EURURO Vol. 72 No. 5

823 860

0.40–0.74) regardless of time since quitting. Current use
was associated with lower risk of lethality (HR 0.68, 95% CI
0.52–0.89) regardless of duration. Past use was associated
with higher overall mortality (HR 1.18, 95% CI 1.09–1.28)
regardless of time since stopping, while current use was
associated with lower overall mortality (HR 0.80, 95% CI
0.74–0.86) regardless of duration ).
3.2. Survival analysis
By 2015 (end of follow-up), 407 men initially diagnosed
with nonmetastatic disease developed lethal PC. At
diagnosis, aspirin users were more likely to be white and
have hypertension . Overall, 8% of cases never used
aspirin, 10% only after diagnosis, 19% only before diagnosis,
and 63% before and after diagnosis.
Current postdiagnostic use was associated with lower
risk (lethality: HR 0.68, 95% CI 0.52–0.90; overall mortality:
HR 0.72, 95% CI 0.61–0.84), regardless of duration,
compared to never users. Past postdiagnostic use was
associated with higher risk (lethality: HR 1.50, 95% CI 1.10–
2.05; overall mortality: HR 1.28, 95% CI 1.08–1.53; .
3.3. Secondary risk analyses
Associations between aspirin and reduced lethality did not
hold among cases diagnosed in the PSA era .
Aspirin was not associated with total, high-grade, or
advanced PC (Supplementary ), but was associated
with lower PC mortality (Supplementary ). Lagging
of aspirin use slightly weakened the associations (Supple-
mentary ).
Aspirin randomization was not associated with lethal PC
(HR 0.92, 95% CI 0.77–1.10). When follow-up was limited to
the randomization period (20 lethal events), a nonsignifi-
cant association was observed (HR 0.69, 95% CI 0.28–1.70),
but the wide confidence interval reflects lack of statistical
power.
Table 2 – Risk analysis. Regular prediagnosis aspirin use nd the risk of lethal prostate cancer nd overall mortality among participants in
the Physicians’ Health Study (n = 22 037 participants)
Lethal prostate cancer Overall mortality
Cases PM AA HR (95% CI) MHR (95% CI) Cases PM AAHR (95% CI) MHR (95% CI)
Aspirin use
Never use 85 112162 1.00 (reference) 1.00 (reference) 929 112261 1.00 (reference) 1.00 (reference)
Past use 124 119492 0.54 (0.40–0.73) 0.54 (0.40–0.74) 2925 119760 1.15 (1.06–1.25) 1.18 (1.09–1.28)
Current use 293 299527 0.66 (0.50–0.85) 0.68 (0.52–0.89) 3455 299846 0.72 (0.66–0.78) 0.80 (0.74–0.86)
Time since stopping
Never 85 112162 1.00 (reference) 1.00 (reference) 929 112261 1.00 (reference) 1.00 (reference)
10 yr 31 23131 0.56 (0.35–0.90) 0.55 (0.34–0.88) 655 23190 1.19 (1.06–1.34) 1.14 (1.02–1.28)
8–9 yr 13 9818 0.59 (0.31–1.09) 0.59 (0.31–1.09) 289 9854 1.28 (1.11–1.48) 1.30 (1.13–1.50)
6–7 yr 17 13642 0.57 (0.33–1.00) 0.57 (0.33–1.00) 441 13669 1.46 (1.29–1.66) 1.50 (1.33–1.70)
4–5 yr 19 20981 0.48 (0.28–0.81) 0.47 (0.28–0.80) 595 21039 1.39 (1.24–1.56) 1.42 (1.26–1.59)
2–3 yr 44 51921 0.53 (0.35–0.79) 0.53 (0.35–0.79) 945 52008 1.06 (0.95–1.17) 1.10 (1.00–1.22)
Duration
Never 85 112162 1.00 (reference) 1.00 (reference) 929 112261 1.00 (reference) 1.00 (reference)
Current, 1–4 yr 38 81352 0.66 (0.44–0.99) 0.69 (0.46–1.03) 407 81391 0.67 (0.60–1.76) 0.73 (0.64–0.82)
Current, 5 yr 255 218175 0.66 (0.50–0.86) 0.70 (0.53–0.92) 3048 218455 0.72 (0.66–1.78) 0.85 (0.78–0.92)
PM = person-months; AAHR = age-adjusted hazard ratio; CI = confidence interval; MHR = multivariate hazard ratio.
a Regular prediagnosis aspirin use is defined as taking >3 tablets/wk for at least 1 year. The risk analysis value is updated until prostate cancer diagnosis.
b Lethal prostate cancer is defined as tumor metastases to bones or other organs or death if the cause of death was prostate cancer. The date was taken for
whichever outcome occurred first. Risk analysis outcomes are defined as time from 1981/82 baseline until outcome.
c Multivariate models adjusted for age (years; continuous), race (white, non-white/missing), body mass index (kg/m2; continuous), height (m; continuous),
smoking status (current, past [quit within 10 yr], never use/remote [quit >10 yr ago]), hypertension (yes, no), and diabetes (yes, no).
Table 1 – Baseline characteristics for the risk analysis according to
regular aspirin use at baseline among all participants in the
Physicians’ Health Study
Baseline characteristics Regular aspirin use at baseline
Yes (>3 d/wk) No ( 3 d/wk)
Participants (n) 12 454 9496
Age (yr) 55.0 55.5
Body mass index (kg/m2 ) 24.8 24.8
Height (cm) 178 178
Race
White 92 93
Non-white 8 7
Smoking status
Never/quit > 10 yr ago 49 50
Past, quit within 10 yr 39 39
Current 11 11
Medical history
Hypertension 27 29
Diabetes 3 4
Vigorous exercise
5 times/wk 12 13
3–4 times/wk 26 25
1–2 times/wk 21 20
< 1 time/wk 41 42
Alcohol intake
Daily 25 25
Weekly 49 49
Monthly 11 11
Rarely/never 15 15
Results are presented as the age-adjusted mean for continuous variables
and as the percentage frequency for categorical variables.
a Baseline ranged from 1981 to 1983.
E U R O P E A N U R O L O G Y 7 2 ( 2 0 1 7 ) 8 2 1 – 8 2 7 823