GLP-1 Receptor Agonists in Cancer: Reduced Mortality Shown

Abstract

Study at a Glance

Study Snapshot

Study Facts Table

What Researchers Actually Did

Mahadevan and colleagues queried TriNetX, a federated global EHR network encompassing more than 120 million patients, to assemble two distinct analytic cohorts. The overall cohort included patients with ICD-10-coded T2D who received any GLP-1RA within 3 months before or after starting systemic cancer therapy, compared against patients receiving metformin in the same window. Metformin users were actively excluded from the GLP-1RA arm to enforce monotherapy comparisons. The new-start cohort isolated patients already established on metformin who initiated either a GLP-1RA or insulin within 3 months of their first systemic cancer therapy. Cancer types spanned the top 10 causes of cancer-related death in men and women, incorporating both solid tumor and hematologic malignancies diagnosed from January 2010 forward.

Prior to survival analysis, cohorts underwent 1:1 propensity score matching on age, sex, race, ethnicity, cancer type, presence of distant metastases, surgery, radiation, systemic therapy type, hyperlipidemia, hypertension, smoking, chronic kidney disease, and obesity. Kaplan-Meier survival analyses were landmarked at 6 months after initiation of systemic therapy to mitigate immortal time bias; all analyses required a minimum 3-month gap between any GLP-1RA or metformin initiation and the start of survival follow-up. HbA1c subgroup analyses used values measured within 3 months of systemic therapy initiation. Multiple comparisons were addressed via the Benjamini-Hochberg procedure.

Key Findings: Primary Outcomes

• All-cause mortality, overall cohort: GLP-1RA monotherapy was associated with a statistically significant reduction in mortality compared to metformin monotherapy (HR 0.875; 95% CI 0.778–0.985; P = .0268). Kaplan-Meier curves were followed from 6 months to 2.5 years after cancer treatment initiation.
• All-cause mortality, new-start cohort: Among patients on metformin who initiated a GLP-1RA at the time of cancer therapy versus those who initiated insulin, mortality was further reduced (HR 0.786; 95% CI 0.662–0.934; P = .0062), reflecting a more pronounced signal when the insulin comparator, rather than metformin, serves as the baseline.
• Post-match cohort balance: After propensity matching, 3,551 patients were compared in each arm of the overall cohort. Mean age was 65.3 years in both arms. The GLP-1RA cohort included a modestly lower proportion of White patients (72.8% vs. 75.0%; P = .0375), the only demographic variable achieving post-match statistical difference.

Key Findings: Secondary Outcomes and Subgroup Analyses

Secondary Outcomes (GLP-1RA vs. Metformin)

• All-cause hospitalization: RR 0.796 (95% CI 0.741–0.855; P = .0004) — a 20% relative reduction.
• Pneumonia: RR 0.586 (95% CI 0.475–0.722; P = .0004) — roughly a 41% relative reduction.
• Sepsis: RR 0.735 (95% CI 0.605–0.894; P = .004).
• Major adverse cardiovascular events (MACE): RR 0.722 (95% CI 0.578–0.901; P = .0053).
• Pulmonary embolism: RR 0.608 (95% CI 0.440–0.840; P = .004).
• Diabetic ketoacidosis: RR 1.176 (95% CI 0.628–2.204) — numerically elevated but not statistically significant.
• Hyperosmolar hyperglycemic state: RR 0.558 (95% CI 0.268–1.162) — not statistically significant.

Subgroup Analyses

• HbA1c-stratified survival: Trends toward improved survival with GLP-1RA were observed across optimal (HbA1c 6.5–7%; HR 0.830; 95% CI 0.675–1.020), moderate (7–9%; HR 0.821; 95% CI 0.665–1.014), and suboptimal (>9%; HR 0.762; 95% CI 0.539–1.077) glycemic control, but none met statistical significance after Benjamini-Hochberg correction (all P = .128).
• Obesity subgroup: Obese patients showed a trend toward improved survival (HR 0.877; 95% CI 0.759–1.013; P = .156) that did not reach significance; non-obese patients demonstrated no survival difference (HR 1.036; 95% CI 0.838–1.279; P = .746).
• The authors attribute the failure of these subanalyses to reach significance primarily to reduced power, as subgroup samples ranged from 11% to 28% of the primary cohort size.

Adverse Events and Safety Profile

Within the secondary outcomes examined, diabetic ketoacidosis occurred numerically more often in GLP-1RA users (RR 1.176; 95% CI 0.628–2.204), though the wide confidence interval and lack of statistical significance preclude firm conclusions. No prior literature has explicitly examined the GLP-1RA-DKA relationship in patients with active cancer, and the authors acknowledge this gap. Hyperosmolar hyperglycemic state rates were similar between groups (RR 0.558; 95% CI 0.268–1.162; NS). The database methodology precludes granular adverse event capture beyond these ICD-10-coded outcomes; patient-reported events, infusion reactions, gastrointestinal toxicity from GLP-1RAs, or events not triggering a coding encounter are undetectable in this dataset.

Statistical Approach and Rigor

The study employed 1:1 propensity score matching on 14 prespecified covariates, which is a defensible strategy in a large observational dataset. The use of 6-month landmark analysis is methodologically important: it limits immortal time bias that would otherwise artificially inflate benefits in the GLP-1RA arm, since patients must survive long enough to be included. The Benjamini-Hochberg false discovery rate procedure was applied for multiple comparisons across secondary and subgroup outcomes. However, several statistical realities warrant scrutiny. The primary cohort is substantially imbalanced before matching (3,747 GLP-1RA users vs. 52,061 metformin users), meaning a large fraction of the metformin pool was discarded. The survival follow-up

Physician-Led, Whole-Person Care

A doctor who takes the time to truly understand you.

Personal care that starts with listening and is guided by experience and ingenuity.

Health, Longevity, Wellness

One-on-One Cannabis Guidance

Metabolic Balance

Leave a Message
Metabolic Care
Medical Consulting
Cannabis Care