CAR T-Cell Trial in Rheumatoid Arthritis Starts Off Strong

LONDON — Most patients with super-refractory, seropositive rheumatoid arthritis (RA) responded well to chimeric antigen receptor (CAR) T-cell therapy in the first part of an ongoing phase I/II trial, a researcher reported here.

Of six patients treated in the study’s phase I portion, all showed reductions in swollen and tender joint counts. Four saw important decreases in overall RA severity as measured with the 28-joint Disease Activity Score (DAS28), according to Fredrik Albach, MD, of Charité-Universitätsmedizin Berlin in Germany, and colleagues.

Three of those patients achieved clinical remission when followed for at least 36 weeks, he added, and autoantibody levels dropped markedly in all six.

Safety data were consistent with those seen in other studies of CAR T-cell therapy in rheumatologic diseases, Albach also told attendees at the European Alliance of Associations for Rheumatology (EULAR) annual meeting.

Tremendously exciting results from those studies have been reported in the past few years, but almost all have involved diseases such as lupus and inflammatory immune myopathies that are primarily driven by B cells, the target for current CAR T-cell therapies.

RA, on the other hand, hasn’t been considered a B-cell disease. But at least in seropositive RA, in which patients develop anti-citrullinated protein antibodies (ACPA) and/or rheumatoid factor (RF), autoantibodies found in most but not all RA cases, B cells are involved. That has led researchers to consider CAR-T treatments in patients with the most recalcitrant disease.

The new study was designed to test the therapy initially in six patients with heavily pretreated RA, mainly to assess safety (phase I), to be followed by a small randomized trial in 10 patients (phase II), half of whom would receive placebo on top of background therapy. This second phase is now ongoing, with Albach reporting results just from phase I.

Three of the initial six participants were men — not entirely surprising even though women comprise 60%-75% of most RA patient samples, as the disease is often more severe in men. Patients in this cohort had previously received three to eight biologic or other targeted agents yet still showed substantial disease activity, with DAS28 scores ranging from 4.7 to 6.4, reflecting moderate to high severity, and with disease durations up to 24 years. Mean age was 52 years but varied considerably. Eligibility included positive tests for ACPA and RF along with ultrasound-confirmed synovitis.

Autologous CAR T-cells were manufactured and administered in the usual way, with a lymphodepletion regimen to prepare for the engineered cells, and previous disease-modifying drugs were stopped.

By week 36, five of the six had at least achieved ACR20 responses (20% reduction in symptom severity by American College of Rheumatology criteria), and four had ACR70 responses (70% reduction).

Albach said his group could not identify any patient- or treatment-related factors at baseline that differentiated the two patients with minor responses; these patients had a smaller decrease in ACPA and RF levels compared with those with stronger clinical improvement.

Ominously, perhaps, one patient followed for 52 weeks who initially had RF IgA levels go to zero showed a partial rebound after week 36, possibly flagging a future loss of effect.

Adverse events were common but none were severe. No cases of the much-feared condition known as immune effector-cell associated neurotoxicity syndrome (ICANS) developed. Of note, CAR T-cell therapy had originated in oncology as a lymphoma treatment, where ICANS has sometimes been life-threatening.

All six patients did develop cytokine release syndrome, a related but milder condition, which was treated successfully with interleukin-6 inhibitors and dexamethasone. Three patients developed neutropenia, as expected from the lymphodepletion regimen, but no severe infections followed. Similarly, gamma globulin levels dropped somewhat in five patients, again with no adverse clinical consequences.

Limitations to the study, as reported at EULAR, included the small number of patients and the lack of a control group.