TPO-ras for persistent thrombocytopenia following CAR-T therapy in multiple myeloma: A multicenter clinical experience

Abstract Introduction: While chimeric antigen receptor T (CAR-T) cell therapies demonstrated significant efficacy in multiple myeloma (MM) patients, 20-40% of patients developed grade 3-4 persistent thrombocytopenia (PT), which might lead to heightened bleeding risk and compromised subsequent therapeutic interventions. Current management of PT post CAR-T therapy remains primarily supportive, with its underlying mechanisms and optimized treatment strategies requiring further investigation. Methods: In this multicenter retrospective study, we collected and analyzed data from MM patients who received CAR-T therapy from August 2022 to October 2024 in Wuhan Union Hospital and Tongji Hospital. PT required confirmed platelet levels <50×10⁹/L on ≥2 consecutive tests after day +30. Univariate and multivariate analyses were conducted between patients with PT (cohort 1) and without PT (cohort 2). Patients with PT were treated with thrombopoietin receptor agonists (TPO-RAs), including: hetrombopag (initial dose: 2.5 mg/day), avatrombopag (initial dose: 20 mg/day), eltrombopag (initial dose: 50 mg/day). Response was defined as transfusion independency along with resolution of platelets > 50×10⁹/L for three consecutive values on different days. TPO receptor agonists were tapered down when response was met. Results: From August 2022 to October 2024, A total cohort of 72 patients were included in this study, of whom 31 developed PT. Univariate analysis showed that R-ISS stage, bone marrow tumor burden, baseline cytopenia and inflammatory marker peaks were associated with PT; Logistic regression analysis showed that the peak level of ferritin was the independent risk factor for PT (P=0.0307). Twenty-two patients were treated with TPO-RAs (hetrombopag, n=7; eltrombopag, n=3; avatrombopag, n=12). Median time from CAR-T infusion to initiation of TPO receptor agonist was 16.5 (range 10-48) days. 36.4% (8/22) of patients achieved recovery from severe thrombocytopenia. During TPO-RAs treatment, platelet counts significantly elevated (P<0.0001), rising from a median of 14×10⁹/L (range: 5-49×10⁹/L) to 41×10⁹/L (range: 20-109×10⁹/L). Concurrently, hemoglobin levels showed a significant increase compared to pre-treatment (P=0.0027), whereas no statistically significant correlations were observed in leukocyte or neutrophil counts. Patients receiving TPO-RAs maintained higher platelet levels than those receiving supportive therapy alone (P=0.0032). Bone marrow biopsies performed post-treatment showed hypercellularity with megakaryocytic presence in 62.5% cases (10/16). No significant hepatotoxicity was observed. Conclusion: Hyperferritinemia post-CAR-T independently predicted PT, while TPO-RAs demonstrated both platelet-boosting effects and marrow recovery. Further prospective studies are warranted to validate these findings and optimize therapeutic protocols.