Genetically Dictated Higher Lipid Levels Seem to Protect Against Type 2 Diabetes
Patients with genetic predispositions to higher levels of LDL and HDL cholesterol and triglycerides carry lower risks of type 2 diabetes, a mendelian randomization study has shown. The results might be relevant when evaluating the effects of statins and other lipid-modifying drugs.
“An interesting interpretation of our findings is that the increased risk of diabetes seen with statin use may well be a general consequence of LDL cholesterol lowering, rather than a unique ‘side effect’ of statin therapy,” senior author Michael Holmes, MD, PhD (University of Oxford, England), told TCTMD in an email. “If so, this has implications for new drugs in use and/or under development that lower LDL cholesterol insofar as clinicians should carefully monitor for risk of diabetes.”
He stressed, however, that the results—published online August 3, 2016, ahead of print in JAMA Cardiology—“once again show very clear and robust evidence that LDL cholesterol is causally related to risk of heart disease and that patients should continue to take their lipid-lowering medications as prescribed by their doctor.”
Commenting on the study for TCTMD, Ronald Goldberg, MD (University of Miami, FL), said that its findings provide the impetus to explore potential mechanisms behind the observed associations.
He noted that LDL cholesterol could potentially be causally related to type 2 diabetes risk through inflammation, which has been associated with insulin resistance. But he stressed that more research is needed.
Overall, Goldberg said, the findings are “not clinically applicable at this time, but certainly worth noting.”
Looking to Genetics for a Clearer Answer
Although it is well established that LDL cholesterol levels are directly related to risks of heart disease and stroke, the link between circulating lipid levels and type 2 diabetes risk remains inconclusive, Holmes noted.
“This is because observational studies can be hampered by types of error that can’t be fully controlled for,” he said, “and previous genetic studies used older approaches to guide the selection of genetic variants in the analysis and the choice of statistical approach.”
Using more up-to-date techniques, Holmes, lead author Jon White, PhD (University College London, England), and colleagues performed a mendelian randomization study with data from genome-wide association studies. They identified genetic variants associated with altered concentrations of circulating blood lipids and created genetic instruments for exposure to each of the individual lipid fractions.
The pooled data set included 188,577 people with measures of blood lipids, 63,158 patients with CAD, and 34,840 with diabetes.
A 1-standard deviation (SD) increase in LDL cholesterol—equal to 38 mg/dL—was associated with an increase in CAD risk (OR 1.68; 95% CI 1.51-1.87). A similar relationship was seen for a 1-SD increase in trigylcerides (OR 1.28; 95% CI 1.13-1.45), but the link between HDL cholesterol and CAD risk was not significant (OR 0.95; 95% CI 0.85-1.06).
In contrast, increases in all three lipid fractions were tied to lower risks of type 2 diabetes:
- LDL cholesterol (OR 0.79; 95% CI 0.71-0.88)
- HDL cholesterol (OR 0.83; 95% CI 0.76-0.90)
- Triglycerides (OR 0.83; 95% CI 0.72-0.95)
“Our data add another piece to the complicated jigsaw on the underlying relationships between circulating blood lipids and risk of heart disease and diabetes,” Holmes said, adding, however, that the study does not address potential mechanisms.
“What is now needed is investigative work into understanding just how LDL cholesterol and HDL cholesterol have these protective effects on risk of diabetes,” he said. “Taking the example of LDL cholesterol, is this a direct effect or is there a pathway (from LDL cholesterol to diabetes) that can be blocked? The implications are potentially huge, as if we can identify a modifiable protein or receptor on the pathway from LDL cholesterol to diabetes, it may be possible to identify new therapeutic approaches not only to combat the increased risk of diabetes associated with lipid-lowering therapy but also to treat diabetes in general.”
Implications for PCSK9 Inhibitors and Other Medications
In an accompanying editorial, Danish Saleheen, MBBS, PhD (University of Pennsylvania, Philadelphia), and colleagues say that there is an active debate about the potential relationship between type 2 diabetes and each of these lipid fractions.
The findings of the current study, they point out, are consistent with some prior investigations but not with others, including two that used the same type 2 diabetes data set. Nevertheless, they add, accumulated evidence suggests that LDL cholesterol might be causally associated with type 2 diabetes.
“This conclusion, extrapolated beyond statins and 3-hydroxy-3-methylglutaryl-CoA reductase, has major implications for PCSK9 inhibitors and other LDL-lowering medicines in development,” Saleheen et al say.
The most counterintuitive finding of the analysis is the link between higher levels of triglycerides and lower risk of diabetes, because prior studies have shown either no relationship or one going in the opposite direction, they say.
If the findings are confirmed, “the implications of this are substantial, including the potential that intervention to reduce [triglyceride] levels could paradoxically increase the risk for [type 2 diabetes],” they write. “That said, there have been a number of large [triglyceride]-lowering clinical trials to date, primarily with fibrates, and in those studies, there is currently no suggestion of increased [type 2 diabetes] risk in those participants randomized to active treatment.”
On the whole, the study findings “will no doubt fuel the controversy” on the link between circulating lipids and type 2 diabetes, the editorialists conclude. “The importance of this issue is clear: it has the potential to provide new insights into the pathogenesis of [type 2 diabetes] and has implications for the effect of specific lipid-altering therapies on the development of [the disease].”
White J, Swerdlow DI, Preiss D, et al. Association of lipid fractions with risks for coronary artery disease and diabetes. JAMA Cardiol. 2016;Epub ahead of print.
Saleheen D, Rader DJ, Voight BF. Disentangling the causal association of plasma lipid traits and type 2 diabetes using human genetics. JAMA Cardiol. 2016;Epub ahead of print.
- White, Holmes, and Saleheen report no relevant conflicts of interest.