Treatment of non-Hodgkin lymphoma with point-of-care manufactured CAR T cells: a dual institution, phase 1 trial

Abstract

Background: Point-of-care manufacture of chimeric antigen receptor (CAR)-T cells can significantly reduce the time from apheresis to infusion. We conducted a dual-institution phase I trial aimed evaluating the safety and feasibility of this manufacturing model.

Methods: CASE 2417 was a phase I clinical trial. Adults with relapsed or refractory CD19 positive non-Hodgkin lymphoma (R/R NHL) treated with ≥2 prior systemic therapies were eligible. MB-CART-19 is an anti-CD19 CAR T-cell product manufactured using the CliniMACS Prodigy device with 4-1BB and CD3ζ costimulatory domains. Lymphodepletion included fludarabine 25 mg/m² for 3 days and cyclophosphamide 60 mg/kg for 1 day. Prophylactic tocilizumab was allowed. Three dose levels (0.5, 1.0 and 2.0 × 10⁶ cells/kg) were tested using a 3 + 3 dose-escalation schema. The primary outcome of this study was to determine the safety as defined by the dose limiting toxicities of MB-CART-19 in patients with relapsed and refractory NHL, co-primary outcome was determining the phase 2 dose of MB-CART-19. Secondary outcomes include defining the toxicity profile and to evaluate the initial efficacy of MB-CART-19 against relapsed or refractory NHL. This study was registered in ClinicalTrials.govNCT03434769.

Findings: Thirty-one patients were enrolled between July 2018 and January 2021. Twenty-four (77%) had aggressive lymphoma, 7 (24%) had indolent lymphoma (follicular lymphoma and marginal zone lymphoma). The median number of previous therapies was 5 (range 2-13, interquartile range [IQR] 3-5). All enrolled patients received MB-CART-19. Median apheresis to infusion time was 13 days (range 9-20, IQR 9-13). One dose limiting toxicity (DLT) was observed in dose escalation (fatal cytokine release syndrome [CRS]), whereas one patient died in dose expansion secondary to hemophagocytic syndrome. Both deaths were considered treatment-related. Twenty (65%) patients had CRS, three (10%) grade ≥3. Ten patients (32%) experienced immune effector cell neurotoxicity syndrome (ICANS), four (13%) grade ≥3. Neutropenia (n = 28, 90%), thrombocytopenia (n = 15, 48%) and anaemia (n = 13, 42%) were the most frequent grade ≥3 adverse events. Twenty-five out of 29 (86%, 95% confidence interval [CI]: 68-96%) response-evaluable patients had disease response and 22 (76%, 95% CI: 56-90%) had complete response; the overall and complete response rates for response-evaluable aggressive lymphoma patients (n = 22) were 82% (n = 18, 95% CI: 60-95%) and 73% (n = 16, 95% CI: 50-89%). Median follow up was 24.5 (IQR 17-32) months, median progression free survival (PFS) was 26 months (95% CI: 19-not reached [NR]) and median PFS was not reached (95% CI: 25 months-NR). Two-year estimates of PFS and overall survival (OS) were 63% (95% CI: 47-83%) and 68% (95% CI: 52-88%), respectively. Median PFS was 26 months (95% CI: 7-NR) for aggressive lymphoma patients with 2-year PFS estimate of 53% (95% CI: 36-78%), while median OS had not been reached for aggressive lymphoma patients (95% CI: 19 months-NR), and 2-year OS estimate was 60% (95% CI: 43-85%).

Interpretation: Point-of-care CAR T-cell manufacture was feasible and replicable across sites. MB-CART-19 has a safety profile comparable to other CAR T-cell products and high response rates. The recommended phase 2 dose is 2 × 10⁶ MB-CART-19 cells/kg. Short CAR T-cell manufacturing time permits treatment of patients with rapidly progressive lymphoma, a group of patients with high risk disease for whom access to autologous immune effector cellular therapies is usually limited.

Funding: This clinical trial was funded through University Hospitals Seidman Cancer Center and Washington University School of Medicine Institutional Funds. Correlative analyses were funded in part by the European Union-Next Generation EU-NRRP M6C2-Investment 2.1 Enhancement and strengthening of biomedical research in the NHS (project #PNRR-MAD-2022-12376059), and the Italian Ministry of Health Ricerca Finalizzata 2019 (project #RF-2019-12370243).

Keywords: Chimeric antigen receptor T cell; Immune effector cell therapy; Lymphoma; Phase I; Point-of-care manufacture.

Fig. 1

CONSORT flow diagram for phase 1 clinical study of MB-CART-19 anti-CD19 chimeric antigen receptor (CAR) T cells for adult patients with relapsed or refractory B cell non-Hodgkin lymphoma. Patients received a single infusion of MB-CART-19 following lymphodepletive chemotherapy. The first cohort received MB-CART-19 at a dose of 0.5 × 10⁶ cells/kg; this dose was escalated in subsequent cohorts using a 3 + 3 design to 1.0 × 10⁶ cells/kg and finally 2.0 × 10⁶ cells/kg.

Fig. 2

Waterfall plot showing best response to anti-CD19 chimeric antigen receptor (CAR) T cells among patients with measurable disease response. Columns show the percentage change from baseline in the product of the sum of diameters. Patients with measurable responses included 28/31 patients with at least one disease response assessment by imaging studies after CAR T-cell infusion. Colours represent the best response assessment based on Lugano response criteria (Red = progressive disease (n = 3); Salmon = stable disease (n = 1); Light blue = partial response (n = 3); Blue = complete response (n = 21). Non-measurable patients included two patients that died before response assessment and one patient who achieved complete response but had disease involving only the bone marrow and was not measurable through imaging studies.

Fig. 3

Swimmer plot representing patient events through follow up. Bar length represents time in months from chimeric antigen receptor (CAR) T-cell infusion, bar shade represents the histologic group of lymphoma; symbols represent adverse events of interest (cytokine release syndrome [CRS], immune effector—associated neurotoxicity syndrome [ICANS]), partial response (PR), complete response (CR) relapse and death.

Fig. 4

Patient outcomes. A. Progression free survival for all trial participants. B. Overall survival for all trial participants. Survival estimates done with the Kaplan Meier method. Gray shaded areas represent 95% confidence intervals.

Fig. 5

Chimeric antigen receptor (CAR) T-cell persistence and expansion. A. Scatterplot of CAR T-cell expansion measured by quantitative polymerase chain reaction (qPCR). Trends generated by locally estimated scatterplot smoothing (loess). Blue dots represent results of individual patients. B. CAR T-cell persistence for each individual patient (n = 31), measured by qPCR. Colours represent CAR T-cell dose infused. qPCR, quantitative polymerase chain reaction.

Fig. 6

Circulating tumour DNA (ctDNA) studies. A. ctDNA (haploid genome equivalents per mL [hGE/mL]) concentration at different timepoints for patients treated with antiCD19 chimeric antigen receptor (CAR) T cells. Day −6: baseline, pre-lymphodepletion, n = 26; Day 0: pre-CAR T-cell infusion, n = 22; Day 30: Day 30 after CAR T-cell infusion, n = 27. Comparisons between baseline group (Day—6) group and Day 0 and Day 30 done using the Wilcoxon signed-rank test. B. Progression free survival for patients with detectable ctDNA on day 30 after anti-CD19 CAR T cells [Day 30 ctDNA (+), blue line], vs. patients without detectable ctDNA on day 30 [Day 30 ctDNA (−), red line]. Survival estimates done with the Kaplan Meier method; comparisons done with Log rank test.

Fig. 7

Baseline and peak laboratory parameters. Boxplots represent the interquartile range (IQR), with the mid-line indicating the median. Vertical lines represent the minimum (Q1−1.5 IQR) and maximum (Q3 + 1.5 IQR). Dots represent individual values. A. Lymphocyte count at baseline (n = 31) prior to lymphodepletion and at peak value after chimeric antigen receptor (CAR) T-cell infusion (n = 31). Paired comparisons done with the Wilcoxon signed rank test. B. Baseline and peak lymphocyte counts in patients with (n = 20) and without (n = 11) cytokine release syndrome (CRS). Comparisons done with Mann–Whitney test. C. Baseline (apheresis) (n = 30) and peak post CAR T-cell infusion of C-reactive protein (CRP) plasma concentration (n = 31). Paired comparisons done with the Wilcoxon signed rank test. D. Baseline and peak CRP plasma concentration in patients with (n = 20) and without CRS (n = 10). Comparisons done with Mann–Whitney test. E. Baseline (n = 31) and peak plasma ferritin (n = 31) concentrations. Paired comparisons done with the Wilcoxon signed rank test. F. Baseline and peak plasma ferritin concentrations in patients with (n = 20) and without CRS (n = 11). Comparisons between groups done with the Mann–Whitney test.