Allogeneic CAR-engineered cellular therapy for relapsed and refractory large B cell lymphoma: a systematic review and meta-analysis

Abstract

Introduction: Relapsed/refractory (r/r) large B-cell lymphoma (LBCL) remains a difficult-to-treat disease with limited treatment options and high unmet clinical need, necessitating the development of new therapies with greater potency and broader applicability. While autologous chimeric antigen receptor (CAR)-T cell therapies have transformed the treatment landscape, 60-65% of patients receiving these therapies eventually relapse, underscoring the need for improved approaches. Allogeneic CAR-T and CAR-NK cell therapies have recently emerged as promising alternatives, offering the potential to shorten manufacturing times, reduce costs, and expand access to a broader patient population. This systematic review and meta-analysis compiles the currently available clinical trial data on the efficacy and safety of these novel therapies in adult patients with r/r LBCL.

Methods: A systematic search of MEDLINE, EMBASE, Web of Science, and the Cochrane Central Register of Controlled Trials was conducted for studies published up to January 12, 2025, involving allogeneic CAR-T and CAR-NK cell therapies in R/R LBCL. The primary outcomes assessed were the best overall response rate (bORR) and best complete response rate (bCRR) at any time point. Secondary outcomes included rates of grade 1-2 and grade 3+ cytokine release syndrome (CRS), grade 1-2 and grade 3+ immune effector cell-associated neurotoxicity syndrome (ICANS), grade 1-2 and grade 3+ infections and incidence of graft-versus-host disease (GvHD).

Results: Nineteen studies met the inclusion and exclusion criteria, encompassing 334 patients (155 CAR-NK; 179 CAR-T) evaluable for safety and 235 patients evaluable for response (77 CAR-NK; 158 CAR-T). The pooled estimates for the best overall response rate (bORR) and the best complete response rate (bCRR) were 52.5% [95% CI, 41.0-63.9] and 32.8% [95% CI, 24.2-42.0], respectively. Safety analysis revealed very low incidences of grade 3+ CRS (0.04% [95% CI 0.00-0.49]) or grade 3+ ICANS (0.64% [95% CI 0.01-2.23]) and only one occurrence of a GvH-like reaction across 334 infused patients enrolled in the included studies, highlighting the remarkable safety profile of CAR-engineered "off-the-shelf" allogeneic approaches. The estimated overall incidence of low-grade CRS was 30% [95% CI, 14-48], while the estimated overall incidence of low-grade ICANS was 1% [95% CI, 0%-4%], markedly lower than current-generation autologous CAR-T cell products. The incidence of low-grade and severe infections was 25% [95% CI 14-36%) (n=252) and 7% [95% CI 2-14%] (n=291), respectively.

Discussion: Together, allogeneic CAR-T and CAR-NK cell therapies demonstrate encouraging efficacy in heavily pretreated patients with r/r LBCL. Coupled with their favorable safety profiles and the potential for off-the-shelf availability, allogeneic cell therapies hold great promise to broaden the reach of live cell-based treatments, delivering impactful results to a wider patient population in the coming years.

Keywords: CRISPR gene editing; adoptive cell therapy (ACT); allogeneic CAR-NK cells; allogeneic CAR-T cells; cellular engineering; clinical trial; large B cell lymphoma; precision gene editing.

Figure 1

Conceptual depicting characteristics of autologous and allogeneic CAR-engineered cellular therapeutics.

Figure 2

PRISMA flow diagram illustrating the identification, screening, exclusion, and inclusion of clinical trials.

Figure 3

Forest plot illustrating the best overall response rate (bORR), stratified by cell source. Pooled estimates were computed for each subgroup and the overall weighted effect.

Figure 4

Forest plot illustrating the best complete response rate (bCRR), stratified by cell source. Pooled estimates were computed for each subgroup and the overall weighted effect.

Figure 5

Forest plot illustrating the incidence of grade 1–2 cytokine release syndrome (CRS), stratified by cell source. Pooled estimates were computed for each subgroup and the overall weighted effect.

Figure 6

Forest plot illustrating the incidence of grade 3+ cytokine release syndrome (CRS), stratified by cell source. Pooled estimates were computed for each subgroup and the overall weighted effect.

Figure 7

Forest plot illustrating the incidence of grade 1–2 immune effector cell-associated neurotoxicity syndrome (ICANS), stratified by cell source. Pooled estimates were computed for each subgroup and the overall weighted effect.

Figure 8

Forest plot illustrating the incidence of grade 3+ immune effector cell-associated neurotoxicity syndrome (ICANS), stratified by cell source. Pooled estimates were computed for each subgroup and the overall weighted effect.

Figure 9

Forest plot illustrating the incidence of grade 1–2 infections, stratified by cell source. Pooled estimates were computed for each subgroup and the overall weighted effect.

Figure 10

Forest plot illustrating the incidence of grade 3+ infections, stratified by cell source. Pooled estimates were computed for each subgroup and the overall weighted effect.

Figure 11

Forest plot illustrating the incidence of graft-versus-host disease (GvHD), stratified by cell source. Pooled estimates were computed for each subgroup and the overall weighted effect.