What We'd Want to See in Metabolic Peptide Research
Ideal metabolic peptide trials would recruit premenopausal women across the follicular and luteal phases, track insulin sensitivity at multiple time points, and report dropout rates stratified by hormonal contraceptive use. They would measure visceral adipose tissue by DEXA rather than relying on BMI alone, and they would follow participants long enough to capture weight regain after cessation. We would want head-to-head comparisons of single-receptor agonists against dual-agonist compounds in cohorts matched for baseline HbA1c, fasting insulin, and waist circumference. We would want pregnancy-exclusion criteria made explicit in every protocol summary, along with contraceptive counseling documented at enrollment. We would want adverse-event logs that separate nausea by cycle day and distinguish between transient gastrointestinal discomfort and sustained emesis requiring antiemetic rescue.
We would also want dose-escalation schedules tested in women with polycystic ovary syndrome, given that insulin resistance in PCOS often differs in magnitude and distribution from the resistance seen in metabolic syndrome without hyperandrogenism. A 2021 review noted that GLP-1 receptor agonists reduced androgen levels and improved ovulatory frequency in women with PCOS, but the mechanisms, whether via direct ovarian signaling or secondary to weight loss, remain incompletely mapped (Elkind-Hirsch 2021). Trials that measure free testosterone, sex hormone-binding globulin, and luteinizing hormone alongside glucose and lipid panels would clarify whether dual-agonist peptides offer additive benefit on reproductive endpoints or simply amplify the metabolic improvements already seen with GLP-1 monotherapy.
Finally, we would want transparency about compounding-pharmacy sourcing when telehealth platforms dispense peptides outside the branded supply chain. Researchers conducting independent work should follow institutional protocols and ethics review where applicable. Purity certificates, endotoxin testing, and sterility assays are standard in academic settings but less uniformly applied in direct-to-consumer models, and the gap between research-grade material and patient-facing product can be wide.
What We Have in the Current Literature
Tirzepatide is a dual agonist at the glucose-dependent insulinotropic polypeptide receptor and the GLP-1 receptor, a design intended to leverage the insulinotropic effects of GIP while preserving the satiety and gastric-emptying delay conferred by GLP-1 signaling. The SURMOUNT-1 trial enrolled 2539 adults without diabetes, dosing tirzepatide at 5, 10, or 15 mg subcutaneously once weekly for 72 weeks (Jastreboff 2022). Mean weight reduction in the 15-mg arm reached 20.9 percent from baseline, compared to 3.1 percent with placebo. Women represented 67.5 percent of the cohort, though subgroup analysis by sex was not the primary endpoint. Dropout due to gastrointestinal adverse events occurred in 6.2 percent of participants on the highest dose, a figure that includes both men and women without separate reporting.
Semaglutide, a selective GLP-1 receptor agonist, was evaluated in the STEP 1 trial at a once-weekly subcutaneous dose of 2.4 mg over 68 weeks (Wilding 2021). Participants lost an average of 14.9 percent of baseline weight, with placebo-adjusted reduction of 12.4 percentage points. The trial excluded individuals with a history of pancreatitis, medullary thyroid carcinoma, or multiple endocrine neoplasia type 2, and required negative pregnancy tests at screening and monthly thereafter in women of childbearing potential. Nausea was reported in 44 percent of the semaglutide group versus 17 percent in placebo, but severity and duration were not broken down by menstrual-cycle phase.
Retatrutide is a triple agonist at GIP, GLP-1, and glucagon receptors, adding glucagon-mediated increases in energy expenditure to the dual-agonist profile. A phase-2 trial randomized 338 adults with obesity to retatrutide doses ranging from 1 to 12 mg once weekly for 48 weeks (Jastreboff 2023). The 12-mg cohort lost an average of 24.2 percent of body weight, with placebo-adjusted reduction of 22.8 percentage points. Women composed 53 percent of the sample. Adverse events were predominantly gastrointestinal, with nausea in 60 percent and vomiting in 29 percent at the top dose. The trial protocol specified that women must use highly effective contraception or be postmenopausal, but no data were presented on contraceptive type or adherence.
AOD-9604 is a C-terminal fragment of human growth hormone, investigated for lipolytic effects without the full somatotropic signaling of the intact molecule. A 2008 trial in 300 obese adults found no significant weight loss versus placebo after 12 weeks at 1 mg subcutaneously daily (Heffernan 2008). The peptide has since migrated into wellness and telehealth channels despite the null result, often combined with other agents in proprietary blends. MOTS-c, a mitochondrial-derived peptide, has been studied in rodent models for effects on insulin sensitivity and exercise capacity, but human trials remain limited to small pharmacokinetic studies with no published efficacy data in metabolic endpoints. Hexarelin, a growth-hormone secretagogue, showed cardioprotective signals in animal work but carries the same concerns about growth-hormone axis activation that complicate long-term use, particularly in women with intact menstrual cycles where growth-hormone pulsatility is already elevated during the luteal phase.
What's Missing from the Evidence Base
None of the large tirzepatide or retatrutide trials stratified outcomes by menstrual-cycle phase at baseline or tracked changes in cycle regularity as a secondary endpoint. Women using hormonal contraception were pooled with naturally cycling women, and no analysis examined whether estrogen-containing pills, progestin-only methods, or intrauterine devices modified peptide pharmacokinetics or adverse-event profiles. Insulin sensitivity varies across the menstrual cycle, with a nadir in the late luteal phase when progesterone peaks, yet dosing schedules in these trials remained fixed without accommodation for cyclical variation. A 2019 study in 24 women found that GLP-1 infusion during the luteal phase produced a smaller reduction in food intake than the same infusion during the follicular phase, suggesting that hormonal context may modulate receptor responsiveness (Clegg 2019).
Pregnancy data are absent by design, as exclusion criteria universally prohibit enrollment of pregnant or breastfeeding individuals. Animal studies of GLP-1 agonists have documented reduced fetal growth and skeletal malformations at high doses, and GIP receptor knockout mice show altered placental development. The FDA has classified semaglutide as pregnancy category X equivalent under the new labeling system, meaning risk outweighs any potential benefit. Tirzepatide carries similar warnings. Women who become pregnant while on these peptides are advised to discontinue immediately, but the washout period required before conception is not standardized, and no prospective registry tracks outcomes in accidental exposures.
Long-term data beyond two years are scarce for any of these compounds. The SURMOUNT-1 extension is ongoing, but published follow-up stops at 72 weeks. Weight regain after discontinuation has been documented in smaller cohorts, with one analysis showing that participants regained two-thirds of lost weight within one year of stopping semaglutide (Wilding 2022). Whether dual or triple agonism confers any advantage in maintaining weight loss after cessation is unknown. Telehealth platforms that prescribe these peptides often do so in open-ended treatment plans without a defined endpoint, a model that prioritizes adherence over exit strategy.
Comparative trials of tirzepatide versus semaglutide in the same population are limited. The SURPASS-2 trial compared tirzepatide to semaglutide 1 mg in adults with type 2 diabetes, finding greater HbA1c reduction and weight loss with tirzepatide 10 and 15 mg (Frías 2021). However, the semaglutide dose used was the diabetes-approved 1 mg, not the 2.4-mg obesity dose, making it difficult to assess whether the dual-agonist mechanism offers true superiority or whether the comparison simply reflected a dose mismatch. No trial has directly compared retatrutide to tirzepatide, and the phase-2 retatrutide data come from a different population and protocol design than SURMOUNT, precluding firm conclusions about relative efficacy.
How to Read the Existing Data
When a trial reports mean weight loss of 20 percent, recognize that the distribution around that mean is wide, and that the median may differ from the mean if a subset of high responders skews the average upward. SURMOUNT-1 showed that 36 percent of participants on tirzepatide 15 mg lost at least 25 percent of body weight, but that also means 64 percent did not reach that threshold. The placebo-adjusted effect is the more informative number, as it isolates the drug effect from the contribution of dietary counseling, self-monitoring, and regression to the mean. In SURMOUNT-1, placebo participants lost 3.1 percent, so the 20.9 percent in the tirzepatide arm translates to a 17.8 percentage-point drug-attributable reduction.
Adverse-event rates should be read in light of the background rate in placebo groups. Nausea occurred in 31 percent of tirzepatide 15-mg recipients and 9 percent of placebo recipients in SURMOUNT-1, yielding a 22 percentage-point absolute increase attributable to the drug. The fact that 9 percent of placebo participants reported nausea underscores the role of expectation, dietary change, and the clinical-trial context itself in symptom generation. Dropout rates are a proxy for tolerability in real-world conditions, where participants are not compensated for visits or monitored as closely. SURMOUNT-1 saw 14.3 percent discontinuation in the 15-mg arm versus 26.4 percent in placebo, a pattern in which higher placebo dropout often reflects disappointment with lack of efficacy rather than intolerability.
Subgroup analyses by sex are often underpowered, as trials are sized for the primary endpoint in the full cohort, not for interaction testing within demographic strata. When a paper states that "no significant interaction by sex was observed," it means the confidence intervals around the male and female point estimates overlapped, not that the effects were identical. Women in SURMOUNT-1 lost slightly more weight than men in absolute kilograms, but when expressed as percentage of baseline, the difference narrowed. Hormonal factors, body composition, and baseline metabolic rate all differ by sex, yet most trials adjust only for baseline weight and diabetes status in their models.
Compounded tirzepatide from telehealth sources may not be pharmaceutically equivalent to branded Mounjaro or Zepbound. Compounding pharmacies operate under Section 503A or 503B of the Federal Food, Drug, and Cosmetic Act, which permits preparation of drugs that are in shortage or for patients with specific needs, but does not require the same manufacturing controls as FDA-approved products. Potency can vary, sterility is not uniformly third-party verified, and excipient profiles may differ. A 2020 analysis of compounded semaglutide samples found peptide content ranging from 82 to 118 percent of label claim across eight pharmacies (Smith 2020). Patients receiving compounded peptides should request certificates of analysis and be aware that insurance coverage, adverse-event reporting, and product liability protections differ from those associated with branded drugs.
The Honest Answer on Dual-Action Peptides
Telehealth platforms are shifting to tirzepatide and retatrutide because the published weight-loss percentages are higher than those seen with semaglutide at the doses tested, and higher numbers are easier to market. Whether the dual or triple agonism is the mechanistic driver or whether the effect simply reflects dose optimization and trial design is less clear. The GIP receptor's role in human metabolism is still debated, with some investigators arguing that GIP agonism enhances insulin secretion and lipid clearance, while others note that GIP receptor knockout mice are protected from diet-induced obesity, suggesting that antagonism, not agonism, might be beneficial. Tirzepatide's dual-agonist activity may work despite GIP agonism rather than because of it, with the GLP-1 component doing most of the heavy lifting.
For women, the absence of cycle-stratified data is a persistent gap. Insulin resistance worsens in the luteal phase, appetite increases in the days before menses, and nausea thresholds may shift with estrogen and progesterone flux. A fixed weekly dose may be suboptimal compared to a cycle-adjusted regimen, but no trial has tested that hypothesis. Women with PCOS may see dual benefits, weight loss plus androgen reduction, but the magnitude of androgenic improvement and its durability after stopping the peptide are not well characterized beyond one or two small studies. Pregnancy risk is real, and the teratogenic signals in animal models are concerning enough that any woman of reproductive potential should be on reliable contraception and have a plan for discontinuation well before any planned conception.
Retatrutide's triple-agonist profile adds glucagon receptor activation, which increases hepatic glucose output and energy expenditure but also raises theoretical concerns about hyperglycemia in the fasted state and about cardiovascular stress in individuals with underlying coronary disease. The phase-2 trial excluded participants with recent cardiovascular events, so safety in that population is unknown. The 24 percent mean weight loss at 12 mg is impressive, but the 60 percent nausea rate and 29 percent vomiting rate are higher than those seen with tirzepatide, and whether patients will tolerate those effects outside a trial setting, where close monitoring and supportive care are standard, is uncertain.
AOD-9604, MOTS-c, and hexarelin appear in telehealth peptide stacks despite minimal or null human efficacy data. AOD-9604's 2008 trial failure has not prevented its inclusion in weight-loss protocols, often at doses and regimens that differ from the studied protocol. MOTS-c has mechanistic plausibility from rodent work, but translation to humans is speculative, and no dose-ranging or safety study has been published in a peer-reviewed journal. Hexarelin's growth-hormone secretagogue activity carries the same risks as other GH-releasing peptides, glucose intolerance, fluid retention, and potential acceleration of occult neoplasms, without the efficacy data to justify those risks in a weight-loss context.
Telehealth prescribing of these peptides operates in a regulatory gray zone. Compounded versions are legal when the branded product is in shortage or unavailable, but the definition of "unavailable" is interpreted variably across states. Some platforms prescribe tirzepatide off-label for weight loss even when Zepbound (the obesity-approved brand) is in stock, relying on the argument that compounded versions are more affordable. Others prescribe retatrutide, which has no FDA approval for any indication, under the rationale that it is investigational and therefore not subject to the same restrictions. This article is strictly informational; possession, sale, or use of the substances discussed may be restricted under federal, state, or local law in your jurisdiction. Consult applicable regulations before any action.
The shift to dual-action peptides is driven by efficacy signals in controlled trials, but those signals come with caveats. The trials are short, the populations are selected, and the real-world adherence and safety profiles are still emerging. For women, the lack of sex-specific dosing guidance, cycle-phase analysis, and pregnancy-outcome data means that clinical use is proceeding ahead of the evidence base. The 20 to 24 percent weight-loss figures are compelling, but they represent averages in trial cohorts, not guarantees for individual patients, and the adverse-event burden is not trivial. Whether the dual-agonist mechanism will prove more durable, safer, or more tolerable than GLP-1 monotherapy in the long term is a question that will take years and larger cohorts to answer, and the telehealth model, where follow-up is often asynchronous and dropout is invisible, may not be the setting in which those answers emerge most clearly. The current data support cautious optimism, but not the certainty that marketing language often implies.
Share This Article