Cartesian Therapeutics has claimed a phase 2b win for its BCMA-directed CAR-T therapy in generalized myasthenia gravis (gMG). But the biotech’s victory came on a primary endpoint it changed in May and in a population that excluded the 25% of participants treated at community clinics.
Investigators randomized 36 patients to receive Descartes-08 or placebo. Cartesian restricted its primary endpoint analysis to 26 people enrolled at academic medical centers who received at least one dose and completed at least one post-baseline assessment of MG Composite (MGC) score. The decision removed nine patients treated at community clinics from the primary analysis. Another subject dropped out.
The biotech switched its primary endpoint from the MG-ADL symptom scale to MGC in May. Argnex, Johnson & Johnson and UCB looked at MG-ADL for the primary endpoints of their pivotal gMG studies.
In Cartesian’s primary efficacy population, 10 out of the 14 recipients of Descartes-08 had a five-point or greater improvement in MGC after three months. Three of the 12 participants on placebo had such an improvement. The difference in response rates, 71% versus 25%, was enough for the clinical trial to meet the MGC endpoint in the modified intent-to-treat efficacy population.
The biotech said a three-point change in MGC is generally seen as clinically meaningful. Cartesian used the “more stringent endpoint of five points” based on the responses seen in an earlier study. The press release lacks details of how Descartes-08 compared to placebo using a three-point response threshold.
Cartesian did share results from a per-protocol population that included 31 patients. In that population, the MGC response rates in the Descartes-08 and placebo arms were 69% and 33%, respectively, resulting in a p-value of 0.048.
The biotech assessed change from baseline in MG-ADL, the original primary objective, as a secondary endpoint. Recipients of Descartes-08 had a mean 5.6-point improvement on the MG-ADL symptom scale after three months. MG-ADL scores in the placebo arm kept pace with the treatment cohort over the first month but then reversed. The placebo improvement was less than two points after three months.
Cartesian reported a significant imbalance in the proportion of people with thymoma, a type of cancer associated with gMG. While 42% of people on placebo had thymoma, none of their counterparts in the Descartes-08 arm had the cancer. There were numeric imbalances on other variables, with more people on placebo having received complement inhibitors or FcRN drugs and had their thymus removed.
The readout is the first big milestone for Cartesian since it joined the Nasdaq through a reverse merger with Selecta Biosciences. Cartesian emerged from the deal with $110 million to test its belief that mRNA can make CAR-T viable in autoimmune disorders by reducing the toxicity associated with DNA cell therapies and by eliminating the need for preconditioning chemotherapy.
Cartesian ended March with almost $105 million, a sum it said would support operations into the second half of 2026. The biotech shared plans to raise $130 million through a private investment in public equity alongside news of the phase 2b results.