Identifying differences between biochemical failure and cure: incidence rates and predictors

Patients treated with radiation therapy (RT) for prostate cancer were evaluated to estimate the length of time required to document biochemical cure (BC) after treatment and the variables associated with long-term treatment efficacy.

2,100 patients received RT alone for localized prostate carcinoma (external-beam RT, n = 1,504; brachytherapy alone, n = 241; or brachytherapy + pelvic radiation, n = 355). The median external-beam dose was 68.4 Gy, and the median follow-up time was 8.6 years. Biochemical failure (BF) was defined according to the Phoenix definition.

Biochemical failure was experienced by 685 patients (32.6%). The median times to BF for low-, intermediate-, and high-risk groups were 6.0, 5.6, and 4.5 years respectively (p < 0.001). The average annual incidence rates of BF for years 1-5, 5-10,11-15, and 16-20 in low-risk patients were 2.0%, 2.0%, 0.3%, and 0.06% (p < 0.001); for intermediate-risk patients, 4%, 3%, 0.3%, and 0% (p < 0.001); and for high-risk patients, 10.0%, 5.0%, 0.3%, and 0.3% (p < 0.001). After 5 years of treatment, 36.9% of all patients experienced BF. The percentage of total failures occurring during years 1-5, 5-10, 11-15, and 16-20 were 48.7%, 43.5%, 6.5%, and 1.3% for low-risk patients; 64.0%, 32.2%, 3.8%, and 0% for intermediate-risk patients; and 71.9%, 25.9%, 1.1%, and 1.1% for high-risk patients, respectively. Increasing time to nadir was associated with increased time to BF. On multivariate analysis, factors significantly associated with 10-year BC included prostate-specific antigen nadir and time to nadir.

The incidence rates for BF did not plateau until later than 10 years after treatment, suggesting that extended follow-up time is required to monitor patients after treatment. Prostate-specific antigen nadir and time to nadir have the strongest association with long-term BC.