Chimeric Antigen Receptor (CAR) T-cell therapy has changed the treatment landscape for patients with Non-Hodgkin Lymphoma (NHL). Despite the excellent responses in relapsed or refractory (R/R) aggressive NHL (aNHL), the outcome of patients (pts) that fail CAR T-cell therapy remains poor, and there is not a clear path for management of their disease.


We conducted a retrospective analysis of aNHL pts treated with axicabtagene ciloleucel (axi-cel) at the Mayo Clinic campuses in Arizona and Florida between June 2018 and August 2020. We evaluated the predisposing factors, management, toxicities, and response after CAR T-cell therapy failure. Statistical calculations using parametric tests were performed, and survival curves were estimated using the Kaplan-Meier method and compared statistically using the log-rank test and Pearson's correlation.


Thirty-four pts with aNHL received axi-cel. The median age was 53 years [IQR 42-63], and 62% were male. All pts received inpatient axi-cel infusions and the median length of hospital stay was 14 days (IQR: 11-17). Cytokine Release Syndrome (CRS) was observed in 91% of pts (3% grade ≥3), while Immune Effector Cell Associated Neurotoxicity Syndrome (ICANS) was observed in 41% (24% grade ≥3).

At day 30 response assessment, 16 pts (47%) had complete responses (CR), 9 (26%) had a partial response (PR), 4 (12%) had stable disease (SD), 4 (12%) showed progression with primary refractory disease (PD) and 1 (3%) died before assessment due to grade 5 ICANS (Table 1).

After a median follow-up of 178 days, we observed PD in 12 (35%) pts. The median time-to-progression was 72 days (IQR 58-93) and most of the pts (83%) progressed during the first 3 months. None of the patients with more than 5 months of sustained response developed progression of disease. The likelihood of progression during the first 6 months after axi-cel infusion was 19%, 57%, 50% for pts that initially achieved a CR, PR, SD, respectively. Expression of CD19 was observed in 66% (4/6) of pts with available biopsies after axi-cel suggesting a failure mechanism other than antigen escape. The mortality rate of the R/R aNHL group was 58% with a median survival time of 83 days (IQR: 50-109).

There was no association between age, stage, number of previous therapies, time from previous therapy to axi-cel infusion, time from apheresis to infusion, use of tocilizumab, or steroids with progression of disease. Of note, no correlation between CRS or ICANS with progression of disease was found (2-way ANOVA test F (1, 4) = 3.802, p=0.1230). Maintaining a response to axi-cel treatment (CR, PR, or SD) for ≥ 3 months was a strong predictor of durable response with an HR of 0.05 (p= <0.0001).

Eleven R/R pts received subsequent therapies with a median time to retreatment of 76 days. Those treatments included: Radiotherapy (n=7), pembrolizumab (n=3), polatuzumab-rituximab with (n=3) and without (n=1) bendamustine, obinutuzumab with (n=1) or without (n=1) lenalidomide, Hyper-CVAD (n=2), R-GemOx (n=1), rituximab with lenalidomide (n=1) and intrathecal methotrexate (n=1). Only 2 (17%) patients have responded to salvage therapy achieving PR (one patient treated with radiotherapy and the other with rituximab-lenalidomide after two other salvage therapies).


Our experience demonstrates the majority of aNHL patients respond to axi-cel. If patients maintain their response for more than 3 months, the likelihood of progression is very low - 15%. Similar to what has been previously reported in the literature, our series showed that 35% of patients progressed after axi-cel, and subsequently have a poor prognosis with median survival after a relapse of only 83 days (IQR: 50-109). Therapy options following axi-cel were limited due to severe cytopenias, only 2 of 11 patients have responded to salvage therapy, suggesting that conventional treatments are probably not effective/safe in this high-risk group of patients. Interestingly, the majority of R/R pts with available biopsies showed persistent CD19 expression suggesting that CAR T-cell exhaustion, poor in vivo expansion, and inhibitory signals of the tumor microenvironment may contribute to resistance. Additional strategies for monitoring of axi-cel persistence and its immunophenotypic profile could be helpful for prognosis and management of

CAR T-cell pts receiving axi-cel.


Kharfan-Dabaja:Daiichi Sankyo: Consultancy; Pharmacyclics: Consultancy. Castro:Fate Therapeutics: Research Funding; Kite Pharma: Research Funding; Pharmacyclics: Research Funding.

Author notes


Asterisk with author names denotes non-ASH members.