Exposure to Agent Orange, a defoliant used during the Vietnam War linked to an increased risk of prostate cancer (PCa) and other genitourinary malignancies, is associated with a decreased risk of death among men receiving androgen deprivation therapy (ADT) for advanced PCa, according to a new study.

Compared with veterans not exposed to Agent Orange (AO), exposed patients had a significant 16% decreased risk of death after propensity score adjustment, a team at the University of Wisconsin in Madison led by Kyle A. Richards, MD, reported in the Journal of Urology. The investigators found no difference in the risk of skeletal-related events (SREs) or cancer-specific survival (CSS) between exposed and nonexposed patients.

“The improved survival observed in men exposed to AO could be a result of the paradoxical antitumor activity of TCDD related to reduced cell proliferation or perhaps to the delayed development of castration resistance,” the investigators wrote.

The compound in AO believed to be associated with various cancers is TCDD. The investigators studied men on ADT because evidence suggests that the compound influences androgen receptor activity.

Using Veterans Affairs (VA) national databases, the investigators identified 87,344 veterans receiving ADT for advanced PCa. Of these, 3475 were exposed to AO and 83,869 were not.

The AO exposure group was younger than the no-exposure group (median age 60 vs 75 years). They also had lower PSA levels at diagnosis and ADT initiation, and they were more likely to receive local therapy and chemotherapy than nonexposed veterans.

Dr Richards and his colleagues said the study, to their knowledge, is unique in evaluating the impact of AO exposure on overall survival, SREs, and CSS in men with advanced PCa on ADT in a large cohort. “The stringent requirements upheld by the VA to claim AO exposure made this population ideal for examining associations between exposure and oncologic outcomes,” they noted. “Additionally, the VA provides continuous care for most veterans, which is monitored through 1 EHR [electronic health record], making outcomes easier to assess.” The study was limited, however, by its retrospective, observational design, suggesting a potential for unmeasured confounding and/or missing variables. The investigators pointed out that they tried to minimize confounding variables using inverse propensity score weighted adjustment.

In editorial comments accompanying the new report, Kirk A. Keegan, MD, of Vanderbilt University Medical Center in Nashville, said Dr Richards’ team “performed an important and appropriate effort at propensity score adjustment.” Nevertheless, he added, members of the cohort had significant differences in age, race, local treatment and systemic therapy, as well as potential variations in TCDD dosing. Despite propensity score adjustment, “there may still have been considerable unmeasured confounders which may have biased the cohorts in the direction of the current findings.

“In addition,” Dr Keegan continued, “while the study is thought provoking and hypothesis generating regarding the potential influence of TCDD on the androgen axis and the subsequent translational impact on PCa progression, we should exercise caution, given the potential policy implications to veterans of a superficial and cursory interpretation of these data.”

In a reply to editorial comments, Dr Richards’ team stated: “Our data should not be used to significantly alter health policy but it provides valuable information to help inform the clinical treatment of patients with prostate cancer who were exposed to Agent Orange. Clinicians should not assume a worse prostate cancer prognosis in patients exposed to Agent Orange and treatment should be tailored to the tumor biology of each patient (grade, stage, PSA, etc).”

References

Etheridge T, Liou JI, Downs TM, et al. The impact of Agent Orange exposure on prostate cancer outcomes. J Urol. 2019;201:742-750.

https://www.auajournals.org/doi/10.1016/j.juro.2018.10.005

Keegan KA. Editorial comment. J Urol. 2019;201:749.

https://www.auajournals.org/doi/10.1097/01.JU.0000554774.94661.e3