Anti-CD38 Antibody CM313 Lifts Platelet Counts in 83% of Refractory ITP Patients

Immune thrombocytopenia (ITP) is an acquired autoimmune disorder in which the immune system mistakenly destroys platelets, leaving patients at serious bleeding risk. While glucocorticoids and intravenous immunoglobulin provide initial relief for many patients, relapse is common and a significant subset progresses to persistent or chronic disease requiring second-line therapies. Existing options — including splenectomy, rituximab, thrombopoietin receptor agonists, and fostamatinib — help some patients but leave others without durable control, motivating the search for novel mechanistic targets.

CD38 is a glycoprotein expressed abundantly on the plasma cells and plasmablasts that produce pathogenic anti-platelet autoantibodies. Anti-CD38 antibodies such as daratumumab have shown early promise in ITP and other autoantibody-driven diseases. CM313 is a new humanized anti-CD38 monoclonal antibody with a distinct complementarity-determining region sequence from daratumumab. Building on encouraging phase 1/2 single-arm results, investigators launched this multicentre, randomized, double-blind, placebo-controlled phase 2 trial at five hospitals in China between January and June 2024.

Forty-five patients were randomized 2:1 — 30 to CM313 (16 mg/kg intravenously weekly for 8 weeks) and 15 to placebo. Participants had heavily pre-treated disease: median prior treatment lines were 3–5, and 100% had previously received both glucocorticoids and thrombopoietin receptor agonists. Baseline platelet counts were critically low (mean 8×10⁹/L in the CM313 arm; 15×10⁹/L in placebo). The primary endpoint — overall response rate at week 8, defined as at least two consecutive platelet counts ≥30×10⁹/L with at least a doubling from baseline and no bleeding — was met by 83% (25/30) of CM313 patients versus 20% (3/15) on placebo, a difference of 63.3% (95% CI 33.7%–81.3%; P<0.001).

Response kinetics were particularly striking. The median time to achieving two consecutive platelet counts ≥50×10⁹/L was just one week in the CM313 group, compared to a median that was never reached in the placebo arm (P<0.001). This rapid onset suggests prompt depletion of autoantibody-producing plasma cells rather than a slow immunomodulatory effect. Importantly, responses were durable: the median cumulative duration of platelet counts ≥50×10⁹/L was 18 weeks in CM313-treated patients versus only 3 weeks in placebo recipients (P=0.004). Complete response rates (platelet count ≥100×10⁹/L) also favored CM313 substantially.

On the safety side, treatment-emergent adverse events occurred in 87% of CM313 patients and 80% of placebo patients — a modest difference given the placebo group's underlying disease severity. Infusion-related reactions and petechiae were the most common adverse events in the CM313 arm. One patient discontinued treatment due to an adverse event. Serious events were not dominant and the profile was broadly consistent with the known class effects of anti-CD38 antibodies. No unexpected safety signals emerged. This phase 2 trial was not powered for definitive efficacy conclusions, and larger confirmatory trials in diverse populations are needed, but the magnitude and speed of response represent a clinically meaningful advance for a population with few remaining options.