Isotype-specific heavy/light chain quantification identifies prognostic biomarkers after ARI0002h BCMA-directed CAR-T cell therapy in multiple myeloma

Abstract

Abstract Background ARI0002h (cesnicabtagene autoleucel) is a second-generation humanized BCMA-directed chimeric antigen receptor (CAR) T-cell therapy developed at Hospital Clínic de Barcelona. In a phase II trial, 60 patients with relapsed or refractory multiple myeloma (RRMM) treated with ARI0002h achieved a 95% overall response rate and a median progression-free survival of 20 months (Oliver-Caldés et al., 2023; ASCO, 2024). In CAR T-cell therapy, where both residual disease and immune reconstitution are clinically relevant, isotype-specific immunoglobulin profiling may offer complementary information. Heavy/light chain (HLC) assays allow sensitive quantification of involved clonal (iHLC) and uninvolved polyclonal (uHLC) immunoglobulins, providing higher specificity than serum M-protein or free light chains (FLC). Derived indices such as dHLC (iHLC – uHLC), rHLC (kappa/lambda), and iHLC/uHLC ratio reflect the clonal-to-polyclonal balance and may inform on both tumor burden and immune status. Although delayed humoral reconstitution after CAR T-cell therapy has been reported (Wang et al., 2021), the utility of isotype-specific HLC monitoring in this setting remains unexplored. This study aimed to characterize early changes in HLC biomarkers before and after CAR T-cell infusion and evaluate their association with treatment response and survival. Methods We retrospectively analyzed 27 RRMM patients treated with ARI0002h between 2020 and 2024, with serum samples available at baseline, day +28, day +100, and follow-up ≥18 months. Patients with non-IgG/IgA subtypes were excluded. HLC markers were quantified via Hevylite® (Binding Site, part of ThermoFisher Scientific). We assessed their association with response and survival using Spearman correlations, Kaplan–Meier analysis, Cox regression, and a maximally selected rank statistic (maxstat) threshold. Response was based on the International Myeloma Working Group (IMWG) 2014 criteria. Results Median age at infusion was 59 years (IQR 54–65); 44% were female. Most patients had IgG-kappa (48%) or IgG-lambda (26%) isotype. At a median follow-up of 19 months, progression-free survival was 18.9 months, while median overall survival was not reached, in line with the clinical trial results. Elevated iHLC levels were observed in 74.1% of patients at baseline, decreasing to 37% at day +28 and 3.7% at day +100. uHLC suppression was present in 93%, 89%, and 85% at each timepoint. Among 19 patients with available samples at later timepoints, 53% showed uHLC normalization, suggesting progressive immune recovery. At baseline, iHLC correlated with serum M-protein (ρ = 0.53, p = 0.004), but not with rFLC or bone marrow plasma cells (BMPC). At day +28, both iHLC and dHLC correlated with serum M-protein (ρ = 0.60 and ρ = 0.55), while dHLC also associated with rFLC ratio and BMPC (ρ = 0.47–0.52). No HLC marker correlated with extramedullary disease. In univariate Cox regression analysis, higher post-infusion dHLC levels were significantly associated with shorter progression-free survival (hazard ratio [HR] = 1.09 per g/L; 95% confidence interval [CI]: 1.00–1.19; p = 0.042) and overall survival (HR = 1.10 per g/L; 95% CI: 1.01–1.21; p = 0.029). Using a maximally selected rank statistic, a dHLC threshold of 11.0 g/L at day +28 was identified as optimal to stratify patients according to progression-free survival (10.3–10.4 vs. 26.6–27.0 months, p < 0.05). At day +28, lower dHLC levels were associated with more favorable responses per IMWG criteria. Moreover, none of the patients above this threshold achieved a complete response (CR) at day +100 or at 6 months. Among patients with dHLC ≤11.0 g/L at day +28 who had not yet achieved a very good partial response (VGPR), 73% subsequently improved to VGPR or better, including 27% who reached a stringent CR. Conclusions Isotype-specific HLC quantification after ARI0002h infusion offers a practical and accessible serum-based tool to assess both residual disease and early immune recovery. Elevated dHLC levels at day +28 identified patients with increased risk of early progression, reduced overall survival, and a lower likelihood of achieving deep responses. Recovery of uninvolved HLCs in a subset of patients suggests that humoral immune reconstitution may progressively occur over time. Further investigation will determine how this immunological recovery relates to CAR T-cell persistence and impacts long-term clinical outcomes.