Statins and type 2 diabetes: genetic studies on target

310 www.thelancet.com Vol 385 January 24, 2015 The topic of the side-eff ects of statin treatment is important and controversial. In The Lancet, Daniel Swerdlow and colleagues used an updated meta-analysis of trials to investigate whether or not statins increased the risk of type 2 diabetes, and a genetic app roach to address how statins might increase the risk of the disorder. Using data from 20 randomised controlled trials, they confi rm fi ndings from previous reports that statin treatment increased the risk of incident type 2 diabetes, with an odds ratio (OR) of 1·12 (95% CI 1·06–1·18) versus controls. In contrast to previous eff orts, they then studied common genetic variants near the gene encoding the HMG-coenzyme A (HMGCoA) reductase protein—the enzyme inhibited by statins to lower LDL cholesterol. Genome-wide association studies had previously identifi ed these variants as associated with altered circulating LDL cholesterol concentrations with robust levels of statistical confi dence. Using these genetic variants and a combination of their own and published data, the investigators provide evidence that reduced HMGCoA activity causes a slight increased risk of type 2 diabetes (rs17238484-G allele OR per allele 1·02, 95% CI 1·00–1·05; rs12916 allele 1·06, 1·03–1·09), and therefore surmise that an increased risk of type 2 diabetes is at least partly conferred by an on-target eff ect of statins. These results are important because they suggest that any attempts to make statins more specifi c and reduce off -target eff ects will not reduce the risk of the diabetogenic side-eff ect. The investigators also provide Statins and type 2 diabetes: genetic studies on target 1 Nordestgaard BG, Chapman MJ, Humphries SE, et al. Familial hypercholesterolaemia is underdiagnosed and undertreated in the general population: guidance for clinicians to prevent coronary heart disease: consensus statement of the European Atherosclerosis Society. Eur Heart J 2013; 34: 3478–90. 2 Benn M, Watts GF, Tybjaerg-Hansen A, Nordestgaard BG. Familial hypercholesterolemia in the Danish general population: prevalence, coronary artery disease, and cholesterol-lowering medication. J Clin Endocrinol Metab 2012; 97: 3956–64. 3 Sjouke B, Kusters DM, Kindt I, et al. Homozygous autosomal dominant hypercholesterolemia in the Netherlands: prevalence, genotype– phenotype relationship, and clinical outcome. Eur Heart J 2014; published online Feb 28. DOI:10.1093/eurheartj/ehu058. 4 Raal FJ, Santos RD. Homozygous familial hypercholesterolemia: current perspectives on diagnosis and treatment. Atherosclerosis 2012; 223: 262–68. 5 Alonso R, Andres E, Mata N, et al, for the SAFEHEART Investigators. Lipoprotein(a) levels in familial hypercholesterolemia: an important predictor of cardiovascular disease independent of the type of LDL receptor mutation. J Am Coll Cardiol 2014; 63: 1982–89. 6 Versmissen J, Oosterveer DM, Yazdanpanah M, et al. Effi cacy of statins in familial hypercholesterolaemia: a long term cohort study. BMJ 2008; 337: a2423. 7 Horton JD, Cohen JC, Hobbs HH. PCSK9: a convertase that coordinates LDL catabolism. J Lipid Res 2009; 50 (suppl): S172–77. 8 Watts GF, Gidding S, Wierzbicki AS, et al. Integrated guidance on the care of familial hypercholesterolaemia from the International FH Foundation. Int J Cardiol 2014; 171: 309–25. 9 Raal FJ, Santos RD, Blom DJ, et al. Mipomersen, an apolipoprotein B synthesis inhibitor, for lowering of LDL cholesterol concentrations in patients with homozygous familial hypercholesterolaemia: a randomised, double-blind, placebo-controlled trial. Lancet 2010; 375: 998–1006. 10 Cuchel M, Meagher EA, du Toit Theron H, et al. Effi cacy and safety of a microsomal triglyceride transfer protein inhibitor in patients with homozygous familial hypercholesterolaemia: a single-arm, open-label, phase 3 study. Lancet 2013; 381: 40–46. 11 Santos RD. Lipid-lowering treatment for homozygous familial hypercholesterolaemia. Lancet 2013; 381: 1182. 12 Ridker PM. LDL cholesterol: controversies and future therapeutic directions. Lancet 2014; 384: 607–17. 13 Stein EA, Gipe D, Bergeron J, et al. Eff ect of a monoclonal antibody to PCSK9, REGN727/SAR236553, to reduce low-density lipoprotein cholesterol in patients with heterozygous familial hypercholesterolaemia on stable statin dose with or without ezetimibe therapy: a phase 2 randomised controlled trial. Lancet 2012; 380: 29–36. 14 Raal F, Scott R, Somaratne R, et al. Low-density lipoprotein cholesterol-lowering eff ects of AMG 145, a monoclonal antibody to proprotein convertase subtilisin/kexin type 9 serine protease in patients with heterozygous familial hypercholesterolemia: the Reduction of LDL-C with PCSK9 Inhibition in Heterozygous Familial Hypercholesterolemia Disorder (RUTHERFORD) randomized trial. Circulation 2012; 126: 2408–11. 15 Stein EA, Honarpour N, Wasserman SM, Xu F, Scott R, Raal FJ. Eff ect of the proprotein convertase subtilisin/kexin 9 monoclonal antibody, AMG 145, in homozygous familial hypercholesterolemia. Circulation 2013; 128: 2113–20. 16 Rader DJ, Cain W, Ikewaki K, et al. The inverse association of plasma lipoprotein(a) concentrations with apolipoprotein(a) isoform size is not due to diff erences in Lp(a) catabolism but to diff erences in production rate. J Clin Invest 1994; 93: 2758–63. 17 Raal FJ, Honarpour N, Blom DJ, et al, for the TESLA Investigators. Inhibition of PCSK9 with evolocumab in homozygous familial hypercholesterolaemia (TESLA Part B): a randomised, double-blind, placebo-controlled trial. Lancet 2014; published online Oct 2. http://dx.doi.org/10.1016/S01406736(14)61374-X. 18 Raal FJ, Stein EA, Dufour R, et al, for the RUTHERFORD-2 Investigators. PCSK9 inhibition with evolocumab (AMG 145) in heterozygous familial hypercholesterolaemia (RUTHERFORD-2): a randomised, double-blind, placebo-controlled trial. Lancet 2014; published online Oct 2. http://dx.doi. org/10.1016/S0140-6736(14)61399-4. 19 Varbo A, Benn M, Tybjaerg-Hansen A, Nordestgaard BG. Elevated remnant cholesterol causes both low-grade infl ammation and ischemic heart disease, whereas elevated low-density lipoprotein causes ischemic heart disease without infl ammation. Circulation 2013; 128: 1298–309. 20 Seed M, Betteridge DJ, Cooper J, et al. Normal levels of infl ammatory markers in treated patients with familial hypercholesterolemia: a cross-sectional study. JRSM Cardiovasc Dis 2012; 1: 1–9. 21 Blom DJ, Hala T, Bolognese M, et al. A 52-week placebo-controlled trial of evolocumab in hyperlipidemia. N Engl J Med 2014; 370: 1809–19. 22 Stroes E, Colquhoun D, Sullivan D, et al. Anti-PCSK9 antibody eff ectively lowers cholesterol in patients with statin intolerance: the GAUSS-2 randomized, placebo-controlled phase 3 clinical trial of evolocumab. J Am Coll Cardiol 2014; 63: 2541–48. 23 Cholesterol Treatment Trialists’ (CTT) Collaboration. Effi cacy and safety of more intensive lowering of LDL cholesterol: a meta-analysis of data from 170 000 participants in 26 randomised trials. Lancet 2010; 376: 1670–81.

Swerdlow and colleagues 1 provide a new angle to the debate about the adverse side-eff ects of statins. They used a genetic approach-mendelian randomisationthat has proven a valuable method to help understand disease mechanisms. 3,4 Mendelian randomisation is based on a fundamental principle of biology-that inherited DNA sequence variation is randomised during meiosis independently of the environment and disease processes. Mendelian randomisation approaches have been useful in understanding causality of potentially modifi able risk factors for cardiovascular disease. For example, several studies have used common genetic variants near the gene encoding C-reactive protein (CRP) to show that raised high-sensitivity CRP is unlikely to aff ect the risk of heart disease causally. 5 More recently, studies of genetic variants altering circulating HDL cholesterol concentrations have provided evidence that higher total HDL cholesterol concentrations are unlikely to reduce the risk of coronary artery disease, independently of any eff ects on LDL cholesterol or triglycerides. 6,7 Of more direct relevance to the debate on statins, common genetic variants that alter LDL cholesterol concentrations are also associated with coronary artery disease. 7,8 The eff ect of common genetic variants is often subtle, but in some instances it provides great potential for improving mechanistic understanding. The genetic variants in the HMGCR gene have small eff ects on circulating LDL cholesterol (0·06 mmol/L) compared with other variants, which alter LDL concentrations by up to 0·36 mmol/L. Yet the variants in HMGCR provide proof of principle that genetics can be used to identify therapeutic targets.
Swerdlow and colleagues 1 used their own set of 26 236 cases and 164 021 controls as well as those from published studies. The results provide an important addition to a cascade of evidence that suggests a slight on-target type 2 diabetes side-eff ect of statins (fi gure). This cascade includes individual randomised controlled trials that provided some evidence of causality, although the diff erent statins and defi nitions of diabetes used made results hard to interpret. 9,10 Observational associations between increased statin use and higher incidence of type 2 diabetes were consistent with the early trial data, but were probably heavily confounded by obesity and other factors. 10 Meta-analyses of randomised controlled trials followed 12 with, for example, an analysis of ten trials 11 showing an increased risk of type 2 diabetes. An analysis of intensive-dose versus moderate-dose statin treat ment showed a doseresponse eff ect, with individuals randomly assigned to intensive-dose treatment having a relative risk of 1·12 (95% CI 1·04-1·22) for diabetes compared with individuals assigned to moderate-dose treatment. 13 However, none of these studies could establish whether or not the diabetogenic eff ect of statins operated through the same pathway as the lipid-lowering HMGCoA-reductase eff ect or an off -target eff ect. Swerdlow and colleagues 1 answer this question-because the genetic variant lies near HMGCR, the diabetogenic eff ect of statins probably operates through the same mechanism as the lipidlowering eff ect. The fi ndings imply that new types of statin that more specifi cally target HMGCoA-reductase would not reduce the adverse side-eff ect of increased risk of type 2 diabetes. Evidence of a causal effect of statins on type 2 diabetes risk. Caveat: results differed-some were positive, some negative, and some null. 10 2. Observational epidemiology. Incidence of type 2 diabetes increased as statin use increased. Caveat: there were several sources of confounding and bias.
3. Meta-analyses of randomised controlled trials (eg, Sattar and colleagues 11 and Rajpathak and colleagues 12 ). Evidence of a causal effect of statins on risk of type 2 diabetes. Caveat: diabetes was not the primary endpoint and different statins and definitions of diabetes were used. 11 4. Meta-analysis of randomised controlled trials-intensive versus normal statin treatment. Dose-response effect. Intensive statin treatment caused greater risk of diabetes than normal-dose treatment. Caveat: is the diabetogenic effect on-target or off-target? 13 5. Mendelian randomisation. Common variants in the HMGCR gene were associated with an LDL-lowering effect, increased body-mass index, insulin resistance, and type 2 diabetes. Suggests an on-target effect. 1 The UK has the second highest rate of tuberculosis among western European countries. 1 Tuberculosis clinics in London manage more cases a year than those in all other western European capital cities put together. Rates of tuberculosis are now nearly fi ve times higher in the UK than in the USA. 2 Lack of progress with tuberculosis control in the UK does not just represent a risk to domestic public health, 3 but also an international embarrassment with examples of cases acquired in the UK leading to infections in other low-incidence countries. In recognition of this unacceptable trend, Public Health England has led a coalition of stakeholders to develop a forum, the national Tuberculosis Oversight Group, where innovation and good practice are shared between local, regional, and national health leaders. These discussions have led to local changes, with several areas establishing tuberculosis control boards and systematic cohort review, and the identifi cation of tuberculosis as a major priority for Public Health England. However, the implementation of improved tuberculosis control measures has not been universal, and there is still unacceptable variation in the quality of clinical and public health measures across England.

A collaborative strategy to tackle tuberculosis in England
There are some limitations to the study. 1 First, the subtle eff ects of the HMGCR variants meant that the investigators had to use large numbers of cases and controls, and the associations between the variants and type 2 diabetes are not statistically beyond reproach-more cases and controls would help confi rm the fi ndings. Second, we cannot be certain that the variants operate directly and solely through the HMGCR gene, although there is some evidence that these variants alter splicing of HMGCR transcripts. 14 Finally, genetic studies are not completely exempt from the confounders and biases of epidemiological studies-survival and index event biases can aff ect genetic studies, and further work with larger numbers of incident cases would provide more reassurance that the genetic associations with type 2 diabetes are real. However, the associations with body-mass index seem to be statistically robust and provide a mechanism downstream of the HMGCoA-reductase eff ect (increased body-mass index leading to increased insulin resistance, and to increased diabetes).
In summary, Swerdlow and colleagues 1 have used naturally occurring human genetic variation to provide another piece of evidence about the side-eff ects of statins, but have not cast any doubt on the evidence that the benefi ts of statins vastly outweigh their risks.