Table 2. Hazard ratio for prostate cancer by amount and duration of use of finasteride and by prostate cancer stage and grade, Finnish Prostate Cancer Screening Trial.
| Overall | Gleason⩽6 | Gleason 7–10 | Organ-confined tumoursa | Advanced tumoursb | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Quantity/duration of medication use | No. of cases | HR (95% CI)c | No. of cases | HR (95% CI) | No. of cases | HR (95% CI) | No. of cases | HR (95% CI) | No. of cases | HR (95% CI) |
| Finasteride | ||||||||||
| Non-users | 1507 | Reference | 1139 | Reference | 338 | Reference | 1364 | Reference | 143 | Reference |
| All users | 87 | 0.87 (0.63–1.19) | 55 | 0.59 (0.38–0.91) | 26 | 1.33 (0.77–2.30) | 81 | 0.89 (0.65–1.24) | 6 | 0.55 (0.14–2.24) |
| Cumulative quantity of finasteride use (daily doses) d | ||||||||||
| 28–180 | 34 | 1.34 (0.74–2.42) | 24 | 0.80 (0.33–1.92) | 6 | 1.17 (0.29–4.74) | 32 | 1.32 (0.70–2.46) | 2 | 1.48 (0.21–10.68) |
| 181–398 | 21 | 0.91 (0.50–1.65) | 14 | 0.76 (0.36–1.60) | 5 | 0.79 (0.20–3.20) | 19 | 1.00 (0.55–1.81) | 2 | — |
| 399–1086 | 17 | 0.57 (0.27–1.19) | 13 | 0.64 (0.29–1.43) | 4 | 0.37 (0.05–2.68) | 17 | 0.61 (0.29–1.28) | 0 | — |
| ⩾1087 | 15 | 0.82 (0.47–1.46) | 4 | 0.28 (0.09–0.87) | 11 | 2.49 (1.27–4.89) | 13 | 0.81 (0.45–1.48) | 2 | 0.96 (0.13–6.94) |
| P trend e | 0.204 | 0.009 | 0.114 | 0.275 | 0.415 | |||||
| Years of finasteride use d | ||||||||||
| 1 | 41 | 0.89 (0.5–1.48) | 30 | 0.62 (0.31–1.24) | 7 | 0.57 (0.14–2.32) | 39 | 0.91 (0.53–1.54) | 2 | 0.66 (0.09–4.71) |
| 2 | 19 | 0.96 (0.50–1.85) | 13 | 0.84 (0.38–1.88) | 5 | 1.02 (0.25–4.13) | 19 | 1.03 (0.53–1.99) | 0 | — |
| 3–4 | 11 | 0.72 (0.39–1.35) | 7 | 0.48 (0.20–1.16) | 4 | 1.60 (0.66–3.91) | 10 | 0.70 (0.36–1.34) | 2 | 1.10 (0.15–7.94) |
| >4 | 16 | 1.00 (0.47–2.11) | 5 | 0.40 (0.10–1.61) | 10 | 2.61 (1.06–6.45) | 13 | 1.07 (0.51–2.28) | 2 | — |
| P trend | 0.411 | 0.019 | 0.057 | 0.524 | 0.429 | |||||
a
Men with T1N0/XM0/X and T2N0/XM0/X tumours combined.
b
Men with stage T3N0/XM0/X, T4N0/XM0/X, T1–4N1M0 or T1–4N0–1M1 tumours combined.
c
From Cox proportional hazard regression adjusted for age, family history of prostate cancer, use of alpha-blockers, number of PSA screens and time period of screening (before or after year 2000).
d
Stratification in quartiles of cumulative quantity/duration of finasteride use.
e
Estimated by including cumulative dose (DDDs) or duration (years) of medication use into Cox regression model as a continuous covariate. All statistical trends are inverse, i.e., indicating a decreased risk with larger amount of medication use.