The headline numbers
- Roughly a third lower odds: Women taking GLP-1 medications had 35.1 percent lower odds of breast cancer in the full cohort and 30.5 percent lower in a tightly matched group.
- Large sample: 111,646 women aged 45 to 80, all with a BMI of 25 or higher.
- Association, not proof: The study was retrospective and observational. It flags a pattern worth chasing, not a reason to prescribe.
- Real gaps: It did not track which GLP-1 drug women used, for how long, their genetic risk, or the cancer type at diagnosis.
The study at a glance
| Parameter | Detail |
|---|---|
| Presentation | 2026 ASCO Annual Meeting, Abstract 10506 (June 2, 2026) |
| Publication | JCO Oncology Practice |
| Design | Retrospective observational cohort |
| Full cohort | 111,646 women, ages 45-80, BMI 25 or higher |
| GLP-1 exposed vs unexposed | 15,264 (13.7%) vs 96,382 (86.3%) |
| Matched cohort | 30,528 women, one-to-one controls |
| Matched on | Age, race, ethnicity, BMI, breast density, diabetes status |
| Result | 35.1% lower odds (full), 30.5% lower odds (matched) |
What the researchers actually did
GLP-1 drugs like semaglutide have already reshaped how doctors treat diabetes and obesity. The open question is what else they touch. This study looked at one specific thing: breast cancer.
Dr. Elizabeth McDonald, a professor of radiology at the University of Pennsylvania Perelman School of Medicine, led a team that pulled records on 111,646 women between the ages of 45 and 80, all carrying a BMI of 25 or higher. Of those, 15,264 had taken a GLP-1 medication and 96,382 had not. The team then built a tighter comparison, a matched cohort of 30,528 women, pairing each GLP-1 user one-to-one with a non-user who lined up on age, race, ethnicity, BMI, breast density, and diabetes status.
In the full group, GLP-1 users had 35.1 percent lower odds of developing breast cancer. In the matched group, where the two sides were more alike to begin with, the gap held at 30.5 percent lower odds. The findings went to the 2026 ASCO Annual Meeting as Abstract 10506 and were published in JCO Oncology Practice.
Why a signal like this is plausible
There is a mechanistic reason not to dismiss the result out of hand. Obesity is an established risk factor for several cancers, breast cancer among them, partly through the hormonal and inflammatory changes that come with excess body fat. GLP-1 drugs drive substantial weight loss and improve metabolic health. So a lower cancer rate in people who lost weight on these drugs is not a surprising direction for the arrow to point.
But plausible is not proven, and this is where the study earns credit for restraint. The matching step matters. By controlling for BMI and diabetes status, the researchers tried to separate the drug’s effect from the simple fact that GLP-1 users differ from non-users. The signal survived that adjustment, which makes it harder to wave away as an artifact of who takes these drugs.
The limits, stated plainly
This is an observational study. It cannot establish that GLP-1 drugs lower breast cancer risk, only that the two moved together in this population. McDonald was direct about that boundary.
“While our study was observational and does not definitively confirm an association between GLP-1 medications and reduced breast cancer incidence, it does add to the growing body of evidence suggesting that it’s worth investigating these weight-loss drugs as potential cancer prevention tools,” she said.
The gaps are specific and they matter. The analysis did not account for which GLP-1 medication each woman used, or for how long. It did not capture genetic risk factors, which weigh heavily in breast cancer. And it did not sort by the stage or type of cancer at diagnosis. Any of those could shift the picture. The honest read is that this is a strong reason to run a proper trial, not evidence to change what anyone does today.
The work was funded by the American College of Radiology Center for Research and Innovation, the Pennsylvania Breast Cancer Coalition, and the Abramson Cancer Center.
What to watch next
The natural next step is a study designed from the start to test this question, one that tracks the specific drug, the dose, and the duration, and follows women forward in time rather than looking back through records. That is the only way to move from a correlation this size to a claim a doctor can act on.
Until then, the practical guidance is unchanged. No one should start a GLP-1 drug to lower cancer risk, because that use is not approved and this study does not support it. What the result does is add a serious, well-matched data point to a question oncologists were already starting to ask.
Sources:
- Penn Medicine: https://www.pennmedicine.org/news/glp-1-use-linked-to-lower-breast-cancer-incidence
- EurekAlert: https://www.eurekalert.org/news-releases/1130136
Disclaimer: This article is for general information only and does not constitute medical advice, diagnosis, or treatment. Always consult a qualified healthcare provider about any medical condition or before making health decisions.