Clonal Hematopoiesis Underlies Prolonged Cytopenias after Anti-BCMA CAR-T Therapy

Le résumé fourni par la source

Background While chimeric antigen receptor (CAR) T-cells are effective therapies for relapsed/refractory multiple myeloma (RRMM), ∼40% of patients experience prolonged cytopenias of unclear origin, with clonal hematopoiesis (CH) postulated as a contributor. Traditionally, CH of indeterminant potential is defined by the presence of somatic mutations in key genes exceeding a variant allele frequency (VAF) of 2%. However, ultra-sensitive methods revealed a high prevalence of CH mutations with lower VAF in aging adults. To explore the contribution of CH as a function of VAF to cytopenias , we first characterized mutations in canonical CH genes in newly diagnosed (NDMM) and RRMM patients. We then focused on patients receiving CAR-T cells (idecabtagene vicleucel, ide-cel) and associated with cytopenias . Methods & Results We performed targeted DNA sequencing of the germline (PBMC) using CAPP-Seq (Cancer Personalized Profiling by Deep Sequencing) from 72 patients with MM, including 17 NDMM and 55 RRMM. CH mutations in 17 canonical genes were detected at VAF≥2% in 5.9% of NDMM vs 20% of RRMM patients (p=0.27). When the VAF threshold was lowered to >0.3%, prevalence of CH rose, with 65% of NDMM vs 56% of RRMM (p=0.56). Among 55 RRMM patients, 37 received CAR-T (ide-cel, n=29; cilta-cel, n=8; Fig 1a). We focused on ide-cel cohort for further analysis due to sample size. We compared baseline characteristics between those with CH mutations with VAF>0.3% (n=16) vs not (n=13). There were no significant differences between cohorts, including age (median 67 vs 65, p =0.47), number of prior lines of therapy (6 vs 6, p =0.58), and pre-treatment tumor burden assessed by circulating tumor DNA (2.2 vs 1.4 log hGE/mL, p =0.27). We then assessed the efficacy and toxicity of CAR-T between these cohorts. There was no significant difference in CAR-T peak expansion level (1.4 vs 1.2 log/uL, p =0.79), the rate of cytokine release syndrome (CRS; 92% vs 82%, p =0.56), immune effector cell-associated neurotoxicity syndrome (ICANS; 14% vs 18%, p =1), or progression-free survival (PFS; median 8.8 vs 6.3 months, p =0.92). We next assessed cytopenias using CTCAEv5 grading prior to lymphodepletion chemotherapy (pre-LD) and at 3 months. We categorized patients into having no detectable CH mutations (CH neg ) vs CH mutations with VAF 0.3-2% (CH 0.3-2 ) vs VAF>2% (CH 2 ). Among CH neg , 7.7% of patients experienced grade (G)≥3 cytopenias of any cell lineages pre-LD compared to 38% among CH 0.3-2 (p=0.16) and 33% among CH 2 (p=0.35). At 3 months post CAR-T therapy, 0% experienced G≥3 cytopenia among CH neg compared to 42% among CH 0.3-2 (p=0.039) and 33% among CH 2 (p=0.23) (Fig 1b). Conclusions CH mutations are pervasive in MM patients at initial diagnosis and relapse. Patients with CH at low VAF are associated with prolonged cytopenias at 3 months following CAR-T. The presence of CH mutations did not affect CAR-T expansion, toxicity or efficacy.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.