Search

Context: Dysregulated cell proliferation caused by inhibitors of programmed cell death (apoptosis) contributes to tumor progression and spread. Aberrant expression of Bcl-2, the most notable inhibitor of apoptosis, has been well characterized in several human malignancies. Recent studies have described a novel apoptosis inhibitor, survivin, in human carcinomas, although its exact role remains to be characterized., Objective: The purpose of this study was to evaluate the immunohistochemical expression of Bcl-2 and survivin proteins in prostate cancer and to correlate the results with clinicopathologic variables., Design: Formalin-fixed, paraffin-embedded tissue sections from 138 cases of prostatic adenocarcinomas (PACs) were immunostained by an automated method using specific antibodies against survivin and Bcl-2. Staining was semiquantitatively scored based on both intensity and distribution, and results were correlated with morphologic and prognostic variables., Results: Of the 138 PACs tested, 113 (82%) expressed survivin. We found no correlation between survivin expression and prognostic variables, including grade, stage, DNA content (ploidy), and recurrence. Bcl-2 expression was positive in 95 (69%) of these 138 cases and correlated with nondiploid DNA content. Fourteen (50%) of 28 nondiploid PACs expressed Bcl-2, compared to 17 (25%) of 68 diploid tumors (P =.02). A trend for association of Bcl-2 expression with tumor stage was noted as follows: 21 (39%) of 54 advanced-stage PACs expressed Bcl-2, in comparison with 20 (24%) of 84 low-stage tumors (P =.07). On univariate analysis, 25 (48%) of the 52 PACs that recurred expressed Bcl-2, as compared with 16 (19%) of the 86 nonrecurrent PACs (P <.001). No correlation was noted between survivin and Bcl-2 expression., Conclusion: Survivin is expressed in a majority of PACs and is not a prognosis-related marker, but may be a potential target for apoptosis-based therapy. Overexpression of Bcl-2 correlates with other prognostic variables and predicts disease recurrence of PACs. These data also suggest that survivin and Bcl-2 may regulate cell proliferation and cell death through different mechanisms.

DNA ploidy analysis of five renal oncocytomas, six pure oncocytic-granular renal cell carcinomas, 15 pure clear cell renal carcinomas, and two cases of mixed oncocytic-granular and clear cell heterogenous renal cell carcinomas were determined on Feulgen-stained paraffin sections using the (CAS-200 image Analyzer). All five renal oncocytomas were diploid, as were six oncocytic-granular renal cell carcinomas. Three of the 15 clear renal cell carcinomas were aneuploid. In one heterogeneous renal cell carcinoma, the oncocytic-granular foci were diploid and the clear cell focus was aneuploid. The other heterogeneous renal cell carcinoma was uniformly diploid. In 21 renal cell carcinomas, one of 14 stage I tumors was aneuploid, all four stages II and III tumors were diploid, and two of three stage IV cases were aneuploid. All stages I, II, and III patients were free of disease 3 to 48 months after surgery. All three stage IV cases were dead of their disease within 36 months of surgery. We conclude that DNA analysis by image cytometry best identifies the heterogeneity of ploidy patterns in mixed cell type carcinomas, generally correlates with the stage of disease, and may be of value in predicting overall prognosis of renal epithelial neoplasms. DNA analysis by image cytometry, however, did not reveal significant differences between oncocytic granular cell carcinoma and oncocytoma, since both lesions were generally diploid. The distinction between the two tumors is best made on morphologic grounds.