GLP-1 Benefits, Risks Detailed in ‘Atlas’ Mapping Outcomes

A study of more than two million people with diabetes builds on evidence of broad-ranging benefits and risks of glucagon-like peptide 1 receptor agonists (GLP-1 RAs) in the clinical setting, providing an “atlas” mapping extensive outcomes and some new insights to potentially explore in more rigorous clinical trials.

“This is the largest study on GLP-1 receptor agonists,” first author Ziyad Al-Aly, MD, chief of research and development at the VA [veterans affairs] St. Louis Health Care System, in St. Louis, told Medscape Medical News regarding the research, published this week, in Nature Medicine.

“The [study] reflects the real experiences of people using GLP-1 RAs [in the VA] clinical setting,” he said.

“Altogether, our discovery approach confirms previous studies and clinical trials and also uncovers previously unreported benefits and risks of GLP-1 RAs,” the authors wrote.

For the comprehensive study, Al-Aly and his colleagues evaluated data from the US Department of Veterans Affairs on more than two million veterans treated for diabetes between October 2017 and December 2023, assessing GLP-1 RA treatment in comparison with other diabetes therapies regarding a striking 175 clinical outcomes.

Of the patients, 215,970 initiated treatment with GLP-1 RAs; 159,465 started sulfonylureas, 117,989 dipeptidyl peptidase 4 inhibitors, and 258,614 were initiated on sodium-glucose cotransporter-2 inhibitors.

The study also included a composite group of the latter three drug groups (n = 536,068), and a control group of 1,203,097 of patients receiving usual care, who were compared with usual care with the addition of GLP-1 RAs.

After inverse probability weighting, the groups were well-balanced in terms of their baseline characteristics. While the majority in the VA cohort overall were White men, the study adjusted for gender, age, race, comorbidities, and an extensive array of covariates.

With an average follow-up of 3.68 years, after the multivariate adjustment, GLP-1 RAs showed “effectiveness and risks that extended beyond those currently recognized,” in comparison with each of the treatment groups and with the main control group of usual care, the authors reported.

For the largest comparison with the main control group of usual care alone, the addition of GLP-1 RAs was associated with a decreased risk in 24% of the outcomes evaluated, and an increased risk in 10.86% of outcomes, with no significant difference for the remaining 65.14% of outcomes.

Of the various benefits, key improvements included a reduced risk for several substance use disorders including alcohol (hazard ratio [HR], 0.89) and opioid (HR, 0.87) use, suicidal ideation, attempt or intentional self-harm (HR, 0.90), seizures (HR, 0.90), neurocognitive disorders including Alzheimer’s disease (HR, 0.88) and dementia (HR, 0.92), coagulation and clotting disorders (HR, 0.92), and cardiac arrest (HR, 0.78).

Further benefits vs usual care alone included a reduced risk for infectious illnesses (HR, 0.88), acute kidney injury (HR, 0.88), and chronic kidney disease (CKD) (HR, 0.97; P

In terms of risks associated with GLP-1 RAs, in addition to the well-known risks for nausea and vomiting, additional increased risks vs usual care included gastrointestinal disorders such as noninfectious gastroenteritis (HR, 1.12), hypotension (HR, 1.06), arthritis (HR, 1.11), tendinitis and synovitis (HR, 1.10), interstitial nephritis (HR, 1.06), nephrolithiasis (HR, 1.15), and the known risk for drug-induced acute pancreatitis (HR, 2.46).

Neuropsychiatric Effects

Among the various benefits in the study, Al-Aly said some of the most intriguing are those involving the brain.

“I am struck by the consistent effects on many neuropsychiatric disorders — this aligns with data showing the presence of GLP-1 receptors in the brain and evidence showing that GLP-1s permeates through the blood brain barrier and acts on the brain to reduce inflammation and oxidative stress, improve neuroplasticity, etc.,” he said.

“Clearly, there is a neurotropic effect. There is also the possibility of an effect on the immune system/fighting infection — with reduced risks of infections, sepsis, etc.”

The reductions in suicidal ideation are encouraging after earlier reports of suicidal thoughts and self-injury among young users of GLP-1 RAs prompted concerns, including a 2023 review of the drug use by the European Medicines Agency that ultimately found no causal association, the authors added. The US Food and Drug Administration also found no association with GLP-1s and suicide risk.

The reductions in addictive behaviors are also encouraging and are consistent with the role of GLP-1 receptors in the brain in terms of impulse control and reward signaling that can relate to addictive behaviors, Al-Aly explained.

The reduced risks for dementia and Alzheimer’s disease are likewise consistent with preclinical studies in animal models of Alzheimer’s disease, as well as clinical studies showing a reduced risk for dementia in patients with type 2 diabetes, the authors noted.

The observed reduced risk for seizures further “adds to an emerging body of knowledge, both mechanistic and early clinical data, indicative of the anticonvulsant properties of GLP-1 RA use,” they added.

“GLP-1 RAs should be further evaluated in future studies as potential adjuvant therapeutics for epilepsy and its associated comorbidities,” the authors suggested.

Kidneys

While the findings support evidence of protective effects of GLP-1 RAs on the kidneys and a reduction in CKD risk, notable risks observed, also involving the kidneys, include nephrolithiasis or kidney stones.

Al-Aly noted the mechanisms with kidney stone formation are very different from CKD, and he speculated that the risk for the former could in fact stem from potentially low hydration with GLP-1 RA use.

“When patients are on GLP-1 RAs, they definitely eat a lot less to lose weight, but they also hydrate themselves less,” he explained in a press briefing. “They drink less water because they feel full very quickly after eating, and I’m just theorizing, but perhaps chronic dehydration [is behind] the increased risk of kidney stones.”

Modest Effects?

While, overall, the benefits of GLP-1 RA drugs showed modest benefits ranging between a 10% and 20% reduction for most outcomes, Al-Aly said those effects are still important.

“The modest effect does not negate the potential value of these drugs, especially for conditions where few effective treatment options exist, for example, dementia,” he said in the press statement.

“This may also imply that these drugs are most beneficial when used in conjunction with other interventions, such as lifestyle changes or other medications.”

Potential Confounders A Concern

Commenting on the study, David M. Nathan, MD, founder of the MGH Diabetes Center and a professor of medicine at Harvard Medical School, in Boston, Massachusetts, noted that, while the study is hypothesis-generating, the key limitation is its observational nature.

“The authors did a perfectly respectable job of doing all you can do to adjust for [confounders], but with these kinds of studies, as much as you try to statistically account for differences in the populations before they were put on the drug, you can never truly adjust for all the potential confounders that may influence the results,” he told Medscape Medical News.

In addition, the 3.8-year follow-up time of the study, as the authors acknowledge, is especially short considering that GLP-1 RAs are generally recommended to be taken indefinitely.

“You have to take these drugs presumably for a lifetime and we have no idea what the longer-term benefits and risks are,” Nathan said.

Nathan, who was among the first investigators to evaluate GLP-1 RAs about 30 years ago, underscored that “I do think that these drugs are generally really spectacular; they’ve taken over the world and they are probably the single greatest pharmaceutical story of the 21st century.”

“But much more rigorous randomized trials would be needed to prove study results that haven’t already been established in previous clinical trials,” he said.

“The types of [randomized] trials that are necessary are very expensive and require a huge amount of work, but at the end of the day, they provide proof as to what does and doesn’t work, and what the true risks are,” he added. “Whether the GLP-RAs will cure all ills and bring about world peace needs to be proved.”

In further comments provided through the Science Media Center, Stephen O’Rahilly, FRS, a professor of clinical biochemistry and medicine and director of the Wellcome-MRC Institute of Metabolic Science-Metabolic Research Laboratories, University of Cambridge, Cambridge, England, echoed Nathan’s concern that “studies such as these have to be interpreted very cautiously as the people studied have not been randomly allocated to GLP-1 RA treatment, so any difference between those taking and not taking the class of drug could potentially be attributable to factors other than the drug.”

He noted, however, that “the study provides useful reassurance about the safety of this class of drugs. The expected benefits on heart disease, stroke and other cardiovascular and most kidney diseases are clearly seen.”

Al-Aly reported being an uncompensated consultant for Pfizer. Nathan, who has previously conducted clinical trials on GLP-1 RAs, currently has no relationships to report. O’Rahilly reported receiving remuneration from several pharmaceutical companies for scientific advice relating to the development of drugs for metabolic diseases, but none involving GLP-1 RAs in the past 3 years.