Long-Term Clinical Outcomes and Correlative Efficacy Analyses in Patients (Pts) with Relapsed/Refractory Follicular Lymphoma (r/r FL) Treated with Tisagenlecleucel in the Elara Trial

Background: Tisagenlecleucel is the first CAR-T cell therapy approved in the United States and European Union for r/r FL in the third-line or later setting. In the primary analysis of the single-arm, open-label, Phase II ELARA trial, tisagenlecleucel demonstrated high response rates and an excellent safety profile in adult pts with high-risk r/r FL. Herein, we report >2-year follow-up data on the continued durability of response (DOR), longer-term safety, as well as biomarker and pharmacokinetic analyses from the ELARA trial.

Methods: Eligible pts had r/r FL (grades 1-3A) following ≥2 lines of therapy that included an anti-CD20 antibody and an alkylating agent or had relapsed after autologous stem cell transplant (SCT). Bridging chemotherapy was permitted. Following lymphodepleting (LD) chemotherapy, pts received single-dose tisagenlecleucel (0.6-6×10⁸ CAR+ viable T cells). Baseline clinical factors, tumor microenvironment (TME) assessment, blood soluble factors, and circulating blood cells were correlated with clinical response and progression-free survival (PFS). Cellular kinetics were determined by measurement of transgene levels by quantitative polymerase chain reaction.

Results: Among 94 pts evaluable for efficacy [median follow-up of 28.9 months (mo)], complete response rate was 68% (95% CI, 57.7-77.3) and overall response rate was 86.2% (95% CI, 77.5-92.4). Median PFS was not reached (NR); the estimated 24 mo-PFS post infusion was 57.4% (95% CI, 46.2, 67.0; Figure). The estimated 24 mo-DOR and OS post infusion were 64.6% (77.8% for pts with CR) and 87.7% (95% CI, 78.3, 93.2), respectively. No new safety signals were reported. Neurological events (NEs) were reported in 12 (12.3%) pts including 1 pt with probable progressive multifocal leukoencephalopathy who had prior grade 4 immune effector cell-associated neurotoxicity syndrome. Three new deaths (total 13 deaths) occurred during this longer-term follow-up period [progressive disease (PD), n=1; serious adverse events (SAEs), n=2, urothelial bladder carcinoma and post allogeneic SCT (alloSCT) complications]. No new deaths were treatment related. A total of 23 (24%) pts received at least one subsequent therapy post infusion. Of these, 4 pts received alloSCT, and one pt received only localized radiotherapy. Persistence of the CAR transgene was observed for up to 925 days (median, 210 days; range, 13 to 925 days). Eight pts had SARS-COV-2 infection at the time of data cut-off.

Among all parameters tested, elevated tumor burden [TB; total metabolic tumor volume (TMTV) ≥240 cm³] at baseline (pre-LD chemotherapy), progression of disease within 2 years of initial chemoimmunotherapy (POD24), and >4 nodal areas at inclusion were clinical factors that correlated significantly with lower efficacy. Furthermore, baseline serum IL-10 and TNF-α levels were highly correlated with TB (Spearman correlation = 0.80 and 0.86, respectively (both P<0.001), with higher levels significantly associated with non-response/relapse and shorter PFS. Additionally, baseline TME was characterized by fluorescence immunohistochemistry, and an increase in LAG3+ exhausted T cells (≥3% of total T cells) was the only factor that significantly associated with shorter PFS (median PFS, 5.9 mo vs NR, P=0.0017). Also, decreased CD8+ cytotoxic T cells and increased CD19+ B cells in peripheral blood at baseline were associated with poor outcome. POD24, >4 nodal areas at inclusion, baseline TB, and tumor infiltrated LAG3+CD3+ cells remained significant in multivariate Cox models for PFS analysis.

Cellular kinetics shows that the mean in vivo maximal expansion (C_max) in pts with no POD24 was higher compared with pts with POD24. Pts with high TB, based on TMTV at baseline, had a shorter DOR relative to pts with low TB.

Conclusions: Updated long-term data from the ELARA trial demonstrated robust durable responses >2 years post infusion, with a continued favorable safety profile. Correlative analyses suggest that high-risk disease characteristics (elevated TB at baseline, POD24, >4 nodal areas at inclusion), exhausted T cells in the TME, and increased IL-10 and TNF-α levels are associated with worse long-term outcomes following tisagenlecleucel in pts with r/r FL.