New oral hypoglycaemics fail to show cardiovascular benefits

BMJ 2013; 347 doi: (Published 06 September 2013) Cite this as: BMJ 2013;347:f5458
  1. Deborah Cohen
  1. 1BMJ

Two trials involving the new oral hypoglycaemic drugs saxagliptin and alogliptin have found that neither reduces the risk of cardiovascular events in the short term, contradicting the findings of an analysis earlier this year.

Saxagliptin and alogliptin are dipeptidyl peptidase-4 (DPP-4) inhibitors and were approved by regulators because they lowered blood glucose in people with diabetes, but it was unclear what effects they might have on the cardiovascular system.

Since 2008, after concerns over the cardiac safety of rosiglitazone, manufacturers of diabetes drugs have to carry out a long term cardiovascular outcome study after approval. One pooled analysis published earlier this year pointed to the possibility that DPP-4 inhibitors lower cardiovascular risk.1

However, both the new trials, published in the New England Journal of Medicine, and conducted as part of the manufacturers’ regulatory obligations, found that this might not be the case.

In the saxagliptin (SAVOR TIMI-53) trial, sponsored by AstraZeneca and Bristol-Myers Squibb, 16 492 patients with a history of, or at risk for, cardiovascular events were randomly assigned to receive saxagliptin or placebo for an average of two years.2

In the alogliptin (EXAMINE) trial, sponsored by Takeda, 5380 patients were randomised to receive alogliptin or placebo after an episode of acute coronary syndrome. They were followed up for an average of 18 months.3

Neither trial showed that the drugs decrease patients’ risk of heart attack, stroke, or death, but nor did they increase it in comparison with placebo (saxagliptin: hazard ratio 1 (95% confidence interval 0.89 to 1.12; P=0.99 for superiority and P<0.001 for non-inferiority); alogliptin: hazard ratio 0.96 (upper boundary of the one sided repeated confidence interval 1.16; P<0.001 for non-inferiority; P=0.32 for superiority)).

Moreover, in the saxagliptin trial there was about a 25% increase in the risk of hospitalisation because of heart failure in the treatment arm compared with placebo (hazard ratio 1.27 (1.07 to 1.51; P=0.007)).

This trial’s lead investigator, Itamar Raz, head of the Center for the Prevention of Diabetes at Hadassah Medical Center in Israel, told the BMJ, “The increase in hospitalisation for heart failure seems real but was not associated with an increase in the primary or secondary endpoint or in death.”

The drop in glycated haemoglobin in the saxagliptin trial compared with placebo at the end of the study was small (7.7% v 7.9%), but it dropped below 7% in more participants in the saxagliptin group (36.2% v 27.9%).

These data were not provided in the alogliptin trial, so the BMJ asked Takeda for them. A spokeswoman told the BMJ, “HbA1c levels were significantly lower in patients on alogliptin than placebo (mean difference of −0.36%, P< 0.001).” She did not give the exact glycated haemoglobin levels at the end of the trial.

Concern has been raised about the potential for DPP-4 treatment to increase the risk of pancreatitis and other pancreatic disease. A BMJ investigation found unpublished data of raised pancreatic enzyme concentrations in humans,4 and unpublished data of pancreatitis in clinical trials. A study published in Diabetes earlier this year found an increase in pancreas size of 40% in people taking another DPP-4 drug, sitagliptin.5

Shortly after publication of the BMJ investigation, the American Diabetes Association asked all companies involved in the development or marketing of such drugs to make patient level data available for an independent review.6

The saxagliptin trial reported definite acute pancreatitis in 17 patients (0.2%) in the treatment arm and nine patients (0.1%) in placebo (P=0.17). Chronic pancreatitis occurred in two patients taking saxagliptin (<0.1%) and six patients in the placebo arm (0.1%) (P=0.18). There were five cases of pancreatic cancer in the saxagliptin group and 12 in the placebo group (P=0.095).

The published alogliptin trial did not report the number of events of pancreatitis—although the authors stated that the “incidences of acute and chronic pancreatitis were similar in the two groups; no cases were fatal.” But such omission can cause problems for future meta-analysis.

Takeda did not provide the number of events of pancreatitis and pancreatic cancer when asked by the BMJ. A spokeswoman said, “Rates of pancreatitis were similar for alogliptin and placebo groups. No events of pancreatic cancer were reported during the trial.”

There were no data about baseline increases in enzymes or changes in pancreas size obtained by scans. The BMJ asked if these data were recorded.

Raz told the BMJ that because it was a cardiovascular outcome study routine scans of the pancreas and enzymes were not done. “We didn’t measure enzymes unless there was a suspicion of pancreatitis but we do have the patients’ serum after two years under the drug and we can do it in the future,” he said.

The spokeswoman from Takeda did not answer the question.

The difference in HbA1c was small, but the study confirms previous concerns that HbA1c is a poor surrogate for cardiovascular risk.

Writing an accompanying perspective article7 in the New England Journal of Medicine, William Hiatt, Sanjay Kaul, and Robert Smith—all members of the FDA advisory committees that debated the risks of rosiglitazone—said, “The evidence does not support the use of glycated haemoglobin as a valid surrogate for assessing either the cardiovascular risks or the cardiovascular benefits of diabetes therapy. The optimal approach to the reduction of cardiovascular risk in diabetes should focus on aggressive management of the standard cardiovascular risk factors rather than on intensive glycaemic control.”


Cite this as: BMJ 2013;347:f5458