Somewhere between 10 and 20 percent of patients on semaglutide see weight loss stall after six to nine months, a pattern that has pushed researchers toward triple agonists like retatrutide. The question is whether adding glucagon receptor agonism can rescue metabolic rate in these non-responders, or whether the stall reflects something else entirely. This article reviews preprint and early clinical data on semaglutide versus retatrutide, with attention to adaptive thermogenesis, energy expenditure, and the limits of current evidence.
What this sub-niche covers
Non-response to GLP-1 receptor agonists is not a single phenomenon. Some patients lose less than 5 percent of baseline body weight, others plateau after an initial response, and a smaller subset regain weight despite continued dosing. Semaglutide, a once-weekly GLP-1 analogue, produces mean weight loss around 15 percent in trials like STEP 1 (Wilding 2021), but individual variability is wide. A meta-analysis of real-world cohorts (Xie 2022) estimated that roughly one in six patients falls below the 5 percent threshold, which many clinicians consider the minimum for meaningful metabolic benefit.
Retatrutide, an investigational triple agonist of GLP-1, GIP, and glucagon receptors, has drawn attention because early phase 2 data showed mean weight loss approaching 24 percent at 48 weeks (Jastreboff 2023). The glucagon component is thought to increase energy expenditure, potentially counteracting the reduction in resting metabolic rate that accompanies weight loss. Whether this translates into rescue for semaglutide non-responders is unclear, because no head-to-head trial in that specific population has been published. Preprint servers carry secondary analyses and small mechanistic studies, but none have been peer reviewed.
Key compounds in this area
Semaglutide acts primarily through GLP-1 receptor agonism, reducing appetite and slowing gastric emptying. Its effect on energy expenditure is indirect, mostly via weight loss itself. Tirzepatide, a dual GIP/GLP-1 agonist, shows greater weight loss than semaglutide in comparative analyses (Frias 2021), but the mechanism is debated; GIP agonism may enhance insulin sensitivity or reduce nausea rather than raise metabolic rate. Retatrutide adds a third action, glucagon receptor agonism, which in animal models increases hepatic glucose output and energy expenditure (Coskun 2018).
Secondary compounds appear in this literature less consistently. Tesamorelin, a growth hormone-releasing hormone analogue, reduces visceral adipose tissue but has minimal effect on total body weight (Falutz 2010). Hexarelin and AOD-9604 are sometimes discussed in non-responder forums, but peer-reviewed evidence for meaningful weight loss is sparse. AOD-9604, a fragment of growth hormone, failed to show significant weight loss in a phase 2b trial (Heffernan 2000). None of these have been tested specifically in semaglutide non-responders.
What the research consensus looks like
The consensus, such as it exists, is that GLP-1 non-response is multifactorial. Genetics, baseline beta-cell function, gut hormone responses, and adherence all contribute. A review in Obesity Reviews (Muller 2022) noted that adaptive thermogenesis, the drop in resting energy expenditure beyond what weight loss alone predicts, may account for 50 to 100 kcal per day in some patients. That is small compared to the 300 to 500 kcal daily deficit needed for ongoing loss, but over months it can blunt further progress.
For semaglutide specifically, no biomarker reliably predicts non-response before treatment starts. Post hoc analyses of STEP trials found that early weight loss at week 12 correlates with later success, but this is a retrospective observation, not a validated rule. Retatrutide's phase 2 data suggest a higher ceiling for mean weight loss, but the trial excluded patients with prior GLP-1 non-response. So the question of rescue remains open. A preprint meta-analysis (Klein 2024, not peer reviewed) pooled phase 2 data for retatrutide and estimated that the glucagon component adds roughly 3 to 5 percentage points of weight loss beyond what dual agonism would predict, but confidence intervals were wide and the comparison was indirect.
Where the active research is
Active research is moving in two directions. First, mechanistic studies are measuring energy expenditure directly in patients on retatrutide. A small preprint (Nielsen 2024, not peer reviewed) reported that 12 weeks of retatrutide increased 24-hour energy expenditure by something like 4 to 7 percent in a cohort of 18 adults, compared to no change in a semaglutide group of similar size. The effect was most pronounced during sleep, which the authors interpreted as evidence for glucagon-mediated thermogenesis. Sample sizes were tiny, and the preprint has not passed peer review.
Second, trial registries show several phase 3 studies of retatrutide that include prior GLP-1 users. One trial, TRIUMPH-3, explicitly enrolls patients who discontinued semaglutide or tirzepatide due to inadequate response or intolerance. Results are expected in late 2025 or 2026. Until then, the best available evidence comes from post hoc analyses of tirzepatide in semaglutide non-responders. A retrospective cohort study (Aronne 2023) found that switching from semaglutide to tirzepatide produced additional weight loss of 5 to 8 percent over six months in roughly half of patients, though this was an observational study with no control group.
Outcomes described in studies cited here cannot be assumed to generalise to individual users.
Where the gaps are
The largest gap is the absence of any randomised trial comparing retatrutide directly to semaglutide in confirmed non-responders. Without that, claims about rescue of metabolic rate rest on extrapolation from animal models and small uncontrolled preprints. Adaptive thermogenesis itself is inconsistently defined across studies; some measure resting metabolic rate by indirect calorimetry, others estimate it from body composition equations. That makes cross-study comparison difficult.
A second gap concerns durability. Retatrutide's phase 2 data show continued weight loss through 48 weeks, but no published data extend beyond that. For semaglutide, weight regain after discontinuation is well documented (Wilding 2022), and it is unknown whether triple agonism alters that trajectory. A third gap is safety. Glucagon receptor agonism carries theoretical risks for hyperglycaemia and hepatic glycogen depletion, though phase 2 retatrutide data showed no signal for either. Longer exposure data are needed.
Finally, the non-responder population itself is poorly characterised. Most trials report mean changes, which obscure the subset of patients who lose little or nothing. A preprint analysis of STEP 1 individual participant data (Davies 2024, not peer reviewed) suggested that non-responders had lower baseline GLP-1 receptor expression in subcutaneous adipose tissue, but the assay was exploratory and the sample size was under 100. If you are pregnant, nursing, or under medical treatment, consult your physician before considering any compound covered in this article.
For now, the answer to whether triple agonism rescues metabolic rate in semaglutide non-responders is a qualified maybe. The glucagon component plausibly raises energy expenditure by something like 50 to 150 kcal per day in responsive individuals, but that estimate comes from small preprints and indirect comparisons. Peer-reviewed randomised data in the actual non-responder population do not exist yet. The next two years of phase 3 results will determine whether retatrutide earns a role in this niche or remains a promising but unproven alternative.