Semaglutide May Boost or Block Chemotherapy Depending on the Drug Pairing

Writing in Frontiers in Pharmacology, researchers report that GLP-1 receptor signaling appears to reshape the levels of three key chemotherapy targets in ways that could either strengthen or undermine cancer treatment. The findings suggest that semaglutide markedly enhances the antitumor effect of gemcitabine while weakening methotrexate and paclitaxel, a divergence with potentially major implications for how GLP-1 receptor agonists are combined with cancer therapy. The research team, led by investigators at Sir Run Run Shaw Hospital affiliated with Zhejiang University School of Medicine in Hangzhou, China, began with a bidirectional Mendelian randomization analysis.

This method uses naturally occurring genetic variants as instrumental variables to test whether a causal relationship exists between an exposure, in this case genetically predicted GLP-1 receptor activity, and an outcome, here the protein levels of chemotherapy drug targets. Because genetic variants are randomly assorted at conception, the approach mimics the logic of a randomized trial and reduces the confounding and reverse-causation biases that plague ordinary observational studies. Using single-nucleotide polymorphisms selected from the IEU OpenGWAS Project at a significance threshold of P < 1 × 10⁻⁵, the team identified 31 variants associated with GLP-1 receptor levels, all with F-statistics above 10 to rule out weak instrument bias, and pruned for linkage disequilibrium.

The inverse variance weighted analysis, the primary method, revealed that higher genetically predicted GLP-1 receptor activity was associated with increased protein levels of dihydrofolate reductase, or DHFR (β = 0.323, P = 6.02 × 10⁻¹⁷) and the apoptosis regulator Bcl-2 (β = 0.259, P = 2.1 × 10⁻¹¹), but with decreased levels of the ribonucleoside-diphosphate reductase large subunit, RRM1 (β = −0.12, P = 2.32 × 10⁻³). Sensitivity analyses, including MR-Egger intercept tests, MR-PRESSO, Cochran's Q tests, and leave-one-out analyses, found no evidence of horizontal pleiotropy or influential heterogeneity. Each of these three targets matters clinically because it anchors the action of a major chemotherapy drug.

Methotrexate, an important chemotherapeutic drug for leukemia, kills cancer cells by inhibiting DHFR, so higher DHFR levels blunt its effectiveness. Paclitaxel works partly through Bcl-2 to trigger apoptosis in tumor cells, making elevated Bcl-2 a liability. Gemcitabine, a deoxycytidine analogue used against tumors including non-small cell lung cancer, covalently traps RRM1, the catalytic engine of ribonucleotide reductase that supplies the deoxynucleotide building blocks for DNA replication, so lower RRM1 leaves tumor cells with less enzymatic reserve to draw upon.

Notably, no significant association was found between GLP-1 receptor activity and DNA topoisomerase 1, the target of doxorubicin, after the strict statistical criteria were applied. The reverse Mendelian randomization direction confirmed a bidirectional relationship. DHFR protein levels were positively associated with genetically predicted GLP-1 receptor activity (β = 0.340, P = 1.62 × 10⁻⁹), Bcl-2 was likewise positively associated (β = 0.241, P = 1.49 × 10⁻⁵), and RRM1 was negatively associated (β = −0.150, P = 8.86 × 10⁻⁵).

The authors interpret the DHFR result as evidence of a positive feedback loop, in which rising GLP-1 receptor levels push DHFR upward and elevated DHFR in turn drives GLP-1 receptor levels higher, though the precise molecular wiring of such a loop remains to be worked out experimentally. Genetics alone cannot reproduce the pharmacology of a real drug, so the team turned to semaglutide, a classic and representative GLP-1 receptor agonist, and tested it in three authenticated human cancer cell lines: A549 lung cancer, MDA-MB-231 breast cancer, and COLO205 colorectal adenocarcinoma. In cell viability assays, adding semaglutide to gemcitabine slashed the IC50, the concentration needed to halve cell viability, dramatically in every line.

In A549 cells the gemcitabine IC50 fell from 94.8 to 23.3 nanomolar, in MDA-MB-231 cells from 2,302 to 325.1 nanomolar, and in COLO205 cells from 588.3 to 180.5 nanomolar. Flow cytometry confirmed that the combination drove substantially more tumor cells into apoptosis than gemcitabine alone. The same cell experiments told the opposite story for the other two drugs.

Semaglutide co-treatment raised the IC50 of methotrexate from 52.4 to 203 nanomolar in A549 cells, from 6,637 to 8,375 nanomolar in MDA-MB-231 cells, and from 368.9 to 485.3 nanomolar in COLO205 cells, and similarly shifted paclitaxel IC50 values upward by several-fold in all three lines. Apoptosis induced by methotrexate or paclitaxel fell significantly when semaglutide was present. Western blotting with densitometric quantification across three biological replicates showed the mechanism in protein terms: semaglutide downregulated RRM1, upregulated Bcl-2, and, when combined with methotrexate, significantly elevated DHFR in all three cell lines, exactly as the genetic analysis predicted.

To test whether these effects hold up in a living system, the researchers ran a subcutaneous tumor model in 80 male BALB/c nude mice injected with Lewis lung carcinoma cells, divided into two parallel cohorts, one for tumor burden over 28 days and one for survival follow-up, with five animals per group per cohort, random allocation, and blinded operators. The in vivo results mirrored the cell work with striking fidelity. Adding semaglutide to gemcitabine nearly halted tumor growth compared with gemcitabine alone (P < 0.05), produced the smallest excised tumors of any group, and extended median survival from 70 to 88 days (P < 0.05).

In contrast, the methotrexate-plus-semaglutide group showed faster tumor expansion than methotrexate alone (P < 0.001), heavier terminal tumors, and shortened median survival of 45 versus 55 days (P < 0.05), while the paclitaxel combination similarly lost tumor control and cut median survival from 65 to 54 days (P < 0.05). Semaglutide alone showed no significant antitumor effect at the dose used. Importantly, the modulatory effects of semaglutide came without apparent added systemic harm.

Across the 28-day treatment course, no treatment group lost more than the monitored thresholds of body weight, suggesting the drug did not exacerbate the systemic toxicity classically associated with cytotoxic chemotherapy. The animal study was approved by the Institutional Animal Care and Use Committee of the Zhejiang Center of Laboratory Animals, and all enrolled animals were included in the final analysis. Mechanistically, the authors propose that semaglutide suppresses RRM1 transcription at its source by repressing NF-κB signaling, since NF-κB positively regulates ribonucleotide reductase expression, and possibly through cAMP/PKA cascade effects triggered by GLP-1 receptor activation, thereby starving tumor cells of the deoxynucleotide pool that gemcitabine's trapped RRM1 already limits.

The study's authors note that these results, if replicated, would argue against concurrent use of GLP-1 receptor agonists with methotrexate or paclitaxel in cancer treatment, while supporting their potential as adjunctive agents alongside gemcitabine. Yet the evidence base has clear boundaries. The GWAS data derived primarily from European-ancestry populations, limiting generalizability to other regions, and residual confounding from unmeasured genetic or environmental factors cannot be excluded.

The findings rest on cell lines and mouse xenografts rather than human patients, and the authors themselves caution that although the phenotypic outcomes coincide with changes in target protein levels, the direct molecular mechanisms remain to be fully elucidated. They call for deeper mechanistic studies and, ultimately, randomized controlled trials to determine whether these preclinical observations translate into clinical practice. For now, the study stands as a first genetic-level warning that GLP-1 receptor agonists are not pharmacologically invisible to oncology, and that drug-drug interactions with chemotherapy deserve careful attention as their use continues to expand.

Subject of Research: Interaction between GLP-1 receptor agonists such as semaglutide and chemotherapy drug targets

Article Title: The impact of GLP-1 receptor agonists on chemotherapeutic efficacy: insights from GWAS and experiments

Article References: Huang, S., Feng, Y., Liu, Y., Mo, X., Hu, H., Chen, H., Han, M., Huang, W., Zheng, W., Hou, X., Wen, R., Liu, C., Wang, J., & Wang, Y. (2026). The impact of GLP-1 receptor agonists on chemotherapeutic efficacy: insights from GWAS and experiments. Frontiers in Pharmacology, 17, Article 1957757. https://doi.org/10.3389/fphar.2026.1957757

Image Credits: AI Generated

DOI: 10.3389/fphar.2026.1957757

Keywords: semaglutide, GLP-1 receptor agonists, chemotherapy, gemcitabine, methotrexate, paclitaxel, Mendelian randomization, RRM1, DHFR, Bcl-2, cancer drug interactions, preclinical study