ACCORD: Do Genes Explain CV Deaths in Intensive-Control Arm?

Liam Davenport

August 23, 2016

Genetic variants in two regions of DNA not previously implicated in coronary artery disease may explain the paradoxical increase in cardiovascular mortality seen with intensive glycemic control in the Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial, new results from a genomewide analysis indicate.

The ACCORD trial, which compared intensive and standard glycemic control among type 2 diabetes patients at risk of cardiovascular disease, had to be stopped early due to an increased mortality risk among those in the intensive-control arm.

In the current analysis, Dr Hetal S Shah of Harvard TH Chan School of Public Health, Boston, Massachusetts, and colleagues found that two single nucleotide polymorphisms (SNPs) were associated with a more than twofold increase in cardiovascular mortality among patients in the intensive glycemic control arm.

Moreover, a score developed from these two SNPs indicated that patients with the highest score had a threefold increased mortality risk, while those with the lowest score had a fourfold reduction in mortality with intensive glycemic therapy. In contrast, the two SNPs were not associated with increased mortality in patients undergoing standard glycemic control.

Personalized Glycemic Control Targets Depending on Genetics?

The research, which was published online recently in Diabetes Care, indicates that the two genetic markers could possibly be used to identify type 2 diabetes patients who would benefit most from intensive glycemic control, although senior author Alessandro Doria, MD, PhD, MPH, associate professor in the department of epidemiology at Harvard TH Chan School of Public Health, urged caution in interpreting the findings.

Speaking to Medscape Medical News, he pointed out that these genetic markers "don't say anything about the cardiovascular risk in people who are not treated in an intensive way" and are "silent, if you will, when you don't apply this pressure."

However, he said that clearly something happens when glucose levels are intensively lowered that could be related to the intervention or to the physiological responses that are regulated by these markers and that the most "interesting" aspect is that these SNPs have not previously been implicated as predictors of coronary artery disease events.

If the markers are validated in further studies as predictors of cardiovascular mortality in patients undergoing intensive glycemic therapy, Dr Doria believes that it could take the treatment of diabetes a step closer to personalized, or precision, medicine.

"We are not there yet, to be able to say: 'Okay, tomorrow we will have new guidelines to use these markers for that purpose,' " he said, noting there are not that many examples of genetic markers that can guide treatment being identified. Nevertheless, "This is a significant step in that kind of direction.

"The idea that one treatment fits everybody is, of course, not right, and we know that. We have tried to use clinical features, so far, to try to identify the best treatment for a specific person. Now, perhaps, we may have tools like these genetic markers; although, again, these are early days."

Results of Genetic Substudy of ACCORD

In the original ACCORD study, 8174 participants from the overall population of 10,251 individuals with type 2 diabetes and high cardiovascular risk who were randomized to intensive and standard glycemic arms consented to give DNA samples.

Here, the researchers performed genomewide genotyping for 6,839,462 million common variants on genomic DNA extracted from white blood cells and examined their association with cardiovascular mortality in 2667 self-reported white subjects from the ACCORD intensive-treatment arm.

Any significant loci that they identified were then tested in the overall ACCORD white genetic data set of 5360 individuals for their impact on the cardiovascular response to the type of glycemic treatment.

In addition, the team used data from the Joslin Kidney Study in Type 2 Diabetes (JSK) on 422 self-reported white participants for whom DNA samples were available to examine the interaction between the significant loci and the effect of long-term glycemic control on cardiovascular mortality.

The team found two loci that reached genomewide significance, one on chromosome 10 and the other on chromosome 5.

The SNP on chromosome 10 (rs9299870) had a minor allele frequency (MAF) of 0.08 and was associated with a significantly increased risk of cardiovascular mortality, at a hazard ratio (HR) of 3.58 (P = 9.77x10-2). The rs57922 SNP on chromosome 5 had an MAF of 0.48 and was associated with an HR for cardiovascular mortality of 2.65 (P = 2.04x10-8).

A further 22 other loci were identified that were linked with cardiovascular mortality in the ACCORD data set, but the associations did not reach significance, including the MLB2 gene, which is associated with coronary artery disease, and the PFKP gene, which is linked to body mass index.

The team then developed a genetic risk score (GRS) using the rs9299870 and rs57922 SNPs.

Among ACCORD patients with a GRS score of 0, representing 22.6% of participants, assignment to intensive therapy was associated with substantial and significant reductions in cardiovascular mortality and nonfatal myocardial infarction (MI), at HRs of 0.24 and 0.56, respectively. Among participants with a GRS score of 1 (47.7% of the ACCORD population), assignment to intensive glycemic control had no impact on cardiovascular mortality but reduced nonfatal MI, at HRs of 0.92 and 0.70, respectively.

Among those with a GRS score of 2, however (29.6% of ACCORD patients), the respective HRs were 3.08 for CV mortality and 0.95 for nonfatal MI, respectively, suggesting a major impact on cardiovascular death with intensive therapy.

The association between GRS score and excess cardiovascular mortality was not affected by controlling for known risk factors, the presence of neuropathy, or the use of aspirin at baseline.

The team were also able to confirm the impact of GRS score on the association between glycemic control and cardiovascular mortality in the JSK cohort (P = .029).

They write: "Our additional analyses suggest that these loci could be potentially used as screening tools to identify subjects with type 2 diabetes who may highly benefit from intensive glycemic control rather than derive harm from it, although further validation is needed.

"These two loci also point to novel candidate pathways linking glycemic control to cardiovascular outcomes, the study of which may lead to the development of new interventions to prevent cardiovascular disease in diabetes."

Dr Dario explained that one of the issues with investigating these associations further is that there is "only one ACCORD trial" and, while there have been similar studies, it remains to be seen whether other clinical trials can identify the same genetic markers.

As a result, he said, it is "early days" for these genetic markers, and they are "not yet ready for prime time" in terms of stratifying patients and deciding who would benefit from intensive therapy.

He continued: "There have been companies, not in this field but in other fields, that, based on results of association studies, are proposing genetic testing and so on and so on. I don't think that we are there yet for complex, multifactorial disorders.

"On the other hand, this is as close as we've gotten at this point, where we're trying to guide therapy based on genetic predictors."

Dr Doria said that the team also intend to examine further the 22 loci that were found to have nonsignificant associations with cardiovascular mortality.

He noted: "Of course, the significance you get is a function of the strength of the signal on the one hand and, on the other, of power; basically, whether you have, for that intensity of the signal, enough observations to be able to reduce the error to a point where you can say this is significant according to those stringent criteria."

Dr Doria added: "Certainly, there is more work to do to mine that section of the data that were suggestive but not significant."

The ACCORD genomewide association analysis was supported by National Institutes of Health (NIH) grants and by an NIH grant (Advanced Genomics and Genetics Core of the Diabetes Research Center at the Joslin Diabetes Center) and the National Center for Advancing Translational Sciences (NCATS), NIH. The following companies provided study medications, equipment, or supplies: Abbott Laboratories, Amylin Pharmaceuticals, AstraZeneca, Bayer, Closer Healthcare, GlaxoSmithKline, King Pharmaceuticals, Merck, Novartis, Novo Nordisk,Omron Healthcare, Sanofi, Schering-Plough, and Takeda Pharmaceuticals. The authors report no relevant financial relationships.

For more diabetes and endocrinology news, follow us on Twitter and on Facebook.

Diabetes Care. Published online August 16, 2016. Abstract


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.