Login to Access Video or Poster Abstract: MP14-03
Sources of Funding: Supported by GlaxoSmithKline and NIH 1K24CA160653

Introduction

Given the prevalence of prostate cancer (PC) and hypercholesterolemia, multiple studies have examined the link between these two conditions with mixed results. These findings may be influenced by studies showing a correlation between higher cholesterol and higher PSA, introducing a bias that may impact the rate of prostate biopsy and cancer detection thereby making high cholesterol appear to be correlated with PC risk. We tested the association between serum lipids and PC in a post-hoc analysis of the REDUCE study, in which subjects underwent study mandated biopsies regardless of PSA, mitigating any bias due to PSA.

Methods

REDUCE was a 4 year multi-center study testing the effect of dutasteride on PC risk in men with a PSA of 2.5-10.0 ng/mL and a negative pre-enrollment biopsy. As part of the study protocol subjects were required to undergo study-mandated biopsies. The associations between baseline serum cholesterol, low-density lipoprotein cholesterol (LDL) and high-density lipoprotein cholesterol (HDL) with overall PC risk and disease grade (Gleason 2-6 vs. 7-10) at the 2-year biopsy was examined with logistic and multinomial logistic regression, adjusted for baseline covariates. Continuous lipid levels were presented in 10 mg/dL increments to help interpret hazard ratios. We excluded men taking statins.

Results

4,904 subjects not taking statins were included. Elevated serum cholesterol was associated with a higher risk of high-grade PC diagnosis on multivariable analysis (OR 1.23, p=0.008), though no association was seen between overall or low-grade PC risk (p >0.137). No association was seen between serum LDL and overall risk of PC or low- or high-grade disease (p >0.138). In contrast, elevated serum HDL was associated with a higher risk of overall PC risk (OR 1.34, p=0.028) and high-grade PC risk (OR 1.74, p=0.013) on multivariable analysis.

Conclusions

In post hoc analysis of REDUCE, both elevated cholesterol and elevated HDL were associated with increased high-grade PC risk. These data support the hypothesis that high cholesterol is linked with aggressive PC. Given recent data questioning the role of HDL as a cardioprotective factor (Ko et al, J Am College of Cardiology, 2016) and a meta-analysis showing drugs that increase HDL do not reduce cardiovascular risk (Keene et al, BMJ 2014), further study is needed to better understand the link between high HDL and increased PC risk.

Funding

Supported by GlaxoSmithKline and NIH 1K24CA160653