Real World Comparison of Axicabtagene Ciloleucel and Lisocabtagene Maraleucel in Relapsed or Refractory Diffuse Large B-Cell Lymphoma

Background: Chimeric antigen receptor T-cell (CAR-T) therapies targeting CD19, such as Axicabatagene ciloleucel (axicel) and Lisocabtagene maraleucel (lisocel), have shown efficacy with a tolerable safety profile in the treatment of relapsed or refractory diffuse large B-cell lymphoma (RR DLBLC) in clinical trials. Limited studies illustrate the comparison in a real-world setting between axicel and lisocel.

Objectives: To provide a real-world comparison between axicel and lisocel in terms of efficacy and safety in clinical practice for RR DLBCL.

Methods: This is a retrospective analysis using TriNetX global research database that provided de-identified patient (pt) information across 89 healthcare organizations in total. Two cohorts were created – cohort 1 with axicel and cohort 2 with lisocel for RR DLBCL. Total of 1279 pts with relapsed DLBCL were analyzed: 1,052 pts in the axicel group and 227 pts in lisocel group. Propensity score matching was conducted on baseline characteristics, resulting in 179 matched pts per cohort. The primary outcome of interest was overall survival (OS). Secondary outcomes of interest were relapse rate, rate of cytopenia 30 days after administration of axicel or lisocel, defined by platelet < 50x103/µL, Hgb < 8 g/dL, and ANC < 500/µL, risk of infection, cytokine release syndrome (CRS), immune effector cell-associated neurotoxicity syndrome (ICANS), and subsequent use of bispecific T-cell engager therapy (BiTE). Outcomes were measured from day 1 after administration of axicel or lisocel to 3 years. Analysis methods included measures of association, number of instances, and Kaplan-Meier survival estimates, with T-test statistics assessing differences between cohorts.

Results and conclusions: Pts were predominantly Caucasian in both cohorts analyzed (85.5% in axicel and 87.7% in lisocel). Using both unmatched and propensity score-matched analyses, there was no significant difference in OS probabilities between the two cohorts. The propensity score-matched median OS in axicel group was 329 days, while in the lisocel group it was 420 days (HR 1.278, 95% CI [0.934, 1.750]; p=0.121). Similar propensity score-matched relapse rate were seen in both axicel and lisocel group with a relapse rate of 27.9% (n=50) and 22.9% (n=41) (p=0.275), respectively. No significant difference in cytopenias or infection risk was seen between the groups. CRS in propensity score-matched analysis showed no difference in all grade CRS between two groups, with CRS noted in 49.7% in axicel group and 41.3% in lisocel group (p=0.111) and grade ≥ 3 CRS rates were similar in both axicel and lisocel group (both around 5.6%). Significantly more ICANS was seen within the axicel group (n=60) compared to lisocel (n=38), with 12.3% more ICANS seen in axicel group compared to lisocel group (p=0.009). While axicel group had significantly more grade ≥ 3 ICANS (14.0%) compared to lisocel group (7.3%) (p=0.039), but 10 pts (5.6%) in lisocel group died of ICANS (grade 5) compared to 0 pt in axicel (p=0.001). No difference in BiTE use seen between both groups.

This result is comparable to each respective trial data in that more ICANS noted in axicel compared to lisocel in any grade as well as those ≥ grade 3. However, more grade 5 ICANS was noted in lisocel group, certainly raising questions for further investigation. Although comparisons for OS cannot be made due to published trial’s data maturity, this study provides real-world evidence for pts with RR DLBCL who underwent treatment with either axicel or lisocel. Further maturation of data from clinical trials is needed to better guide treatment for these groups, but comparison data from our analysis looks promising and doesn’t show any significant difference between the two products.