Peptides for Long COVID, ME/CFS, and Post-Viral Fatigue: What the Evidence Actually Shows

The NIH's RECOVER-TLC program — the largest coordinated Long COVID treatment initiative in history, around $1.15 billion in funding — picked its first treatment arms in 2025. The choices: stellate ganglion blocks, low-dose naltrexone, baricitinib, and a GLP-1. No peptides made the cut.

That is a striking data point because the peptide clinic industry has moved aggressively into the Long COVID and ME/CFS space. Hudson Medical, Peak Health, SSP Research, and dozens of functional-medicine practices now offer peptide protocols for post-viral fatigue, brain fog, and dysautonomia. The most common entries are BPC-157, KPV, thymosin alpha-1, Selank, and sometimes intranasal VIP.

The gap between clinic protocols and RECOVER's evidence-weighted picks isn't malpractice. It reflects the thin RCT base for peptides in this population. The one peptide to reach a large NIH-funded COVID trial (aviptadil) failed. The peptide with the best Long COVID data (thymosin alpha-1) has exactly one ex vivo study on isolated lymphocytes. The mitochondrial peptides with the strongest mechanistic rationale (SS-31, MOTS-c) have zero completed trials in ME/CFS or Long COVID. The rest of this piece is a peptide-by-peptide walk through what each one actually has behind it.

Thymosin alpha-1 (Tα1, marketed as Zadaxin outside the US) is a 28-amino-acid immune modulator that regulates Th1/Th17 balance, reduces T-cell exhaustion, and drives CD4+ maturation. It's approved in ~35 countries for hepatitis B/C and as a cancer adjuvant, though not FDA-approved in the US.

The Long COVID data point: Matteucci et al. (International Immunopharmacology, March 2023) took lymphocytes from PASC patients and treated them with Tα1 in vitro. Tα1 shifted CD4+ central-memory to effector-memory cells, reduced PD-1+ CD8+ exhaustion markers, and dampened inflammatory gene expression. The effect was strongest in patients who had required respiratory support and those with neuropsychiatric PASC. This is the best human PASC data available for any peptide.

It's also not an RCT. It's not an in vivo study. It's lymphocyte culture from ~20 patients. The mechanistic implications are genuinely intriguing, but anyone citing this as proof that Tα1 treats Long COVID is overstating what was measured.

The deflating counterpoint: The TESTS trial (Lancet Respiratory Medicine, January 2025), which randomized 1,089 sepsis patients to Tα1 or placebo, showed no mortality benefit (23.4% vs 24.1%). Earlier acute-COVID meta-analyses suggested mortality benefit in severe disease, but the sepsis null result tempers broad enthusiasm about Tα1 as a general immune reset.

Bottom line on Tα1 for Long COVID: the strongest mechanistic and immune-biomarker story of any peptide in this space, paired with zero interventional trials.

Aviptadil — synthetic vasoactive intestinal peptide (VIP), sold as Zyesami or RLF-100 — had the highest-profile peptide trial in COVID history.

NRx's Phase 2b/3 (NCT04311697, IV aviptadil in critical COVID) initially reported a win: 71% recovery vs 48% in the high-flow nasal cannula subgroup. Stock surged. The NIH funded a confirmatory trial. Then ACTIV-3b (NCT04843761) stopped for futility after ~460 patients — day-60 mortality was 37% on aviptadil vs 36% on placebo. I-SPY COVID (inhaled/nebulized aviptadil) also stopped for futility at ATS 2022.

NRx transferred all aviptadil assets to Relief Therapeutics in December 2022. NRx pivoted to bipolar depression. Relief continues aviptadil development for rare pulmonary indications. There are no active Long COVID aviptadil trials.

The Shoemaker intranasal-VIP protocol, which predates COVID, is an entirely separate conversation. Ritchie Shoemaker uses nasal VIP (50 mcg/spray, 4-8x/day) as the final step of his Chronic Inflammatory Response Syndrome (CIRS) protocol for mold and biotoxin illness. His published data is observational across ~10,000 patients in his own practice — no RCT exists. Some clinicians extrapolate from CIRS to Long COVID because of cytokine and autoimmune overlap, but no Long COVID VIP trials have been conducted.

ME/CFS has a real mitochondrial dysfunction story. Robert Naviaux's 2016 metabolomics work identified a 'dauer-like' hypometabolic signature. A 2024 validation study replicated 77% (23/30) of the original metabolic pathway abnormalities. A February 2025 Physiology review and a 2025 Cell Reports Medicine paper confirmed abnormal energy metabolism across the condition.

That creates real therapeutic rationale for mitochondrial peptides. Rationale is not evidence.

SS-31 (Elamipretide) became the first FDA-approved mitochondrial peptide in September 2025, for Barth syndrome. It binds cardiolipin, stabilizes the inner mitochondrial membrane, and restores electron transport chain efficiency. The MMPOWER-3 trial (NCT03323749) in primary mitochondrial myopathy didn't improve 6-minute walk distance or fatigue at 24 weeks vs placebo. Short-term crossover data showed an exercise benefit; long-term confirmation was negative. No ME/CFS or Long COVID trials have been completed.

MOTS-c is a mitochondrial-derived peptide that activates AMPK (the exercise pathway). Rinaldi et al. (Journal of Translational Medicine, 2019) found lower MOTS-c transcript and plasma levels in Q-fever fatigue and CFS patients than in controls. That's correlational, not interventional. The closest MOTS-c analog to reach humans was CohBar's CB4211 in NAFLD/obesity (NCT03998514), a 20-patient Phase 1; CohBar wound the program down. A July 2025 preprint argued for MOTS-c in ME/CFS, but it's a hypothesis piece, not a trial.

Humanin has correlational fatigue data (Rinaldi 2019) and no interventional studies in post-viral conditions.

The pattern across all three is the same: the scientific case for mitochondrial peptides in ME/CFS and Long COVID is the strongest of any peptide class, and no one has run the trial.

BPC-157 (Body Protection Compound-157) is the most commonly prescribed peptide in Long COVID and ME/CFS clinic protocols. It also has the thinnest human evidence and — notably — the loudest negative anecdote signal in exactly this population.

The evidence base: One theoretical paper (Deek, Medical Hypotheses 2022) proposed BPC-157 for COVID. That's a hypothesis, not data. No RCTs exist in Long COVID, ME/CFS, or any post-viral condition.

The community signal: The Phoenix Rising forum — the largest ME/CFS community in the English-speaking world — has several long-running BPC-157 threads. The reports are mixed, with concerning negative outliers:

• Multiple users describe becoming housebound after starting BPC-157 despite having been working and walking beforehand. This is the classic ME/CFS crash pattern.
• One user reported "the worst cluster headache of my life" at 250 mcg BID.
• Reports of anhedonia and dopaminergic shifts.
• Some users do report post-exertional malaise reduction, but these are selected anecdotes.

The mechanistic case for BPC-157 in Long COVID centers on its documented vascular and gut-repair effects. The pattern in real-world ME/CFS use is different — hypersensitive post-viral patients appear to crash on peptides that healthy athletes tolerate well. This is one of the clearest examples in peptide medicine of a population mismatch between the rationale and the real-world result.

BPC-157 re-entered and then exited FDA Category 2 during 2024-2026; as of April 2026 it's back in Category 1 pending the July PCAC review.

Selank and Semax are intranasal peptides approved for medical use in Russia and a handful of post-Soviet states — nowhere else. They're heavily marketed in the US by functional-medicine clinics for Long COVID brain fog and fatigue, entirely on mechanistic extrapolation.

Semax is a 7-amino-acid fragment of ACTH. Russian trials in acute ischemic stroke at 9-18 mg/day intranasal have reported neurological recovery benefits. Zero Western trials have been conducted. Zero Long COVID trials have been conducted. Claims of BDNF upregulation, GABAergic and dopaminergic modulation, and cognitive recovery all come from Russian animal and small-human studies not replicated elsewhere.

Selank is a tuftsin analog studied in Russia for anxiety. Same pattern: no Western trials, no Long COVID trials.

It's worth stating plainly: these peptides have no Western regulatory approval for any indication, have never been evaluated for Long COVID in any trial, and their use rests entirely on extrapolation from Russian literature and clinician preference. That doesn't mean they don't work — it means no one has tested whether they do in the target population.

Consolidated view of the evidence base for peptides in Long COVID, ME/CFS, and post-viral fatigue:

The meaningful takeaway: Thymosin α-1 has the best immune-focused mechanistic data. Mitochondrial peptides (SS-31, MOTS-c) have the best metabolic-fatigue rationale. Neither has an RCT in the target population. BPC-157 has the loudest community enthusiasm paired with the loudest negative signal.

If you're considering peptide treatment for Long COVID, it's worth knowing what's actually in randomized trials:

• Patterson/HealthBio Phase 3 of maraviroc + atorvastatin in PASC (252 patients, 32 weeks, started March 2025). No peptides in the protocol. Built on a case series of maraviroc 300 mg BID + pravastatin 10 mg QD.
• ImmunityBio ANKTIVA — nogapendekin alfa, an IL-15 superagonist fusion protein — Phase 2 Long COVID trial (NCT07108036), 40 patients, 600 μg SC, running August 2025 through July 2026. This is technically a protein therapeutic, not a classical peptide, but it's the closest thing to a "biologics" trial in Long COVID.
• RECOVER-TLC treatment arms: stellate ganglion blocks, LDN, baricitinib, GLP-1 (semaglutide). The LDN pediatric arm enrolls summer 2026.
• RSLV-132 (RNase Fc fusion): Phase 2 Long COVID, NCT04944121.

The absence of peptides from these flagship trials is telling. It's not that investigators dismiss the mechanistic rationale — it's that the preclinical and small-scale human data aren't strong enough yet to justify the cost and complexity of a large Long COVID RCT.

For patients and clinicians considering peptide protocols for Long COVID, ME/CFS, or post-viral fatigue, a few things follow from the evidence landscape.

Start with mast cell and autonomic assessment. Roughly 40-50% of Long COVID patients have overlapping MCAS and/or POTS. Any peptide trial in this population should identify those comorbidities first, because they change which peptides are appropriate. CJC-1295/Ipamorelin cause histamine release and are commonly problematic in MCAS. GH secretagogues can destabilize blood pressure in POTS.

Microdose and track. The Phoenix Rising crash reports suggest dose-dependent tolerance issues. Starting at 10-25% of standard doses with a wearable tracking HRV, sleep, and resting heart rate is what most Long COVID-aware clinicians do.

Pick the peptide that matches the dominant phenotype. Immune-forward PASC (positive autoantibody or inflammatory biomarker profile) has the strongest Tα1 rationale. Fatigue-dominant ME/CFS with a clear metabolic signature has the strongest mitochondrial peptide rationale. Gut-dominant presentations may warrant KPV or BPC-157, with caution. 'All peptides for all presentations' is unsupported.

Source matters more than usual here. The peptide quality crisis applies with extra force to post-viral populations. A counterfeit or contaminated vial in an MCAS-prone patient isn't just ineffective; it can produce severe reactions. Third-party HPLC verification and endotoxin testing are worth the premium.

Plan for an exit. If symptoms worsen on peptide therapy, reversibility depends on the peptide. BPC-157 has a short half-life and washes out in days. Cerebrolysin effects may persist for weeks. Immunomodulators like Tα1 can have longer-lasting downstream effects on T-cell populations.

Don't replace evidence-based care. RECOVER-TLC enrolls eligible patients. LDN is affordable, compounded, and has a real Long COVID efficacy signal. Stellate ganglion blocks are covered by some insurance. Peptides should be additive to interventions with actual trial data, not a replacement for them.

Four years into widespread Long COVID research, and an even longer history of ME/CFS science, no peptide has an RCT-grade efficacy signal in this population. The best immune evidence is a single ex vivo study. The best mitochondrial rationale has zero interventional data. The most-used peptide (BPC-157) has concerning crash reports in the exact population using it most.

That isn't an argument that peptides don't help anyone. The functional-medicine clinics that prescribe them have patients who report genuine improvement. It's an argument that the evidence does not support the confident 'this is what works for Long COVID' framing that dominates clinic marketing.

What's actually true: Thymosin α-1 has the most credible immunological mechanism for PASC, with no interventional trial. Mitochondrial peptides have the strongest metabolic-fatigue rationale, with no interventional trial. BPC-157 and intranasal VIP are the most community-used, with the thinnest human evidence base. Russian peptides (Semax, Selank) have no Western or Long COVID trials at all. The one peptide that did reach a large NIH COVID trial failed.

If you're a patient, that's the picture your clinician should be able to articulate. If a protocol is being pitched with the confidence of established treatment, the confidence is ahead of the data. That doesn't mean it won't help. It means the decision is a judgment call in conditions of real uncertainty, not a standard of care.