Objective: To evaluate the risk of diagnosing prostate cancer on repeated biopsies in patients with fluctuating PSA values compared to patients with stable or regularly increasing PSA values.
Material: Retrospective study conducted on the 2000-2003 databases of 2 French teaching hospitals. Selected patients had a first negative prostatic biopsy, then at least one other series of prostatic biopsies. "Sawtooth" PSA was defined by a PSA value less than that of the previous assay. Other cases were described as "stepwise" PSA.
Results: 191 patients were included: 79 in the "sawtooth" group and 112 in the "stepwise" group. Prostate cancer was diagnosed in 53 patients (27.7%), on the second prostatic biopsy 39 cases. Prostate cancer was detected in 17 (21.5%) of the 79 patients of the "sawtooth" group. This proportion was not significantly different (p = 0.14) from that observed in patients of the "stepwise" group: 36/112 (32.1%). No significant difference in terms of age, stage, Gleason score and initial PSA was observed between patients with a diagnosis of prostate cancer in the "sawtooth" and "stepwise" groups.
Conclusion: In our study, the risk of diagnosing prostate cancer on repeated prostatic biopsies was not greater in patients with "stepwise" PSA compared to patients with "sawtooth" PSA.