Kyverna closed CD19 CAR T in lupus nephritis without an efficacy readout

What was tried

Kyverna Therapeutics ran KYSA-1 (NCT05938725), an open-label, non-randomized, multicenter Phase 1/2 study of KYV-101, an autologous fully-human anti-CD19 chimeric antigen receptor T-cell therapy now named mivocabtagene autoleucel, in adults with refractory lupus nephritis. Entry required systemic lupus erythematosus by 2019 EULAR/ACR criteria, biopsy-proven proliferative Class III or IV nephritis by 2018 ISN/RPS criteria, and serologic activity (antinuclear antibody titer of 1:80 or above, anti-dsDNA at 30 IU/mL or above by ELISA, or anti-Smith). Participants received standard lymphodepletion followed by a single CAR T infusion.

The Phase 1 primary endpoints were adverse events, laboratory abnormalities and dose-limiting toxicities over two years. The Phase 2 primary endpoint was complete renal response rate at 52 weeks. The trial opened on 28 April 2023, recorded an actual enrollment of 6, and closed on 29 May 2026. ClinicalTrials.gov gives the reason as "Study discontinued due to sponsor decision."

Two sibling trials closed on the same posting date with the same stated reason. KYSA-3 (NCT06342960), also in refractory lupus nephritis, enrolled 2 and closed on 22 December 2025. KYSA-5 (NCT06400303), in diffuse cutaneous systemic sclerosis, enrolled 3 and closed on 16 June 2026.

The biological hypothesis

CD19 (ENSG00000177455) is a B-cell surface antigen and, per Open Targets, an approved-drug-tractable antibody target. The autoimmune CAR T thesis is that deep, tissue-penetrating B-cell depletion achieves something antibody-mediated depletion does not: elimination of autoreactive B cells in lymphoid tissue, followed by reconstitution from naive precursors that reconstructs a non-autoreactive repertoire. The clinical case originates with a single lupus patient treated in Erlangen (Mackensen et al., Nat Med, 2022) and the follow-up case series reporting drug-free remission across lupus, myositis and systemic sclerosis (Müller et al., N Engl J Med, 2024). Reviews have consolidated the mechanism and its open questions (Nat Rev Immunol, 2024,, and Nat Rev Rheumatol, 2024).

Lupus nephritis was the natural first indication. It carries high unmet need, an objective and biopsy-anchored endpoint, and the largest share of the published autoimmune CAR T experience. Open Targets scores the CD19 to lupus nephritis association at 0.0915 for MONDO_0005556, composed of a clinical datatype score of 0.1444 and a literature datatype score of 0.1225, with no genetic evidence datatype returned.

What actually happened

The trials were not stopped for toxicity or for a failed endpoint. No safety finding is posted on any of the three records, and no efficacy result from KYSA-1 has been released.

The sequence is visible in the sponsor's own disclosures. At the J.P. Morgan Healthcare Conference on 12 January 2026, Kyverna listed lupus nephritis among indications in development and stated it expected to report Phase 1 lupus nephritis data during 2026. Its 2026 priorities were a stiff person syndrome BLA filing in the first half, launch readiness by year end, and execution of the Phase 3 generalized myasthenia gravis trial. By the second-quarter update on 11 August 2026, the pipeline was described around three RMAT-designated indications: stiff person syndrome, generalized myasthenia gravis and non-active secondary progressive multiple sclerosis. Lupus nephritis and systemic sclerosis are absent from that framing. Cash was $199.4millionat30June2026,downfromapproximately$279 million at 31 December 2025. The three trial records were posted as terminated on 31 August 2026.

Total enrollment across the three closed studies was 11 participants over roughly three and a half years of open trials.

Failure mechanism, best guess

This is a strategic reprioritization, and the mechanism is enrollment economics rather than biology.

The proximate constraint is accrual. KYSA-1 enrolled 6 patients in 37 months, KYSA-3 enrolled 2 in 36 months, and KYSA-5 enrolled 3 in 22 months. Autologous CAR T in autoimmune disease competes for a narrow patient pool: patients must be refractory enough to justify lymphodepletion and apheresis, well enough to tolerate it, and willing to accept an oncology-grade procedure for a non-malignant disease that has several approved biologics. Lupus nephritis is the indication where that competition is stiffest, because belimumab, voclosporin and anti-CD20 agents already occupy the refractory line.

The distal constraint is regulatory leverage. The three retained indications share a feature the two dropped ones lack: RMAT designation and, in stiff person syndrome, a near-term filing path in a disease with no approved therapy. A company with $199.4 million and a BLA to file allocates apheresis slots, manufacturing runs and clinical operations to the indication that reaches market first. Lupus nephritis, with a 52-week complete renal response endpoint and a crowded comparator field, is the slowest of the set.

Nothing here indicates that CD19 CAR T does not work in lupus nephritis. It indicates that a single-product cell therapy company could not run five indications at once, and that the two it dropped were the two furthest from a filing.

How to prevent this next time

No endpoint-level data was released, so no quantitative reanalysis is possible and none is attempted. Three qualitative levers apply.

Feasibility modeling before indication launch is the first. An accrual rate of roughly 0.16 patients per month, computed from 6 participants over 37 months of open enrollment in KYSA-1, does not support a Phase 2 readout on any planned timeline. A prospective site-level feasibility estimate against the refractory lupus nephritis pool, net of competing trials, would have surfaced that before three parallel protocols were opened.

Portfolio sequencing is the second. Running lupus nephritis, systemic sclerosis, myasthenia gravis, multiple sclerosis and stiff person syndrome concurrently on one autologous manufacturing chain spread a fixed slot capacity across indications with very different regulatory returns. Sequencing by time-to-filing rather than by scientific appeal would have concentrated that capacity earlier.

Base-rate adjustment is the third. The published autoimmune CAR T experience is dominated by small academic series in highly selected patients. Using those rates to size an industry-sponsored program understates both screening burden and withdrawal rate in a commercial trial network.

The single highest leverage change would have been to model refractory lupus nephritis accrual against competing trials and available apheresis capacity before opening a second and third parallel protocol in the same indication.

What this means for similar programs

The Claidex Mechanism Risk Score for CD19 is 58 out of 100, in the orange band, up from 38 after this insert. The increase is driven by saturation and by a second documented failure rather than by any new safety or efficacy finding. CD19 now carries two Claidex-documented failures in autoimmune indications: the LY3541860 interim futility in rheumatoid arthritis, an efficacy failure, and this reprioritization.

Those two failures are not the same signal and should not be read as one. The rheumatoid arthritis result tested a non-depleting antibody and failed on effect. This trial tested deep depletion and never produced an effect estimate. The operational lesson transfers further than the biological one: any autologous cell therapy entering a crowded autoimmune indication should treat apheresis slot capacity and refractory-population size as primary design constraints.

Open questions

What did the 6 KYSA-1 participants show at 52 weeks, and will those data be published? Was accrual, cost, or an internal efficacy read the binding constraint? Does Kyverna retain lupus nephritis as a post-approval expansion, or is the indication released? Do in vivo CAR approaches (Science, 2025) remove the capacity constraint that closed these trials?

Sources

1. ClinicalTrials.gov, NCT05938725, KYSA-1, Phase 1/2, terminated. https://clinicaltrials.gov/study/NCT05938725.

2. ClinicalTrials.gov, NCT06342960, KYSA-3, terminated. https://clinicaltrials.gov/study/NCT06342960.

3. ClinicalTrials.gov, NCT06400303, KYSA-5, terminated. https://clinicaltrials.gov/study/NCT06400303.

4. Kyverna Therapeutics, "Provides Corporate Update and Outlines 2026 Strategic Priorities at the J.P. Morgan Healthcare Conference," 12 January 2026.

5. Kyverna Therapeutics, "Reports Pipeline Progress and Second Quarter 2026 Financial Results," 11 August 2026.

6. Open Targets Platform API v4, target ENSG00000177455, disease MONDO_0005556, accessed 2 September 2026.

7. Mackensen A, et al. Anti-CD19 CAR T cell therapy for refractory systemic lupus erythematosus. Nat Med. 2022.

8. Müller F, et al. CD19 CAR T-cell therapy in autoimmune disease, a case series with follow-up. N Engl J Med. 2024.

9. Schett G, et al. CAR T-cell therapy in autoimmune diseases. Lancet. 2023.

10. Chung JB, et al. Chimeric antigen receptor T cell therapy for autoimmune disease. Nat Rev Immunol. 2024.

11. Schett G, et al. Advancements and challenges in CAR T cell therapy in autoimmune diseases. Nat Rev Rheumatol. 2024.

12. Hunter TL, et al. In vivo CAR T cell generation to treat cancer and autoimmune disease. Science. 2025.