A large retrospective cohort study has found that glucagon-like peptide-1 receptor agonists (GLP-1RAs) are associated with significantly lower mortality than dipeptidyl peptidase-4 inhibitors (DPP4is) in older adults with cancer and type 2 diabetes (T2D). Published in JAMA Network Open, the study provides new real-world evidence suggesting a potential survival advantage with GLP-1RAs in this high-risk population. By contrast, no significant mortality difference was observed when comparing GLP-1RA users with those prescribed sodium-glucose cotransporter-2 inhibitors (SGLT2is).
Study Background and Rationale
GLP-1RAs are commonly prescribed for glucose control and weight reduction in T2D, but they have recently attracted attention for their potential anticancer effects. These medications reduce glucose levels, adiposity, insulin resistance, and systemic inflammation—factors implicated in cancer progression. While earlier observational data on glucose-lowering drugs and cancer survival have been mixed, this study offers focused evidence on GLP-1RA use in a cancer cohort.
Study Methods
The researchers used Medicare data from 2013 to 2020 to identify adults aged 66 and older with both T2D and one of nine common cancers, including colorectal, lung, breast, prostate, and pancreatic cancer. Participants had to initiate either a GLP-1RA, SGLT2i, or DPP4i and survive at least one year after their cancer diagnosis.
To reduce confounding, 1:1 propensity score matching was performed. Cox proportional hazards models were used to estimate hazard ratios for all-cause mortality under an intention-to-treat framework. Subgroup analyses assessed effect modification by age, sex, race and ethnicity, obesity status, and cancer type.
Key Findings
The comparison between GLP-1RA and DPP4i users included 2,564 matched pairs. GLP-1RA use was associated with significantly lower all-cause mortality, with a hazard ratio (HR) of 0.60 (95% CI: 0.51–0.70). This survival benefit was consistent across key subgroups, including different age groups, sexes, obesity status, and in cancers such as colorectal, lung, and breast. The E-value for this association was 2.73, suggesting moderate robustness to unmeasured confounding.
In contrast, the GLP-1RA versus SGLT2i comparison, which included 2,553 matched pairs, showed no significant difference in mortality. The HR was 1.03 (95% CI: 0.85–1.23), indicating that survival outcomes between these two drug classes were comparable.
Interpretation and Clinical Implications
These findings suggest that GLP-1RAs may offer a survival advantage over DPP4is in older adults with cancer and T2D. Both GLP-1RAs and DPP4is act on the incretin pathway, but GLP-1RAs have stronger effects on glycaemic control, weight loss, and reductions in inflammation and insulin levels—all of which could plausibly contribute to reduced cancer progression and improved overall survival.
Genetic testing can transform surgical decisions and family risk assessment. New tools are making it easier for all clinicians to integrate testing without specialist referrals.
Listen Now | Exclusive to The Oncology Network
The lack of a mortality difference between GLP-1RAs and SGLT2is may reflect shared cardiometabolic benefits, though more mechanistic research is needed to understand how these medications interact with tumour biology.
Conclusion
This Medicare-based cohort study found that GLP-1RA use was associated with significantly lower mortality than DPP4i use in older adults with cancer and type 2 diabetes. The survival benefit persisted across key demographic and clinical subgroups. While no difference was observed when compared with SGLT2i use, these findings highlight the need for prospective trials to confirm the potential role of GLP-1RAs in cancer survivorship strategies.
Limitations
As an observational study using administrative data, there remains the potential for residual confounding. Claims-based data also lack cancer staging, treatment regimens, and cause-specific mortality, which limits insight into whether the mortality reductions were cancer-related. Several subgroup analyses may have been underpowered, and causality cannot be inferred from this design.
Paper: Radwan RM, Lu Y, Dai H, et al. GLP-1 RA Use and Survival Among Older Adults With Cancer and Type 2 Diabetes. JAMA Netw Open. 2025;8(7):e2521887. doi:10.1001/jamanetworkopen.2025.21887
