VA Study Finds Potential OS Benefit for Enzalutamide in Older mCSPC Patients

Enzalutamide (Xtandi) and abiraterone (Zytiga) yielded comparable overall survival (OS), time to treatment switch or death (TTS), and prostate cancer survival (PCS) in a large population with metastatic castration-sensitive prostate cancer (mCSPC), with a modest OS advantage seen among patients 75 years or older, according to results from a nationwide Veterans Affairs (VA) study published in JCO Clinical Cancer Informatics.¹

The retrospective cohort study used the VA Corporate Data Warehouse to identify 5135 veterans with mCSPC who initiated treatment with abiraterone (n = 3332) or enzalutamide (n = 1803) between January 1, 2020, and December 31, 2023. After inverse probability of treatment weighting (IPTW) to balance baseline characteristics, the 3-year restricted mean survival time (RMST) difference for OS was 0.72 months (95% CI, –0.06 to 1.50) favoring enzalutamide; for TTS, the difference was 0.53 months (95% CI, –0.45 to 1.51); and for PCS at 1 year, the difference was –0.12 months (95% CI, –0.35 to 0.11). Among patients 75 years or older, however, the 3-year OS RMST difference was 1.65 months (95% CI, 0.41-2.89) favoring enzalutamide, a signal not observed in patients younger than 75 years (RMST difference, 0.13 months; 95% CI, –0.86 to 1.11).

“For years, physicians have largely relied on their own historical precedent rather than direct empiric evidence when deciding which of these therapies to prescribe first,” Wadih Arap, MD, PhD, chief of oncology/hematology at Rutgers New Jersey Medical School, who coauthored an editorial accompanying the study, stated in a news release..^2,3 “Large studies like this give clinicians better information to individualize treatment based on the patient sitting in front of them.”

Retrospective Study Design

No randomized trials have directly compared enzalutamide and abiraterone in mCSPC, and prior indirect comparisons via meta-analysis of clinical trial data have produced conflicting results. To address this gap, the study authors applied 2 previously validated informatics tools to the VA Corporate Data Warehouse: a natural language processing algorithm to ascertain metastatic disease and a rule-based classifier to identify castration resistance.¹ The median patient age was 74.33 years; 58.0% of patients were non-Hispanic White, 28.2% were non-Hispanic Black, and 5.6% were Hispanic. Before weighting, patients who received enzalutamide were more likely to have heart disease and diabetes than those who received abiraterone; after IPTW, baseline characteristics were well balanced, with a standardized mean difference of less than 0.1 for all covariates.

Potential Mechanism for the Age-Related Signal

The study authors, as well as Arap and his colleagues, proposed that the age-stratified OS benefit for enzalutamide may relate to corticosteroid exposure rather than greater anticancer activity.^1,2 Abiraterone inhibits CYP17A1, which lowers cortisol synthesis and typically requires concomitant low-dose prednisone to prevent mineralocorticoid excess, whereas enzalutamide, which blocks androgen-receptor signaling directly, requires no steroid coadministration. In patients over 75 with a high burden of cardiovascular disease and diabetes, cumulative glucocorticoid exposure may elevate noncancer mortality risk. The study authors noted that the benefit for enzalutamide was observed in OS and TTS but not in PCS, a pattern consistent with a toxicity-driven rather than efficacy-driven explanation, although the study could not directly confirm this because prednisone use was not reliably captured in VA pharmacy data. Arap stated the survival gap is small but potentially clinically actionable.³

Context Among VA Comparative Effectiveness Studies

In the editorial, Arap and his coauthors situated the new findings alongside 2 earlier VA retrospective analyses.² A prior VA study conducted in the metastatic castration-resistant setting among 5822 veterans, similarly reported an OS advantage for enzalutamide overall and among patients 75 years or older, with the largest effect seen in veterans with cardiovascular disease, diabetes, or high comorbidity burden.⁴ A second VA study which evaluated 1569 veterans in the castration-sensitive setting did not report a significant age-stratified difference, which the editorial attributed to differences in primary end point, cohort window, and sample size.^2,5 Arap and his colleagues noted that the recent study assembled the largest reported VA mCSPC cohort to date.

Study Limitations

The study authors acknowledged that IPTW can balance only measured covariates and cannot adjust for unmeasured confounders such as tumor volume, Gleason score, or metastatic burden. Prednisone use among patients receiving abiraterone could not be confirmed in structured pharmacy data, despite its being standard clinical practice, which the study authors and the accompanying editorial both identified as a meaningful gap given the proposed steroid-related mechanism.^1,2 Fewer patients remained at risk for the PCS end point compared with OS in later follow-up because the VA Mortality Data Repository, the source for cause-of-death data, lags behind overall mortality reporting. The authors also noted that comorbidity burden is generally high among veterans, so findings may not generalize less comorbid populations, and that chemotherapy-exposed patients were excluded from the analysis because of limited sample size.

Investigator Perspective

In the editorial, Arap and his coauthors wrote that for veterans 75 years or older initiating first-line therapy for mCSPC, particularly those with cardiovascular disease, diabetes, or other comorbidities that could magnify glucocorticoid-related risk, enzalutamide may be the more prudent first-line androgen-receptor pathway inhibitor choice, but cautioned that the observational nature of the data and the modest absolute difference are not sufficient to justify contraindicating abiraterone in the older patient population.² The cost difference for enzalutamide also could be resolved soon, as enzalutamide generics are expected to become available in 2027. They added that a randomized trial focused on this age group, particularly one with robust data on prednisone exposure, is needed for a definitive answer.

Arap also highlighted the VA’s role as a research infrastructure. “The VA is one of the closest things the United States has to the national health care databases used for large population studies in countries like the UK or Sweden, Norway and Denmark,” Arap stated in the news release.³ “That makes it possible to answer important clinical cancer management questions by using real-world data at a population scale that’s hard to achieve elsewhere.”

REFERENCES

1. La J, Wang L, Corrigan JK, et al. Abiraterone or enzalutamide for patients with metastatic castration-sensitive prostate cancer: a nationwide Veterans Affairs study. JCO Clin Cancer Inform. 2026;10:e2500213. doi:10.1200/CCI-25-00213

2. Arap W, Pasqualini R, Chen I. Abiraterone or enzalutamide in metastatic castration-sensitive prostate cancer: has the Veterans Affairs corporate data warehouse spoken? JCO Clin Cancer Inform. 2026;10:e2600178. doi:10.1200/CCI-26-00178

3. How a Veterans Affairs Study Can Help Guide Prostate Cancer Treatment. News release. Rutgers Cancer Institute. August 10, 2026. https://tinyurl.com/rd5n5n9p

4. Schoen MW, Carson KR, Eisen SA, et al. Survival of veterans treated with enzalutamide and abiraterone for metastatic castrate resistant prostate cancer based on comorbid diseases. Prostate Cancer Prostatic Dis. 2023;26(4):743-750. doi:10.1038/s41391-022-00588-5

5. Leuva H, Zhou M, Teply BA, et al. Abiraterone vs enzalutamide among US veterans with metastatic hormone-sensitive prostate cancer. JAMA Netw Open. 2025;8(11):e2540730. Published November 3, 2025. doi:10.1001/jamanetworkopen.2025.40730