GLP-1s Grow Up: What the Latest Research Says About Weight Regain, Brain Health, and Your Eyes

Bariatric surgery is one of the most effective interventions in modern metabolic medicine — and also one of the most quietly frustrating. Weight regain is common, sometimes years later, and until recently the playbook for what to do about it has been thin. A Swiss retrospective study of 40 post-bariatric patients, published in BMC Endocrine Disorders, asked a practical question: can a GLP-1 actually rescue the regain?

The answer, at 12 months, was a striking yes. Patients started a GLP-1 a median of roughly six years after surgery, after regaining about 15% of their body weight. After a year on either liraglutide 3.0 mg daily or semaglutide 1.0 mg weekly, they had lost a median of 10.5 kg — essentially all of the weight they had regained, with semaglutide outperforming liraglutide on BMI reduction. It's a single center and a small cohort, so don't read it as the last word. Read it as the first solid real-world signal that the regain phase has a serious pharmacologic option.

If you've felt like the headline numbers from the GLP-1 trials don't quite match the stories you hear from friends, you're not imagining it. A new narrative review in Diabetes, Obesity & Metabolism pulled together the contemporary real-world evidence on liraglutide, semaglutide and tirzepatide and found a consistent pattern: weight loss in clinical practice tends to be lower than in the randomized trials, with discontinuation rates of 20%–50% in the first year and many patients on doses well below those studied.

The same review offers a measured read on safety: gastrointestinal side effects are common, as expected, but observational data so far show no clear signal for pancreatitis, pancreatic cancer, thyroid disorders, or depression and self-harm in obesity or type 2 diabetes populations. The authors flag eye disease and other rare outcomes as still needing more evidence — which sets up the next finding.

The neuroprotection story is the most exciting thread in the new data — and the one most easily overstated. A network meta-analysis in BMC Medicine pooled randomized trials of GLP-1 receptor agonists and SGLT2 inhibitors and looked at the incidence of seven neurodegenerative diseases, from Parkinson's and Alzheimer's to ALS. The authors framed it as a pharmacodynamics-based evaluation of prophylactic benefit across major neurodegenerative diseases, suggesting these drugs may carry preventive potential beyond their metabolic effects.

Here's the careful read. Almost none of the included trials were designed to test prevention of dementia or Parkinson's — those outcomes were captured as adverse events or secondary signals. A network meta-analysis can surface a trend that warrants dedicated studies; it can't substitute for them. So: a real and biologically plausible signal, in line with years of preclinical neuroprotection work, and absolutely worth watching. Just not, yet, a reason to start a GLP-1 to protect your brain.

The most discussed safety story of the spring came from a pharmacovigilance analysis, also in BMC Medicine, that mined the FDA Adverse Event Reporting System (FAERS) for vision-impairment reports linked to semaglutide. Compared with other GLP-1 receptor agonists, semaglutide had a reporting odds ratio of 1.95 for vision impairment (95% CI 1.75–2.17), with higher ratios still versus DPP-4 inhibitors, SGLT2 inhibitors, metformin, and orlistat. Topiramate was the only comparator that out-reported it.

What FAERS can and can't do matters here. It's a passive reporting system, vulnerable to media-driven reporting waves and unable on its own to prove that a drug caused an event. What it does well is raise hypotheses worth testing. The authors are explicit that their findings warrant further investigation and vigilant post-marketing surveillance, and the real-world review echoed that eye outcomes need more evidence. If you're on semaglutide and notice changes in your vision, that's a call to your prescriber, not a reason to panic.

Read together, these four papers tell a coherent story about where GLP-1s are heading. They are maturing from a single-outcome weight drug into a broader metabolic-neuro toolkit, with a credible new use case in post-bariatric regain, a tantalizing brain-protection hypothesis, and a vision-safety question that deserves serious follow-up rather than dismissal or alarm. The drugs are not miracles, and they are not menaces. They are, increasingly, a long-term relationship — one in which the right questions, the right monitoring, and the right prescriber matter more than any single headline.