Introduction:

Scant data exist on outcomes of PSA screening in African American (AA) men in the US, with only about 4% AA men included in the US-based Prostate, Lung, Colorectal and Ovarian (PLCO) trial. While prior studies have compared rates of PSA screening for AA vs White men in the US, it is unknown if varying PSA screening intensity has a differential impact on prostate cancer (PCa) diagnoses in AA men.

Methods:

Henry Ford Health System is a large, racially diverse integrated healthcare system in Southeast Michigan. Over the study period (1995-2019), men aged 45+, who self-identified as AA or White, without a prior PCa diagnosis and had at least one PSA and remained in the system for at least 6 years were included in our analyses.  Screening intensity was assessed based on a 5-year window and defined as routine: at least one PSA q1-2 years, occasional: at least one PSA q3 yearly and sporadic: no PSA for three consecutive years, averaged over the duration of the follow up within the study. Outcome variables were time to diagnosis of PCa, PSA, Gleason score and stage at diagnosis.

Results:

Overall, 46,693 (34.5%), 44790 (33.1%) and 44,015 (32.5%) men were included in the sporadic, occasional and routine PSA screening groups, respectively. AA men represented 34,623 (25.6%) of all patients. AA men were younger at time of first PSA (median age 54.9 vs. 57.8 years), but less likely to undergo routine screening (29.4% vs. 33.6%, both p<0.001). Over a median (IQR) follow-up of 9.5 (5.9 - 15.3) years, cumulative incidence of PCa in AA vs White men was 3% vs. 1% in sporadic, 5% vs 2% in occasional and 9% vs 4% in routine screening groups (all p<0.001). For White men in occasional vs. sporadic groups, median PSA at diagnosis was 5.6 vs. 6.3ng/mL, clinical N+ 1.5% vs. 3.5% and metastasis 9.1% vs. 13% (all p<0.01). However, for AA men in these groups, median PSA was 6.8 vs. 8.8ng/mL, clinical N+ 2.1% vs. 4.2% and metastasis 9.9% vs. 17.7% (all p<0.001).

Conclusion:

Results from a large, equal access, racially diverse US-based healthcare system demonstrated that over a nearly 20-year period, fewer AA men underwent PLCO-defined routine PSA screening. Equally important, AA men who underwent occasional or sporadic PSA testing had worse tumor characteristics at PCa diagnosis than their White counterparts with similar intensity of PSA screening, suggesting intrinsic differences in tumor biology between AA vs White men.

Funding: N/A