Science1 distinct publisher3 min readPublished
The Berlin phase 1 trial was built to measure cytokine release syndrome rather than remission, and its six-patient denominator carries no comparator, so the ten-patient second phase against an approved B-cell drug is the one that settles anything.
The Scientist · Science desk
Compiled by The ScientistSomething wrong?How this is made
The trial is testing whether CD19 CAR-T cells can rebuild the B-cell system, not just suppress it more strongly. The Charite group's argument is that CD19 CAR-T cells can hit the B cells keeping the disease persistent, clear the pathological B-cell memory, and let the B-cell system start fresh [20]. The corroborating measurement is the sharp decline in the autoantibodies characteristic of RA after infusion [8], which is what you would expect if the compartment producing them had been emptied. But that decline shows depletion, not durability: a titre can fall and the memory that regenerates it can come back.
Six is the denominator. The endpoint structure matters more than the remission count. The primary endpoints were the incidence and severity of cytokine release syndrome, ICANS and adverse events in the first four weeks [7], which makes every efficacy figure in this trial a secondary or exploratory reading [17]. There was no comparator arm; comparison arrives only in the second phase [19]. Three of six, half the cohort, were off all RA medication at last follow-up [6][14]. Read alongside the single patient whose disease returned after an initial medication-free period [10], that implies at least four of the six reached a drug-free interval and one did not keep it [15]. Follow-up ran 36 to 52 weeks [5], roughly eight months to a year [18], against a disease that otherwise requires lifelong medication [21].
The second phase is the more informative experiment. Ten additional patients, and the comparator is a drug already approved for RA that also depletes B cells [12]. That controls the obvious confound, because anything that removes B cells will help RA for a while; the live question is whether a one-time reset beats continuing depletion on depth and duration, which is precisely what the investigators say they are testing [12]. Sixteen patients across both phases [16] are enough to test whether "reset" is a mechanism or a label, not enough to tell anyone who should get this.
What rheumatology would inherit, if the effect holds up, is an oncology workflow. The product is autologous [3], so every dose is a manufacturing run for one person. Patients stopped all disease-modifying antirheumatic drugs and received standard lymphodepletion before the infusion [5]. All six developed CRS, temporary and mild to moderate, reported as readily manageable, with no severe neurological complications and rare infections [9]. That record was produced at Charite's Hematology Early Clinical Trial Unit, whose medical director, Marie Luise Huetter-Kroenke, described it [9]. How a unit that sees this kind of CRS only a few times a year, rather than most weeks, would manage it is a separate question this record cannot answer. The washout carries its own exposure: a patient off all DMARDs and through lymphodepletion needs somewhere to go if the reset does not take, and there is no long-term experience with this therapy to draw on [5][11].
Gerhard Kroenke, who leads the joint clinical rheumatology group at Charite and the German Rheumatology Research Center [13], keeps the framing narrow, and it is worth keeping it there: selected patients who have not responded adequately to existing treatment, in whom it may in future be possible to reset immune memory rather than suppress inflammation continuously [22].
Ranked by verification strength, evidence, and original report placement.
Six patients with severe rheumatoid arthritis received CD19 CAR T-cell therapy at Charite-Universitaetsmedizin Berlin, described by the publisher as the world's first clinical trial of the approach in RA.
The work is published in Nature Medicine as "CD19 CAR-T cell therapy for treatment-refractory seropositive rheumatoid arthritis: a Phase 1 trial."
The COMPARE study evaluates the safety and efficacy of mivocabtagene autoleucel (miv-cel), an autologous fully human CD19 CAR T cell therapy, in rheumatoid arthritis.
The cohort was three women and three men aged 31 to 69 who had received up to eight targeted or biologic therapies over the previous ten years, none of which was sufficiently effective.
All six patients had treatment-refractory, anti-citrullinated protein antibody (ACPA)-positive RA and were followed for 36 to 52 weeks after a single infusion of miv-cel, given after stopping all disease-modifying antirheumatic drug treatment and after standard lymphodepletion therapy.
Gerhard Kroenke reports that disease activity decreased markedly in all six patients and that during follow-up of up to one year, three patients were in sustained remission without any medication for rheumatoid arthritis.
Distinct publishers with included, body-backed reporting in this cluster.
1 article · August 28, 2026
Follow any of these and your For You feed starts watching them — no settings page required.
science
Columbia says depression stalls new neurons. The subtype it implies has no test yet2 distinct publishers
science
Oxford's Zostavax-to-Shingrix switch closes the healthy-vaccinee loophole, and opens another2 distinct publishers
build
First-turn evals test the safest part of your product, a 90,000-exchange audit finds1 distinct publisher
science
A spreading PX1 cluster in Uganda dulls both drugs in Africa's front-line malaria pill1 distinct publisher
Evidence-backed comparisons of source perspectives and observed adoption signals. Read the methodology
Which Builder, Operator, and Investor concerns the observed source mix emphasized—not a truth score.
Evidence, demonstrated adoption, hype gap, incentives, and confidence are assessed independently, each on its own current evidence. How these are measured.
One outlet, one institution's account
The six patients, the 36-to-52-week window, the three drug-free remissions — all of it reaches a reader through a single GEN write-up assembled from Charité's announcement and quotes from the two clinicians who ran the trial. What lifts this above the usual single-source health item is that the Nature Medicine paper is named in full, so the primary record is locatable. What holds it down is that nobody outside Berlin has read that record back to us, and the efficacy figures arrive as sentences rather than as data.
Six patients, one hospital
Six people at one Berlin hospital received an investigational cell product that no regulator has approved for arthritis. That is the entire footprint. The only forward commitment on the record is ten more patients in a planned second phase, which has produced nothing yet. Anyone reading this as a treatment option rather than an experiment is reading past the denominator.
Remission headline over a safety trial
GEN's title says CAR T-cell therapy puts severe RA into remission. The trial's own primary endpoints, quoted three paragraphs down, were cytokine release syndrome, neurotoxicity and adverse events in the first four weeks — which makes every remission figure a secondary or exploratory observation in an uncontrolled six-patient series where one patient relapsed. The gap is one of framing rather than concealment: the relapse, the incomplete responders and the experimental status are all in the piece. They just sit downstream of a promise the title already made.
Both voices belong to the trial team
The only two people quoted are Charité investigators — Krönke, who leads the joint Charité/DRFZ rheumatology group that ran the study, and Hütter-Krönke, who directs the hematology unit that delivered it — describing a result their own institution announced as a world first. No outside clinician appears. And in a story built around a proprietary named cell product, who funds COMPARE and who owns or supplies mivocabtagene autoleucel never comes up at all; that omission is the sharper one.
Plausible, checkable, unreplicated
Middling, and for structural reasons rather than doubts about the science. The mechanism is coherent, the peer-reviewed paper is identified, and the reporting keeps its own caveats visible instead of burying them. Against that: one publisher, no independent clinician, a denominator of six, and a durability question the story openly says cannot yet be answered.