Impact of Pre-Treatment Liver-Related Factors on Clinical Outcomes after CAR T-Cell Therapy for Lymphoma

Background: Previous research by the US Lymphoma Chimeric Antigen Receptor T-cell (CAR-T) consortium identified several factors to be associated with outcomes and complications from standard-of-care CAR-T for large B-cell lymphoma (LBCL) (Nastoupil et al. Journal of Clinical Oncology, 2020). Among these factors, elevated total bilirubin (> 1.5 g/dL) before CAR-T was found to be associated with both reduced overall survival (OS) and with increased risk of severe cytokine release syndrome (CRS). However, the impact of pre-existing liver-related factors on outcomes after CAR-T is not well elucidated and this study aims to evaluate this gap in knowledge. Methods: The study cohort consisted of all adult patients who received standard-of-care CAR-T for lymphoma at our institution between January 2018 and July 2023. The lymphoma histologies (and corresponding CAR-T products) included LBCL (axi-cel, tisa-cel or liso-cel), follicular lymphoma (FL; axi-cel) and mantle cell lymphoma (MCL; brexu-cel). Health record review was done for 5 different pre-CART liver-related factors including cirrhosis (as determined by either liver biopsy or documented gastroenterology clinical assessment), liver fibrosis (as determined by ultrasound elastography), presence of lymphomatous liver lesions or ascites on baseline PET/CT scan prior to CAR-T and total bilirubin at time of initiation of lymphodepletion chemotherapy. Primary endpoints were OS and progression-free survival (PFS) after CAR-T as well as response determined by complete response (CR) rate from PET/CT scan at 90 days after infusion. CRS and ICANS were graded per the ASTCT Consensus Grading system. Results: There were 522 patients including 436 with LBCL, 56 with MCL and 30 with FL. Cohort demographics included 356 male patients (68%), 414 (81%) White/Caucasians, 27 (5%) Asians, 27 (5%) Black/African Americans; 74 patients (15%) were of Hispanic ethnicity. 6 patients (1.2%) had liver fibrosis, 9 patients (1.7%) had elevated total bilirubin (≥ 1.5 g/dL) and 7 patients (1.3%) had cirrhosis; none had decompensated cirrhosis. Baseline PET/CT prior to CAR-T detected lymphomatous liver lesions in 42 patients (8%) and ascites in 18 patients (3.5%). 70 patients (13%) had at least one of the five liver related factors present. With a median follow-up time after CAR-T of 23 months, the median PFS and OS times after CAR-T for the whole cohort were 8.5 and 25 months respectively. Univariate analysis found significantly lower PS and OFS in those patients with pre-treatment lymphomatous liver lesions (both p-values < 0.001) and pre-treatment ascites (both p-values < 0.001). These two liver-related factors continued to be significantly associated with decreases in both OS and PFS in multivariate analyses incorporating other pre-treatment factors such as sex, type of CAR-T product, lymphoma histology and LDH at time of lymphodepletion. After adjusting for these other factors, risk of death was 2.06 (p < 0.001) and 2.53 times (p < 0.001) higher for patients with lymphomatous liver lesions or ascites respectively compared to other patients. Univariate analysis found significant decrease in OS (p = 0.013) but not PFS for patients with elevated total bilirubin; multivariate analysis did not find significant association with decreased OS. There were no significant differences in OS or PFS for patients with cirrhosis or liver fibrosis. None of the 5 liver-related factors had a significant impact on the CR rate at day+90 after CAR-T or on the incidence of severe ICANS of grade ≥ 3. The incidence of severe CRS of grade ≥ 3 was seen to be significantly higher (p < 0.01) in patients with lymphomatous liver lesions (26%) compared to rest of cohort (11.6%); this finding was not observed in patients with the other liver-related factors. Conclusions: To the best of our knowledge, this is the first systematic investigation of the impact of pre-treatment liver-related factors on clinical outcomes after anti-CD19 targeted CAR-T therapy for lymphoma. The response to therapy was not significantly impaired by presence of liver related factors. However, the presence of either lymphomatous lesions or ascites on baseline imaging prior to CAR-T were both poor prognostic factors for disease progression and for overall survival. Further studies are needed to elucidate the underlying mechanisms for these findings which have implications for counselling prior to CAR-T and monitoring after therapy.