The assumption baked into most obesity-adjacent thinking right now is that GLP-1s are the endpoint. That if the drugs get good enough, everything else becomes obsolete. Cleveland Clinic published a study this week that quietly complicates that: a large cohort of people who either could not tolerate GLP-1s or got almost nothing from them, and went on to lose substantial weight from surgery anyway.
The study, published in Annals of Surgery on 3 November, followed 4,929 adults with obesity who had bariatric surgery at Cleveland Clinic hospitals in Ohio between 2018 and 2024. The comparison is not GLP-1 versus surgery. It is surgery outcomes in people with a documented history of GLP-1 non-response or intolerance versus surgery outcomes in GLP-1-naive patients.
What the Research Says
- The cohort split into 4,661 patients who had never taken semaglutide or tirzepatide, 61 who stopped due to intolerable side effects, and 207 who had inadequate weight loss while on the medications. Median body weight at surgery ranged from roughly 273 to 287 pounds (about 124 to 130 kg) across the three groups.
- Two years after surgery, patients who stopped GLP-1s for side effects lost about 23% to 26% of total body weight (lower figure gastric sleeve, higher figure gastric bypass).
- Patients who had inadequate weight loss on the medications lost about 18% to 33%.
- Patients who had never been on a GLP-1 lost about 24% to 33%.
- In several groups, weight loss was comparable to GLP-1-naive patients, meaning prior medication failure did not predict worse surgical outcomes.
- Lead author Ali Aminian, director of Cleveland Clinic’s Bariatric and Metabolic Institute, framed the finding directly: GLP-1s do not work for everyone, and surgery can be the next step.
Why the numbers matter more than they look
A 2% weight loss on a GLP-1 versus an 18% to 33% loss from surgery in the same patient two years later is a hard contrast to argue with. The group that got almost nothing from the medications did not do worse with surgery. If anything, the non-responder group’s sleeve outcomes overlap the naive group’s, which suggests that GLP-1 response and surgical response are driven by different biology rather than one being a degraded version of the other.
That maps onto what the pharmacology already hints at. GLP-1 response varies significantly between individuals for reasons that are still not fully mapped, whether receptor density, genetic variation in the appetite-regulation pathways, or something else entirely. Surgery works through mechanical and hormonal mechanisms that do not depend on the same receptors. A partial or non-responder to one is not predicted to fail the other, and this dataset is the largest real-world confirmation of that to date.
The honest caveats
This is observational, not randomised. Nobody was assigned surgery versus GLP-1 in a controlled way, and the patients who ended up at surgery after a failed medication round were a self-selected, referred, insured subset. The 61-patient intolerance group and 207-patient inadequate-response group are small slices of a much larger narrative, and the study cannot say what percentage of GLP-1 non-responders would benefit from or even want surgery. Sixty-one patients is also a tiny denominator for a statement about tolerance-driven discontinuation, and the wide confidence intervals are doing real work in that subgroup.
There is also an access reality. In Australia, bariatric surgery sits largely outside the reach of most people without significant private cover or tens of thousands of dollars in out-of-pocket costs, while compounded-adjacent GLP-1 access has been the subject of the TGA compounding crackdowns we have covered. The US data says “surgery is the fallback.” Australian access says “good luck with that.” That gap is a policy question, not a research one, and nobody in Canberra is currently answering it.
Why this is the counter-narrative
The field’s momentum is all incretin. Lilly and Novo between them are running dozens of phase 3 programmes, and the community conversation, including our own obesity drug showdown coverage, skews hard toward which molecule loses the most weight. A study saying a large minority get little benefit from the molecules and do fine with the scalpel cuts against the investment narrative, and it is getting a fraction of the attention the phase 3 topline results get.
For the research-chemistry community, the relevance is indirect but real: non-response is a genuine biological phenomenon, not a compliance failure, and any framework that treats GLP-1 outcomes as deterministic is missing the actual data.
Our semaglutide and retatrutide pages cover the response-distribution data in the major trials for anyone who wants the variance picture.
The Grey Highway Telegram group has an active thread on non-responders and what the community’s own observations suggest about the response distribution.
Sources
Source: Cleveland Clinic, “Cleveland Clinic Study Finds Bariatric Surgery Effective for Substantial Weight Loss After GLP-1 Intolerance or Limited Weight Loss”, 3 November 2026: https://newsroom.clevelandclinic.org/2026/11/03/cleveland-clinic-study-finds-bariatric-surgery-effective-for-substantial-weight-loss-after-glp-1-intolerance-or-limited-weight-loss
Source: Aminian A, et al. “Bariatric surgery as a rescue strategy after semaglutide or tirzepatide failure”, Annals of Surgery, 3 November 2026, DOI: 10.1097/sla.0000000000007251: https://www.ovid.com/jnls/annalsofsurgery/fulltext/10.1097/sla.0000000000007251~bariatric-surgery-as-a-rescue-strategy-after-semaglutide-or
Disclaimer: This article is for educational and informational purposes only. It does not constitute medical advice, therapeutic recommendations, or endorsements of any compound. Grey Highway is a research-education community. We do not sell, supply, or promote the use of research compounds. Always consult a qualified healthcare professional regarding health decisions. For Australian regulatory information, visit the TGA website.