Metformin Use Linked to Fewer Asthma Attacks

Treatment with metformin was associated with a reduction in asthma attacks among patients with asthma and type 2 diabetes, with additional reductions observed with the use of GLP-1 receptor agonists, according to a self-controlled case series and population-based cohort study.

Among over 12,000 patients, use of metformin was associated with fewer asthma attacks in both the self-controlled case series analysis (incidence rate ratio [IRR] 0.68, 95% CI 0.62-0.75) and an analysis of a new user cohort with inverse probability of treatment weighting (IPTW; HR 0.76, 95% CI 0.67-0.85), reported Chloe I. Bloom, MSc, PhD, of Imperial College London, and colleagues in JAMA Internal Medicine.

Of the other antidiabetic medications evaluated, only GLP-1 receptor agonists appeared to have an additive association (IRR 0.60, 95% CI 0.49-0.73).

“Metformin had a surprisingly strong effect, reducing asthma attacks by about 30%. Furthermore, it didn’t seem to be affected by metabolic factors [body mass index (BMI), glycemic control] or asthma phenotype [type 2 inflammation, asthma severity],” Bloom told MedPage Today.

“We also saw a huge effect from adding GLP-1s, with an additional 40% reduction in asthma attacks. These two diabetes medications together therefore appear to be more effective than asthma biologics,” she added.

Many asthma patients also have type 2 diabetes, and are at higher risk for asthma attacks. Previous studies have suggested that metformin and GLP-1 receptor agonists may reduce these attacks.

“Many asthma patients have comorbid metabolic conditions; we need to be thinking about this in our routine asthma care,” Bloom said. “The landscape for treating metabolic conditions is rapidly changing, [and] many of these medications could treat both asthma and their metabolic condition. Metformin is cheap, easily available, and an oral medication; these factors make it a very attractive drug to be repurposed for asthma.”

In an accompanying editorial, Katherine N. Cahill, MD, of Vanderbilt University Medical Center in Nashville, Tennessee, and Dinah Foer, MD, of Brigham and Women’s Hospital in Boston, wrote that the study findings suggest that asthma’s response to metformin in those with diabetes was independent of asthma inflammatory phenotypes, which predict clinical response to current therapies like biologics.

“In this regard, the [current] analyses … go beyond supporting the preferential use of these medications in patients with comorbid type 2 diabetes and asthma,” Cahill and Foer wrote, “raising the question if metformin and GLP-1RAs [receptor agonists] should be considered for asthma therapy without type 2 diabetes.”

They noted that there have been no similar phase III randomized trials of type 2 diabetes medications in a cohort with diabetes that evaluated chronic respiratory disease as a primary or secondary outcome. “This oversight may reflect our siloed biomedical enterprise, competing pharmaceutical industry interests, and the strong headwinds investigator-initiated trials face” in getting their trials off the ground, they wrote.

Such trial data are needed, they emphasized, although they cautioned that “the enthusiasm for repurposing medications, especially GLP-1RAs, should not only be tempered by concerns about cost and equitable access, but also by cautionary tales from past efforts to borrow from the diabetes armamentarium for asthma,” like thiazolidinediones and peroxisome proliferator-activated receptor γ agonists, which “failed to demonstrate a significant benefit” for asthma in clinical trials.

For this study, Bloom and colleagues used data from the U.K. Clinical Practice Research Datalink Aurum, a database of primary care electronic health records, which were linked to Office of National Statistics mortality data and Hospital Episode Statistics English hospital admission data from 2004 to 2020. Patients with a diagnosis of type 1 diabetes, chronic obstructive pulmonary disease, and chronic kidney disease were excluded.

In the self-controlled case series, each patient acted as their own control to account for any confounding, including genetics, socioeconomic status, or metabolic dysfunction. Eligible participants had to have had a 12-month “washout period” free of asthma attacks, followed by an observation period of 12 months before starting metformin, and 12 months after, that included an asthma attack within those 24 months.

This analysis also evaluated other antidiabetic medications, including DPP-4 inhibitors, sulfonylureas, SGLT2 inhibitors, and insulin, but none of these showed associations with a reduction in asthma attacks.

In the IPTW analysis, the researchers included patients who were either exposed to metformin within the first 12 months of cohort entry, or unexposed to metformin. Patients were only eligible to be included in the unexposed group if they were eventually prescribed metformin to reduce indication bias. In this analysis, participants were censored at 12 months, change of exposure status, first asthma attack, or cohort end date, whichever came first.

They included 4,278 patients in the self-controlled case series (mean age 52.9, 61.2% women) and 8,424 in the IPTW cohort (55.7% women, mean age 61.6 in the unexposed group and 59.7 in the exposed group). In the self-controlled case series, 70.8% of participants had a BMI over 30, as did about two-thirds of patients in the IPTW cohort.

The researchers noted that medication adherence, metformin dose, and changes in weight could not be assessed. In addition, some patients may have been misdiagnosed with asthma, and there was no information available on prescription of asthma biologics.