Examining the Relationship between Cholesterol and Cancer
Does HDL-C Have A Protective Role?
By Genna Rollins
An inverse relationship between cholesterol levels and cancer mortality and incidence has been noted in many studies. However, there has been conflicting evidence and few large, long-term prospective trials to clarify whether the association is causal or due to reverse causality. Now, new research that addresses this question has been published and is described in this issue of Strategies.
A series of studies dating back to the 1980s found an association between low circulating levels of total cholesterol and increased cancer incidence and mortality. This inverse relationship largely had been attributed to reverse causality, meaning that undiagnosed cancer triggered lower serum cholesterol levels. However, there was conflicting evidence, so researchers had not been able to entirely exclude the possibility that low total cholesterol levels in some way led to cancer. For instance, one study found an association between low cholesterol levels and elevated cancer incidence even after excluding the first 6 years of follow-up, while another found a small but persistent association between low cholesterol levels and modestly increased cancer mortality after excluding the first 5 years of follow-up.
In addition to the questions about total cholesterol, little was known about any association between high-density lipoprotein cholesterol (HDL-C) and cancer risk. However understanding these relationships is important in the context of current cardiovascular disease (CVD) prevention strategies, according to Demetrius Albanes, MD, senior investigator at the National Cancer Institute. “These findings about total cholesterol from well-conducted studies were troubling in that public health recommendations related to cardiovascular disease have lowering serum cholesterol as a key component,” he explained. Albanes was the senior investigator of a recent study that prospectively examined the cholesterol-cancer association(Cancer Epidemiol Biomarkers Prev 2009;18:2814-21).
Albanes and his colleagues evaluated the relationship between serum total cholesterol and HDL-C and risk of overall and site-specific cancer among more than 29,000 participants in the Alpha-Tocopherol, Beta-Carotene Cancer Prevention (ATBC) study. ATBC tested whether α-tocopherol or β-carotene supplementation would reduce the incidence of lung or other cancers in a population of middle-aged male smokers in Finland. Baseline fasting serum cholesterol levels were available from virtually all participants (99.9%), and measurements taken at the time of third-year follow-up visits were available from more than three-quarters. Average total cholesterol at baseline was 241.2 mg/dL, while average HDL-C was 46.3 mg/dL. Participants were placed in quintiles based on total cholesterol and HDL-C levels, with lowest total cholesterol quintile <209.9 mg/dL and highest >276.7 mg/dL, and lowest HDL-C quintile <36.2 mg/dL and highest >55.3 mg/dL. Using data from the Finnish Cancer Registry, which documents approximately 100% of cancers nationwide, the researchers identified more than 7,500 cancers among ATBC participants over 18 years of follow-up.
In their initial analysis, Albanes and his colleagues found that higher serum total cholesterol was associated with decreased cancer incidence, with a relative risk of 0.85 comparing highest to lowest quintiles. However, when they conducted a lag analysis that excluded cancer cases diagnosed in the first 9 years of follow-up, this inverse association attenuated significantly and was no longer statistically significant. The inverse relationship between serum total cholesterol and cancer risk was particularly evident in lung and liver cancers, but these associations also became statistically insignificant after exclusion of cases diagnosed during the first 9 years of follow-up. In addition, the analysis revealed that there was a greater decline between baseline and third-year follow-up total cholesterol levels in cancer cases that were diagnosed within 9 years of follow-up.
Given ATBC’s large study population, prospective serum measurements, and long and complete follow-up for cancer incidence, “our finding supports the idea that the lower serum cholesterol levels we detected as a possible cancer risk factor may actually have been the result of undiagnosed cancer,” said Albanes. This finding is particularly significant in terms of prevention messages involving cholesterol levels and cancer, according to Eric Jacobs, PhD, strategic director of pharmacoepidemiology at the American Cancer Society, who was not involved in the study. “The results should help dispel any lingering concerns anyone might have had that having lower cholesterol levels could cause cancer,” he observed.
Albanes explained that there are numerous mechanisms by which cancer could impact total cholesterol levels. “Cholesterol is a key biochemical component essential to cell membrane structure, and it’s a very important precursor to steroid metabolism. So there are a large number of ways that a growing tumor could have metabolic effects that could alter cholesterol profile,” he said.
While the ATBC analysis appears to close the door on the issue of low total cholesterol levels playing a pathogenic role in cancer, the results suggest that there may be a protective role for HDL-C. The investigators found that higher serum HDL-C levels were modestly but significantly associated with decreased cancer incidence overall, and that this inverse association remained significant after they excluded cancer cases diagnosed during the first 12 years of follow-up. Most of this association was attributable to lung and liver cancer, with relative risk after 12 years of exclusion of 0.84 and 0.49, respectively. The results remained “essentially unchanged” when the researchers analyzed HDL-C levels measured at the time of third year follow-up visits, or when they used an average of baseline and third-year HDL-C levels. Additionally, “in contrast to the time trend for total cholesterol, in this case excluding the early years of follow-up made the HDL association stronger rather than weaker,” noted Albanes. As with total cholesterol, there are several plausible mechanisms for an HDL-C-cancer association, including HDL-C regulation of cell cycle entry and apoptosis, modulation of cytokine production, and anti-oxidative effects, according to the researchers.
Albanes emphasized that the results need further validation in other populations, especially women and non-smokers. He also cautioned against any implication that lipid-lowering medications might decrease cancer risk. “It may be premature to read from our findings in ATBC about efforts to actively lower cholesterol levels to achieve cancer benefits. Our data don’t speak to that and we would need more research looking at lipid lowering drugs and cancer outcomes,” he said.
Even with these caveats, the ATBC study “raises an interesting question about whether HDL cholesterol might lower the risk of some cancers. This is a very new, exciting question, but we need to do more research before we have any clear answers,” Jacobs observed.