Virchows Arch (2003) 443:524–527 DOI 10.1007/s00428-003-0862-7
ORIGINAL ARTICLE
Boo KruÐlin · Davor Tomas · Hermann Rogatsch · ˇ upic´ · Mladen Belicza · Irena Novosel · Hrvoje C Ognjen Kraus · Gregor Mikuz
Periacinar retraction clefting in the prostatic needle core biopsies: an important diagnostic criterion or a simple artifact? Received: 17 April 2003 / Accepted: 16 June 2003 / Published online: 24 July 2003 Springer-Verlag 2003
Abstract The diagnosis of prostatic adenocarcinoma in needle core biopsy is based on major and supportive criteria. One of the supportive criteria is the presence of retraction clefting around neoplastic glands. We analyzed a series of 137 prostatic cancer cases diagnosed by needle core biopsy to determine the frequency, extent and criteria for periacinar retraction clefting. Clefting was analyzed on ten neoplastic and ten normal glands in three different high power fields. One-third or more glands with clefts affecting more than 50% of circumference were significantly more common in tumors (51.8%) than in benign glands (8%) (P<0.0001). A stricter criterion that designated as positive the cases with at least 50% of neoplastic glands (15 of 30) with clefts that affected more than 50% of circumference revealed clefts in only 15.3% of the malignant cases but none in benign cases (0%) (P<0.0001). Regardless of their extension, 15 or more glands with clefts were also more prominent in malignant cases (86.9%) than in benign cases (20.4%) (P<0.0001). We conclude that periacinar retraction clefting represents a reliable criterion for diagnosis of the prostatic adenocarcinoma, especially in cases with clefts affecting more than 50% of circumference in at least 50% of suspicious glands. B. KruÐlin ()) · D. Tomas · H. Cˇupic´ · M. Belicza Ljudevit Jurak University Department of Pathology, Sestre milosrdnice University Hospital, Vinogradska 29, Zagreb, Croatia e-mail:
[email protected] Tel.: +385-1-3787177 Fax: +385-1-3787244 I. Novosel Department of Pathology, Dr Ivo PediÐic´, County Hospital Sisak, Croatia H. Rogatsch · G. Mikuz Department of Pathology, University of Innsbruck, Muellerstrasse 44, Innsbruck, Austria O. Kraus Department of Urology, Sestre Milosrdnice University Hospital, Vinogradska 29, Zagreb, Croatia
Keywords Prostate adenocarcinoma · Histological criteria · Retraction clefting · Needle core biopsy
Introduction Prostatic adenocarcinoma is the most common noncutaneous malignant neoplasm in humans, with an estimated prevalence—based on histological studies—of over 30% in males older than 50 years [12]. The diagnosis of prostatic carcinoma is complex and is based on a constellation of three major histological criteria: the infiltrative growth pattern, the absence of a basal cell layer and the presence of macronucleoli (nucleoli larger than 1 mm in diameter) [11, 12]. The valid clinical strategy of performing biopsies in patients with elevated serum prostate-specific antigen without suspicious digital rectal examination and the exchange of the traditional 14-gauge biopsy needle with an 18-gauge biopsy needle yield less tissue for examination and small foci of cancer, which are often presented with only few glands. Therefore, the histological diagnosis of prostatic adenocarcinoma becomes even more difficult [11]. Several supportive diagnostic criteria have been proposed, but only a minority of these supportive features is specific for tumor. Proposed additional criteria are: marginated nucleoli, multiple nucleoli, wispy, bluish mucinous secretions, intraluminal crystalloids, intraluminal amorphous eosinophilic material, collagenous micronodules, glomerulations, periacinar cleft-like spaces and others, but many of these supportive criteria may also be present in benign glands or in some non-neoplastic proliferative acinar conditions [1, 2, 4, 11, 12]. Few authors paid attention to the so-called periacinar halos, retraction clefting or cleft-like spaces within neoplastic tissue [3, 5, 6, 7, 11]. The neoplastic cells of prostatic cancer often appear pulled away from the surrounding stroma, leaving halos around the acini [5]. Only a limited number of reports have compared and validated retraction clefting as supportive diagnostic
525
features in needle core biopsies in tumor tissue and benign glands [3, 7, 11]. In this study, we analyzed the amount and frequency of retraction clefting in tumors and non-tumorous prostatic tissue in needle core biopsies to estimate diagnostic importance and to define the criteria for the determination of retraction clefting.
Materials and methods The study group consisted of 137 patients who underwent prostatic needle core biopsy during the period from 1 January 2000 to 30 June 2002. Biopsy specimens (18-gauge prostate needle core) diagnosed as cancer from 100 patients at the Department of Pathology, Leopold Franzens University, Hospital Innsbruck, and 37 biopsy specimens diagnosed as a cancer at the Ljudevit Jurak University Department of Pathology, Sestre Milosrdnice University Hospital, were histologically analyzed. The diagnosis of adenocarcinoma was confirmed in all cases, and all tumors consisted of at least 30 or more neoplastic glands. The urologists of both hospitals pull needle core biopsies from different areas of the prostate into 6 to 16 parts, usually maintaining orientation of the side and the part of prostate where biopsies were taken. An improved pre-embedding method was applied in both institutions as recently described [8, 9]. We evaluated each case as a whole, without regard to separate parts. The age range of the patients was between 41 years and 87 years (median age 63.5 years). Level of prostate-specific antigen (PSA) in all patients varied from 0.9 ng/ml to 31.7 ng/ml. At the time of diagnosis, all patients were without lymph-node or distant metastases. Specimens from both hospitals were fixed overnight in 10% buffered formaldehyde, embedded in paraffin, and cut at 5-mm thickness and routinely stained with hematoxylin and eosin. In some cases the material was stained by high molecular weight
Table 1 Distribution of Gleason score among 137 prostatic needle biopsies with prostatic adenocarcinoma Gleason grades with scores
Number of cases
%
4 (2+2) 5 (2+3) 5 (3+2) 6 (2+4) 6 (3+3) 6 (4+2) 7 (3+4) 7 (4+3) 8 (3+5) 8 (4+4) 9 (5+4) Total
3 4 16 1 69 1 25 11 2 4 1 137
2.2 2.9 11.8 0.7 50.4 0.7 18.2 8.1 1.4 2.9 0.7 100
cytokeratin, PSA (MSIP; Dako, Copenhagen, Denmark) and alcianperiodic acid-Schiff. All cancers were graded using the Gleason grading system. Periacinar retraction clefting were graded as a percentage of gland circumference separated from the stroma in the three categories: glands without clefts, glands with clefts up to 50% of circumference and glands with clefts which affected more than 50% of the circumference. Clefting at the edges of a biopsy and cleft-like spaces in glands with incomplete circumference were excluded from examination, as at least some of it was likely to be a consequence of sectioning. We examined ten neoplastic glands in three different fields in needle core biopsy specimens under high magnification (400). We used the same material as a control group, examining ten non-neoplastic glands in three different fields under the same magnification, and compared number of glands with clefts. Cases with ten or more glands with clefts, which affected more than 50% of glands circumference, were considered as positive. After needle core biopsy, the patients underwent prostatectomy and the diagnosis of adenocarcinoma was confirmed by pathohistological analysis in all cases. The difference between the two proportions was tested using test of proportion. For all analyses, type-I error of 5% was considered statistically significant.
Results The Gleason score distribution is shown in Table 1. Of the tumors, 3 (2.2%) were well differentiated, 91 (66.5%) moderately differentiated (Gleason score 5–6) and 43 (31.3%) were poorly differentiated (Gleason score 7–9). However, there were 36 (26.3%) tumors with a Gleason score of 7 and 7 (5%) cases with a Gleason score of 8–9. The most common Gleason grade was 3. In 127 (92.7%) tumors, one or both grades were 3. Both Gleason grades of 3 were found in 69 (50.4%) cases. The distribution of the various degrees of periacinar retraction clefting in the malignant and benign cases is shown in Table 2 and Fig. 1A, B and Fig. 2A, B. Glands (10 or more) with clefts that affected more than 50% of circumference were significantly more common in tumors (51.8%) than in benign glands (8%) (P<0.0001). Glands (10 or more) with clefts, regardless of their extension, were also more frequent in malignant cases (93.4%) as opposed to benign cases (57.7%) (P<0.0001). A stricter criterion that designated as positive cases with at least 50% of glands (15 of 30) with clefts that affected more than 50% of circumference revealed clefts in the malignant cases only (15.3%) but not in benign cases (0%) (P<0.0001). Glands with clefts (15 or more), regardless of their extension, were also more
Table 2 Distribution of periacinar retraction clefting in the malignant and benign cases (n=137) Frequency in malignant cases Glands (10/30 or more) with clefts that affected more than 50% of circumference Glands (10/30 or more) with clefts regardless of their extension Glands (15/30 or more) with clefts that affected more than 50% of circumference Glands (15/30 or more) with clefts regardless of their extension Cases with clefts that affected more than 50% of circumference regardless of the number of positive glands
71 (51.8%)
Frequency in benign cases
Statistical significance (test of proportion)
11 (8%)
Yes (P<0.0001)
128 (93.4%) 21 (15.3%)
79 (57.7%) 0
Yes (P<0.0001) Yes (P<0.0001)
119 (86.9%) 125 (91.2%)
28 (20.4%) 113 (82.5%)
Yes (P<0.0001) Yes (P=0.034)
526 Fig. 1 A Periacinar retraction clefting affecting more than 50% of circumference of neoplastic acini in prostatic adenocarcinoma [hematoxylin and eosin (H&E) 400]. B Neoplastic acini in prostatic adenocarcinoma without retraction clefting (H&E 400)
Fig. 2 A Periacinar retraction clefting affecting more than 50% of circumference of normal prostatic acini [hematoxylin and eosin (H&E) 400]. B Normal prostatic acini without retraction clefting (H&E 400)
prominent in malignant cases (86.9%) than in benign cases (20.4%) (P<0.0001). In all cases showing retraction clefting, at least one Gleason grade was 3. Comparison between malignant and benign cases with clefts which affected more than 50% of circumference regardless of the number of positive glands showed more frequent appearance of cleft in tumor (91.2%) than in non-tumor (82.5%) specimens (P=0.034). There was not a relationship between PSA value and the extent of retraction clefting.
Discussion There are three histological criteria that are considered as diagnostic for prostatic adenocarcinoma on needle biopsy: perineural invasion, mucinous fibroplasias and glomerulations [1]. However, these features are very rarely seen. According to Baisden et al. [1], perineural invasion was observed in 9 of 1480 consult cases (0.6%), fibroplasias were observed in 0.1% of cases and glomerulations were not seen. Other authors reported perineural invasion in 2– 3% of cases [4, 10]. Thorson et al. [10] reported the most common morphological features of minimal carcinomas (less than 1 mm) in needle biopsy to be nucleomegaly (96%), infiltrative growth pattern (88%), intraluminal mucinous secretions (78%) and prominent nucleoli (64%). Helpful diagnostic features reported by Epstein
[4] were nuclear enlargement in 77%, prominent nucleoli in 76% and pink acellular luminal secretions in 53% of cases. Mean and median number of glands in Epstein’s study were 31 and 20, respectively (range 2–300). Many additional criteria for the diagnosis of prostatic carcinoma-like marginated nucleoli, multiple nucleoli, intraluminal crystalloids, intraluminal amorphous eosinophilic material, and others are used in routine diagnostic procedure [2, 11, 12]. One of the proposed additional criteria is a so-called retraction clefting, appearing more frequently about neoplastic glands in comparison with benign glands; but only few authors paid attention to this phenomenon [5, 6, 11]. Retraction clefting might also be observed in other tumor types, representing a well-known diagnostic criterion between basal cell carcinoma and adnexal skin tumors. Halpert et co-workers were the first to briefly describe the clefts in prostatic adenocarcinoma [5, 6]. In their autopsy studies of prostatic adenocarcinoma, halos around neoplastic acini were observed. In a recent study by Amin and his group, retraction clefting was found in 38.6 cases of prostatic cancer diagnosed by needle core biopsy and in 7% of non-neoplastic glands [11]. They graded retraction clefting as a percentage of gland circumference separated from the stroma using a scale from 0 to 4. The same authors considered grades 3 (50–75%) and 4 (more than 75%) as positive. However, the authors did not specify the number of neoplastic glands showing clefting
527
that was regarded as positive. They also did not mention under which magnification the clefts were analyzed. In the current study, we analyzed clefting under high magnification (40 objective) counting ten glands in three different fields. Clefting could be observed under lower magnification but it is obvious that the same magnification should be used by different observers for the sake of comparison. Our results showed that clefting affecting more than 50% of circumference in more than 50% of analyzed glands was found in neoplastic glands only. Such strict criteria were fulfilled in 15.3% of all cancer cases and were not observed in normal glands, suggesting the potential use of this criterion in routine diagnostics. The application of less stringent criteria (one-third of glands with more than 50% of circumference) revealed positive clefting in 51.8% of carcinoma cases and 8% of non-neoplastic glands (P<0.0001). The presence of clefting in prostatic needle core biopsy is usually attributed to the retraction artifact. Therefore, the quantity of clefting is considered as an indicator of quality of technicians and laboratory staff. In this study, we tried to quantify the extent of periacinar clefting. To minimize the influence of laboratory procedure, an improved pre-embedding method of prostatic needle core biopsy specimens was employed in both institutions as recently described by Rogatsch et al. [8, 9]. The origin of clefting in tumor specimens is unknown but may be related to an abnormality in the expression of collagenases or some other enzymes required for invasion. As it is known, malignant glands lack basal cells, and this also may be one of the reasons for appearance of clefts. The most prominent clefts were found in Gleason grade-3 prostatic carcinoma, especially in Gleason grade 3A, as was also observed in our study [7, 12]. Our criteria may be applied in a similar manner to carcinoma cases composed of few neoplastic glands. However, the cut-off value should be determined in the future studies. We propose that the presence of retraction clefting in at least one-third of glands affecting more than 50% of circumference along with one of other supportive criteria and/or clefting in at least 50% of glands with more
than 50% of circumference should be considered as a criterion for the diagnosis of prostatic adenocarcinoma. Similar clefting changes should not be present in adjacent normal glands within the same specimens. Further studies regarding the determination of reliability of the proposed criterion and its refinement are certainly welcome.
References 1. Baisden BL, Kahane H, Epstein JI (1999) Perineural invasion, mucinous fibroplasia and glomerulations: diagnostic features on prostate needle biopsy. Am J Surg Pathol 23:918–924 2. Bostwick DG, Dundore PA. (1997) Biopsy pathology of the prostate. Chapmann & Hall Medical, London 3. Durham JR, Lee MW, Amin MB, Shulz DS, Stetler-Stevenson WG (1994) Clefting in adenocarcinoma of the prostate gland (abstract). Am J Clin Pathol 102:536 4. Epstein JI (1995) Diagnostic criteria of limited adenocarcinoma of the prostate on needle biopsy. Hum Pathol 26:223–229 5. Halpert B, Schmalhorst WR (1966) Carcinoma of the prostate in patients 70 to 79 years old. Cancer 19:695–698 6. Halpert B, Sheehan EA, Schmalhorst WR, Scott RJ (1963) Carcinoma of the prostate: a survey of 5000 autopsies. Cancer 16:736–742 ˇ upic´ H, Spajic´ B, Belicza M 7. KruÐlin B, Novosel I, Reljic´ A, C (2002) Periacinar cleft-like spaces in prostatic needle core biopsies (abstract). Acta Clin Croat 41:175 8. Rogatsch H, Mairinger T, Horninger W, Gschwendtner A, Bartsch G, Mikuz G (2000) Optimized preembedding method improves the histologic yield of prostatic core needle biopsies. Prostate 42:124–129 9. Rogatsch H, Moser P, Volgger H, Horninger W, Bartsch G, Mikuz G, Mairinger T (2000) Diagnostic effect of an improved preembedding method of prostate needle biopsy specimens. Hum Pathol 31:1102–1107 10. Thorson P, Vollmer RT, Arcangeli C, Keetch DW, Humphrey PA (1998) Minimal carcinoma in prostate needle biopsy specimens: diagnostic features and radical prostatectomy follow-up. Mod Pathol 11:543–551 11. Varma M, Lee MW, Tamboli P, Zarbo RJ, Jimenez RE, Salles PGO, Amin MB (2002) Morphologic criteria for the diagnosis of prostatic adenocarcinoma in needle biopsy specimens. Arch Pathol Lab Med 126:554–561 12. Young RH, Srigley JR, Amin MB, Ulbright TM, Cubilla AL (1998) Atlas of tumor pathology. AFIP, Washington, D.C., pp 124–171