Long-Term Peptide Safety Data Gaps: What We Still Don't Know

Compounded Peptide Safety

47 FDA-approved peptides

An analysis of every peptide approved by the FDA from 1998 to 2019 found no specific regulatory guidance exists for peptide safety assessment.

Mitra et al., Regulatory Toxicology and Pharmacology, 2020

Mitra et al., Regulatory Toxicology and Pharmacology, 2020

View as image

Between 2016 and 2024, peptides accounted for over 11% of all new pharmaceutical entities authorized by the FDA.^[1] That number keeps climbing. But here is the problem with long-term peptide safety: for the vast majority of these compounds, researchers do not know what happens after years of use. Clinical trials that support FDA approval typically last months to a few years. What happens at year five, year ten, or beyond often remains a blank space in the scientific record. For compounded and unregulated peptides, the picture is even more incomplete, because most have never been through a controlled human trial of any length.

This article maps what is known, what is not, and where the specific data gaps sit across different peptide categories.

The Regulatory Guidance Gap

Peptides sit in an awkward regulatory space. They are too large to be classified as traditional small molecules but too small and structurally simple to fit neatly into the biologics framework. This ambiguity has real consequences for how their long-term safety is evaluated.

Mitra et al. (2020) analyzed 47 peptides approved by the FDA between 1998 and 2019, including 22 chemically synthesized, 6 semi-synthetic, 18 recombinant, and 1 naturally derived compound.^[2] Their central finding: there is no specific regulatory guidance for peptide development. Sponsors rely on associating existing small molecule guidance (ICH M3(R2)) with biologic guidance (ICH S6(R1)), making case-by-case determinations on requirements like genotoxicity testing and the acceptable length of general toxicology studies.

Zane et al. (2021) reached the same conclusion from a joint industry-FDA symposium, noting that "disparities in interpretation and application of existing regulatory guidances to innovative synthetic and conjugated peptide assets have resulted in challenges for both regulators and sponsors."^[3] The symposium specifically flagged genotoxicity testing, impurity assessment, and safety evaluation of peptides containing non-proteogenic amino acids as areas where regulatory frameworks remain undefined.

Colalto (2024) added a European perspective, arguing that peptide-related impurities "cannot follow the small molecule approach" because hazardous interactions with biological systems, including specific interactions with cellular or membrane targets similar to natural peptide toxins, cannot be ruled out without dedicated testing.^[4] The practical result: each new peptide therapeutic enters a regulatory process where the rules for evaluating its long-term safety are still being written.

What Long-Term Safety Data Actually Exists

The gap between approved and unapproved peptides is enormous. For FDA-approved peptide drugs, post-marketing surveillance collects safety signals through FAERS (the FDA Adverse Event Reporting System), physician reporting, and periodic safety update reports. These systems catch acute and medium-term problems but depend on voluntary reporting and are poorly designed to detect gradual, cumulative effects.

Fosgerau and Hoffmann (2015) characterized the general landscape: approved peptide therapeutics show "high selectivity and efficacy" with "relatively safe and well tolerated" profiles, but this conclusion is based primarily on controlled trial data lasting months to a few years.^[5] At the time of their review, approximately 140 peptide therapeutics were in clinical trials, and the longest safety datasets came from established drugs like insulin analogs and GnRH agonists rather than newer peptide classes.

For compounded peptides, the situation is fundamentally different. Products from compounding pharmacies are not required to undergo clinical trials, and post-marketing surveillance is inconsistent. The FDA's 2023 decision to recategorize more than a dozen peptides (including BPC-157 and several growth hormone secretagogues) as ineligible for compounding was driven partly by the absence of adequate safety data, not by documented harms.

The Immunogenicity Unknown

One of the most underappreciated long-term risks with peptide therapeutics is immunogenicity: the formation of antidrug antibodies (ADAs) that can neutralize the drug, alter its clearance, or trigger immune reactions.

Achilleos et al. (2025) reviewed this problem in detail, noting that immunogenicity "can potentially limit the efficacy and safety of peptide-based therapeutics" and that the risk can be triggered by the peptide itself or by impurities introduced during production or formulation.^[1] Current FDA guidelines require immunogenicity risk assessment in market authorization applications, including identification of drug impurity levels above 0.1%.

The challenge with long-term use is that ADA formation is often a slow process. A patient may tolerate a peptide drug for months before their immune system begins producing antibodies against it. Clinical trials designed to last 6 to 12 months may never detect this response. And because peptide synthesis is shifting toward greener chemistries to reduce the 3 to 15 tons of waste generated per kilogram of peptide yield, new impurity profiles are being introduced that require fresh immunogenicity assessment.

Khan et al. (2018) cataloged additional long-term toxicity concerns for biologically active peptides, including intestinal wall disruption, erythrocyte and lymphocyte toxicity, free radical production, and cytotoxicity.^[6] Their review emphasized that in silico toxicity prediction methods, while improving, cannot fully replace long-term in vivo observation.

Growth Hormone Secretagogues: The Cancer Question

Growth hormone releasing peptides (GHRPs) and other secretagogues like ipamorelin and ibutamoren stimulate GH release, which in turn raises insulin-like growth factor 1 (IGF-1) levels. Elevated IGF-1 has been associated in observational studies with increased cancer risk, creating a theoretical concern that sustained use of these peptides could promote tumor growth.

Sigalos and Pastuszak (2018) reviewed the available evidence and found that "few long-term, rigorously controlled studies have examined the efficacy and safety of GHSs."^[7] Available studies indicated that GHSs are "well tolerated, with some concern for increases in blood glucose because of decreases in insulin sensitivity." But the review explicitly called for evaluation of "cancer incidence and mortality" with long-term use, data that still does not exist for any growth hormone secretagogue peptide.

The IGF-1 and cancer connection is not settled science. Studies of recombinant growth hormone in childhood cancer survivors have not shown excess de novo cancer rates. But these populations were medically monitored, received standardized doses, and were followed by endocrinologists. The bodybuilding and wellness communities using growth hormone secretagogues at self-selected doses without medical supervision occupy an entirely different risk landscape, one where long-term safety data simply does not exist.

GLP-1 Agonists: Where the Data Is Strongest

Among all peptide drug classes, GLP-1 receptor agonists have the most extensive long-term safety dataset, driven by massive cardiovascular outcome trials and years of widespread clinical use for diabetes and obesity.

Huang et al. (2024) published a global retrospective cohort study following 12,123 propensity-matched individuals with obesity (without type 2 diabetes) on GLP-1 RAs for up to 5 years.^[8] They reported a lower risk of all-cause mortality (hazard ratio 0.23, 95% CI 0.15-0.34), reduced cardiovascular complications, and lower risk of acute kidney injury. These protective effects held across subgroups and geographic regions.

Even here, gaps exist. Nagendra et al. (2023) conducted a systematic review and meta-analysis of 37 randomized controlled trials and 19 real-world studies examining semaglutide and cancer risk.^[9] They found no increased risk of thyroid cancer (OR 2.04, 95% CI 0.33-12.61, P = 0.44) or pancreatic cancer (OR 0.25, 95% CI 0.03-2.24, P = 0.21) compared to placebo. But these confidence intervals are wide, and the follow-up periods in most included trials were 1-3 years.

A Danish nationwide emulated trial following GLP-1 RA users for up to 10 years found that among sustained users, 4.1 more patients per 100 developed cancer compared to DPP-4 inhibitor users. Fewer patients died without prior cancer in the GLP-1 group, and overall "death or cancer" rates were similar. This is the kind of nuanced, long-duration data that most other peptide classes lack entirely.

The thyroid cancer question also illustrates the gap between animal and human data. Rodent studies show GLP-1 receptor activation causes thyroid C-cell tumors, but humans have far fewer GLP-1 receptors on C-cells. Cardiovascular outcome trials showed stable calcitonin levels over 3 years. Whether 10 or 20 years of exposure changes this picture remains unknown. For more on what years of liraglutide safety monitoring have revealed, see our dedicated analysis.

Research Peptides: The Complete Data Void

For peptides that have never been through the FDA approval process, the term "long-term safety data" is essentially meaningless because short-term safety data barely exists either.

BPC-157 is the most prominent example. Despite hundreds of animal studies, only a handful of human studies have been published. The most recent is a pilot study by Lee and Burgess (2025) that administered intravenous BPC-157 to 2 participants (a 58-year-old male and a 68-year-old female) over 3 days at doses of 10 mg and 20 mg.^[10] No adverse effects on heart, liver, kidney, thyroid, or blood glucose biomarkers were detected. The study concluded that BPC-157 was "well-tolerated," but a 3-day observation window in 2 people tells us nothing about what happens after months or years of repeated use.

TB-500 (thymosin beta-4) has zero published human clinical trials. Its safety in humans is completely unknown. Pharmacokinetic data (absorption, distribution, metabolism, elimination) remains uncharacterized for both BPC-157 and TB-500, which means even the most basic dose-response relationships needed for safety assessment have not been established.

This is not a gap that more animal data can fill. Rodent models miss immunogenicity signals that only appear in humans. They do not detect the psychological and behavioral changes that peptide users report anecdotally. And they cannot model the real-world pattern of use: intermittent self-administration of varying purity products at doses extrapolated from animal studies without allometric scaling.

Impurity and Degradation Risks Over Time

Long-term peptide use introduces compounding risks beyond the peptide itself. Achilleos et al. (2025) noted that peptide synthesis generates significant waste and that new impurity profiles emerge as manufacturing processes evolve.^[1] Any impurity above 0.1% in an approved peptide product requires comparative immunogenicity risk assessment.

Colalto (2024) emphasized that the analytical investigation of peptide-related impurities requires approaches specific to their complex mechanisms of action.^[4] Direct genotoxic mechanisms from peptide impurities "cannot be excluded," and hazardous interactions with biological systems, similar to those seen with natural peptide toxins, remain a concern that current testing frameworks address incompletely.

For users of compounded peptides, these risks multiply. Independent testing has found that a significant fraction of peptides sold online show purity discrepancies compared to supplier claims. Methionine and tryptophan oxidation from poor handling or storage creates degradation products whose long-term effects have not been studied. A user taking the same peptide from the same supplier over months may be exposed to varying impurity profiles with each new batch, creating a toxicological scenario that no clinical trial has ever modeled.

The Bottom Line

The long-term safety profile of peptide therapeutics ranges from reasonably well-characterized (GLP-1 agonists with 5-10 year datasets) to completely unknown (BPC-157, TB-500, and most growth hormone secretagogues). The regulatory framework itself is a gap: no peptide-specific guidance exists for safety assessment, immunogenicity monitoring beyond trial periods is sparse, and the theoretical cancer risks associated with sustained IGF-1 elevation remain untested. For research peptides, the absence of controlled human trials of any duration makes long-term safety assessment impossible with current data. These gaps are not speculative concerns. They are documented, quantified, and acknowledged by the same regulatory scientists responsible for approving peptide drugs.

Frequently Asked Questions

Are peptides safe for long-term use?

Safety depends entirely on which peptide. FDA-approved peptides like semaglutide have multi-year safety data from large clinical trials and post-marketing surveillance showing generally favorable profiles. Research peptides like BPC-157 and TB-500 have no controlled long-term human safety data. A 2020 analysis of 47 FDA-approved peptides found no dedicated regulatory guidance exists for peptide-specific safety assessment, meaning each product's long-term monitoring framework is determined on a case-by-case basis.^[2]

What are the long-term side effects of peptide therapy?

For approved peptides, documented long-term effects vary by class. GLP-1 agonists show gastrointestinal side effects that typically diminish over time, with rodent thyroid tumor signals that have not been confirmed in human studies up to 5 years. Growth hormone secretagogues show decreases in insulin sensitivity with sustained use. For compounded and research peptides, long-term side effects are unknown because no studies longer than a few days have been conducted in humans.^[7]

Why is there so little long-term safety data on peptides?

Three factors converge. First, no peptide-specific regulatory guidance exists, so safety testing requirements are inconsistent across products. Second, clinical trials for drug approval typically last months to a few years, too short to detect cumulative effects. Third, for research peptides sold as "not for human consumption," no regulatory body requires human safety studies, and manufacturers have no financial incentive to conduct them.^[3]

Can peptide injections cause cancer?

No peptide injection has been proven to cause cancer in humans. The concern is theoretical, centered on peptides that elevate IGF-1 (like growth hormone secretagogues), since elevated IGF-1 correlates with increased cancer risk in observational studies. A 2023 meta-analysis of 37 RCTs found semaglutide was not associated with increased cancer risk (all neoplasms OR 0.95, P = 0.82), but most trials lasted 1-3 years. Long-term cancer incidence data for growth hormone secretagogues does not exist.^[9]

How long has semaglutide been studied for safety?

Semaglutide has the most extensive dataset among newer peptide therapeutics. Randomized controlled trials span 1-3 years, cardiovascular outcome trials extend to 3-5 years, and a Danish nationwide observational study followed GLP-1 RA users for up to 10 years. A 2024 retrospective cohort of 12,123 matched patients tracked outcomes over 5 years, finding reduced all-cause mortality (HR 0.23) and cardiovascular protection.^[8]

What is immunogenicity and why does it matter for peptide safety?

Immunogenicity is an unintended immune response to a peptide drug that produces antidrug antibodies (ADAs). These antibodies can neutralize the drug, alter how the body clears it, or cause allergic reactions. The risk increases with long-term use because ADA formation can be slow, taking months to develop. A 2025 review found that immunogenicity testing remains a priority as peptide manufacturing shifts to new chemistries that introduce novel impurity profiles requiring reassessment.^[1]