Serum lipids and risk of obesity-related cancers in postmenopausal women.

PURPOSE

Obesity, which is commonly accompanied by dyslipidemia, is associated with an increased risk of certain cancers. However, the association of serum lipids with specific obesity-related cancers is unclear.

METHODS

We examined the association of baseline lipids (total cholesterol, low-density lipoprotein cholesterol [LDL-C], high-density lipoprotein cholesterol [HDL-C], and triglycerides) with risk of developing seven obesity-related cancers in a subcohort of 24,208 participants in the Women's Health Initiative. We used Cox proportional hazards models to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) for quartiles of lipids with cancers of the breast, colorectum, pancreas, endometrium, ovary, and kidney, and multiple myeloma.

RESULTS

Total cholesterol and LDL-C showed no association with these outcomes. HDL-C was inversely associated, and triglycerides were positively associated, with several cancers. However, after adjustment for other lipids or insulin, consideration of preclinical disease, and exclusion of women taking statins, most associations were attenuated and no longer significant. Only the inverse association of HDL-C with pancreatic cancer (HR for highest vs. lowest quartile 0.52, 95% CI 0.32-0.85, p for trend 0.007) and the positive association of triglycerides with kidney cancer (HR for highest vs. lowest quartile 3.21, 95% CI 1.63-6.33, p for trend = 0.0001) remained significant. However, the inverse association of HDL-C with pancreatic cancer was no longer significant when women who lost substantial weight before diagnosis were excluded.

CONCLUSIONS

Our results suggest that when possible sources of confounding and bias are taken into account there are few robust associations of lipids with obesity-related cancers.