Study: Diabetes drug could reduce death rate in overweight people

Obesity, especially abdominal obesity, is associated with various health problems such as diabetes, high blood pressure and cardiovascular disease. Various weight loss methods have been tried over the years, but most have not been effective.

Recently, a class of drugs called GLP-1 agonists, originally used to treat diabetes, has attracted attention for their role in weight loss. These drugs mimic the gut hormone GLP-1, which increases insulin secretion and slows digestion, promoting a feeling of fullness. Among these drugs, semaglutide has been used to treat diabetes since 2017. In higher doses, it is now used as a weight loss drug in Western countries. A recent study published in Journal of the American College of Cardiology showed promising results when used in people without diabetes.

The SELECT trial, funded by the makers of semaglutide, involved 17,604 obese or overweight participants with cardiovascular disease but no diabetes. They were randomly assigned to receive weekly injections of either semaglutide or a placebo and followed for three years. The goal was to see if deaths, heart attacks and strokes decreased because the drug is known to reduce weight. During follow-up, 833 people (4.7%) died. Compared to the placebo group, participants who received semaglutide had lower rates of death from all causes, including cardiovascular, non-cardiovascular and also COVID-19 deaths.

The main results of the SELECT study have been published previously The New England Journal of Medicine in December 2023. The highlight was a 19 percent reduction in cardiovascular mortality, heart attacks and strokes in the semaglutide group.

But these results also raised questions. For example, why was there a reduction in deaths so early—before participants had even lost weight? It usually takes longer for the health benefits of weight loss to become apparent. For example, a Swedish study of bariatric surgery by Sjöström et al., which also showed a reduction in deaths from weight loss, had an average follow-up of 10.9 years. Although weight loss occurred in the first year after surgery, the reduction in deaths did not occur until much later.

Second, the participants in the SELECT trial were not diabetic, meaning that the known antidiabetic effects of GLP-1 agonists cannot account for the lower mortality. Finally, there was an unexpected reduction in COVID-19 deaths. The vaccination status of the participants was not mentioned in the study. This leaves unanswered questions about how GLP-1 agonists might affect COVID-19 mortality.

When a clinical trial finds an unexpected benefit in the treatment group compared to placebo, two possibilities arise: either the treatment is actually better, or the participants in the placebo group were in worse health at the start of the trial. The surprising results of this trial were the earlier-than-expected reduction in deaths and the apparent impact on COVID-19 mortality. This warrants a closer look at whether there were important differences between the two groups at baseline.

Large randomized trials such as SELECT can minimize such discrepancies. Accordingly, there were no major differences between the semaglutide and placebo groups in terms of age, gender, HbA1c, blood pressure, cholesterol levels, BMI or waist circumference.

However, the supplementary tables comparing baseline parameters of those who died with those of survivors show a striking difference in the use of loop diuretics. Of the survivors, only 11.7% used loop diuretics at the start of the study, compared with 35.9% of those who later died of cardiovascular causes. Loop diuretics are commonly prescribed for advanced heart, liver, and kidney disease, and their use could serve as an indirect indicator of the severity of participants' health status. This suggests that many participants who died during the course of the study may have had more advanced heart disease to begin with.

Although randomization generally ensures a balanced distribution of participants, some unevenness can still occur. The total number of people with heart failure was comparable in both groups, which included different levels of severity. However, the paper does not indicate whether the placebo group had a higher proportion of people on loop diuretics, which could indicate more advanced heart failure. Such a discrepancy could potentially have contributed to higher death rates during follow-up. This could also explain the unexpected difference observed in COVID-19 deaths. It is also possible that properties of GLP-1 agonists other than weight loss and control of diabetes are at play here.

Research is essential to advancing medical knowledge. Although unexpected results are not uncommon, it is important to explore all possible explanations before drawing conclusions. GLP-1 agonists are already recommended for people with diabetes, and this study suggests that they may also be of benefit to overweight and obese individuals without diabetes. These findings could have implications for medical practice, particularly if further studies confirm the results.

(Rajeev Jayadevan is Chairman of the Research Cell of IMA in the state of Kerala)