Introduction Brexucabtagene autoleucel (BA) cellular therapy is the most recent milestone for treating patients with mantle cell lymphoma (MCL) and the only CAR-T cell therapy approved by the FDA for MCL. However, relapses to BA inevitably occur in some patients and are associated with poor patient outcomes with median survival of only 4.1 months. Furthermore, the mechanisms underlying BA relapse in MCL have not been characterized extensively to date.

Methods We performed single-cell RNA sequencing for 39 longitudinal samples from 15 patients sequentially treated with Bruton's tyrosine kinase inhibitors (BTKi) and CAR-T therapy. An additional cohort of 10 patients treated with BTKi only (3 BTKi-resistant and 7 BTKi-sensitive) were included for comparison. We focused our bioinformatics analysis on tumor B cells in this study. Differentially expressed genes, gene set enrichment analysis (GSEA), and trajectory analysis were performed to identify gene signatures and cancer hallmarks associated with CAR-T resistance as well as potential early drivers of CAR-T resistance. Functional studies were performed to assess the potential of targeting the HSP90-MYC-CDK9-axis to overcome dual resistance to BTKi and CAR-T therapy in MCL.

Results CAR-T resistance-associated gene signatures and cancer hallmarks were identified by comparing the samples collected from BTKi-resistant (BTKi-R) patients and from BTKi/BA dual-resistant (Dual-R) patients. Cyclin-dependent kinase 9 (CDK9) and polymerase (RNA) II subunit C (POLR2C) were two of the top differentially expressed genes in Dual-R patients. MYC-targets_v1 and _v2 were the predominant cancer hallmarks dramatically dysregulated further in Dual-R patients. CDK9 is the gatekeeper for transcription elongation, while POLR2C is one important subunit of RNA polymerase II, the primary enzyme for genome-wide transcription. CDK9-mediated transcription elongation was reported to be required for disease maintenance in MYC-overexpressing cancer cells. Furthermore, several early driver genes near the bifurcation point of the major trajectories were identified, with the top early driver being HSP90AB1 (Heat Shock Protein 90 Alpha Family Class B Member 1), the constitutive form of the chaperone protein HSP90. Analysis of loss-of-function screens from the publicly available Depmap resource revealed that HSP90 and MYC are synthetic lethal and specific to lymphomas among many other cancer types. Consistent with this, HSP90AB1 expression correlated well with cancer hallmarks MYC-targets in BTKi-R and Dual-R MCL cells. HSP90AB1 itself is a transcriptional target of the oncogene MYC, and MYC, in turn, is a client of heat shock protein 90 (HSP90). These findings led to our new hypothesis that the HSP90-MYC-CDK9 axis is a positive feedback loop and acts as a major tumor-intrinsic factor in driving dual resistance. Indeed, targeting CDK9 with AZD4573 induced impressive anti-MCL activity in vitro and overcomes the dual resistance in vivo in patient-derived xenograft models. Targeting HSP90 with two independent inhibitors PU-H71 or 17-AAG was also potent in MCL preclinical models. Furthermore, dual targeting of CDK9 and HSP90 achieved remarkable anti-MCL synergy in MYC-driven cell lines by suppressing cancer hallmark MYC targets and potently inducing cell apoptosis beyond any single agent.

Conclusion Our data demonstrated that an unexpected but novel axis led by HSP90-MYC-CDK9 drives the dual resistance to BTKi and CAR T therapies. Our study sheds light on the underlying mechanism of CAR-T resistance in addition to BTKi resistance in MCL and provides compelling preclinical evidence for therapeutic targeting of the HSP90-MYC-CDK9 axis to overcome the dual resistance in patients with MCL.

Wang:Oncology Specialty Group: Honoraria; Studio ER Congressi: Honoraria; Kite Pharma: Consultancy, Honoraria, Research Funding; Physicians Education Resources (PER): Honoraria; OncLive: Honoraria; Moffit Cancer Center: Honoraria; Meeting Minds Experts: Honoraria; Janssen: Consultancy, Honoraria, Research Funding; Pharmacyclics: Consultancy, Honoraria, Research Funding; Pepromene Bio: Consultancy; Oncternal: Consultancy, Research Funding; Milken Institute: Consultancy; VelosBio: Consultancy, Research Funding; Merck: Honoraria; Lilly: Consultancy, Research Funding; Celgene: Research Funding; Genmab: Research Funding; Eastern Virginia Medical School: Honoraria; Loxo Oncology: Research Funding; Molecular Templates: Research Funding; Leukemia & Lymphoma Society: Consultancy, Honoraria; LLC TS Oncology: Honoraria; IDEOlogy Health: Honoraria; Dava Oncology: Honoraria; Genentech: Consultancy, Research Funding; Medscape: Honoraria; MJH Life Sciences: Honoraria; Practice Point Communications (PPC): Honoraria; Juno Therapeutics: Consultancy, Research Funding; Vinverx: Research Funding; InnoCare: Consultancy, Research Funding; Deciphera: Consultancy; BioInvent: Consultancy, Honoraria, Research Funding; BeiGene: Consultancy, Honoraria, Research Funding; AstraZeneca: Consultancy, Honoraria, Research Funding; Acerta Pharma: Honoraria, Research Funding; AbbVie: Consultancy.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution