Real-World Data Shows Higher OS in African American Patients with Prostate Cancer

Oliver Sartor, MD

African American men with prostate cancer may fare better in terms of overall survival (OS) than Caucasian men, according to the results of a pooled analysis of phase III data. The data were obtained through the PROCEEED registry, which includes over 1900 patients with metastatic castration-resistant prostate cancer (mCRPC) sipuleucel-T (Provenge), created at the recommendation of the FDA.¹

“Racial differences in the effectiveness of treatments in mCRPC are well documented,” said Oliver Sartor, MD, co-lead author of the publication, associate dean for oncology at Tulane University School of Medicine and medical director of the Tulane Cancer Center. “The high percentage of AA men enrolled in PROCEED compared to other trials provided a unique opportunity to evaluate the effectiveness of immunotherapy in extending survival in this patient population.”

A total of 1976 patients were enrolled in the registry and other those patients, 1902 received at least once cycle of sipuleucel-T. Of the treated patients, 1649 were Caucasian and 221 were African American, and 32 patients identified as other races. The majority of patients were treated in oncology practices (66%), while less than half of the patients were treated in urology practices (34%). A larger proportion of patients were treated at 140 community clinics (79%) versus the 52 academic centers that were included in the registry.²

In terms of patient characteristics, prostate specified antigen was significantly higher in African American patients at baseline, compared with Caucasian patients (33.0 vs 13.9 ng/mL). Additionally, investigators noticed disparities in African American patients, which included lower hemoglobin levels, longer time from diagnosis to treatment, and the reduced probability that patients had prior chemotherapy. These gaps were true for the overall population as well as the PSA-matched cohorts. The study excluded 2 African American patients from the PSA-matching process due to a lack of PSA at baseline and high PSA value.

Overall, the median OS in African American patients was 35.2 versus 29.9 in Caucasian patients. In the PSA-matched set, the OS differed between the 2 groups showing an HR of 0.70 (95% CI, 0.57-0.86; P =.001) in African Americans versus an  of HR of 0.81 for Caucasian men (95% CI, 0.68-0.97; P =.03). The median OS in the PSA-matched set was 35.3 for African American patients and 25.8 for Caucasian patients.

The investigators determined that survival outcomes in all patients were impacted baseline PSA. Specifically, the variations in the patients had statistical significance at a lower baseline PSA, higher baseline PSAs had no statistical significance.

In the PSA-matched set invariable analyses were conducted and showed 12 out of 18 evaluated baseline characteristics held statistical significance in their association with OS. These characteristics included age, body weight, ethnicity, ECOG performance status, PSA, alkaline phosphatase, hemoglobin, acetate dehydrogenase (LDH), lymph node only metastases, prior prostatectomy, prior treatment with abiraterone (Zytiga) plus enzalutamide (Xtandi), and prior treatment with docetaxel plus cabazitaxel (Jevtana).

The final analysis looked at all characteristics except LDH, which was excluded due to extensive missing data. The final primary multivariable analysis showed that the African American race is an independent prognosticator for patients with prostate cancer, which can indicate the median OS following treatment with sipuleucel-T. A similar result developed from the sensitivity analysis, where African American race was determined to be an independent predictor of OS following treatment with sipuleucel-T.

Post-sipuleucel-T, the survival-prolonging anticancer interventions that were most commonly used include abiraterone, enzalutamide, and docetaxel for both race groups. Although all patients had mCRPC, hormone therapy was more often administered to African Americans whereas most Caucasian patients received chemotherapy. However, the number of ACIs administered were equal between the groups.

The safety analysis for the PSA-matched set found that 21% of African American patients experienced serious adverse events (SAEs) compared with 14.6% of Caucasian patients. Additionally, 15% of African American subjects hade grade-5 SAEs compared with 11% of Caucasian subjects. For all-grade and grade 3-5 SAEs, occurrences were similar between races.

The key objective of developing the multicenter, open-label, observation registry (PROCEED), was to observe the different survival outcomes between races. For the registry, patients were followed for at least 3 years and OS data were assessed from the time of each patient’s first infusion of sipuleucel-T until time of death, withdrawal from the study, or the close of the study. The registry also studied efficacy in terms of race, the number of OS-prolonging ACIs after sipuleucel-T, the types of OS-prolonging ACIs after sipuleucel-T.

Sartor et al note that more research around the difference in efficacy among African American and Caucasian patients is warranted.