The Prognostic Value of p16 Expression in Locally Advanced Prostate Cancer: A Study Based on RTOG 92-02

Purpose/Objective: The retinoblastoma (RB) pathway is known to be deregulated in virtually all known human tumors. p16, the upstream regulator of RB, is among the most commonly affected members of this pathway. Previously we reported that loss of p16 expression was associated with adverse clinical outcome (J.Clin. Oncol., 21:3328–34, 2003). In the present study, we examined the prognostic value of p16 expression in men with locally advanced prostate cancer who were enrolled on RTOG 9202. Materials/Methods: Of the 1514 eligible cases, 612 patients had adequate tumor material for p16 analysis. Of these, 327 were randomized to LTAD (long-term androgen deprivation)+RT and 285 were randomized to STAD (short-term androgen deprivation) +RT. Expression levels of p16 were determined by immunohistochemical staining using anti-p16 antibody (Santa Cruz Biotechnology). An image-analysis system (ACIS, Chromavision) was used to measure the percentage of cells with nuclear staining. p16 mean index % was dichotomized as (1) ≤81.3% (percent positive staining, PPV) vs. >81.3% PPV, (2) ≤20% (PPV)vs. >20%. Cox proportional hazards models were utilized to identify the impact of p16 expression as both a continuous and categorical variable to overall survival, prostate cancer survival, distant metastasis, local progression, and biochemical progression. Actuarial estimates for overall and prostate cancer survival were calculated using the Kaplan-Meier method and the cumulative incidence method was used to estimate the local progression, distant metastasis, and biochemical progression failure rates. Results: On multivariate analysis, after adjusting for pretreatment clinical characteristics and assigned treatment, p16 mean index > 81.3% (higher levels of p16) was statistically significantly associated with decreased rate of distant metastases (hazard ratio= 0.60, 95% confidence intervals= 0.38 to 0.96, p= 0.0332). The multivariate analysis of assigned treatment by p16 index revealed that for patients with a mean p16 index > 81.3%, LTAD+RT significantly improved prostate cancer survival (PCS) over STAD+RT (unadjusted 5-year PCS: 97.8% vs. 89.2%, respectively, p=0.0008) and reduced the frequency of distant metastasis DM (unadjusted 5-year rate of DM: 7% vs. 16%, respectively, p=0.0069) over STAD+RT. In contrast, for patients with tumors demonstrating a mean p16 index ≤81.3%, LTAD+RT failed to improve these outcome parameters compared to STAD+RT, but did appear to decrease the frequency of local progression (p=0.02). Further, within the LTAD+RT arm, patients with tumors demonstrating a mean p16 index > 81.3% compared to ≤81.3% were found to have improved prostate cancer survival (unadjusted 5-year PCS: 97.8% vs. 94.1%, respectively, p=0.05) and reduced frequency of distant metastasis (unadjusted 5-year rates: 7.0% vs. 13.0%, respectively, p=0.02). p16 mean index did not appear to be of prognostic value within the STAD+RT arm. Conclusions: p16 mean index % ≤81.3% (indicating a greater degree of p16 loss) on ACIS appears to be significantly associated with higher risk of distant metastases on multivariate analysis in patients with locally advanced prostate cancer who were treated on RTOG 92–02. The patients who appeared to derive the greatest benefit from LTAD+RT were those with tumors demonstrating higher levels of p16 expression (e.g. a mean p16 value of >81.3%).