Applying the Novel EHA/EBMT Grading System for Icaht Following CAR-T Therapy: Comparative Incidence across Disease Entities and Association with Infections and Mortality

Abstract

Background: Hematological toxicity is a frequent side effect of CAR-T therapy and predisposes for severe infectious complications. Recently, the EHA/EBMT consensus grading for Immune Effector Cell Associated HematoToxicity (ICAHT) was developed ( Rejeski et al. Blood 2023). In contrast to existing grading systems such as the CTCAE, the novel ICAHT grading integrates both the depth and duration of neutropenia. We hypothesized that ICAHT severity is associated with key clinical sequelae such as transfusion requirements, hospitalization, severe infections, and adverse treatment outcomes. Methods: We applied the novel grading of early ICAHT (day 0-30) to a multicenter retrospective real-world cohort of 549 patients treated with BCMA- or CD19-directed CAR T-cells for relapsed/refractory B-cell malignancies (334 large B-cell lymphoma [LBCL], 112 multiple myeloma [MM], 103 mantle cell lymphoma [MCL]) across twelve international sites (Germany, France, Spain, USA). Hematotoxicity endpoints included the total cumulative duration of severe neutropenia (ANC <500/µL, day 0-60) and phenotypes of neutrophil recovery ( Rejeski et al. Blood 2021). Infections were studied until day +90 after CAR-T and were graded on a 5-grade scale as mild, moderate, severe, life-threatening, or fatal. Non-relapse mortality (NRM) was defined as death following CAR-T without evidence of relapse or progression. Kaplan-Meier estimates of progression-free (PFS) and overall survival (OS) were compared using log-rank test. Results: Our study cohort included 424 patients without severe ICAHT (absent or grade 1-2) and 125 patients with severe ICAHT (grade 3-4). The distribution of mild (1°), moderate (2°), severe (3°) and life-threatening (4°) ICAHT was 36%, 29%, 17%, and 5%. Seventy patients (13%) did not develop ICAHT of any grade. When comparing ICAHT grades across disease entities, we noted a numeric trend towards increased severe ICAHT in MCL compared to LBCL or MM patients (Grade ≥3: 28% vs. 23% vs. 15%). Patients that subsequently developed severe ICAHT displayed elevated serum inflammatory markers and pronounced cytopenia at baseline - reflected by increased CAR-HEMATOTOX scores (median 3 vs. 1, p<0.0001). On multivariate binary logistic regression, co-stimulatory domain (CD28z), low platelet count, low ANC, low hemoglobin, increased serum ferritin, and the presence of bone marrow infiltration were independent risk factors for severe ICAHT. We noted a strong positive correlation between ICAHT severity and the cumulative duration of severe neutropenia (r=0.92, p<0.0001). The aplastic phenotype of neutrophil recovery was enriched in patients with grade 3 (73%) and especially grade 4 ICAHT (100%). Severe ICAHT was also associated with coincident multilineage cytopenias, resulting in increased transfusion requirements for both platelets (85% vs. 21%, p<0.0001) and red blood cells (86% vs. 41%, p<0.0001). In terms of management of hematotoxicity, the severe ICAHT group more commonly received G-CSF (72% vs. 44%, p=0.01), TPO mimetics (12% vs. 2%, p<0.0001) and hematopoietic stem cell boosts (6% vs. 0.2%, p<0.0001). Importantly, patients with severe ICAHT displayed a significantly increased rate of severe infections (49% vs. 13%, p<0.0001, Fig. 1a) and severe bacterial infections (36% vs. 8%, p<0.0001). This translated into a markedly increased 1-year NRM rate (14% vs. 4.5%, logrank p<0.0001, Fig. 1b), which was primarily attributed to fatal infections of fungal or bacterial origin. Of note, the ICAHT grading showed superior discrimination for severe infections (C-index 0.73, 95%CI: 0.67-0.78) compared to CTCAE grading (C-index 0.55, n.s.). Prolonged hospitalization was noted in case of severe ICAHT (median 21 vs. 16 days, p<0.0001) and intensive care admission was more common (23% vs. 4%, p<0.001). Finally, we observed inferior PFS (1-year PFS 35% vs. 51%, median 4 vs. 14 months, p<0.0001) and inferior OS (1-year OS 52% vs. 73%, median 12 months vs. not-reached, p<0.0001) in the patients that developed severe ICAHT. Conclusions: These data highlight the clinical relevance of the novel EHA/EBMT ICAHT grading system. Importantly, ICAHT severity was closely associated with transfusion dependence, increased infection incidence, and adverse treatment outcomes with high NRM. Taken together, these findings argue for reporting ICAHT grades in clinical trials evaluating established and novel CAR constructs.