香蕉直播

Key Points

• In one of the largest real-world data-based studies to date, GLP-1 receptor agonist use was associated with improved overall survival in patients with cancer in a large, nationally representative community oncology setting.
• GLP-1 receptor agonist use was associated with improved overall survival in fully adjusted observational analyses.
• Because this was an observational study, the findings should be interpreted with appropriate caution and validated through additional studies, including replication in additional real world data sources that engage robust causal inferential methods, and those that leverage randomized trial data.

Our team at Ontada conducted one of the largest real-world studies so far to explore whether using GLP-1 receptor agonists affects overall survival for cancer patients. We analyzed electronic health records from a large community oncology group in the U.S. to see if this topic deserves more research. Our results show there may be an important link, but they also highlight the need for careful analysis and additional studies.

Why did we study GLP-1 receptor agonists in oncology?

Glucagon-like peptide-1 receptor agonists, or GLP-1 receptor agonists, are medicines that mimic a naturally occurring hormone involved in blood sugar regulation and appetite. These therapies are used to treat type 2 diabetes, support chronic weight management and reduce cardiovascular risk.

In recent years, interest in these therapies has grown beyond metabolic health as researchers explore whether they may also influence outcomes in other disease areas, including cancer.

At the same time, large-scale studies across multiple cancer types have remained limited, especially in community oncology settings where many patients receive care.

That gap led our research team to ask a straightforward question: In a large real-world oncology population, is GLP-1 receptor agonist use associated with differences in overall survival among patients diagnosed with cancer?

How did we study GLP-1 receptor agonist use in cancer care?

To examine that question, we conducted a retrospective observational cohort study within The US Oncology Network, using structured electronic health record data from , an oncology-specific system that captures outpatient encounters.

The analysis included adult patients diagnosed from January 2021 through October 2024 with one of six solid tumor types: breast, colorectal, hepatocellular, lung, prostate or renal cell carcinoma. Patients who had a prescription for a GLP-1 receptor agonist documented within the oncology electronic health record during the study observation period were classified as users. Among GLP-1 receptor agonist users in the cohort, semaglutide was the most commonly prescribed therapy. They were matched to non-users on age quintile, cancer type and diagnosis year.

We then used Cox proportional hazards models to evaluate the association between GLP-1 receptor agonist use and overall survival. The multivariable and propensity score models adjusted for age at diagnosis, sex, race, ethnicity, geographic region, rural versus urban location, body mass index, stage at diagnosis and cancer type.

What did the analysis find?

Across models, GLP-1 receptor agonist use was associated with improved overall survival. In the fully adjusted analyses, GLP-1 receptor agonist use among patients diagnosed with cancer was associated with a 34% lower rate of death compared with non-users.

That pattern was not limited to a single modeling approach. The unadjusted, multivariable adjusted and propensity score adjusted models all favored GLP-1 receptor agonist use, which demonstrates that the observed association was consistent across the analytic framework used in this study.

Our team recently presented these findings at ASCO, but the larger story is not the conference itself. It is what real-world evidence can help surface: meaningful signals at scale — often much more quickly than traditional evidence generation approaches — that may help the research community identify where deeper investigation is warranted.

Why do these findings matter?

For researchers, the results add to ongoing discussion about whether GLP-1 receptor agonists may be associated with outcomes beyond the indications for which they are currently approved.

For oncology, that raises important questions about whether some patients could benefit differently based on tumor type, treatment context or other clinical factors.

For the broader healthcare community, the study also reinforces the value of real-world data. Community oncology data can help generate evidence from routine care settings across diverse patient populations, complementing what we learn from traditional clinical trials.

When used thoughtfully, these data can help researchers spot patterns earlier, refine hypotheses and guide where more rigorous study should follow.

What are the limitations of this research?

The findings should be interpreted with appropriate caution. This was an observational study, which means it can show associations but cannot establish causality.

Residual and unmeasured confounding remain possible despite multiple approaches to confounding control. A healthy user effect may still be present. The analysis was also limited to structured data within an oncology-specific electronic health record system, which likely means GLP-1 receptor agonist use was underestimated. In addition, although the study design assigned non-users to an anchor date to address immortal time bias, residual bias is still possible.

Although these limitations don't render the signal insignificant, they indicate it should serve as a starting point for more research rather than as proof of treatment effectiveness. It is important to note that GLP-1 therapies have not been authorized for use in cancer treatment.

What comes next for GLP-1 and oncology research?

The next step is to test these findings across additional data sources and study designs. Future research should encompass diverse data sources, including randomized trials and claims data, use causal inference methods for observational data and identify patient subgroups who may benefit most if a true effect exists.

This is where real-world evidence can continue to play an important role. By helping researchers isolate clinically meaningful questions in large, representative populations, real-world data can accelerate the path from signal to understanding. That is ultimately the value of this work: not overstating what we know today but advancing the questions that matter for tomorrow.