Impact of Salvage Versus Palliative Dose Bridging Radiation Therapy on Local Control and Overall Survival in Patients with DLBCL Receiving CD19 CAR T Cell Therapy.

Abstract

In patients with relapsed/refractory diffuse large B-cell lymphoma (DLBCL), radiation therapy (RT) can be used to "bridge" patients during the period of chimeric antigen receptor T cell (CAR T) manufacturing. Although RT has been shown to improve local control (LC) in patients who receive RT versus do not, it is unknown whether there is an optimal dose. This was a retrospective study of patients with DLBCL who received bridging RT and CAR T cell therapy between 8/2021 and 12/2022 at a single institution (IRB #202103122). Patient, disease, and treatment characteristics were abstracted. PET/CT and laboratory data were collected at various time points. RT data including dose, planning target volume (PTV), and body volume receiving 2Gy and 4 Gy were tested for associations with LC, overall survival (OS), cytokine release syndrome (CRS) and immune effect cell-associated neurotoxicity syndrome (ICANS). Statistical tests were performed using Python libraries. Twenty-two patients with DLBCL were included for analysis. The median patient age was 68 (range 35-82) and the median prior lines of failed therapies was two. All patients had successful apheresis and subsequent CAR T manufacturing. No patients died before CAR T infusion. All patients had a pre-radiation PET/CT scan that showed Deauville 5 disease. Patients with limited stage disease received salvage dose RT (EQD2∼40Gy, n = 8) while patients with advanced disease received palliative dose RT (n = 14). The median PTV volume was 1332 cm3. With a median follow-up of 156 days (IQR 63 - 252 days) from CAR T infusion, 10 patients had died and six progressed at an RT target. A higher salvage RT dose was not associated with LC at the day 30 PET or day 90 PET time points. In a Cox regression model, no independent variable, including RT dose, was significantly associated with LC or OS, but post-RT residual target SUV trended toward an association with OS (HR 6.2, 95% CI 0.9 - 471.0, p = 0.06). Post-RT LDH was associated with CRS grade (ρ = -0.39, p = 0.05). Other non-significant associative trends were noted between CRS grade and post-RT absolute lymphocyte count (ALC) (ρ = - 0.33, p = 0.09) and residual target SUV max (ρ = - 0.37, p = 0.08) and between ICANS grade and pre- to post-RT LDH change (ρ = -0.38, p = 0.06) and post-RT lymphocyte count (ρ = 0.36, p = 0.07). In an exploratory analysis of body volume receiving integral doses of 4Gy or 2Gy and pre- and post-RT ALC change, it appeared that higher integral dose did lead to larger decreases in ALC (ρ = -0.32, p = 0.15 and ρ = -0.3, p = 0.18, respectively). In this small series of patients with r/r DLBCL treated with bridging RT prior to CAR T therapy, RT dose was not associated with LC. However, higher post-RT target residual SUV was associated with worse OS. The impact of post-RT LDH on CRS and ICANS grades should be explored further.