Batoclimab cleared its Phase 3 endpoint in myasthenia gravis and was shelved anyway

What was tried

Immunovant Sciences GmbH ran IMVT-1401-3101 (NCT05403541), a Phase 3, randomized, quadruple blind study of batoclimab in adults with generalized myasthenia gravis. It lists 233 participants enrolled, start 27 June 2022, primary completion 10 January 2025, study completion 7 August 2026, and on 14 September 2026 moved to status TERMINATED with the posted reason "Sponsor Decision".

Period 1 randomized participants 1:1:1 to batoclimab 680 mg subcutaneously once weekly, batoclimab 340 mg subcutaneously once weekly, or matching placebo. The primary endpoint was change from baseline in the Myasthenia Gravis Activities of Daily Living (MG-ADL) score at Week 12 in acetylcholine receptor antibody seropositive participants. Period 2 re-randomized batoclimab recipients to 340 mg weekly, 340 mg every two weeks, or placebo. Entry needed Myasthenia Gravis Foundation of America class II, III or IVa disease, a Quantitative Myasthenia Gravis score of at least 11, and an MG-ADL score of at least 5. Participants, investigators, outcomes assessors and sponsor staff were masked.

Batoclimab is a fully human IgG1 monoclonal antibody against the neonatal Fc receptor, listed in ChEMBL as CHEMBL4650514, maximum phase 3, synonyms HBM-9161 and HL161BKN. The target gene is FCGRT (ENSG00000104870).

The biological hypothesis

FcRn binds IgG inside the acidified endosome and returns it to the circulation instead of routing it to lysosomal degradation, which is what gives IgG a half-life measured in weeks. In generalized myasthenia gravis most patients carry autoantibodies against the acetylcholine receptor at the neuromuscular junction, and disease activity tracks with that circulating IgG pool. Blocking FcRn short-circuits the loop, IgG catabolism accelerates, and pathogenic autoantibody titre falls along with total IgG.

The hypothesis is pharmacological rather than genetic, and the Open Targets evidence has that shape. The FCGRT to myasthenia gravis association scores 0.6066 overall, built from a clinical precedence score of 0.9742 and a Europe PMC literature score of 0.4706, with no genetic association evidence at all. Three FcRn blockers were already approved in the indication when this study read out: efgartigimod, rozanolixizumab and nipocalimab. The target was about as validated as a target gets without human genetics, and that validation is also why a fourth entrant had a hard argument to make.

What actually happened

The trial worked. Immunovant reported on 19 March 2025 an MG-ADL improvement from baseline at Week 12 of 5.6 points on 680 mg weekly and 4.7 points on 340 mg weekly, against 3.6 points on placebo, with mean IgG reductions of 74 percent and 64 percent. The company described all comparisons for the 680 mg arm at p of 0.001 or lower. In the same release it stated that it does not intend to seek regulatory approval for batoclimab in myasthenia gravis or in chronic inflammatory demyelinating polyneuropathy, and would instead use the data to accelerate IMVT-1402, a second generation anti-FcRn antibody.

A separate Phase 3 conducted in China (NCT05039190) had already reported consistent results. Sustained MG-ADL improvement occurred in 58.2 percent of batoclimab patients (39 of 67) against 31.3 percent on placebo (20 of 64), odds ratio 3.45, 95 percent CI 1.62 to 7.35, P = .001. Those investigators concluded that the clinical effect and the extent of IgG reduction were similar to what had been reported for efgartigimod and rozanolixizumab.

No results have been posted to ClinicalTrials.gov for NCT05403541 as of 17 September 2026, and openFDA holds no adverse event records for batoclimab.

Failure mechanism, best guess

This was not an efficacy failure, a safety halt or an enrollment collapse. Enrollment reached 233 and the primary endpoint was met.

The binding constraint is a mechanism linked laboratory liability. FcRn recycles serum albumin as well as IgG, so deep blockade lowers albumin, and falling albumin moves lipids. The published thyroid eye disease program shows what that costs. Kahaly and colleagues report that the randomized trial there was terminated because of an unanticipated increase in serum cholesterol, leaving 65 of a planned 77 patients analysable, with albumin reductions and lipid increases that reversed on discontinuation.

Set that against three approved competitors delivering comparable IgG reduction without an equivalent albumin signal, and the case for filing weakens even on a positive trial. Immunovant's own framing supports the reading, presenting the batoclimab results as proof of concept for IMVT-1402. The archetype is strategic_reprioritization, and the decision looks like portfolio arithmetic rather than a scientific surprise.

How to prevent this next time

Endpoint level data for NCT05403541 are not public, so no posterior, power estimate or subgroup effect can be computed from sourced inputs. The supportable levers are qualitative.

Treat a mechanism linked laboratory liability as a program level gate before the pivotal trial starts, not as a tolerability footnote inside it. The albumin and lipid behaviour of batoclimab was published in 2023, two years before this Phase 3 read out.

Run the base rate adjustment on competitive position, not on efficacy alone. When three approved drugs already share the mechanism, the probability that an undifferentiated fourth entrant reaches patients is far lower than the probability its trial succeeds, and programs routinely forecast only the second number.

Red team the positive trial with no filing scenario at the go decision. If the honest answer to "what do we do if this works" is "advance the next molecule instead", the pivotal study is a biology experiment carrying a Phase 3 price tag. Where a next generation molecule already sits in the portfolio, decide between the two on the differentiating biomarker, here albumin and LDL cholesterol, in early phase.

The single highest leverage change would have been to convert the published batoclimab albumin and lipid signal into a formal go/no-go gate on the myasthenia gravis Phase 3 before Period 1 randomization, instead of carrying it forward as a tolerability observation.

What this means for similar programs

Open Targets lists five clinical stage programs against FCGRT: orilanolimab at Phase 2, batoclimab at Phase 3, and efgartigimod, rozanolixizumab and nipocalimab at approval. The Claidex Mechanistic Risk Score for FCGRT now stands at 33 of 100, band yellow, with saturation contributing 9.48 of 15, genetic deficit 5.90 of 15, and phase burden 11.80 of 30 from a single Phase 3 event. The score is moderate because the target genuinely works. What it flags is crowding rather than biology.

For a fifth FcRn program the efficacy question is settled and the live ones are albumin behaviour, dosing interval and route. FAERS holds 3,855 reports naming VYVGART, of which 3,552 are coded serious and 344 carry a death outcome, led by myasthenia gravis, fatigue and myasthenic crisis. Those are spontaneous reports rather than a disproportionality analysis, but they mark where the class safety conversation sits.

Open questions

Will Immunovant post results for NCT05403541? Study completion is recorded as 7 August 2026 and the registry entry still shows none, so the Period 2 randomized withdrawal data have no public counterpart.

Does the albumin effect scale with IgG suppression depth, or can the two be separated? IMVT-1402 is the company's answer and no head to head comparison has been published.

Immunovant said it would defer a final decision on submissions until the thyroid eye disease Phase 3 reads out. Whether a lower dose changes the calculus where fewer competitors are approved remains open.

Sources

1. • ClinicalTrials.gov, NCT05403541, IMVT-1401-3101 record, accessed 17 September 2026. https://clinicaltrials.gov/study/NCT05403541 - Immunovant, "Immunovant Announces Positive Results for Batoclimab Myasthenia Gravis (MG) and Chronic Inflammatory Demyelinating Polyneuropathy (CIDP) Studies", 19 March 2025. https://www.immunovant.com/investors/news-events/press-releases/detail/71/immunovant-announces-positive-results-for-batoclimab - FierceBiotech, "Immunovant won't seek approval for autoimmune drug despite phase 3 win", 19 March 2025. https://www.fiercebiotech.com/biotech/immunovant-posts-phase-3-autoimmune-win-wont-seek-approval - Yan C, et al. Batoclimab vs Placebo for Generalized Myasthenia Gravis: A Randomized Clinical Trial. JAMA Neurology, 2024. https://- Kahaly GJ, et al. Proof-of-concept and Randomized, Placebo-controlled Trials of an FcRn Inhibitor, Batoclimab, for Thyroid Eye Disease. J Clin Endocrinol Metab, 2023. https://- Open Targets Platform, FCGRT (ENSG00000104870) association and clinical candidate data, accessed 17 September 2026. https://platform.opentargets.org/target/ENSG00000104870 - ChEMBL, batoclimab (CHEMBL4650514), accessed 17 September 2026. https://www.ebi.ac.uk/chembl/compound_report_card/CHEMBL4650514/ - openFDA drug adverse event API, queried 17 September 2026. https://api.fda.gov/drug/event.json.