8 of 12 people with autoimmune disease got a single infusion of CAR-T cells and went into lasting remission

This is a preprint, not a peer-reviewed paper yet. Its authors (CD19-CAR T-cell therapy resets the EBV+ autoreactive B cell repertoire in autoimmune disease, posted on bioRxiv in late October 2026) report that sustained drug-free remission occurred in 8 of 12 patients treated with CD19 CAR-T cells, versus 0 of 15 treated with rituximab. The CAR-T group included 6 with systemic lupus, 3 with inflammatory myopathy and 3 with systemic sclerosis.

The tweet's main claim is a mechanism. The authors sequenced EBV-infected B cells and found that CAR-T cells depleted them, while rituximab did not. After rituximab, the surviving EBV-infected cells were class-switched, heavily mutated and autoreactive. In lab tests, the EBV-infected cells resisted anti-CD20 and anti-CD19 antibodies but were killed by CAR-T cells. That is a plausible link, but it is correlation in a small group, not proof that clearing these cells causes remission.

Two cautions. The groups are small, and the two arms were not randomized against each other, so differences in disease type, prior treatment and severity could matter. The tweet also says "years, no drugs" and "the gap comes down to one kind of cell", which goes beyond what a 27-patient study can show. Treat it as a hypothesis worth testing.