Serial biopsy results in prostate cancer screening study

PURPOSE: We evaluated prostate biopsy results in men with elevated prostate specific antigen (PSA) levels and/or suspicious digital rectal examination whose initial biopsies did not reveal cancer. MATERIALS AND METHODS: A total of 2,526 volunteers 40 years old or older underwent 1 or more prostate biopsies for serum PSA concentrations greater than 4.0 ng./ml. (before May 1995) or greater than 2.5 ng./ml. (after May 1995), or digital rectal examination suspicious of cancer. We evaluated compliance with the biopsy recommendation and the cancer detection rate with regard to digital rectal examination results and increasing PSA levels. RESULTS: Of the men who underwent up to 10 biopsy procedures the serial cancer detection rates were 29%, 17%, 14%, 11%, 9% and 7%, respectively, on biopsy procedures 1 through 6. No significant difference in the yield of cancer on serial biopsies was observed between the groups using the greater than 4.0 ng./ml. and greater than 2.5 ng./ml. cutoff. There was a trend for more cancers detected through serial screening to be organ confined compared with those detected on initial screening (78% versus 69%, p = 0.05). Also, more cancers detected using the greater than 2.5 ng./ml. cutoff were organ confined (80% versus 66%, p = 0.004). Only approximately 1% of the cancers fulfilled the published criteria for clinically insignificant tumors. CONCLUSIONS: Nearly a quarter of prostate cancers detected in this screening study were missed by the initial biopsy. Of the 962 prostate cancers detected 77% were detected with 1, 91% with 2, 97% with 3 and 99% with 4 biopsy procedures. Serial biopsies detect more organ confined cancers without over detecting clinically unimportant tumors. Future studies are needed to determine whether obtaining more biopsy cores initially would provide earlier prostate cancer detection and avoid unnecessary repeat biopsies.