Prognostic Impact of Clinical Factors and Cell Composition of Apheresis Samples on BCMA CAR-T Cell Therapy for Relapsed or Refractory Multiple Myeloma

Background: Despite the development of novel treatments, patients with triple-class-exposed (TCE) relapsed or refractory multiple myeloma (RRMM) have particularly poor outcomes. Idecabtagene vicleucel (Ide-cel), a BCMA-targeting CAR-T cell therapy, has dramatically improved the outcome of TCE RRMM patients, as shown by the results of the KarMMa-3 trials. However, most patients achieving a favorable response to Ide-cel therapy eventually relapse, but predictive and prognostic factors for clinical outcomes are poorly established. Here, we conducted a comprehensive analysis of baseline clinical characteristics and cellular composition of apheresis samples to predict the outcome of Ide-cel therapy.

Methods: Patients with TCE RRMM (n=28) underwent lymphocyte apheresis (AP) for manufacturing Ide-cel between July 2022 and May 2024. Of these, 3 could not receive Ide-cel infusion due to disease progression and 2 due to manufacturing failure or out of specification. Of the 23 patients who did receive Ide-cel therapy, 22 had AP samples preserved for research use which were analyzed in this study. Immune markers related to T cell populations and maturation were evaluated by flow cytometry. Briefly, AP samples were incubated in Fc blocking reagents after thawing, stained with antibodies to CD3, CD4, CD8, CD45RO, CCR7, together with live/dead staining at 4℃ for 15 minutes. T cell maturity was defined according to staining for CCR7 and CD45RO on CD4 or CD8 T cells by gating control.

Results: Baseline characteristics of the 22 patients were as follows: median age at apheresis was 64 years (range: 46-77), median time from diagnosis to apheresis was 65 months (range: 9-296), median number of prior lines of treatment was 6 (range: 3-13), including all patients with TCE, 6 with penta-refractory disease (27%) and 4 (18%) with prior bispecific antibody treatment (two targeting GPRC5D and two BCMA). Before AP, 10 (45%) and 9 patients (41%) received IMiD- and/or PI-based chemotherapy, respectively. Fourteen patients (64%) had cytogenetic abnormalities associated with poor prognosis. Median follow-up was 172 days (range: 27-545) and median PFS was 118 days (range: 20-543). Patients were divided into two groups with short PFS (n=8) or long PFS (n=14), defined by a cut-off value of 180 days for PFS events. The short PFS group had more poor cytogenetic abnormalities (100% vs 52.9%, p=0.018), had a shorter time from diagnosis (38 months vs 95 months, p<0.001), and had received more alkylating agents in prior therapies (50% vs 7%, p=0.039). Patients with a history of bispecific antibody treatment during the previous 3 months or treatment including alkylating agents before apheresis had a worse PFS (p=0.001, p=0.004, respectively). The short PFS group exhibited a lower proportion of CD8⁺ early memory (CCR7⁺CD45RO^-, 0.43% vs 1.25%, p=0.048) or central memory (CCR7⁺CD45RO⁺, 0.30% vs 0.65%, p=0.041) T cell populations in the AP samples. Particularly in patients receiving bispecific antibodies in the 3 months prior to AP, they had a lower proportion of CD8⁺ central memory cells (0.14% vs 0.60%, p=0.004).

Conclusions: The frequency of naïve or central memory CD8⁺ T cells in AP samples is closely related to clinical outcome after treatment with CAR-T cells targeting BCMA. Prior use of specific treatment options such as bispecific antibodies and alkylating agents just before lymphocyte apheresis may impair the efficacy of CAR-T therapy by decreasing the quality of the CAR-T cell product or weakening the anti-tumor microenvironment. To optimize the efficacy of CAR-T cell therapy, the frequency of specific T cell subsets, such as early memory or central memory cells, in PBMC may serve as a possible predictive marker of the outcome of Ide-cel therapy.