Will Peptides Work for Me? What Genetics, Sex, and Phenotype Tell Us

A landmark 2026 Nature study of 27,885 people found genetic variants that predict GLP-1 drug response. But genetics is just one piece — here's the full picture of why peptides work differently for different people.

Research Digest8 min readApril 8, 2026

The Response Spectrum

Not everyone responds to peptide therapy the same way. This isn't a vague disclaimer — it's a quantifiable reality. Clinical trial data from the STEP 1 semaglutide trial shows a clear distribution: approximately 50% of patients achieved 15% or greater weight loss, but about 14% were non-responders (less than 5% weight loss on the same 2.4mg dose). A retrospective cohort of 483 GLP-1 patients found 33.8% were hyper-responders (>15% weight loss) while 17.8% were non-responders. This means the same drug, at the same dose, produces dramatically different outcomes in different people. Understanding why — and whether you can predict which group you'll fall into — is one of the most important questions in modern peptide therapy.

The Genetics: A Nature Study of 27,885 People

Published in Nature on April 8, 2026, the largest pharmacogenomic study of GLP-1 response to date analyzed genetic data from 27,885 people taking GLP-1 receptor agonists. The 23andMe Research Institute-led study identified a missense variant in the GLP1R gene significantly associated with weight loss efficacy — carriers experienced an additional 0.76 kg of weight loss per copy of the effect allele. The study also found genetic variants in both GLP1R and GIPR associated with nausea and vomiting side effects. Notably, the GIPR association for vomiting was restricted to tirzepatide users (since tirzepatide activates both GLP-1 and GIP receptors, while semaglutide only activates GLP-1). This is a genuine breakthrough, but context matters: a 2025 study of 10,960 individuals across 9 biobanks found no significant genetic predictors of GLP-1 weight loss response. The Nature study's much larger sample size was needed to detect the signal — suggesting the genetic effect is real but individually modest. Genetics contributes to response variation, but it's not destiny.

Sex: The Strongest Predictor We Have

Across all the data, biological sex is the most consistent and significant predictor of GLP-1 weight loss response. A 2025 systematic review of 64 randomized controlled trials (19,906 patients) found women lost an average of 10.9% body weight compared to 6.8% for men — and this was the only subgroup variable that showed statistically significant heterogeneity of treatment effect. Age, race, ethnicity, baseline BMI, and HbA1c did not predict response. A large Italian cohort study (n=7,847, median 4-year follow-up) confirmed: 66.5% of women achieved at least 5% weight loss versus 58.0% of men, and 40.0% of women achieved 10% versus 30.7% of men. Why? The mechanisms aren't fully understood, but likely involve sex differences in body fat distribution, hormonal milieu, GLP-1 receptor expression, and central appetite regulation. For men, this doesn't mean GLP-1 drugs don't work — it means expectations should be calibrated differently.

Diabetes Blunts the Response

Type 2 diabetes consistently predicts a lower weight loss response to GLP-1 agonists. The contrast is stark: in the STEP 1 trial (patients without diabetes), semaglutide 2.4mg produced 14.9% mean weight loss. In STEP 2 (patients with type 2 diabetes), the same dose produced 9.6% — roughly one-third less. A meta-analysis confirmed this gap across multiple studies: patients without diabetes lost an average of 11.57% body weight versus 6.34% for those with type 2 diabetes on the same GLP-1 therapy. The explanation involves insulin resistance, hyperinsulinemia, and compensatory glycemic mechanisms that partially counteract the metabolic effects of GLP-1 agonism. For patients with diabetes, the weight loss is still clinically meaningful, and the glycemic benefits are an additional advantage — but the magnitude of weight reduction is genuinely lower.

Your Obesity Phenotype Matters

One of the most promising approaches to predicting response comes from obesity phenotyping — the idea that obesity isn't one disease but several, each driven by different physiological mechanisms that respond to different interventions. Researchers at Mayo Clinic identified four phenotypes present in 85% of patients: "hungry brain" (impaired satiation — difficulty feeling full), "emotional hunger" (hedonic eating), "hungry gut" (rapid gastric emptying — hunger returns quickly after meals), and "slow burn" (decreased metabolic rate). In a pragmatic trial of 312 patients, phenotype-guided treatment selection produced 15.9% mean weight loss compared to 9.0% for non-guided treatment. 79% of the phenotype-guided group achieved over 10% weight loss versus just 34% without phenotyping. A 2025 study in Cell Metabolism extended this with a machine-learning genetic risk score for satiation. The "hungry gut" phenotype responded significantly better to liraglutide, while the "hungry brain" phenotype did better with phentermine-topiramate. Matching the mechanism to the phenotype nearly doubled efficacy.

Early Response Predicts Long-Term Success

If you're already on a GLP-1 drug and wondering whether to continue, early weight loss is a strong predictor. A post-hoc analysis of the SURMOUNT-1 trial found that 82% of tirzepatide patients were "early responders" (achieving at least 5% weight loss by 12 weeks). The 18% who were "late responders" were more likely to be male, heavier, and with higher baseline BMI. Critically, 90% of those late responders eventually achieved meaningful weight loss by week 72 with continued treatment. The message: if you haven't responded strongly by 12 weeks, it doesn't necessarily mean the drug isn't working — but it does warrant a conversation with your clinician about dose adjustment, switching to a dual agonist like tirzepatide, or adding complementary strategies.

Beyond GLP-1: The Evidence Evaporates

For GLP-1 agonists, we have pharmacogenomic data from nearly 28,000 people, sex-stratified meta-analyses, and phenotype-guided trials. For every other class of therapeutic peptide, the data on individual response variation is sparse to nonexistent. BPC-157 has only one published human study — a safety pilot in 2 patients. There is no data on who responds and who doesn't. GH secretagogues (ipamorelin, CJC-1295, MK-677) show substantial individual variation driven by age, sex, and body fat composition — one study found BMI accounted for 38-65% of response variability — but no pharmacogenomic studies exist. Nootropic peptides (semax, selank) have zero published data on individual response variation. This doesn't mean these peptides don't work. It means we can't yet predict for whom they'll work best. The personalized medicine revolution is arriving for GLP-1 drugs first, with other peptide classes likely years behind.

What You Can Do Now

While pharmacogenomic testing for peptide response isn't yet standard clinical practice, several practical strategies can improve your odds: Work with an experienced clinician who will titrate your dose based on your individual response, not just follow a standard protocol. Give GLP-1 drugs adequate time — slow dose escalation over 16-20 weeks reduces side effects and some patients need the full escalation before seeing significant weight loss. Track your early response: if you've achieved meaningful weight loss by 12 weeks, you're likely in the responder group. If not, discuss with your clinician whether to increase dose, switch agents, or add complementary therapies. Consider your phenotype: if you identify with the "hungry gut" pattern (eating fills you up but you get hungry again quickly), GLP-1 agonists that slow gastric emptying may be particularly well-suited. If your pattern is more "hungry brain" (difficulty feeling full during meals), a different mechanism may be more effective. And manage expectations: women, patients without diabetes, and patients with lower starting BMI tend to see larger percentage weight loss. These are population-level patterns, not individual guarantees — but they help frame what's realistic.

Key Findings

A Nature 2026 GWAS of 27,885 people identified a GLP1R missense variant associated with -0.76 kg additional weight loss per allele copy
Non-responders to semaglutide (<5% weight loss) represent 14-18% of patients across multiple cohorts
Women consistently lose more weight on GLP-1 drugs: 10.9% vs 6.8% for men across 64 RCTs — the only subgroup variable with statistically significant heterogeneity
Type 2 diabetes reduces GLP-1 weight loss response by roughly one-third: 14.9% without diabetes vs 9.6% with diabetes on the same semaglutide dose
Phenotype-guided treatment selection produced 15.9% vs 9.0% weight loss — nearly doubling efficacy by matching mechanism to patient
90% of 'late responders' to tirzepatide eventually achieved meaningful weight loss by 72 weeks with continued treatment
Beyond GLP-1 drugs, essentially no data exists on individual response variation for BPC-157, GH secretagogues, or nootropic peptides

Limitations & Caveats

The Nature GWAS used self-reported weight loss data from 23andMe — less precise than clinical measurement
A contrasting 2025 study of 10,960 people found no significant genetic predictors, suggesting the effect size is modest
Phenotype-guided treatment data comes from a single pragmatic trial — larger confirmatory studies are needed
Sex-based differences may partially reflect reporting bias, compliance differences, and body composition variation
For non-GLP-1 peptides, the lack of response-prediction data is a data gap, not evidence of uniform response
Pharmacogenomic testing for peptide therapy is not yet clinically validated or widely available