Emeline R. Chong

ABSTRACT: Lymphodepletion (LD) is an integral component of chimeric antigen receptor T-cell (CART) immunotherapies. In this study, we compared the safety and efficacy of bendamustine (Benda) to standard fludarabine/cyclophosphamide (Flu/Cy) LD before CD19-directed, CD28-costimulated CART axicabtagene ciloleucel (axi-cel) for patients with large B-cell lymphoma (LBCL) and follicular lymphoma (FL). We analyzed 59 patients diagnosed with LBCL (n = 48) and FL (n = 11) consecutively treated with axi-cel at the University of Pennsylvania. We also analyzed serum samples for cytokine levels and metabolomic changes before and after LD. Flu/Cy and Benda demonstrated similar efficacy, with complete remission rates of 51.4% and 50.0% (P = .981), respectively, and similar progression-free and overall survivals. Any-grade cytokine-release syndrome occurred in 91.9% of patients receiving Flu/Cy vs 72.7% of patients receiving Benda (P = .048); any-grade neurotoxicity after Flu/Cy occurred in 45.9% of patients and after Benda in 18.2% of patients (P = .031). In addition, Flu/Cy was associated with a higher incidence of grade ≥3 neutropenia (100% vs 54.5%; P < .001), infections (78.4% vs 27.3%; P < .001), and neutropenic fever (78.4% vs 13.6%; P < .001). These results were confirmed both in patients with LBCL and those with FL. Mechanistically, patients with Flu/Cy had a greater increase in inflammatory cytokines associated with neurotoxicity and reduced levels of metabolites critical for redox balance and biosynthesis. This study suggests that Benda LD may be a safe alternative to Flu/Cy for CD28-based CART CD19-directed immunotherapy with similar efficacy and reduced toxicities. Benda is associated with reduced levels of inflammatory cytokines and increased anabolic metabolites.

Introduction: Anti-CD19 chimeric antigen receptor-modified T cell therapy (CAR T) is indicated for diffuse large B-cell, high grade B-cell, transformed follicular and primary mediastinal B-cell lymphomas. To date, the longest reported median follow-up is 19 months (mo) for tisagenlecleucel (Schuster et al., Blood 2018) and 24 mo for axicabtagene ciloleucel (Locke et al., Lancet Oncol 2019). We previously reported two-year follow-up for 28 patients (pts) who received CTL019 (Schuster et al., NEJM 2017). We now report our four-year experience with CAR T (CTL019, now tisagenlecleucel). Methods: We enrolled pts with relapsed/refractory CD19+ diffuse large B cell lymphoma (DLBCL) or follicular lymphoma (FL) from February 2014 to September 2017 on a single institution trial of CTL019. Eligible pts had no curative treatment options, prognosis of <2 years survival, and less than complete response to last therapy. Pts underwent leukapheresis, bridging therapy at the investigator's discretion, then lymphodepleting therapy followed by a single dose of 1e8 to 5e8 CTL019 cells. Outcomes were analyzed as of February 25, 2019. Results: Median follow-up is now 49 mos. 49 pts, 24 with DLBCL and 15 with FL, were enrolled. Median age at enrollment was 56 years (range: 25 – 77) and 41% were women. The median number of prior therapies was 5 (range: 2-10). Median ECOG PS was 1 (range: 0-1). 78% had advanced stage lymphoma at enrollment. 38 pts received the protocol-specified dose of CTL019. 46% of DLBCL pts and 71% of FL pts had a complete response (CR). For DLBCL, median progression free survival (PFS) is 5.8 mo (95%CI: 1.6 mo-NE), median overall survival (OS) 22.2 mo (95%CI: 10.9-45.6mo). For FL, median PFS is 32.4 mo (95%CI: 3.5 mo-NE), median OS was not reached (95%CI: 27.2mo-NE). For responding patients with DLBCL or FL, median response duration (RD) was not reached at 49 mo follow-up (DLBCL RD, 95%CI: 3.2 mo-NE; FL RD, 95%CI: 9.5 mo-NE). The patient with the longest follow-up remains in remission from DLBCL (double-hit) at 60 mo after CTL019. Of 21 pts in CR, only 6 received IVIG for recurrent or severe infections and 2 pts developed late myelodysplasia. Four patients relapsed more than 12 months from CTL019 infusion (PFS: 16.3 mo, 26.2 mo, 32.4 mo, 35.2 mo). At the time of these late relapses, CAR T cells persisted by quantitative PCR and there was no loss of CD19 expression by tumor cells. Keywords: diffuse large B-cell lymphoma (DLBCL); follicular lymphoma (FL); T-cells. Disclosures: Chong, E: Consultant Advisory Role: Novartis. Svoboda, J: Consultant Advisory Role: Kite Pharma; Research Funding: Novartis. Nasta, S: Research Funding: Roche, Incyte, Debiopharm, Rafael, Aileron, Takeda/Millennium. Landsburg, D: Consultant Advisory Role: Curis, Celgene; Research Funding: Curis, Takeda, Triphase; Other Remuneration: Seattle Genetics =. Porter, D: Employment Leadership Position: Genentech; Consultant Advisory Role: Novartis, Kite Pharma, Incyte, Glenmark; Stock Ownership: Genentech; Research Funding: Novartis; Other Remuneration: Novartis. Levine, B: Employment Leadership Position: Tmunity Therapeutics; Consultant Advisory Role: CRC Oncology, Cure Genetics, Novartis, Terumo, Avectas, Brammer Bio, Incysus, Vycellix; Other Remuneration: Novartis. June, C: Consultant Advisory Role: Novartis, Immune Design, Viracta, Carisma; Stock Ownership: Tmunity Therapeutics; Research Funding: Novartis; Other Remuneration: Novartis. Schuster, S: Consultant Advisory Role: Pfizer, Nordic Nanovector, Celgene, Gilead, Merck, Novartis, Pharmacyclics, Loxo Oncology, Acerta, AstraZeneca; Research Funding: Acerta, Celgene, Genentech, Gilead, Merck, Novartis, Pharmacyclics; Other Remuneration: Novartis.

ABSTRACT: Mosunetuzumab and other CD20/CD3 bispecific antibodies (BsAbs) have efficacy in B-cell lymphomas relapsing after or refractory to CD19-directed chimeric antigen receptor (CAR)-modified T cells (CAR-T). The optimal timing of BsAbs and biomarkers of BsAb response after CAR-T are unknown. We addressed these questions using clinical data and blood samples from patients previously treated with CAR-T and subsequently treated on a phase 1/2 study of mosunetuzumab. Thirty patients had paired samples at baseline and after 1 cycle of mosunetuzumab. The median time from CAR-T to mosunetuzumab was significantly longer for responding than for nonresponding patients (P = .006, unadjusted for multiple comparisons). Most patients (20/30) did not receive intervening therapy between CAR-T administration and mosunetuzumab. The remainder of patients received 1 intervening therapy after a protocol-mandated drug washout. After mosunetuzumab, responding patients had higher lymphocytes (995 vs 400 cells per μL; P = .02) and greater increases in CD4 and CD8 cells (median change, 73 vs -90 cells per μL [P = .005] and 243 vs -103 cells per μL [P = .004], respectively). Additionally, responding patients had an increase in activated CD8 cells (median fold change, 1.7; P = .02). Nonresponders had a relative decrease in CAR transgene levels (n = 16; P = .04). This is, to our knowledge, the first study to assess changes in lymphocytes, T cells, and CAR transgene levels in patients treated with BsAbs after CAR-T. These findings suggest an interaction between prior CAR-T and BsAb outcomes and have implications for optimal timing of BsAb after CAR-T. The trial was registered at www.ClinicalTrials.gov as #NCT02500407.