Beyond Weight Loss: What the Latest GLP-1 Evidence Says About Liver, Kidney, Heart, and Cost

Non-alcoholic fatty liver disease (NAFLD) is the quiet epidemic running alongside the obesity one. It can sit silently for years, then progress to steatohepatitis (NASH), fibrosis, and eventually cirrhosis. In many countries there is still no approved medical therapy for it.

A 2025 systematic review and meta-analysis in the Journal of the Canadian Association of Gastroenterology pooled six randomized trials covering 478 patients and found that, compared with standard care, GLP-1 receptor agonists were associated with markedly higher odds of NASH resolution on biopsy (odds ratio 4.45, 95% CI 1.92–10.3) and a roughly five-percentage-point reduction in liver steatosis on imaging. Liver stiffness — a proxy for fibrosis — barely moved.

The authors are appropriately measured: GLP-1 therapy probably improves NASH over at least six months. The trials are few, the patient numbers modest, and stiffness, the outcome most tied to long-term liver damage, did not meaningfully change. The takeaway is encouraging without being conclusive.

Most of the cardiovascular headlines around semaglutide have come from the injectable form. But adherence with weekly injections is imperfect, and many patients would prefer a tablet. A 2025 systematic review in the Journal of Clinical Medicine looked specifically at oral semaglutide's effects on cardiovascular risk factors — blood pressure and lipid profile — in people with type 2 diabetes.

The signal, taken across the included studies, points in a familiar direction: oral semaglutide moves cardiovascular risk markers favorably when added to standard care. That matters because, until recently, it was an open question whether the oral formulation would carry the same metabolic halo as the injection. The review does not deliver a definitive head-to-head verdict, and it is built on heterogeneous studies, but it strengthens the case that the route of delivery is not the whole story.

People living with both type 2 diabetes and chronic kidney disease occupy one of medicine's hardest corners. Their cardiovascular risk is high, their medication lists are long, and historically their options for slowing renal decline have been narrow.

A 2025 meta-analysis in Annals of Medicine and Surgery pooled three randomized trials covering 10,013 such patients. Compared with placebo or standard care, semaglutide was associated with a 29% relative reduction in cardiovascular mortality (RR 0.71, 95% CI 0.52–0.97), a 20% reduction in major adverse cardiovascular events, and a 20% drop in kidney-related adverse events. The authors also noted a reduced need for additional cardiovascular medications.

These are relative reductions in a high-risk group, which tends to magnify the absolute benefit a single patient might see — but it also means the results may not generalize neatly to lower-risk readers. And with only three trials in the pool, even an encouraging signal is one or two negative studies away from looking different. Still, for the specific population of type 2 diabetes with CKD, this is among the more coherent stories the GLP-1 class has produced.

The most uncomfortable conversation in this space is not clinical but financial. Anti-obesity medications are expensive, often not reimbursed, and used by a population large enough to shift national drug budgets. The countervailing argument has always been that obesity itself is costly — in medical care, in lost productivity, in years of life lived in poorer health.

A 2025 modeling study in Scientific Reports took a societal-perspective look at this trade-off in Austria. The authors modeled treating 50% of adults with class II and class III obesity (excluding people with diabetes) for 68 weeks with semaglutide alongside lifestyle intervention. In the model, class II obesity prevalence fell from 4% to 2.74% and class III from 1.45% to 0.97%, producing a 12.9% reduction (about €108.7 million per year) in expenses tied to those classes; over the life cycle, dropping an obesity class reduced costs by roughly 40% per patient.

Modeling studies are not real-world evidence. They depend on assumed adherence, assumed durability of weight loss after treatment, and assumed health-system structures. The Austrian figures will not translate cleanly to other countries. But the directional point — that the population-level math can favor treatment in higher-risk groups — is worth taking seriously.

The 2025 evidence does not vindicate the more breathless claims made about GLP-1 drugs, but it does push the conversation beyond the bathroom scale. The liver data are promising and incomplete. The oral cardiovascular data are reassuring and indirect. The kidney data are the strongest of the four, but limited to a specific high-risk group. And the economic data are a model, not a measurement.

For readers already on a GLP-1, or considering one, the most useful posture is neither evangelism nor suspicion. It is curiosity, paired with a clinician who knows your numbers. The questions worth asking are specific: what does my liver look like; what is my kidney function; what is my cardiovascular risk; what would a year, three years, or longer on this class plausibly do for each of them — and what would stopping look like?

The class is doing more than we thought it would. It is also being asked to do more than the evidence yet supports. Holding both ideas at once is, for now, the honest read.