Association of prostate-specific antigen doubling time and cancer in men undergoing repeat prostate biopsy

Abstract

Objectives: To analyze the association between prostate-specific antigen doubling time with prostate cancer risk and grade among men with prostate-specific antigen levels ≥4.0 ng/mL undergoing repeat prostate biopsy.

Methods: A total of 286 patients with prostate-specific antigen ≥4 ng/mL and available prostate-specific antigen doubling time data, who underwent repeat prostate biopsy from 1996-2009, were included in this analysis. Prostate-specific antigen doubling time was divided into three groups: >9 years, 3-9 years and <3 years. Multivariate analyses of prostate-specific antigen doubling time with cancer risk and grade (≤3 + 4 vs ≥4 + 3) were carried out using logistic regression adjusting for prebiopsy prostate-specific antigen, race, age, digital rectal examination, year of biopsy and number of prior negative biopsies.

Results: The median prostate-specific antigen doubling time before biopsy was 4.5 years (interquartile range = 2.5-10). Shorter prostate-specific antigen doubling time was associated with higher prostate-specific antigen (P < 0.001), but it was unrelated to age, digital rectal examination or race. Shorter prostate-specific antigen doubling time as a continuous variable was associated with greater prostate cancer risk in both uni- (hazard ratio = 0.99, 95% confidence interval = 0.98-0.99, P = 0.001) and multivariate analysis (hazard ratio = 0.99, 95% confidence interval = 0.98-0.99, P = 0.004). The prevalence of cancer among prostate-specific antigen doubling time groups (>9, 3-9, <3 years) was 17%, 37% and 40%, respectively. Shorter prostate-specific antigen doubling time groups were associated with higher cancer risk (P = 0.001). Stratified by grade, short prostate-specific antigen doubling time as a continuous variable significantly predicted both low- (P = 0.010) and high-grade disease (P = 0.049). The inclusion of prostate-specific antigen doubling time groups in a multivariate model to predict biopsy positivity increased its accuracy from 0.69 to 0.74.

Conclusion: Prostate-specific antigen doubling time seems to provide further cancer risk assessment in men undergoing repeat biopsy for prostate-specific antigen ≥4.0 ng/mL. If validated in future studies, the present findings support the use of prostate-specific antigen doubling time in the risk stratification of this patient population.