GLP-1s Beyond the Scale: Muscle, Brain Pressure, and the Next Wave of Formulations

Four studies dropped in the last cycle that, taken together, sketch a credible picture of the post-semaglutide landscape. One is a tightly controlled mouse experiment on what GLP-1s actually do to skeletal muscle. One is a real-world cohort study showing GLP-1 receptor agonists may reduce the need for brain surgery in a specific neurological condition. One is an exposure-response analysis of a long-acting Fc-fused analogue. And one is the first-in-human readout of an oral small-molecule GLP-1. None of them is a final answer. All of them move the needle.

The loudest objection from lifters has always been the same: if GLP-1s cause weight loss this aggressive, how much of that scale drop is lean tissue? A 2025 study in The Journal of Physiology tackled the question with the kind of controlled design human trials can't easily replicate — mice with diet-induced obesity given either semaglutide or a calorie-matched diet for four weeks, then tracked through a six-week withdrawal period.

The result is the nuance the discourse has been missing. Semaglutide produced greater weight loss than caloric restriction at matched energy intake, and was apparently more effective at promoting fat loss specifically. But on muscle, the drug behaved like dieting: semaglutide reduced muscle size and strength to the same extent as caloric restriction, with distinct transcriptomic signatures but a similar functional outcome. After discontinuation, lean and fat mass rebounded to baseline, and muscle size and strength normalized across groups.

Translation for the gym floor: the muscle loss on a GLP-1 in this model looks like the muscle loss from any aggressive deficit — not a unique drug-driven catabolic insult. That's a meaningful reframe. It also means the standard lifter playbook for a cut — high protein, hard resistance training, don't crash the deficit — is probably still the right playbook on these drugs. This is a mouse study. Don't extrapolate it past what it is. But it's the cleanest data we have on the question so far.

Idiopathic intracranial hypertension (IIH) — elevated pressure inside the skull, most common in women with obesity — is a condition most lifters have never heard of. It matters here because some patients end up needing real neurosurgery: ventriculoperitoneal shunts, venous sinus stenting, optic nerve sheath fenestration. Weight loss is the cornerstone of treatment, which made GLP-1s an obvious thing to study.

A propensity-matched, multi-institutional cohort study published in the Journal of NeuroInterventional Surgery compared 2,690 IIH patients on GLP-1 RAs to 2,690 matched controls. At both five weeks and six months, GLP-1 RA use was associated with significantly lower odds of undergoing venous sinus stenting and ventriculoperitoneal shunting. That's not a soft endpoint — that's avoided brain surgery.

The caveats matter. This is a retrospective cohort, not a randomized trial; propensity matching is a powerful tool, not a magic wand. And the headline odds ratios as reported are unusual in direction-of-effect framing, so the paper is worth reading carefully rather than memeing. Still: if this signal holds up in a prospective trial, GLP-1s graduate from "weight-loss drug with side benefits" to a tool a neurosurgeon might genuinely reach for first.

Semaglutide is once-weekly and peptide-based. The next iteration of injectables is trying to push pharmacokinetics further and engineer cleaner exposure-response relationships. Efsubaglutide alfa is one example: a long-acting GLP-1 RA built by fusing two human GLP-1 molecules to an IgG2 Fc via a natural immunoglobulin hinge — basically borrowing antibody half-life biology for a peptide drug.

An exposure-response analysis from the YN011-302 trial, in 406 type 2 diabetics on stable metformin, found a robust inverse correlation between efsubaglutide alfa exposure and improvements in HbA1c, fasting plasma glucose, body weight, and BMI, plus a positive correlation with C-peptide AUC suggesting better beta-cell function. The modeled effect: doubling steady-state trough concentrations reduced HbA1c by 0.211%, and every 100 ng/mL increase in average concentration corresponded to 0.5 kg of weight loss at week 24. Baseline HbA1c predicted response.

That's not a blockbuster magnitude on its own. But the cleanliness of the exposure-response curve is the point: it means dosing can be tuned with confidence, which is exactly what you want from a next-gen molecule.

The bigger structural shift may come from oral small molecules. Peptide GLP-1s are hard to absorb through the gut and expensive to manufacture at population scale. A small molecule that hits the GLP-1 receptor changes both equations.

DA-302168S is one such candidate, and its first-in-human Phase I — single ascending dose, multiple ascending dose, and a 28-day weekly-titration study in overweight and obese adults — just read out. It was generally well tolerated, with nausea as the most common adverse event, dose-proportional pharmacokinetics, and mean weight loss of −5.67% to −7.26% versus −2.90% on placebo over four weeks, alongside reductions in HbA1c, glucose fluctuations, and improved lipids.

Four weeks is not a Phase III readout. A few dozen subjects is not a population. But for a first-in-human study, those weight-loss numbers are striking, and the safety profile is what you'd want to see before greenlighting a larger trial.

Zoom out and the through-line is clear: GLP-1 receptor signaling is turning out to be a much broader lever than the obesity story alone suggested, and the molecules hitting that lever are getting more sophisticated fast. The muscle question has a more honest answer than the hype cycle gave it. The neurological signal is real enough to take seriously. The pipeline is no longer just "semaglutide, but weekly-er." Stay skeptical of single studies, stay curious about the trajectory, and — if any of this is personally relevant — have the conversation with an actual clinician, not a group chat.