Introduction: In a population-based case-control study (PROtEuS), we examined the association between prostate cancer (PCa) and (1) benign prostatic hypertrophy (BPH) history at any time prior to PCa diagnosis, (2) BPH-history reported at least 1 year prior to interview/diagnosis (index date) and (3) exposure to BPH-medications.
Methods: Cases were 1933 men with incident prostate cancer diagnosed across Montreal French hospitals between 2005 and 2009. Population controls were 1994 men from the same age distribution and residential area. In-person interviews collected socio-demographic characteristics and medical history, e.g., BPH diagnosis, duration and treatment, as well as on PCa screening. Logistic regression analyses tested overall and grade-specific associations, including subgroup analyses with frequent PSA testing.
Results: A BPH-history was associated with an increased risk of PCa (OR 1.37 [95 % CI 1.16-2.61]), more pronounced for low-grade PCa (Gleason ≤6: OR 1.54 [1.26-1.87]; Gleason ≥7: OR 1.05 [0.86-1.27]). The association was not significant when BPH-history diagnosis was more than 1 year prior to index date, except for low-grade PCa (OR 1.29 [1.05-1.60]). Exposure to 5α reductase inhibitors (5α-RI) resulted in a decreased risk of overall PCa (OR 0.62 [0.42-0.92]), particularly for intermediate- to high-grade PCa (Gleason ≤6: OR 0.70 [0.43-1.14]; Gleason ≥7: OR 0.43 [0.26-0.72]). Adjusting for PSA testing frequency or restricting analyses to frequently screened subjects did not affect these results.
Conclusion: BPH-history was associated with an increased PCa risk, which disappeared, when BPH-history did not include BPH diagnosis within the previous year. Our results also suggest that 5α-RI exposure exerts a protective effect on intermediate and high-grade PCa.
Keywords: Benign prostatic hypertrophy; Case–control study; Diagnosis; Prostate biopsy; Prostate cancer.