CAR-T疗法在类风湿性关节炎试验中效果不一
CAR-T Therapy: Mixed Results in Rheumatoid Arthritis Trial
在类风湿性关节炎的I/II期临床试验中,6名患者中有3人对CAR-T细胞治疗有良好反应,但其他3人效果不佳,且一人出现疾病复发。研究显示个体反应差异显著。
— Six patients treated: three show strong responses, three not so much
by John Gever, Contributing Writer, MedPage Today
November 10, 2026 • 3 min read
ORLANDO -- Rheumatoid arthritis (RA) may not be as amenable to chimeric antigen receptor (CAR) T-cell therapy as has been the case in lupus and certain other autoimmune conditions, research presented here suggested.
Among six patients with very refractory RA treated in the first part of a phase I/II trial called COMPARE, three achieved low disease activity within 6 months after a single infusion of CAR T cells, and no severe adverse reactions were seen, according to David Simon, MD, of Charité University Hospital in Berlin, speaking at the American College of Rheumatology annual meeting.
That was the good news. The not-so-good was that the other three did not achieve low disease activity by standard criteria, and one of those who did experienced a disease flare during the final months of the yearlong study.
This was a more "heterogeneous" response pattern than has been seen with CAR-T therapies in lupus, Simon said. It may not be all that surprising, however, because RA itself is heterogeneous (not all patients develop autoantibodies, for example) and the role of B cells -- CAR-T therapy's main target -- is not as prominent as in lupus and some other autoimmune disorders.
One reason for trying this approach has been that the anti-B-cell drug rituximab (Rituxan) has been helpful in difficult RA cases. That suggests that, at least for some patients, B cells do help drive RA and that the "immune reset" concept with CAR-T treatment, which aims to wipe out B cells' autoimmune memory, could work effectively in RA.
Indeed, these initial results indicate that it's a realistic proposition -- it's just that the pathology differs enough between patients that the efficacy will be uneven.
The phase I portion of COMPARE recruited patients with heavily pretreated RA. Simon said the six who were enrolled had previously tried a mean of 6.5 targeted therapies, in addition to conventional drugs such as methotrexate, without achieving adequate disease control. Disease Activity Score values across 28 joints (DAS28) ranged from 4.6 to 6.8 among the six patients, who were in their early 30s to late 60s. Disease duration spanned 4.5 to 24 years. Three were women and three were men.
As a phase I study, safety was a top priority to examine. Simon said all patients experienced cytokine release syndrome, which is nearly universal with CAR T-cell therapy in oncologic as well as rheumatologic use. None, however, developed more severe reactions such as immune effector cell-associated neurotoxicity syndrome (ICANS) or a condition called LICATS (local immune effector cell-associated toxicity syndrome) that resembles ICANS but affects other organ systems besides the nervous system. As well, no infections more severe than grade 2 occurred; the only adverse effect rated as serious was elevated liver enzymes in one patient.
All patients showed deep depletion of B cells, as expected, almost immediately after CAR T-cell infusions. But other effects were more variable. Four of the six patients showed abolition of autoantibodies, and as noted above, only three achieved DAS28 scores below 2.8, the threshold for low disease activity; none got to zero, whereas full remission has been common with CAR-T therapy in lupus.
Simon spent several minutes discussing the patient who flared. This was a 42-year-old woman with a DAS28 score of 5.5 at baseline. Crucially, and unlike the other five patients, analysis of CAR T-cell counts in circulation showed very poor expansion in the first weeks after the infusion -- short of 100 cells per μL at the maximum, compared with approximately 150-400 cells per μL in the others. B-cell reconstitution was already well underway at week 8, whereas it was week 16 for the next earliest, and still later for others. Analysis of reconstituted B cells indicated that autoreactivity had not been abolished, thus paving the way for relapse that began sometime after week 36.
"Whether the observed immune remodeling translates into durable drug-free disease control remains uncertain," Simon concluded. "Follow-up is limited to approximately 1 year, and re-emergence of autoreactive B-cell populations in one relapsing patient suggests that immune reconstitution may influence long-term outcomes."
Ten more patients will be recruited for COMPARE's phase II portion, which, as the name suggests, will include a control group treated with rituximab.
来源:MedPage Today · medpagetoday.com