Bethania Santos, MD, UT Southwestern Medical Center, Houston, Texas, discusses a retrospective study investigating the impact of glucagon-like peptide-1 receptor (GLP-1R) agonist use on response to chemoimmunotherapy for patients with early-stage triple-negative breast cancer. The pathologic complete response (pCR) of patients who had been exposed to GLP-1 drugs was 30.8%, compared to 64.4% of those who were not exposed to such agents (P = .001). 

Dr Santos concluded, “I think it's too early to say that we should change our practice, but I think the clinicians should be aware of this potential risk. And that's the importance of a multidisciplinary team, so we can discuss with endocrinologists, maybe choose another option to treat diabetes, another option for weight loss.”

These data were first presented at the 2025 American Society of Clinical Oncology (ASCO) Annual Meeting in Chicago, Illinois.

Transcript:

My name is Bethania Santos. I'm a medical oncologist from Brazil and I've been working at UT Southwestern, with the breast cancer research group since 2023. Our group is focused on triple-negative breast cancer, and especially in immunotherapies and mechanisms of resistance to immunotherapy. We would like to understand how metabolic conditions —BMI, diabetes, and other comorbidities— could impact the response to the treatment. 

In collaboration with 2 other institutions, we checked patients with early-stage triple-negative breast cancer, because there's some publications… It’s controversial, but some publications show that patients with higher BMI responded better to chemoimmunotherapy, maybe because of an impact on the tumor microenvironment. There's some different effect in the tumor microenvironment and the fat tissue. And another question that came was, in these patients that are losing weight during the treatment, is there something happening with these patients? 

This was a retrospective study, with 3 institutions, of patients with early-stage triple-negative breast cancer treated between 2021 to 2023. These patients received at least 4 cycles of treatment. Patients should be using GLP1-agonists since the beginning of the treatment to the end of the treatment. The patients that started or stopped, interrupted the GLP1-agonist, during the chemoimmunotherapy, we didn't include these patients because we would like to see what happened with these patients that were using the drug during all the treatment. 

We divided the patients in 3 cohorts: one cohort is the non-diabetic patients, the second cohort is diabetic patients using other classes of diabetes drugs, and the third cohort was patients using GLP1-agonist and not just GLP1-agonists, we call GLP1-targeted agents because we included DPP4 inhibitors because DPP4 inhibitors increase the levels of circulating GLP1. Since we would like to know how did GLP1 affect the response to treatment, we thought that was interesting to include this other class of drug, DPP4 inhibitors. 

And the difference of PCR rate between these 3 groups was, in non-diabetic patients the PCR rate was 65%, 64% or 65%. Diabetes patients using other class of medications, the PCR rate was 63%. And patients using GLP1 drugs, the PCR rate was 30%, 30.8%. It’s a huge difference. And then we want to understand why did it happen? How are these drugs affecting the response to the treatment? 

The first experiment was in immunohistochemistry to check the presence of GLP1 receptors in 100 triple-negative samples. We found the presence of GLP1 receptors in tumors in 30% of the samples. And in tumor microenvironment, like stoma, immune cells, around 80% of the samples showed some GLP1 expression. It helped us to determine, okay, definitely there's something happening. We can see the expression of GLP1 in the tumor and in the tumor microenvironment. 

The second step was spatial transcriptomic, in collaboration with another institution. And we sent samples of triple-negative patients that were exposed to GLP1, which occurred before starting chemoimmunotherapy and after chemoimmunotherapy. The proportion of GLP1 positive cells before the treatment was a bit lower than after the treatment. And we compared cases that responded better to treatment with cases that didn't respond to the treatment. And the cases that didn't respond to the treatment, the GLP1 expression was higher. What we could see with the transcriptomics, this is because the GLP1 can affect the tumor, maybe activating sign pathways that leads to tumor growth, tumor proliferation, and immune cells. It seems that the GLP1 works as a break to immune response. It’s not affecting just the effect of immunotherapy, but the effect of chemotherapy. 

Because the study that approved this protocol, this regimen, KEYNOTE-522, showed a response rate in patients using chemo and immunotherapy of 64%. That is the same that we found in patients using other classes of diabetes medication and patients [who are] non-diabetic. And in KEYNOTE-522 patients using just chemotherapy without immunotherapy, the pCR rate was 51%. If the GLP1 was affecting just the immunotherapy, at least the pCR rate should be 51%. And it was 30%. So the GLP1 is acting both in tumor and immune cells. 

I think it's too early to say that we should change our practice, but I think the clinicians should be aware of this potential risk. And that's the importance of a multidisciplinary team, so we can discuss with endocrinologists, maybe choose another option to treat diabetes, another option for weight loss. Of course, we already know that exercise and dietary impact positively the response to treatment. The clinicians should be aware of this risk. Not just as a clinician, but I think as a patient, because I am a cancer patient, I would like to know if this drug can be a risk for my treatment. And if I can avoid, why not? 

Our focus was triple negative breast cancer. I think it's important to check and understand how the GLP1 agonist can affect HER2-positive tumors, hormone receptor-positive tumors. We should explore better these other types of tumors. 

And another limitation of our study, since it was a retrospective study and the number is low and we didn't stratify the patients between patients using GLP1 to treat diabetes or treat obesity, so we don't know if it can affect in both scenarios. That’s something important to keep investigating.  I think it's important to say that we know about the risk of developing cancer in obese patients, and we know how these drugs are helping to treat obesity and preventing new cancer. But we are talking about a different thing, we are talking about patients that already have cancer and are treating [that cancer], so it's different. 

KEYNOTE-522 shows that patients that had a pCR, the overall survival and event-free survival is better. So, if this medication is impacting the response to the treatment, consequently it can impact the overall survival. We still don't have this data because the follow-up time was short, but it's another thing to explore that is in the plan.

Source:

Santos B, Durdana I, McArthur H, et al. Concurrent GLP1R-agonist use with chemoimmunotherapy for early-stage triple-negative breast cancer. Presented at the 2025 ASCO Annual Meeting. May 30-June 3, 2025; Chicago, IL. Abstract #1115