Long-term outcomes of patients with large B-cell lymphoma treated with standard-of-care axicabtagene ciloleucel: Results from the us lymphoma CAR-T cell consortium

ABSTRACT

Introduction: Axicabtagene ciloleucel (axi-cel) is an autologous anti-CD19 Chimeric Antigen Receptor (CAR) T-cell therapy that induces durable responses in patients with relapsed or refractory large B-cell lymphoma. At a median of 27.1 months follow-up on the ZUMA-1 trial, median overall survival (OS) was 25.8 months with 39% progression free survival (PFS) at 2 years post-infusion (Locke, Lancet Onc 2019). We previously reported outcomes of axi-cel patients treated with standard of care therapy at a median follow up of 12.9 months, including 42% who did not meet eligibility criteria for ZUMA-1 based on co-morbidities (Nastoupil, JCO 2020). Here we report results from this cohort at a median follow up of 32.4 months, as well as late outcomes of interest including cytopenias, infections and secondary malignancies. Methods and Results: The US Lymphoma CAR-T Consortium comprised of 17 US academic centers who contributed data independent of the manufacturer. Two hundred and ninety-eight patients underwent leukapheresis with intent to manufacture standard of care axi-cel as of September 30, 2018. In infused patients (n=275), OS and PFS were calculated from date of infusion. After median follow-up of 32.4 months (95% CI 31.1 - 34.3), median OS was not reached (95% CI 25.6 - not evaluable) (Figure 1A) with 1-, 2- and 3-year OS of 68.5% (95% CI 62.6-73.7), 56.4% (95% CI 50.1-62.2) and 52.2% (95% CI 45.7-58.2%), respectively. Median PFS was 9 months (95% CI 5.9-19.6) (Figure 1B);1-, 2- and 3-year PFS was 47.4% (95% CI 41.4-53.2), 41.6% (95% CI 35.6-47.5) and 37.3% (95% CI 31.3-43.2), respectively. Twenty-seven PFS events occurred at or after 1 year post infusion;19 events were progressive lymphoma, with the latest relapse observed 28 months after axi-cel infusion. Eight patients died while in remission from their lymphoma 4 from secondary malignancy, 3 from infection, and 1 from unknown causes. Results of multivariable modeling were similar to our prior analysis: factors associated with both a shorter PFS and shorter OS included male sex, elevated pre-lymphodepletion LDH, and poor ECOG status. Complete blood count and B- and T-cell recovery data were collected at 1 and 2-years post-infusion, excluding patients who had relapsed or been treated for secondary malignancy at time of collection (Table 1). Rates of neutropenia (absolute neutrophil count ≤1000) at 1- and 2- years were 9.2% (10/109) and 11.2% (9/80) and rates of CD4 count ≤200/ul were 62% (23/37) and 27% (7/26). Recovery of B cells was seen in 54% (15/28) and 57% (13/23) at 1-and 2-years post infusion. Infections were reported in 31.2% (34/109) patients between 6- and 12-months post infusion, and 17% (18/109) were severe, requiring either hospitalization and/or IV antibiotics. Twenty-one patients (24%, 21/89) had an infection between 1- and 2- years, 11% of which were severe. Twenty percent (10/49) of patients between 2- and 3-years had an infection and 4 (8%) were severe. Neutropenia, low CD4 counts, and IgG levels were not associated with infection, though patients with infection between 6-12 months were more likely to have received IVIG (p<0.001). No patient in this cohort died of COVID-19. Twenty-two of 275 (8%) patients were diagnosed with subsequent malignancy after axi-cel treatment 14/275 (5%) patients were diagnosed with myeloid malignancies (MDS (n=12), AML (n=1), CMML (n=1));other malignancies included squamous cell carcinoma of skin (n=3);sarcoma (n=1);endometrial (n=1);lung (n=1);mesothelioma (n=1) and AITL (n=1). Patients with myeloid malignancy had a median age of 62 at axi-cel apheresis (IQR 56-67), 64% were male and median lines of prior therapy was 4 (IQR 3-6), including 36% with a prior autologous stem cell transplant. Eleven patients were in remission from lymphoma at myeloid malignancy diagnosis, while 3 were diagnosed after progression and interval therapy. Conclusion: This multi-center retrospective study showed similar long-term results to the ZUMA-1 trial, despite including patients who did not meet ZUMA-1 eligibility criteria ba ed on comorbidities. Sixteen percent of PFS events were seen after 1 year, largely due to disease progression. Late infection was common but was not explained by persistent neutropenia or low CD4 counts. Subsequent malignancy, including MDS, occurred in 8% of patients and require further study to better identify patients at risk. (Figure Presented).