Abstract A108: Serum retinol and risk of prostate cancer in the alphatocopherol, beta-carotene (ATBC) cancer prevention study

Abstract

Abstract Purpose: Retinol (vitamin A) may prevent prostate cancer by promoting cell differentiation and apoptosis, although under some conditions it could stimulate growth and de-differentiation. We examined serum retinol, measured at baseline and three years, in relation to prostate cancer risk in the ATBC Study. Methods: The ATBC Study is a randomized controlled trial conducted to determine the effects of -tocopherol and s-carotene supplements on cancer incidence. 29,133 Finnish male smokers were recruited from 1985–1988 and were assigned to groups using a 2×2 factorial design: 1) -tocopherol, 2) β-scarotene, 3) both supplements, or 4) placebo. Fasting blood samples were collected at baseline and after 3 years, and as of April 30, 2006, 2,041 prostate cancer cases occurred during 417,532 person-years. Cox proportional hazards models were used to estimate the relative risk (as hazard ratio, HR) of total and aggressive (n=461) prostate cancer by quintile of baseline and 3-year serum retinol concentrations, as well as by change in serum retinol. No confounding by factors associated with either prostate cancer or retinol was observed. Results: Men in higher quintiles of baseline retinol were more likely to develop prostate cancer compared with men in the lowest quintile (age-adjusted HR, 95%CI vs. Q1; Q2: 1.05, 0.91 – 1.21; Q3: 1.06 0.92 – 1.22; Q4: 1.13, 0.98 – 1.30; Q5: 1.19, 1.03 – 1.36; p-trend = 0.009). The association was similar for aggressive prostate cancer and when the 3-year retinol was used. Compared to men whose retinol did not differ between baseline and 3 years, men with either higher or lower retinol at follow-up were at similar risk of prostate cancer. When we jointly categorized men based on quintiles of baseline and 3-year retinol, we noted that men who were in the highest quintile of retinol at both time points had the greatest increased risk (baseline/3-year Q5/Q5 vs. Q1/Q1 HR: 1.31 95% CI: 1.08 – 1.59). Other joint categories were also at higher risk, particularly among men who did not receive the trial β-carotene supplement. Men randomized to the β-carotene arm had generally higher risk for any baseline/3-year retinol category compared to men in the no β-carotene arm, and those whose serum retinol decreased substantially from baseline Q4 or Q5 appeared to have a lower prostate cancer risk (HR, 95% CI vs. Q1/Q1; Q5 to Q1–3: 0.60, 0.35 – 1.03; Q4 to Q1–2: 0.85, 0.54 – 1.33). Conclusion: In this largest study to date of serum retinol and prostate cancer that incorporates retinol measured at two time points three years apart, higher serum concentrations were associated with elevated risk, with sustained high exposure conferring the greatest risk. Distinct patterns of change in serum retinol over time may relate to differing risks of developing prostate cancer. Citation Information: Cancer Prev Res 2010;3(1 Suppl):A108.