Abstract B99: Calcium and phosphorus intake and risk of prostate cancer: A 22-year follow-up study

Abstract

Abstract High calcium intake has been associated with increased risk of advanced and high-grade prostate cancer. Phosphorus intake has been studied less frequently, but several prospective studies have found a positive association with prostate cancer risk. We used 22 years of data from the Health Professionals Follow-up Study to examine more closely the independent roles of calcium and phosphorus in prostate cancer. Diet data was collected from 47,896 men every four years from 1986 to 2008. During that time, 5656 cases of prostate cancer were identified, including 670 lethal and 571 high-grade (Gleason score 8–10) cases. We used multivariable Cox proportional hazards models to examine the association between category of mineral intake and prostate cancer risk, adjusting for other risk factors. Mean calcium intake in 1986 was 898 mg/day, and mean phosphorus intake was 1395 mg/day. Calcium came mainly from dairy foods and supplements. Phosphorus came mainly from dairy, meat, fish, and eggs. The Spearman correlation between the two minerals was 0.68. Very high intakes of calcium, over 2000 mg per day, were associated with greater risk of overall prostate cancer and particularly with lethal and high-grade cancers. The relative risk of total prostate cancer was 1.23 (95% confidence interval: 1.01–1.50, p-value for linear trend=0.05) for men consuming 2000 mg or more per day compared to those consuming 500–750 mg per day. The relative risk for lethal prostate cancer was 1.87 (95% CI: 1.21–2.89, p-trend=0.06), and for high-grade was 1.70 (95% CI: 0.96–3.02, p-trend=0.04). These associations were independent of dairy intake. However, the calcium associations were attenuated and no longer statistically significant when phosphorus intake was also adjusted for. The calcium associations were limited to the highest category of intake, and calcium intake in quintiles was not associated with prostate cancer risk. Phosphorus intake was associated with greater risk of overall prostate cancer and lethal and high-grade cancers. The relative risk for the highest versus lowest quintile of phosphorus intake was 1.11 (95% CI:1.02–1.22, p-trend=0.004) for total prostate cancer, 1.35 (95% CI: 1.04–1.75, p-trend=0.03) for lethal prostate cancer, and 1.38 (95% CI: 1.05–1.83, p-trend=0.003) for high-grade cancer. The phosphorus associations were independent of calcium intake and intakes of red meat, white meat, dairy, and fish. Neither mineral was associated with risk of localized or low-grade prostate cancer. In latency analysis, calcium and phosphorus had independent effects for different time periods between exposure and diagnosis. High calcium intake was associated with increased risk of advanced and high-grade disease 12 to 16 years after exposure, whereas high phosphorus intake was associated with increased risk of advanced and high-grade disease 0 to 8 years after exposure. Our findings suggest that phosphorus intake is important in prostate cancer development, and that calcium may not have a strong independent effect on prostate cancer risk except with long latency periods. Citation Information: Cancer Prev Res 2011;4(10 Suppl):B99.