Bridging Radiotherapy to Axicabtagene Ciloleucel (axi-cel) Is Associated with Favourable Outcomes for Large B-Cell Lymphoma (LBCL) in Third-Line and Beyond: An Australian Single-Centre Real-World Experience

Aim: Autologous anti-CD19 CAR T-cell therapy is an established standard-of-care (SOC) for relapsed/refractory (R/R) LBCL, but successful delivery requires disease control through the manufacturing period. The effects of bridging radiotherapy (RT) on outcome are unclear and impacted by disease factors such as bulk and distribution. We sought to compare outcomes of patients receiving RT-alone versus systemic-based bridging therapies (BT) delivered after apheresis and prior to axi-cel infusion. Method: Data were collected retrospectively for all patients with R/R LBCL treated with SOC axi-cel between 2020-2024 with ≥3 months of follow-up post CAR T-cell infusion. Patients were included in the RT-alone group if they received radiotherapy and no systemic therapy (apart from corticosteroids) as their BT. All other patients were included in the systemic-based bridging (ST) group. The primary endpoints were complete response (CR) rate and progression-free survival (PFS) measured from date of axi-cel infusion. We performed univariable and multivariable Cox proportional hazards (CoxPH) modelling of candidate predictive factors at two time points: (1) apheresis, and (2) pre-LD chemotherapy. Results: 98 patients received axi-cel with a median follow-up of 12.9 months. 35 patients received RT-alone and 63 patients received ST (including 22 patients in the ST group who received systemic and RT to a dominant mass as BT). Baseline characteristics were similar between RT-alone and ST groups at both the pre-apheresis and pre-LD timepoint. Median age pre-LD was 63 (range 22-81), RT median age 62 (range 42-80); ST median age 64 (range 22-81)), pre-LD ECOG ≥1 n=38, 39% (RT n=12, 34%; ST n=26, 41%), 54% had elevated LDH pre-apheresis (RT n=21, 62%; ST n=31, 49%), 42% had elevated LDH pre-LD (RT n=15, 43%; ST n=26, 41%), median total metabolic tumour volume (TMTV, ml) pre-apheresis was 48 (range 0 - 2809, RT median 46 (range 1 -1562); ST median 48.5 (range 0 - 2809)), median TMTV pre-LD was 21.5 (range 0 - 2777, RT median 22 (range 0 - 948); ST median 21 (range 0 - 2777)), 64% had stage 3-4 disease pre-apheresis (RT n=16, 46%; ST n=45, 75%) and 56% of patients had stage 3-4 disease pre-LD (RT n=19, 54%; ST n=36, 57%). Within the RT group, the median dosage was 30 (range 8-40) Gy delivered in a median of 10 fractions (range 1-20), with volumetric modulated arc therapy (VMAT) utilised in the 14/17 patients with assessable data. The CR rate post axi-cel infusion was 76%: 85% in the RT and 71% in the ST group (p=0.21). PFS at 12 months was 58% (95% CI: 0.49 - 0.69) for all patients: 75% (95%CI: 0.61 - 0.92) in the RT and 49% (95% CI: 0.38 - 0.64) in the ST group (p=0.02, log-rank test). Within the ST group, those who received concurrent RT (n=22) had near identical 12-month PFS when compared to the remaining patients in the ST group (n=41) - 12-month PFS 53% (95% CI: 0.35 - 0.79) versus 53% (95% CI: 0.37 - 0.75). On univariable CoxPH modelling, elevated LDH (Hazard Ratio (HR) 2.29, 95% CI 1.22 - 4.29) and TMTV continuous per 100ml (HR 1.10, 95% CI 1.05 - 1.15) were negatively associated with PFS at the pre-LD time point. This association was not seen at apheresis (elevated LDH HR 1.15 (95% CI 0.61 - 2.14) and TMTV HR 1.03 (95% C1 0.97 - 1.08)) suggesting that the impact of these risk factors is modifiable with BT. On multivariable CoxPH modelling of PFS adjusted for pre-LD ECOG, LDH, TMTV, and stage, RT alone in comparison to ST was associated with improved PFS (HR 0.41, 95% CI 0.19 - 0.88). Rates of grade 3 or above CAR T-cell associated toxicity were similar between groups, namely cytokine release syndrome n=2, 2% (RT n=1, 3%; ST n=1, 2%), immune effector cell associated neurotoxicity syndrome n=15, 15% (RT n=6, 17%; ST n=9, 14%), month 3 neutropenia n=13, 17% (RT n=7, 21%; ST n=6, 13%) and thrombocytopenia n=7, 9% (RT n=1, 3%; ST n=6, 15%). Conclusion: Bridging with radiotherapy-alone to axi-cel was associated with improved CR rates and PFS for LBCL in third-line and beyond, over systemic therapy-based BT. Multivariable analysis, adjusted for predictors of poor PFS, suggests the effect is not simply explained by better disease control at infusion. In this retrospective analysis it was not possible to separate the impact of disease distribution that is amenable to radiotherapy from the impact of radiotherapy itself. However, given the favourable outcomes seen, radiotherapy should be considered as part of the bridging strategy wherever feasible.