UCSF

To investigate the association between Agent Orange (AO) exposure and long-term prostate cancer (PC) outcomes. Material and Methods: Data from 1,882 men undergoing radical prostatectomy for PC between 1988 and 2011 at Veterans Affairs Health Care Facilities were analyzed from the Shared Equal Access Regional Cancer Hospital database. Men were stratified by AO exposure (binary). Associations between AO exposure and biopsy and pathologic Gleason sum (GS) and pathologic stage were determined by logistic regression models adjusted for preoperative characteristics. Hazard ratios for biochemical recurrence (BCR), secondary treatment, metastases, and PC-specific mortality were determined by Cox models adjusted for preoperative characteristics. Results: There were 333 (17.7%) men with AO exposure. AO-exposed men were younger (median 59 vs. 62y), had lower preoperative prostate-specific antigen levels (5.8 vs. 6.7ng/ml), lower clinical category (25% vs. 38% palpable), and higher body mass index (28.2 vs. 27.6kg/m2), all P<0.01. Biopsy GS, pathologic GS, positive surgical margins, lymph node positivity, and extracapsular extension did not differ with AO exposure. At a median follow-up of 85 months, 702 (37.4%) patients had BCR, 603 (32.2%) patients received secondary treatment, 78 (4.1%) had metastases, and 39 (2.1%) died of PC. On multivariable analysis, AO exposure was not associated with BCR, secondary treatment, metastases, or PC mortality. Conclusions: AO exposure was not associated with worse preoperative characteristics such as elevated prostate-specific antigen levels or biopsy GS nor with BCR, secondary treatment, metastases, or PC death. Thus, as data on AO-exposed men mature, possible differences in PC outcomes observed previously are no longer apparent.