HistopathoZogu 1992,21,249-255

A major solid undifferentiated carcinoma pattern correlates with tumour progression in locally advanced prostatic carcinoma E.F.GAFFNEY, S.N.O’SULLIVAN & A.O’BRIEN* Department of Histopathology, St James’s Hospital and Trinity College Medical School and *Urology Department, Meath Hospital, Dublin, lreland Date of submission 18 November 199 1 Accepted for publication 16 March 1992

GAPFNEY E . P . , O’SULLIVANS . N . & O’BRIEN A .

(1992) Histopathology 21, 249-255

A major solid undifferentiated carcinoma pattern correlates with tumour progression in locally advanced prostatic carcinoma Solid undifferentiated carcinoma was the major microscopic pattern in 24 prostatic carcinomas, 12 of which were clinically recurrent. Tumour cells were uniform, with moderately hyperchromatic nuclei and indistinct cytoplasm, and were arranged in solid or focally irregular aggregates. In areas, the tumour cells were large with vesicular nuclei, nucleoli and more abundant cytoplasm. In previous specimens, solid undifferentiated carcinoma was absent or was a minor pattern. Twenty of 23 cases showed prostate specific antigen and prostatic acid phosphatase immunoreactivity, and nine of 1 7 cases contained scattered argyrophilic or chromogranin-immunoreactivecells. On proliferating cell nuclear antigen immunostainiig of 12 specimens, the mean tumour proliferative fraction in solid undifferentiated carcinoma (range: 10.5-18%) was greater than in areas of grade 3 prostatic carcinoma (range: 3-6%). In all 22 stage C and D cases, there was a close correlation with clinical evidence of tumour progression, and the overall 2-year survival rate was only 16.7%.It is concluded that a major solid undifferentiated pattern correlates with increased biological aggressiveness and a poor prognosis in locally advanced prostatic carcinoma. Keywords: prostate, solid undifferentiated carcinoma, tumour progression, tumour proliferative fraction, PCNA (proliferating cell nuclear antigen)

Introduction In the pathological evaluation of prostatic carcinoma, new approaches including volumetric mea~urementsl-~, computer-assisted morphometric techniques4,quantitative nuclear measurements5 and DNA ploidy analysis6 have all been reported to provide more accurate prognostic information than histopathological grading alone. However, the evaluation of histopathological pattern is particularly important in the identification of certain unusual and distinctive types of prostatic carcinorna7v8 that are consistently associated with a poor prognosis. We report the clinical and pathological findings in a group of patients whose prostatic tumours consisted predominantly of solid undifferentiated carcinoma. We Address for correspondence: Dr E.P.Gaffney,Department of Histopathology, St James’sHospital, Dublin 8. Ireland.

found that the solid undifferentiated pattern correlated with tumour progression and a poor prognosis in locally advanced prostatic carcinoma.

Materials and methods We examined the microscopic slides of 500 consecutive prostatic carcinomas, transurethrally resected at the Meath and St. James’sHospitals, and accessionedby The Central Pathology Laboratory, St James’s Hospital, during 1984-1988. There were no radical prostatectomy specimens. Needle biopsy specimens were not evaluated in the selection of cases. In 2 4 cases, solid undifferentiated carcinoma, Gleason grade SA, was the major microscopic pattern, comprising at least 50% of each tumour. The slides of previous transurethral prostatic resection specimens were also reviewed. From selected blocks, further sections were cut and stained 249

250

E.F.Guflney, S.N.O'Sullivun and A.O'Brien

with the Churukian-Schenk argyrophil staing. Sections were also stained immunocytochemically with antisera to prostate specific antigen (PSA) and prostatic acid phosphatase (PSAP),both from Dakopatts, Denmark, and to chromogranin from Lipshaw, Detroit, USA. Two cases, initially fixed in 10%formalin, were subsequently placed in 2% glutaraldehyde and processed for electronmicroscopy. Paraffin sectionsfrom 12 specimens were stained with a monoclonal antibody to proliferating cell nuclear antigen (PCNA) (Dakopatts,Denmark), using a dilution of 1:100, and incubating sections overnight at 4OC. A standard avidin-biotin complex method (individual reagents from Dakopatts, Denmark) was used, and sections were counterstained with extracted methyl green. Four of the blocks selected were from previous transurethral prostatic resection specimens which showed focal grade 3 prostatic adenocarcinoma. Tumour proliferative fractions were estimated by counting the total number of positively stained nuclei in five different fields of 100 tumour cells using the 40 x objective and an eyepiece with a central frame. Adenocarcinoma in previous TURP Case no.

Age

Stage

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

68 93 88 79 84 86 83 70 62 69 76 58 86 74 89 65 70 66 69 57 59 72 73 77

-

Major grade

D B A

-

D C

C

-

C C A

D C D

"URP =transurethral resection of prostate.

The clinical records of all 24 patients were reviewed. Specificallyrecorded were the date and clinical stageloat identification of a major solid undifferentiated carcinoma pattern (and of initial prostatic adenocarcinoma where applicable), management, clinical course and current status. Patients' general practitioners and families supplied additional information. Autopsy slides were reviewed in one case.

Results CLINICAL FINDINGS

All 24 patients with solid undifferentiated carcinoma presented with severe or total outflow obstruction requiring transurethral resection of the prostate. There were two stage B, nine stage C and 13 stage D cases (Table 1).Adenocarcinoma of the prostate had already been diagnosed in 12 cases, 1-110 months (mean: 35 months) previously. Eight were stage C or D initially, and showed no increase in stage at recognition of solid

Interval (months) before recognition of SUC

SUC stage

B D B D D D D C C D C D C D C C C D C D

D D C

D

survival (months) alive (66) died (7) alive (14) died (23) died (2) died (6) died (2) died (29) died (25) died (5) died (6) died (13) died (2) died (5) died (8) died (20) died (18) died (6) died (6) died (2) died (13) died (17) died (12) died (7)

Table 1. Clinical findings in 24 patients with solid undifferentiated carcinoma (SUC) of the prostate.

Tumour progression in prostatic carcinoma

undifferentiated carcinoma, and four showed a progression in clinical stage. In the remaining 12 cases, solid undifferentiated carcinoma was present initially. Eleven of the 12 patients with a previous diagnosis of prostatic adenocarcinoma were managed initially by hormonal manipulation (either stilboestrol or castration). Case 20, who had a stage A tumour initially, did not receive hormonal treatment. Following transurethral resection for solid undifferentiated carcinoma, 2 1 patients received some form of hormonal manipulation: stilboestrol (15), bilateral orchiectomy (3), leuprolelin acetate (l),cyproterone acetate (1) and flutamide (1). Two patients also had external beam irradiation to the prostate and seven patients required a total of 1 2 further transurethral resections for complete outflow obstruction due to tumour progression. Three patients received no specific treatment. Twenty-one of 24 patients died of their disease and one patient with extensive residual disease died of a stroke. The two stage B patients are alive and asymptomatic, at 14 and 66 months. The 2-year survival rates for stage C and stage D solid undifferentiated carcinoma patients were 22% and 0 respectively, and the overall 2year survival rate for the 24 patients was 16.7%.

Figure 1. Solid undiererentiated prostatic carcinoma. Monotonous picture of evenly-spaced turnour cells with uniform moderately hyperchromatic nuclei. H & E. x 450. Figure 2. Endocrine-lie irregular tumour cell aggregates beneath normal prostatic urethral rnucosa. H & E. x 450.

2 51

PATHOLOGICAL FINDINGS

In all cases, the predominant or only prostatic carcinoma pattern consisted of closely apposed solid aggregates of undifferentiated tumour cells (Figure 1)which at low magnification simulated either transitional cell carcinoma or neuroendocrine carcinoma. Tumour cell aggregates were usually round or oval, but in most cases there were focal irregular or insular clusters (Figure 2). There was no desmoplastic reaction or infiltration by individual tumour cells. Diathermy-induced retraction spaces between tumour cell aggregates, and the suscep tibility of nuclei to a 'stringing' artifact were common (Figure 3). The evenly-spaced tumour cells presented a monotonous appearance, with uniform moderately hyperchromatic nuclei, and a small amount of indistinct cytoplasm. However, in 20 cases there were occasional foci in which the tumour cells had more eosinophilic cytoplasm and formed abortive microacini (Figure 4). There were occasional abrupt transitions to small or large zones in which tumour cells had more copious cytoplasm, vesicular nuclei and prominent nucleoli. Mitoses, which were generally infrequent, were more numerous in such areas (Figure 5). In all cases, the

252

E.F.Gaffneu, S.N.O’Sullivan and A.O’Brien

Figure 3, Diathermy-inducedretraction spaces and ‘stringing’artifact. H & B. x 250. Flgure 4. Abortive micmacinl in solid undifferentiatedcarcinoma. H & B. x 400.

Figure 5. Usual nuclear morphology (left) with abrupt transitionto area containing large tumour cells with prominent nucleoli. Note mltoses. right. H & E. x 400. Figure 6. Prostate specific antigen Lwnunoreactivity mainly conhed to central tumour cells in this field. PAP. x 400.

prostatic urethral and bladder neck urothelium was normal (Figure 2). except where ulcerated by invasive solid undifferentiated carcinoma. None of the cases had evidence of transitional cell carcinoma of the bladder, prostatic urethra or large prostatic ducts, In two cases, solid undifferentiatedcarcinoma was the only microscopic pattern identified, and all cases had a combined Gleason grade of 8-10. In 14cases, there was focal comedo carcinoma or micro-acinar cribiformcarcinoma, and in three, a minor component of small cell carcinoma. In eight of 12 previous adenocarcinomas, the major Gleason pattern was a different grade 5 pattern. In six cases, solid undifferentiated carcinoma constituted 5 2 5 % of the original tumour, and in a seventh, 25-50%. Specimens in which solid undifferen-

tiated carcinoma was the main Gleason pattern were comprised almost entirely of tumour, with two exceptions: in specimens from the two stage B patients, tumour comprised only 50% of the specimen. There was focal or weak PSA and PSAP immunoreactivity (Figure 6) in 20 of 23 cases (blocks from case 1 were not available). The majority of tumour cells were negative. In six of 17 cases, occasional tumour cells exhibited argyrophilia with the Churukian-Schenk stain, and six of 12 contained scattered tumour cells immunoreactive for chromogranin (Figure 7). Overall, nine of 17 showed either argyrophilia or chromogranin immunoreactivity. In sections stained immunocytochemicallyfor PCNA, the mean turnour proliferative fraction in solid undiffer-

Tumour progression in prostatic carcinoma

2 53

conspicuous and carcinoma in situ may be identified. The presence of small foci of a recognizable prostatic carcinoma pattern such as cribriform carcinoma, and PSA or PSAP immunoreactivity, which is not always present in poorly differentiated prostatic carcinoma12J9, are helpful in distinguishingthese tumours. Clinical and cytoscopic findings also help to clarify doubtful cases. Small cell carcinomas are typically patternless, but small trabecular or insular foci may be seen8.20.21.Zonal necrosis, and necrosis of individual tumour cells with Figure 7. Scattered argyrophilictumour ceh and chromograninkaryorrhexis are prominent features. The nuclei are hrnunoreactivetumour cells in solid undifferentiated carcinoma. deeply hyperchromatic, often appearing to overlap21. a Churukian-Schenk stain. x 250. b Chromogranin. PAP. x 320. Argyrophilic cells are commonly absent in prostatic small cell carcinoma20-22 and immunocytochemical exentiated carcinoma ranged from 10.5 to 18%. The pression of neuroendocrine differentiation is variable22. percentage of positive nuclei varied only slightly from The homogeneous microscopic appearance of solid area to area in each section. The few areas of grade 3 undifferentiated carcinoma may suggest a prostatic carcinoma that were assessed had a mean proliferative carcinoma with neuroendocrine feature^^^-^^. However, fraction that ranged from 3 to 6%. Staining in grade 3 the scattered argyrophilia and chromogranin immunoareas was more variable than in solid areas. These reactivity that we identified in 9/17 cases is within the recorded differences in proliferative fractions could be range previously reported in prostatic carcinomas of all appreciated at intermediate magnification. grades26-28.Furthermore, it is well known that certain Ultrastructural examination showed polygonal cells prostatic carcinomas have a carcinoid-lie appearance, with closely apposed cell membranes and no evidence of but lack immunocytochemical or ultrastructural eviacini or neurosecretory granules. dence of neuroendocrine differentiation’’. At autopsy, case 14,which had initially been misA predominant solid Undifferentiated carcinoma patdiagnosed as transitional cell carcinoma of the bladder, tern was identied initially in 1 2 of our cases, and at the had extensive residual tumour in the prostate, trigone, time of clinical recurrence (after a median interval of 35 posterior wall of bladder and around both ureteric months) in the remaining 12 cases. Gleason maintained orifices, associated with obstructive uropathy. There that the grade of a tumour rarely alters with tiie13, but were metastases in para-aortic lymph nodes, liver, lungs, Brawn found that most prostatic carcinomas tend to progress to a higher grade with tie29*30. The present ribs and vertebrae. series confirms that such altered differentiation may indeed occw in certain cases, and furthermore suggests Discussion that the solid undifferentiated pattern may be a marker of ‘functional dedifferentiation’, with (acquired) lack of In this report, we have examined the clinical and pathological features of a group of patients whose hormonal responsiveness and altered, more aggressive prostatic cancers consisted predominantly of solid undifbiological behaviour. The presence of solid undifferenferentiated carcinoma. The tumour cells typically had tiated carcinoma can be rationalized by the genetic l . ~ ~potentially may instability of tumour ~ e l l s ~which uniform, moderately hyperchromatic nuclei and indistinct cytoplasm, as previously illustrated by others11*12. proliferate as heterogeneoussubpopulations, and exhibit variable or selective metastatic potential33.The emerIn areas, too, the tumour cells had vesicular nuclei, gence of an aggressive clone in an enlarging tumour nucleoli and conspicuous cytoplasm: these cytological may be predicated by gene and increased features predominate in most illustrations of the solid EGP receptor mRNA levels, favouring autocrine tumour undifferentiated (Gleason grade 5A) patternI3s1*. cell growth34. Previous therapy alone is an unlikely Solid undifferentiated carcinoma of the prostate must s t i m u l ~ s ’ ~ . However, ~~. the factors involved in the be distinguished from transitional cell carcinoma, small altered biological aggressiveness of prostatic carcinoma cell carcinoma of the prostate and bladder, and prostatic are poorly understood. carcinoma with neuroendocrine features. The tumour We found that the tumour proliferative fraction in cells in transitional cell carcinoma, including the controsolid undifferentiated carcinoma, as determined by the versial primary transitional cell carcinoma of the propercentage of PCNA-immunoreactive hunour cells in state’ 5-18, are large and usually have pleomorphic nuclei paraffin sections, was greater than in areas of grade 3 and abundant eosinophilic cytoplasmll. Mitoses are

254

E.F.Gaffey, S.N.O’Sullivan and A.O’Brien

prostatic adenocarcinoma. Although the proliferative fraction represents the percentage of proliferating cells and is not synonymous with a tumour’s growth rate3’, the increased proliferative fraction observed in solid undifferentiated carcinoma is consistent with its undifferentiated morphology and poor prognosis. This contrasts with the better differentiation, slow tumour growth and hormone-responsiveness of most prostatic carcinomas. Our PCNA staining results are not strictly comparable with growth fraction measurements using Ki-67 on frozen sections, but an increased growth fraction has been reported with higher grade tum0u1-s~~ and with more rapid tumour progression3’. It is suggested that an increased growth fraction correlates with a lack of response to hormonal manipulation and that low values are associated with hormonally-responsive t~mours~~. In this series, solid undifferentiated prostatic carcinoma was temporally associated with tumour progression in all but the two stage B cases. Most patients with a previous diagnosis of prostatic carcinoma had initially responded to hormonal manipulation, and had been stable prior to clinical recurrence. However, there was a negligible response to hormonal manipulation following the second transurethral resection in these 12 cases, all then stage C or D: likewise, the 10 stage C or D cases with a predominant solid component in their initial resections did not respond to hormonal therapy. The 2-year survival figures reported herein-22% for stage C and 0 for stage D-are much poorer than those previously reported for stages C and D prostatic carcinoma patients However, a major overall: 80%and 64% respe~tively~~. limitation with clinical reviews of prostate cancer is that patients are seldom stratified by parameters other than clinical stagem. Our results c o n h that consideration of histological features (including grade) together with clinical stage provides a more realistic assessment of the risk of tumour progre~sionll-’~*~~. In conclusion, we found that a predominant solid undifferentiated carcinoma pattern was associated with a poor prognosis in prostatic carcinoma. However, this microscopic pattern is neither required for, nor invariably associated with, tumour progression. Future studies might examine additional stage B cases and tumours in which the solid pattern is a minor component. Another approach would be to examine prostatic carcinoma specimens obtained at the onset of, or during, tumour progression. Although numerous c l i c a l trials are currently evaluating the efficacy of second-line hormonal manipulation, chemotherapy and biological response modifiers in such patients, there is little information on the pathology of prostate cancer in the preterminal stage.

Acknowledgements The authors thank Orla Sheils for performing argyrophil and immunocytochemicalstains, David Grehan for antiPCNA stains and Ronan Ward for photographic assistance. Mrs Kathryn Kiely prepared the manuscript. Supported in part by a summer student grant to Sinkad O’Sullivan from The Health Research Board (Ireland).

References 1. Partin AW, Epstein JI. Cho KR et al.Morphometric measurement of tumor volume and per cent of gland involvement as predictors of pathological stage in clinical stage B prostate cancer. 1. Urol. 1969: 141; 341-345. 2. Humphrey P, Vollmer RT. The ratio of prostate chips with cancer: a new measure of tumor extent and its relationship to grade and prognosis. Hum.Pathol. 1988: 19; 41 1-418. 3. McNeal JE. Vlllers, AA. Redwine EA et al. Hlstologicdifferentiation. cancer volume, and pelvic lymph node metastasis In adenocarcinoma of the prostate. Cancer 1 9 9 0 66; 1225-1233. 4. Foucar E. Haake G. Dalton L et al. The area of cancer in transurethral resection specimens as a prognostic indicator in carclnoma of the prostate. A computer-assisted morphometric study. Hum. Pathol. 1 9 9 0 21; 586-592. 5. Epstein JI. Berry SJ,Eggleston JC.Nuclear roundness factor: a predictor of progression in untreated stage A2 prostate cancer. Cancer 1984: 54; 1666-1971. 6. Badalament RA. OToole RB, Young DC et al. DNA ploidy and prostate-specific antigen as prognostic factors in clinically resectable prostate cancer. Cancer 1991: 67; 3014-3023. 7. Ro JY,el-Naggar A, Ayala AG et al. Signet ring carcinoma of the prostate. Electron-mtcroscopic and immunohistochemicalstudies of eight cases. Am.1. Surg. Pathol. 1988: 12: 453-460. 8. Tetu B. Ro IY.Ayala AG et al. Small cell carcinoma of the prostate: Part I. A cllnicopathologic study of 2 0 cases. Cancer 1987; 59; 1803-1809. 9. Churukian CJ.Schenk EA. A modificationof Pascual’s argyrophil method. 1. Hfstotechnol. 1979: 2; 102-103. 10. Whitmore WF. Jr. Natural history and staging of prostate cancer. Urol. Clin. North Am. 1984: 2: 205-220. 11. Epstetn JI. Prostate BiopsM Interpretation. New York Raven Press, 1989: 39-149.169-193,205-217. 12. Feiner H. Gonzalez R. Carcinoma of the prostate with atyplcal immunohistological features. Clinical and histologic correlates. Am. 1. Surg. Pathol. 1 9 8 6 10; 765-770. 13. Gleason DF, The Veterans Administration CooperativeUrological Research Group. Histologic grading and cllnlcal staging of prostatic carcinoma. In Tannenbaum M ed. Urologic Patholog#: The Prostate. Philadelphia: Lea & Pebiger, 1977: 171-198. 14. BrandesD. Prostate carcinoma. In Hill GS ed. Uropatholog~.Vol2. New York Churchill Livingstone, 1989: 1203-1258. 15. Ende N, Woods LP. Shelley HS. Carcinoma originating in ducts surrounding the prostatic urethra. Am. 1. Clfn. Pathol. 1963; 40; 183-189. 16. johnson DE, Hogan JM.Ayala AG. Transitional cell carcinoma of the prostate. A clinical morphological study. Cancer 1972: 29: 28 7-29 3. 17. Sawczuk I, Tannenbaum M, Olsson CA. de Vere White R. Primary transitional cell carcinoma of prostatic periurethral ducts. UroIogg 1985: 25; 339-343.

Tumour progression in prostatic carcinoma

18. Matzkin H, SolowayMS,Hardeman S. Transitionalcell carcinoma of the prostate. 1. Urol. 1991: 146: 1207-1212. 19. Svanholm H. Evaluation of commercial immunoperoxidase kits for prostate specffic antigen and prostate specffic acid phosphatase. Acta Path. Microbiol. Jmmunol. Scad. ( A ) 1986: 94; 7-12. 20. Schron DS, Gipson T, Mendelsohn G. The histogenesisof small cell carcinoma of the prostate. Ah immunohistochemical study. Cancer 1984; 53: 2478-2480. 21. Mills SE, Wolfe IT, Weiss MA et d. Small cell undifferentiated carcinoma of the urinary bladder. A light-microscopic, immunocytochemical, and ultrastructural study of 12 cases. Am. 1. Surg. Pathol. 1987: 11: 606-617. 22. Ro JY, Tetu B, Ayala AG, Ordonez NG. Small cell carcinoma of the prostate: Part II. Immunohistochemicaland electron microscopic studies of 18 cases. Cancer 1987: 59; 977-982. 23. Ghali V. Garcia RL. Prostatic adenocarcinoma with carcinoidal features producing adrenocorticotropic syndrome. Immunohistochemical study and review ofthe literature. Cancer 1984: 54; 104 3-1 048. 24. Almagro UA. Argyrophilicprostatic carcinoma. Case report with literature review on prostatic carcinoid and ‘carcinoid-like’prostatic carcinoma. Cancer 1985: 55; 608-614. 25. RojasCorona RR. Chen L, Mahadevia PS. Prostatic carcinoma with endocrine features. A report of a neoplasm containing multiple immunoreactive hormonal substances. Am. 1. Clin. Pathol. 1987: 88: 759-762. 26. di Sant’AgnesePA, de Mesy Jensen KL. Neuroendocrine differentiation in prostatic carcinoma. Hum. Pathol. 1987: 18: 849-856. 27. Abrahamsson P-A. Wadstrom LB. Alumets J et al. Peptide hormone- and serotonin-immunoreactive tumour cells in carcinoma of the prostate. Path. Res. Pract. 1987: 182: 298-307. 28. &hen RJ, Glezerson G. Haffejee Z. Neuro-endocrinecells-a new prognostic parameter in prostate cancer. Br. 1. Urol. 1991; 68: 258-262.

255

29. Brawn PN. The dedmerentiation of prostate carcinoma. Cancer 1983: 52; 246-251. 30. Brawn PN. Speights VO. The demerentiation of metastatic prostate carcinoma. Br. 1. Cancer 1989: 59: 85-88. 31. Nowell PC. The clonal evolution of tumor cell populations. Science 1976: 194; 23-28. 32. Hill RP. Tumor progression: potential role of unstable genomic changes. Cancer Metast. Rev. 1990: 9: 137-147. 3 3. Guthman DA. Farrow GM, Myers RP et al. Adenocarcinomaof the prostate involving 2 cell types(prostatespeciEc antigen producing and carcinoembryonicantigen producing)with selective metastatic spread. J. Urol. 1991: 146: 854-856. 34. Moms GL. Dodd JG. Epidermal growth factor receptor mRNA levels in human prostatic tumors and cell lines. 1. Urol. 1 9 9 0 143: 1 2 72-1 2 74. 35. Hall PA. Levison DA. Review: Assessment of cell proliferation in histological material. 1. Clin. Pathol. 1990; 43; 184-192. 36. Raymond WA, Leong AS-Y, Bolt J et a!. Growth fractions in human prostatic carcinoma determined by Ki-67 immunoStaining. 1. Pathol. 1988; 156: 161-167. 37. Sadi MV. Barrack ER. Determination of growth fraction in advanced prostate cancer by Ki-67 immunostaining and its relationship to the time to tumor progression after hormonal therapy. Cancer 1991: 67; 3065-3071. 38. Oomens EHGM. van Steenbrugge GJ. Van der Kwast Th H Schroder FH. Application of the monoclonal antibody Ki-67 on prostate biopsies to assess the proliferativecell fraction of human prostatic carcinoma. J. Urol. 1991: 145; 81-85. 39. Blackard CE, Doe RP. Mellinger GT. Byar DP. Incidence of cardiovasculardisease and death in patients receiving diethylstilbestrol for carcinoma of the prostate. Cancer 1970: 26: 249-256. 40. Walsh PC. Lepor H. The role of radical prostatectomy in the management of prostatic cancer. Cancer 1987: 6 0 526-537.

A major solid undifferentiated carcinoma pattern correlates with tumour progression in locally advanced prostatic carcinoma.

Solid undifferentiated carcinoma was the major microscopic pattern in 24 prostatic carcinomas, 12 of which were clinically recurrent. Tumour cells wer...
2MB Sizes 0 Downloads 0 Views