WHAT CLINICAL FACTORS PREDICT OUTCOMES IN RECURRENT METASTATIC CASTRATE-SENSITIVE PROSTATE CANCER? RESULTS FROM THE SEARCH DATABASE

Introduction

Approximately one third of patients with prostate cancer (PC) undergoing radical prostatectomy (RP) develop biochemical recurrence within 10 years. PC in these men is heterogenous, with most following an indolent course, however up to one third will progress to metastatic castrate sensitive disease (mCSPC). Patient and clinical factors associated with progression from non-metastatic PSA recurrence to mCRPC or death include increased age, shorter time from surgery to recurrence, higher Gleason grade (PC specific tumor grade), and shorter PSA doubling time (PSADT). The factors that influence disease progression once patients develop mCSPC are unknown.

Methods

After obtaining IRB approval, the SEARCH database was created by combining data on 9,928 patients who underwent RP from 1982 to 2020 at eight VA Medical Centers across the U.S. Patients with recurrent mCSPC after RP were selected for analysis. Those treated with preoperative ADT or radiation therapy were excluded, as were patients who developed CRPC prior to metastasis. Patients who received ADT more than 3 months prior to metastasis were excluded. Of the 278 men who met these criteria, 153 were excluded due to missing data. We used multivariable cox proportional hazards regression modeling to assess the association between clinical variables at the time of mCSPC and the primary outcome of overall mortality (OM).

Results

Of 125 patients with recurrent mCSPC, median age at metastasis was 70. Median time from surgery to metastasis was 61 months. During follow-up, 60 patients (48%) progressed to mCRPC while 56 (45%) died from PC and 81 (65%) died from any cause. Shorter PSADT was the only clinical or pathologic feature associated with progression to OM (HR = 0.98, p < 0.0001). Patients were divided into groups based on PSADT widely used in non-metastatic biochemical recurrence (<3, 3-8.9, >9 months). On multivariable analysis, PSADT <3 months and 3-8.9 months were associated with worse overall survival (OS) than PSADT >9 months (Table 1). Median time to OS and 5-year OS rates were 93 months and 64% for PSADT >9 months, 47 months and 41% for 3-8.9 months, and 25 months and 12% for <3 months.

Conclusions

Outcomes for patients with recurrent mCSPC after RP were driven primarily by PSADT rather than other clinical features, including Gleason score. Our data demonstrates that patients with PSADT < 9 months have rapidly progressive disease that warrants aggressive treatment and inclusion in clnical trials. In contrast, those with PSADT > 9 months have a more indolent course despite the presence of metastases and may be candidates for deintensificaiton strategies.