AMY-101, a peptide inhibitor of complement C3, reduced astrocyte loss, AQP4 depletion, and demyelination in a mouse model of neuromyelitis optica spectrum disorder, and it blocked complement-mediated killing of AQP4-expressing cells in vitro. Reported in Annals of Medicine, the findings support…
AMY-101 , a peptide that inhibits complement C3 , reduced astrocyte loss, aquaporin-4 AQP4 depletion, and demyelination in a mouse model of neuromyelitis optica spectrum disorder NMOSD , and inhibited complement-mediated killing of AQP4-expressing cells in vitro. The findings appear in a report published December 7, 2026, in Annals of Medicine. The authors state that complement C3 inhibition with AMY-101 effectively ameliorates key pathological features of NMOSD in experimental models, supporting its potential as a therapeutic strategy and providing a rationale for further preclinical and clinical investigation.
NMOSD is an autoimmune inflammatory demyelinating disease of the central nervous system, characterized primarily by optic neuritis and transverse myelitis. In most patients the disease is driven by immunoglobulin G autoantibodies against AQP4, the water channel concentrated on astrocyte foot processes at the blood-brain barrier. When AQP4-IgG binds its antigen, it activates complement, and the resulting membrane attack complex MAC lyses the astrocytes. Those cells are the primary target of the disease; their loss is the initiating event that leads to inflammatory demyelination and the necrotic, cavitating spinal and optic nerve lesions that distinguish NMOSD from multiple sclerosis.
The new data matter because they point to C3, rather than the downstream C5 protein that an approved NMOSD therapy blocks, as the intervention point. A peptide that inhibits C3 could interrupt the cascade at its convergence point, preventing MAC formation and the inflammatory fragments generated upstream of it. The evidence is experimental, and the published report carries no quantitative detail, but the study extends the C3-inhibitor concept into a disease space dominated by monoclonal antibodies.
The report covers two experimental systems. In vitro, the investigators exposed AQP4-transfected cells to AQP4-IgG and human complement, with two endpoints:
AMY-101 significantly attenuated the AQP4-IgG- and complement-mediated cytotoxicity and inhibited MAC formation.
In vivo, the investigators used an intracerebral injection mouse model of NMOSD, comparing AMY-101-treated animals with untreated controls. The endpoints were:
In treated animals, all three markers of injury were markedly reduced compared with untreated controls.
What the report does not include is as notable as what it does. No sample size, no duration, no dose, no route of administration, no dosing regimen, no statistical values, and no effect sizes appear in the published account. The directional claims, protection across three pathological hallmarks in vivo and inhibition of MAC formation in vitro, rest on results described without their quantitative support. The study duration in the mouse model is likewise not reported.
The intracerebral injection model is a standard tool in NMOSD research. AQP4-IgG, often accompanied by human complement, is injected directly into the mouse brain, and the characteristic astrocyte-destructive lesion develops within days. The model is well matched to the question this study asks: whether a candidate drug blocks the effector arm of the disease, the binding of antibody to AQP4, the activation of complement, and the resulting death of astrocytes. On exactly those endpoints, AMY-101 performed well.
The model is far less suited to other questions that matter for a therapeutic claim. It does not test whether a drug prevents relapses over months, whether it modifies the adaptive immune response that generates AQP4-IgG, whether it preserves function across repeated attacks, or whether sustained complement suppression is tolerable when the complement system is needed for host defense. The in vitro system, AQP4-transfected cells confronted with antibody and human complement, isolates the same effector step with full experimental control, but it is a static, short-term snapshot of a single cell type.
The design can establish target engagement and proof of mechanism in a controlled setting. It cannot establish clinical efficacy. The report contains no human clinical data, a limitation the published report flags explicitly: only experimental in vitro and mouse-model results are described. No comparison with existing NMOSD therapies is provided, so the study offers no evidence about whether a C3 inhibitor would outperform or complement the treatments already in use.
The complement system is a cascade of proteolytic reactions. All three initiation pathways, classical, lectin, and alternative, converge on C3, the most abundant complement protein in plasma. C3 convertases cleave C3 into C3a, a potent inflammatory anaphylatoxin, and C3b, which attaches covalently to target surfaces, opsonizes them for phagocytosis, and feeds the alternative pathway amplification loop. C3b also assembles into the C5 convertase, the enzyme that cleaves C5 into C5a and C5b; C5b nucleates the assembly of C6 through C9 into the membrane attack complex, a pore that lyses cells.
In NMOSD, AQP4-IgG bound to astrocyte membranes recruits C1q and activates the classical pathway. The cascade runs through C3 and C5 to MAC, and complement-dependent lysis is thought to be a principal cause of astrocyte death in early lesions. Astrocyte injury depletes AQP4 from the tissue, the effect measured as AQP4 depletion, and the ensuing inflammation injures oligodendrocytes and myelin, the effect measured as demyelination. By binding C3 and preventing its cleavage, AMY-101 blocks C3b deposition, prevents C5 convertase assembly, and stops MAC formation, while also suppressing the generation of C3a and, downstream of it, C5a.
That positioning is mechanistically distinct from blocking C5. C5 inhibitors such as the approved NMOSD therapy eculizumab prevent MAC formation and C5a generation but leave C3b deposition, C3a generation, and the amplification loop intact. Inhibiting C3 is upstream of all of those activities. The breadth is also the liability. C3 is required for opsonization and clearance of microbes and immune complexes, for the lytic defense against encapsulated bacteria, and for anaphylatoxin-driven inflammation. Systemic C3 blockade therefore carries a predictable risk of infection and impaired immune-complex handling, a tradeoff that any C3-directed program must confront and that the current data do not address.
For NMOSD researchers, the report is a proof-of-mechanism data point. It demonstrates that a peptide inhibitor of C3 can attenuate complement-mediated cytotoxicity and key neuroinflammatory pathology in experimental NMOSD, and it supports the concept of targeting C3 with peptides to block the complement cascade upstream of MAC formation. It also provides a rationale for advancing AMY-101 as a therapeutic candidate in neuroinflammatory disease, not only NMOSD. The immediate next experiments are visible in the report's omissions: dose-response and pharmacokinetic studies, quantitative histology with statistics, and head-to-head comparison with existing NMOSD therapies.
For clinicians, nothing in this report changes practice, and it should not. The relevant context is that AQP4-IgG seropositive NMOSD is already treated with agents that reduce AQP4-IgG, such as the B-cell-depleting antibody inebilizumab and the interleukin-6 receptor blocker satralizumab , and with the C5 inhibitor eculizumab, which reduces relapse risk by blocking MAC formation while leaving upstream complement fragments active. A C3 inhibitor would act earlier in the pathway. Whether earlier blockade preserves more tissue, reduces disability accrual, or changes the character of lesions is entirely untested. The clinical question that matters, if the molecule advances, is whether protection in an acute model translates into fewer relapses and less cumulative injury in patients over years.
For peptide scientists and the supply chain, the study is a demonstration that a synthetic peptide can engage a central complement protein and protect central nervous system tissue in a disease model. That is not trivial: the complement inhibitor space has been dominated by antibodies, and a peptide offers defined chemical synthesis, potential specificity for a single protein, and a manufacturing path that does not depend on biologic production infrastructure. The same properties create the open problems. Peptides are typically cleared quickly from circulation, so dosing frequency and route of administration become central design questions, and the report gives no information on either. Formulation, stability, scale-up, purity, and distribution of the peptide to the central nervous system, or to sites of blood-brain barrier breakdown, are all unaddressed. Those questions become urgent only if the molecule survives the next stage of preclinical scrutiny, but they will decide whether AMY-101 becomes a drug candidate or remains a laboratory reagent.
The report is best read as hypothesis-generating, with real but narrow evidentiary reach. Only experimental in vitro and mouse-model data are described; no human clinical data are reported. Dose, route, sample size, statistical values, and effect sizes are not reported, so the magnitude of the protective effect, and the confidence that can be placed in it, cannot be assessed from the published account. No comparison with existing NMOSD therapies is provided. These gaps cap what the study can claim, though they do not necessarily reflect flaws in the experiments themselves.
The open questions follow directly. What dose, route, and dosing regimen of AMY-101 were used in the mouse model? What quantitative outcome measures and statistical results support the reported efficacy? How does AMY-101 compare with existing NMOSD therapies, such as complement C5 inhibitors or agents that reduce AQP4-IgG? What additional preclinical and clinical data are needed before AMY-101 can be tested in patients with NMOSD? Each question is answerable with additional reporting, additional experiments, or both.
What would settle the matter: a full report with quantitative histology, sample sizes, and statistics; dose-ranging and pharmacokinetic studies of the peptide; toxicology that addresses the predictable infectious risks of C3 blockade; and, eventually, a controlled trial in AQP4-IgG seropositive patients with relapse rate and disability as outcomes. None of that exists in the current publication. The claim that complement C3 inhibition with AMY-101 effectively ameliorates key pathological features of NMOSD in experimental models is consistent with the direction of the reported effects. The stronger claim, that AMY-101 is a therapy for NMOSD in patients, is not supported by these data, and the authors do not make it; they argue for further preclinical and clinical investigation. That is precisely the appropriate conclusion from evidence of this kind.
Related reading: Patent Argument Offers Rationale, Not Evidence, for Unapproved Peptides, UK Inquest Exposes Diagnostic Trap: Mounjaro Side Effects Mimic Bowel Perforation, Lilly Confirms Q1 2027 Retatrutide BLA After Phase 3 Wins, Tirzepatide tied to lower predicted 10-year CVD risk over 3 years.