CORRELATION OF SURVIVAL WITH QUANTITATIVE TISSUE-STAINING OF PROSTATE-SPECIFIC ACID-PHOSPHATASE IN PATIENTS WITH PROSTATE CARCINOMA BY USING MICROSCOPIC IMAGE-ANALYSIS - A PRELIMINARY-REPORT OF CORRELATIVE STUDIES ON RTOG-PROTOCOL-75-06, RTOG-PROTOCOL-77-06, AND RTOG-PROTOCOL-83-07

Purpose: We have previously shown that the intensity (graded semiquant itatively as 1-4+) of tissue prostate-specific acid phosphatase (PSAP) staining determined immunocytochemically in a cohort of prostate carc inoma patients from Radiation Therapy Oncology Group Protocols (RTOG) protocols 75-06 and 77-06 correlated with survival, The extent of this staining was heterogeneous and was estimated. The extent of staining was not found to be significantly associated with survival. We underto ok the present quantitative study to see if the improved precision and reliability of measurement of the intensity and extent of prostate sp ecific acid phosphatase staining would confirm and extend our previous observations. Methods and Materials: Patient cohorts representative o f the entire group were obtained from RTOG 75-06 plus 77-06 and 83-07. The RTOG 77-06 plus 75-06 patients (No-Hormone population) did not re ceive preradiation hormonal therapy. RTOG 83-07 patients (Prehormone p opulation) received one of two types of preradiation chemical androgen ablation. In this study, histologic slides of tumors were immunocytoc hemically stained for PSAP by the peroxidase-antiperoxidase (PAP) tech nique using diaminobenezidene (DAB) as a substrate and hematoxylin as a nuclear counterstain. The intensity and extent of immunocytochemical PSAP staining (IPSAP stain) was quantified using our dual wavelength and batch mode image process technique. Results: Our study of 151 case s confirmed that overall survival of patients in both populations was positively correlated with the intensity and extent of IPSAP stain. Re sults of the two studies were similar, The statistical significance of the relationship of both extent and intensity was greater in the coho rt from protocol 83-07, which was the patient group receiving pretreat ment with hormones. In a Cox multiple regression analysis including cl inical stage, Gleason and M. D. Anderson grades, and the cohort of pat ients (Prehormone or No-Hormone group) as covariables, both the intens ity and extent of the IPSAP stain significantly correlated with surviv al along with M. D. Anderson grade of the tumor. Conclusion: Quantitat ive image analysis of the IPSAP stain predicts survival in patients tr eated with external beam radiotherapy with and without prior hormonal therapy.