Outcomes with bridging radiation therapy prior to chimeric antigen receptor T-cell therapy in patients with aggressive large B-cell lymphomas

Abstract

Background: Select patients with relapsed/refractory aggressive B cell lymphoma may benefit from bridging radiation (bRT) prior to anti-CD19-directed chimeric antigen receptor T cell therapy (CAR-T). Here, we examined patient and treatment factors associated with outcomes and patterns of failure after bRT and CAR-T.

Methods: We retrospectively reviewed adults with diffuse large B-cell lymphoma (DLBCL) who received bRT prior to axicabtagene ciloleucel, tisagenlecleucel, or lisocabtagene maraleucel between 11/2017-4/2023. Clinical/treatment characteristics, response, and toxicity were extracted. Survival was modeled using Kaplan-Meier or Cox regression models for events distributed over time, or binary logistic regression for disease response. Fisher's Exact Test or Mann-Whitney U methods were used.

Results: Of 51 patients, 25.5% had bulky disease and 64.7% had Stage III/IV disease at the time of RT. Comprehensive bRT alone to all disease sites was delivered to 51% of patients, and 29.4% were additionally bridged with systemic therapy. Median follow-up was 10.3 months (95% CI: 7.7-16.4). Overall response rate (ORR) was 82.4% at 30 days post-CAR-T infusion. Median overall survival (OS) was 22.1 months (6.6-not reached) and the median progression-free survival (PFS) was 7.4 months (5.5-30). OS/PFS were 80% (66-99)/78% (64-87) at 1-year, and 59% (44-71)/54% (40-67) at 2-years, respectively. Comprehensive RT to all sites of disease correlated with improved PFS and OS, p ≤ 0.04. Additionally, ECOG ≥2 and Stage III/IV disease predicted poor OS (p ≤ 0.02). Disease bulk, IPI ≥3, and non-GCB histology were poor predictors for disease-specific survival (DSS), p<0.05. The latter two, as well as bRT dose of ≤30 Gy predicted worse PFS (p<0.05). Among patients with advanced stage disease, comprehensive bRT to all sites of disease (n=10) was not associated with improved OS and PFS compared to focal bRT (n=23), p>0.17. No difference was seen in bridging RT vs. chemoRT. Twenty-six patients developed relapse (50.9%), of which 46% was in-field. Risk of in-field relapse correlated with bulky disease (OR=7, 95% CI: 1.2-41, p=0.03) and lack of response at 30 day post-CAR-T evaluation (OR=16.8, 95% CI: 1.6-176, p=0.02), but not with bRT dose (p=0.27).

Conclusion: bRT and CART is a good treatment strategy for select patients with aggressive B cell lymphoma. Comprehensive bRT including all sites of disease is associated with improved outcomes.

Keywords: CAR-T; DLBCL; bridging RT; chimeric antigen receptor; radiation therapy.

Figure 1

Patient outcomes post-CAR-T infusion. Swimmer’s plot depicting the course of individual patients following CAR-T infusion (A) for the full study cohort (n=51). The median OS was 10.3 months (B), median PFS was 7.4 months (C), and median DSS was 8.9 months (D).

Figure 2

Incidence and nature of relapse and relationship with dose or extent of RT. Relapse was evident in n=26 patients (51%), with type of relapse shown in proportionally-accurate diagrams (A). RT dose is plotted for individual patients, categorized by whether they experienced any relapse (B). Only among patients who developed relapse, RT dose was plotted against emergence of in-field vs. distant relapse (C). Among patients who relapsed following receipt of comprehensive bRT to all sites of disease, the category of relapse is depicted (D). The incidence of relapse in patients who received comprehensive or focal bRT is also shown (E).

Figure 3

Univariate analysis of disease- and patient-related factors associated with survival. OS (first column), PFS (second column), and DSS (third column) were computed for patients who got bridging RT or chemoradiation prior to CAR-T cell therapy. Patients were stratified by patient or disease-related factors including tumor bulk (A), IPI score (B), disease stage (C), and ECOG status (D), with * denoting p<0.05.

Figure 4

Univariate analysis of treatment-related factors associated with survival. OS (first column), PFS (second column), and DSS (third column) are depicted with treatment-related factors, such as disease category as localized vs. extensive (A), receipt of bRT comprehensively to all sites of disease or to focal areas of disease (B), bridging strategy (C), and bRT dose (D), with * denoting p<0.05.

Figure 5

Univariate analysis of survival stratified by comprehensiveness of bRT fields in patients with advanced stage disease. OS (first column), PFS (second column), and DSS (third column) are depicted comparing receipt of bRT comprehensively to all sites of disease or to focal areas of disease in patients with advanced stage disease.