 2 and 3 when reacted with two different brain extracts (A). The second antibody, obtained from U.S. Bioscience (B), did show single bands but of a lower molecular weight than would be anticipated [31], as indicated by the solid arrow at the left. No activity was seen with Adnab-9 and these extracts (C). A specific single band was obtained at 87 kDa when Adnab-9 was reacted with an adenomatous polyp extract (D), as indicated by the open arrow at the right

DCC and Adnab-9 ELISA results, lack of standardization for two DCC monoclonals when reacted with specific control reagents precluded successful reproducibility testing. In contrast, Adnab-9 was strongly positive and specific when ELISA was performed on effluent samples and IHC on tissue sections. These positive attributes also extended to Western blotting using adenoma extracts and to IHC using adenoma sections. The optimal fixation time appears to be 20 min. The time to penetrate to the center of a large adenoma would obviously be increased compared to that for a thin tissue section, however, it is routine to bisect polyps before fixation and section the most superficial layers first. In addition, even at increased fixation times, labeling is still clearly apparent and it is the intensity rather than the absolute labeling that is affected. This is offset by the lack of background staining seen at fixation times of 30 min or longer. One of the problems confronting TAA testing is the lack of standardization of antibodies. DCC is one of the popular TAAs that has been used for CRC prognostication. However, questions have been raised by others regarding the standardization of DCC MAbs and the techniques employed [23]. This issue was also raised in a recent study showing marked variability in staining results using different COX-2 antibodies, demonstrating that the lack of standardization for other tumor markers is also a problem [24]. One of the authors of the DCC publication [14] that described the relation between DCC and CRC prognosis has expressed the view that available DCC MAbs are not necessarily the gold standard (Dr. I. Summerhayes, personal communication). Our study showed a lack of standardization, therefore, we suggest that Springer

2218

industry standards need to be developed for DCC and other candidate surveillance MAbs. In this study, differing bowel preparations did not appear to interfere with Adnab-9 ELISA results, unlike data obtained with other MAbs [18]. Further studies with larger samples need to be done to validate this finding as part of an effort to determine optimum conditions for TAA surveillance testing. Such a prospective study to examine this and other issues raised in this paper is under way. Adnab-9 surveillance testing appears to be a promising prognostic effluent marker given the strong trend in this limited sample. Effluent samples are easily obtained at the time of initial colonoscopy and an ELISA can be done to determine the presence of Adnab-9 binding. Accordingly, the recurrence of CRN is potentially predictable and this knowledge may guide decisions regarding the need and the time of follow-up surveillance colonoscopic examinations and may be important in instituting cost-saving strategies. This prognostic ability has also been demonstrated in a previous study that enrolled both genders, where the Adnab-9 labeling of the index severely dysplastic polyp was significantly correlated with the multiplicity of adenomas as well as the recurrence of adenomas [13]. All the patients in our study are men; thus, these results in colonic effluent may not be applicable to the general population. Another issue is the presence of other tumors that may shed Adnab-9-defined antigens into the upper gastrointestinal tract [11, 25, 26] that may theoretically be found in the effluent aspirate. There were no known tumors documented in our patients that might have influenced the outcome of the effluent test but they were generally not specifically screened for these tumors, far less common than colorectal cancers within the population context. In order to increase the sensitivity of this test, more than one TAA can be used simultaneously in battery testing, as recently reported for stool [27, 28]. However, the prognostic utility of these tests has not yet been demonstrated. We are optimistic regarding DCC’s role as another candidate that may have prognostic utility, but given the experience in our hands, we feel that formal standards for these and other available DCC MAbs [29, 30] should be developed to allow for what we believe will be an effective application. Acknowledgments The authors would like to acknowledge the technical assistance of Chaya Kenkre, MD, Himabindu Boda, MD, Lucyna Zawadzki, and Tashia Orr, BS. This study was supported in part by grants from the Veterans Administration and the Kaiser Permanente Research Foundation. This study is dedicated to the memory of Joe Faigin.

References 1. Jemal A, Tiwari RC, Murray T, Ghafoor A, Samuels A, Ward E, Feuer EJ, Thun MJ (2004) Cancer statistics 2004. CA Cancer J Clin 54:8 Springer

Dig Dis Sci (2006) 51:2213–2219 2. Tobi M (1999) Polyps as biomarkers for colorectal neoplasia. Frontiers Biosci 4:1–8 (online journal: www.bioscience.org) 3. Winawer SJ, Stewart ET, Zauber AG, Bond JH, Ansel H, Waye JD, Hall D, Hamlin JA, Schapiro M, O’Brien MJ, Sternberg SS, Gottlieb LS (2000) A comparison of colonoscopy and doublecontrast barium enema for surveillance after polypectomy. N Engl J Med 342:1766–1772 4. Zauber AG, Winawer SJ, Loeve F, Boer R, Habbema D (2000) Effect of initial polypectomy versus surveillance polypectomy on colorectal cancer incidence reduction: Micro-stimulation modeling of national polyp study data. Gastroenterology 118:A187 (abstract) 5. Loeve F, van Ballegooijen, Snel P, Habbema JD (2005) Colorectal cancer risk after colonoscopy: A population–based study and literature search. Eur J Cancer 41:416–422 6. Tobi M, Prabhu S, Gage RE, Orr T, Lawson MJ (2002) Colorectal cancer risk: The impact of evidence of a field effect of carcinogenesis on blinded diagnosis using an anti-adenoma antibody test performed on colonoscopic effluent. Dig Dis Sci 47:317–321 7. Srivastava S, Verma M, Henson DE (2001) Biomarkers for early detection of colon cancer. Clin Can Res 7:1118–1126 8. Tobi M, Darmon E, Phillips T, Heller T, Rozen P, Nochomovitz L, Steinberg W (1990) Increased expression of a putative adenomaassociated antigen in pre-colonoscopic effluent of patients with colorectal cancer. Cancer Lett 51:21–25 9. Tobi M, Maliakkal B, Zitron I, Alousi M, Goo R, Nochomovitz L, Luk G (1992) Adenoma-derived antibody, Adnab-9 recognizes a membrane-bound glycoprotein in colonic tissue and effluent material from patients with colorectal neoplasia. Cancer Lett 67:61–69 10. Tobi M, Maliakkal BJ, Alousi MA, Voruganti V, Shafiuddin M, Yang S, Gesell MS, An T, Hatfield JS, Fligiel S, Kaila V, Goo RH, Luk GD (1992) Cellular distribution of a colonic adenomaassociated antigen as defined by monoclonal antibody Adnab-9. Scand J Gastroenterol 27:737–742 11. Tobi M, Kaila V, Hassan N, Gallinger S, Fligiel S, Hatfield J, Gesell M, Sakr W, Luk G, Odze RD (1999) Monoclonal antibody Adnab-9 defines a preneoplastic marker in epithelium at risk for adenocarcinoma of the small intestine. Hum Pathol 30:467–473 12. Tobi M, Elitsur Y, Moyer MP, Halline A, Deutsch M, Nochomovitz L, Luk GD (1993) Mucosal origin and shedding of an early colonic tumor marker defined by Adnab-9 monoclonal antibody. Scand J Gastroenterol 28:1025–1034 13. Sheikh RA, Min BH, Yasmeen S, Teplitz R, Tesluk H, Ruebner BH, Tobi M, Hatfield J, Fligiel S, Lawson MJ (2003) Correlation of Ki67, p53, and Adnab-9 immunohistochemical staining and ploidy with clinical and histopathologic features of severely dysplastic colorectal adenomas. Dig Dis Sci 48:223–229 14. Shibata D, Reale MA, Lavin P, Silverman M, Fearon ER, Steele G Jr, Jessup JM, Loda M, Summerhayes IC (1996) The DCC protein and prognosis in colorectal cancer. N Engl J Med 335:1727–1732 15. Cho KR, Vogelstein B (1992) Genetic alterations in the adenomacarcinoma sequence. Cancer 70:1727–1731 16. Thiebault K, Mazelin L, Pays L, Llambi F, Joly M-O, Scoazec J-Y, Saurin J-C, Romeo G, Mehlen P (2003) The netrin-1 receptors UNC5H are putative tumor suppressors controlling cell death commitment. Proc Natl Acad Sci USA 100:4173–4178 17. Hjorth JT, Gad J, Cooper H, Key B (2001) A zebrafish homologue of deleted in colorectal cancer (zdcc) is expressed in the first neuronal clusters of the developing brain. Mech Dev 109:105–109 18. Tobi M, Darmon E, Rozen P, Epstein N, Konikoff F, Stadler J, Harpaz N, Fink A, Bentwich Z, Maliakkal B, Luk G (1991) Oral colon lavage solutions containing polyethylene glycol may interfere with ELISA detection of tumor-associated antigens in colonic effluent. Dig Dis Sci 36:1448–1452 19. Ullah N, Qureshi K, Hatfield J, Sochacki P, David D, Albataineh H, Mejia L, Kenkre C, Lawson M, Tobi M (2004) Small early tubular adenomas and mixed colonic polyps found on screening flexible

Dig Dis Sci (2006) 51:2213–2219

20.

21.

22.

23. 24.

25.

sigmoidoscopy do not predict proximal neoplasia in males. Clin Gastro Hepatol 2:246–251 Winawer SJ, Zauber AG, Ho MN, O’Brien MJ, Gottlieb LS, Sternberg SS, Waye JD, Schapiro M, Bond JH, Panish JF (1993) Prevention of colorectal cancer by colonoscopic polypectomy. N Engl J Med 329:1977–1981 Traicoff JL, Willson JK, Markowitz SD (2002) Early loss of deleted in colorectal carcinoma gene transcript detected in a group of benign colon adenomas. J Biomed Sci 9:716–720 Huerta S, Srivatsan ES, Venkatasan N, Livingston EH (2001) Human colon cancer cells deficient in DCC produce abnormal transcripts in progression of carcinogenesis. Dig Dis Sci 46:1884– 1891 Swaim MW (1997) The DCC protein and colon cancer. N Engl J Med 336:1456 (authors’ reply 1456–1457) Garewal H, Ramsey L, Fass R, Hart NK, Payne CM, Bernstein H, Bernstein C (2003) Perils of immunohistochemistry: Variability in staining specificity of commercially available COX-2 antibodies on human colon tissue. Dig Dis Sci 48:197–202 Tobi M, Hatfield J, Adsay V, Galagan K, Kozarek R, Inagaki M, Kasai S, Tokusashi Y, Obara T, Hruban RH, Lough J, Barkun AN, Jabbari M, Sheikh R, Ruebner B, Lawson MJ, Ben-Josef E, Fligiel S (2001) Prognostic significance of the labeling of Adnab-9 in pancreatic intraductal papillary mucinous neoplasm. Int J Pancreatol 29:141–150

2219 26. Qiao S, Yuan M, Liu Y, Lin X, Zhang X, Tobi M (2003) Detection of gastric cancer and premalignant lesions by novel marker glycoprotein 87 using monoclonal antibody Adnab-9. Cancer Epidemiol Biomark Prev 12:1095–1099 27. Alquist DA, Skeletsky JE, Boynton KA, Harrington JJ, Mahoney DW, Pierceall WE, Thibodeau SN, Shuber AP (2000) Colorectal cancer screening by detection of altered human DNA in stool: Feasibility of a multitarget assay panel. Gastroenteology 119:1219– 1227 28. Imperiale TF, Ransohoff DF, Itzkowitz SH, Turnbull BA, Ross ME, Colorectal Cancer Study Group (2004) Fecal DNA versus fecal occult blood for colorectal-cancer screening in an averagerisk population. N Engl J Med 351:2704–2701 29. Deiner MS, Sretavan DW (1999) Altered midline axon pathways and ectopic neurons in the developing hypothalamus of Netrin-1- and DCC-deficient mice. J Neurosci 19:9900– 9912 30. Finger JH, Bronson RT, Harris B, Johnson K, Przyborski, Ackerman SL (2002) The netrin 1 receptors Unc5h3 and Dcc are necessary at multiple choice points for the guidance of corticospinal tract axons. J Neurosci 22:10346–10356 31. Meyerhardt JA, Caca K, Eckstrand BC, Hu G, Lengauer C, Banavali S, Look AT, Fearon ER (1999) Netrin-1: Interaction with deleted in colorectal cancer (DCC) and alterations in brain tumors and neuroblastomas. Cell Growth Different 10:35–42

Springer