The association between presentation PSA and race in two sequential time periods in prostate cancer patients seen at a university hospital and its community affiliates

Abstract

Purpose We sought to determine whether African American men diagnosed with prostate cancer in the prostate-specific antigen (PSA) era differed in initial presenting serum PSA levels ( iPSA) compared to white men. Recent retrospective studies have demonstrated higher iPSA within the African American men than in white men at the time of diagnosis, suggestive of more advanced disease in African American men. Both biologic differences and/or sociologic factors have been postulated as explaining the noted differences in iPSA. We reviewed our institution's PSA-era experience to determine any association between race and iPSA. Methods and materials Between January 1990 and September 2001, 4519 patients representing a broad demographic sample were seen in the radiation oncology department of a university hospital or one of its four community affiliates. A total of 2332 eligible patients, with data on race, age, year of diagnosis, Gleason score, T stage, and iPSA, were analyzed. The patients were separated into the two following time periods for analysis, based on the new American Cancer Society screening guidelines: ( 1) 1991 to 1996 and ( 2) 1997 to 2001. The relationships between race and iPSA, T stage, Gleason score, and age are explored. Results Of the 2332 patients analyzed, there were 1968 white men and 364 African American men. For the time period 1990 through 1996, the expected average (median) iPSA level was 10.5 (10.2) and 14.6 (15.8) for white men and African American men, respectively. For 1997 to 2001, the expected average iPSA level was 9.5 (8.4) and 10.8 (9.8), respectively. T stage distributions improved, independent of race, toward earlier stage at presentation. Gleason score distribution remained unchanged. African American men are 2.5–3.1 years younger than white men at diagnosis. Conclusion An overall decline in iPSA has occurred in both racial groups over time. More importantly, racial differences in iPSA among men diagnosed in the later time period (1997 to 2001) are less pronounced compared to men diagnosed in the earlier time period (1990 to 1996). This racial convergence in iPSA over time suggests improved penetrance of PSA screening in the African American population. Our findings also suggest that studies comparing racial differences in iPSA should consider time period of diagnosis and possible sociologic changes during that period (i.e., access to medical care, socioeconomic status, and educational level). The American Cancer Society guideline to begin screening African Americans at an earlier age is appropriate.