CAR T-cell expansion: harmful or helpful?

In this issue of Blood Advances, Hamilton et al report results from a prospective investigation of the relationship between peripheral blood chimeric antigen receptor (CAR) T-cell levels with outcomes after standard-of-care anti-CD19 CAR T-cell therapy for different types of B-cell lymphoma.¹ Flow cytometry–based quantification of CAR T cells was performed on days 7, 14, 21, and 28 after infusion in blood samples from patients who received either brexucabtagene autoleucel (brexu-cel) for mantle cell lymphoma (MCL) or axicabtagene ciloleucel (axi-cel) for large B-cell lymphoma (LBCL) or follicular lymphoma (FL). The expansion kinetics of CAR T cells during the first month after infusion were compared across the different B-cell lymphoma subtypes and also across patients with varying grades of treatment-related toxicities, including cytokine release syndrome (CRS), immune effector cell associated neurotoxicity syndrome (ICANS), and prolonged neutropenia.

One key finding of this study is the strikingly more robust expansion of CAR T cells in patients with MCL compared with those with LBCL or FL with significantly higher levels of peripheral blood CAR T cells in the former group across the different time points. The authors provide a potential mechanistic explanation by analyzing a separate messenger RNA microarray data set, which revealed significantly higher CD19 expression in patients with MCL compared with those with LBCL or FL. In addition to proportionally greater CD19 target–driven expansion of CAR T cells, other factors to consider include product differences between axi-cel and brexu-cel and the possible potentiating effect of prior Bruton tyrosine kinase inhibitor use on T-cell function in patients with MCL. A likely clinical consequence of the more robust CAR T-cell expansion in patients with MCL was the significantly higher rate of severe (grade 3 or higher) ICANS in this subset of patients compared with those with LBCL or FL as well as a corresponding increase in cumulative doses of dexamethasone for ICANS treatment and length of hospitalization. These findings are in line with higher rates of ICANS associated with brexu-cel in both clinical trial² and real-world data³ in comparison with other standard-of-care CAR T-cell therapies in LBCL or FL.⁴ 

There was also a significant association between increased CAR T-cell expansion and risk for severe ICANS across the whole data set, which provides additional evidence for this link between CAR T-cell expansion and neurotoxicity that has now been described across different types of CAR T-cell products and lymphoma subtypes in the contexts of both clinical trial^5-8 and standard-of-care use.⁹ These cumulative findings (see table) illustrate the potential for clinical use of peripheral blood CAR T-cell levels as a biomarker for early prediction of severe neurotoxicity, for which there is currently a lack of other clinically validated, predictive laboratory parameters. Earlier identification of patients at risk for severe ICANS could facilitate investigation of novel prophylactic measures, closer monitoring, and/or initiation of more aggressive therapeutic interventions. Longitudinal monitoring of peripheral blood CAR T-cell levels could also guide the titration and tapering of therapies for ICANS such as corticosteroids. However, to facilitate this potential application, the precise impact of corticosteroids on CAR T-cell kinetics and function requires further elucidation by future studies. Another practical consideration requiring additional investigation is the optimal methodology for routine monitoring of CAR T-cell expansion kinetics. Of note, the advantages of flow-cytometry–based quantification of peripheral blood CAR T cells, as used in this study by Hamilton et al, over higher-resolution methodologies such as cell-free DNA include wider accessibility across more centers, faster turn-around time, and relatively lower cost associated with assay implementation and data analysis. The correlation between CAR T-cell levels in blood and those in other tissues such as tumor or cerebrospinal fluid needs to be better understood as well as the potential role for monitoring CAR T cells in cerebrospinal fluid in patients with ICANS.

In addition to being correlated with risk for severe ICANS, increased CAR T-cell expansion was also associated with significantly increased risk for other therapy-related toxicities including severe CRS and prolonged neutropenia with continued requirement of granulocyte colony-stimulating factor therapy after day 14 post-infusion. Interestingly, this study did not find a clear association between blood CAR T-cell levels and response to treatment, as assessed by progression-free survival or durable ongoing response, within the subset of patients with LBCL. These findings are consistent with findings from the ZUMA-7 trial⁸ and from an earlier standard-of-care study with axi-cel⁹ but differ from other pivotal clinical trials (see table) and a recent study in the standard-of-care setting by Blumenberg et al.¹⁰ Potential factors that might explain the differing findings include interstudy differences in the CAR T-cell products, CAR T-cell quantification assays, lymphoma histologies, and in baseline characteristics such as disease burden and number of prior lines of treatment. Further studies with larger sample sizes are required to better delineate the relationship between CAR T-cell expansion kinetics and response to therapy.

In summary, this study by Hamilton et al identifies increased peripheral blood CAR T-cell expansion to be correlated with higher risk for therapy-related complications including severe ICANS, CRS, and prolonged neutropenia in patients with different subtypes of B-cell lymphoma. Of note, patients with MCL were found to have more robust CAR T-cell expansion and correspondingly more severe therapy–related toxicities compared with patients with LBCL and FL. These findings add to emerging evidence for the role of peripheral blood CAR T-cell monitoring as a biomarker with prognostic utility for different therapy-related toxicities and with the potential to guide prophylactic measures and therapeutic interventions for these complications. However, larger and more standardized studies across different centers are needed to definitively determine the relative impact of CAR T-cell expansion on toxicity vs efficacy.

Conflict-of-interest disclosure: S.S.N. received research support from Kite/Gilead, Bristol Myers Squibb (BMS), Allogene, Precision Biosciences, Adicet Bio, Sana Biotechnology, and Cargo Therapeutics; served as advisory board member/consultant for Kite/Gilead, Merck, Sellas Life Sciences, Athenex, Allogene, Incyte, Adicet Bio, BMS, bluebird bio, Fosun Kite, Sana Biotechnology, Caribou, Astellas Pharma, MorphoSys, Janssen, Chimagen, ImmunoACT, Orna Therapeutics, Takeda, Synthekine, Carsgen, Appia Bio, and GlaxoSmithKline; has stock options from Longbow Immunotherapy, Inc; and has intellectual property related to cell therapy. A.C.L. declares no competing financial interests.