Prostate cancer death of men treated with initial active surveillance: clinical and biochemical characteristics.

PURPOSE: Active surveillance for favorable risk prostate cancer is an approach that may reduce the risk of overtreatment of clinically insignificant prostate cancer. In fact, some patients with favorable risk disease at diagnosis harbor more aggressive disease and may be at risk for prostate cancer mortality despite close monitoring. This is a detailed report of 5 of 453 patients on surveillance who died of prostate cancer.

MATERIALS AND METHODS: A large phase 2 prospective trial of active surveillance in patients with favorable risk prostate cancer was initiated in 1995. Eligible patients had favorable risk prostate cancer (prostate specific antigen 10 ng/ml or less, Gleason 6 or less, T1c/T2a). Epstein criteria for clinically insignificant prostate cancer (a third or less of cores positive, 50% or less involvement of any 1 core, and prostate specific antigen density less than 0.15) were used for men younger than 55 years. Patients were followed with serial prostate specific antigen determinations every 3 months for 2 years and then every 6 months if stable. Biopsies were performed at 1 year and then every 3 to 4 years. Radical intervention was offered if prostate specific antigen doubling time was less than 3 years or Gleason 3 + 4 pattern disease was identified on repeat biopsy. For the first 5 years of the study patients older than 70 years were eligible if they had Gleason 3 + 4 or less, or prostate specific antigen less than 15 ng/ml.

RESULTS: The rate of intervention with radiation or surgery was 38% at 10 years (actuarial). All 5 patients had a prostate specific antigen doubling time of 1.6 years or less triggering a recommendation of radical therapy. Radical intervention was performed in 3 of the 5 patients. Patients 1 and 4 received radiation and patient 3 underwent radical prostatectomy. Of the 2 patients who did not receive definitive treatment 1 was lost to followup (patient 2) and was treated conservatively by his family doctor. Patient 5 elected androgen deprivation therapy rather than radical treatment.

CONCLUSIONS: The low prostate cancer mortality in our surveillance cohort provides support for an active surveillance approach to favorable risk prostate cancer. Only 1 of the 5 patients presented with favorable disease and experienced a theoretically preventable death. The absence of preventable deaths suggests that the basic approach is sound. Two patients had a trigger for intervention but did not receive it. This reinforces the importance of close monitoring and of definitive treatment for those in whom disease is reclassified as higher risk over time.