Variation in patterns of practice in diagnosing screen-detected prostate cancer

OBJECTIVE: To determine the practice pattern of repeat prostate biopsies to detect prostate cancer, as there is growing evidence to support the recommendation that a repeat prostate biopsy should be taken after an initially negative prostate biopsy, the rate of cancer detection then being approximately 30%. PATIENTS AND METHODS: We examined the practice patterns of taking a repeat prostate biopsy after an initial negative biopsy and the predictors for cancer at repeat biopsy among 1536 patients who had an initial prostate biopsy because of an elevated prostate-specific antigen (PSA) level (>4.0 ng/mL) or abnormal digital rectal examination. RESULTS: Of the 1536 men, 712 (46.4%) had cancer detected on the first biopsy; of the remaining 824 with no cancer detected, 268 (32.5%) had a repeat biopsy within a year, and 68 of these (25.4%) had cancer detected. Of the cancers detected at repeat biopsy, 31% were high-grade. Men with abnormal histology (prostatic intraepithelial neoplasia or atypia) had an odds ratio of 3.2 (P < 0.001) for having a repeat biopsy. For men with normal initial prostate histology, those with an initial PSA of 10.0-20.0 and >20.0 ng/mL had an odds ratio of 3.6 and 4.5 (both P < 0.001), respectively, for a repeat prostate biopsy, compared with patients with a PSA of <10.0 ng/mL. However, the PSA level was not predictive of prostate cancer at repeat biopsy, but age and prostate volume were. CONCLUSIONS: A third of patients had a repeat biopsy after a negative biopsy. The most important factors influencing whether a patient was to have a repeat biopsy were initial biopsy histology and PSA level. However, the latter was not an important factor for predicting prostate cancer at repeat biopsy.