AUTHORS
Jing Ruan, DO, Dao-Bin Zhou2, Yan Zhang3, Danqing Zhao4, Chong Wei5, Ke Hu6, Fuquan Zhang7, Xiaorong Hou8, Jian Li, MD9, Wei Zhang, MD10
AFFILIATIONS
1 Peking Union Medical College Hospital, Beijing, China
2 Peking Union Medical College Hospital, Beijing, China
3 Peking Union Medical College Hospital, Beijing, China
4 Peking Union Medical College Hospital, Beijing, China
5 Peking Union Medical College Hospital, Beijing, China
6 Peking Union Medical College Hospital, Beijing, China
7 Peking Union Medical College Hospital, Beijing, China
8 Peking Union Medical College Hospital, Beijing, China
9 Peking Union Medical College Hospital, Beijing, China
10 Peking Union Medical College Hospital, Beijing, China
ABSTRACT
Object:
Radiotherapy is a promising strategy for bridging CAR-T therapy while the best regimen remains unclear. Hyper-fractionated radiotherapy (HFRT), different from regular RT, characterized by lower fractional doses and higher treatment frequency, may preserve tumor vasculature and exert immunomodulatory effect. Our preliminary data proved the efficacy and safety of HFRT as a bridging strategy for CAR-T. This study reported an expanded cohort with extended follow-up duration, to investigate the prognosis of bridging HFRT in combination with commercial CD19 CAR-T therapy for R/R DLBCL. The potential mechanisms were also explored.
Methods:
This was a single-center, prospective clinical study (NCT05514327) enrolling R/R DLBCL patients with measurable lesions. HFRT was administered following T-cell apheresis. A maximum of 3 lesion sites was irradiated per patient. A total dose of 30 Gy was prescribed to all patients with 1.5 Gy twice daily over 10 treatment days. Patients with central nervous system (CNS) involvement received whole brain radiotherapy (WBRT). Commercial CAR-T cell infusion started within 1-2 weeks after the completion of radiotherapy. The efficacy and adverse reactions were monitored. Immune cell subsets before and after radiotherapy were analyzed. OCI-LY1 subcutaneous xenograft tumor model was established in NSG mice with PBMC infusion. Transcriptome of the tumor tissues were compared after HFRT and regular RT.
Results: Between June 2022 and Sep 2025, 30 R/R DLBCL patients have been enrolled in our study. The median age was 55.5(27-76) and 22(73.3%) of them were males. 12 (40%) were germinal center (GCB) type, 14 (46.7%) were double expressor, and 6 (20%) had TP53 mutations. The international prognostic index scores (IPI) were 1-2 in 17 (56.7%) patients and 3-5 in 13 (43.3%) patients. 15 (50%) of them were primary refractory. The median number of prior lines of therapy was 2 (1-5). At apheresis, 6 (20%) had bulky diseases (≥7.5cm tumor mass). Focal radiotherapy was delivered in 14 patients (46.7%), and comprehensive regional radiotherapy in 16 patients (53.3%). The median time of apheresis to CAR-T infusion was 32 (21-58) days while the time interval between the completion of radiotherapy and CAR-T cell infusion was 7-33days (median 12 days). All of them achieved partial response (PR) after the completion of hyper-fractionated radiotherapy. 13 (43.3%) patients had grade 1 cytokine release syndrome (CRS), 9 (30%) had grade 2 CRS and 4 (13.3%) had grade 3 CRS. 6 (20%) ICANS were recorded, 2 (6.7%) were grade 1, 2 (6.7%) were grade 3 and 2 (6.7%) was grade 4. 27 (90%) had grade 3-4 granulocytopenia and the median duration time was 5.5 (1-16) days. Till submission, the median follow-up time was 17 (1-44) months. At 1-month follow-up, 10/30 (63.3%) patients achieved complete remission (CR), 5/30 (16.7%) had PR. The CR rate reached 22/30 (73.3%) at 6-month follow-up. The estimated 2-year PFS was 62.9% and the 2-year OS was 72.2%. For the 10 patients with CNS involvement, the CR at 3-month was 80%. These 8 patients have maintained CR to date, among whom seven have been followed up for more than 1 year. Patients with TP53 mutations had lower 1-year OS (25% vs 83.3%, p=0.013). Cox analysis showed that the ratio of higher exhausted T cells (CD3+PD1+TIM3+T and CD8+PD1+T) before CAR-T infusion were associated with inferior PFS and OS. Patients with CD8+PD1+T ratio>8.5% had significantly lower 1-year PFS (0.125 vs 1, p<0.001) and 1-year OS (0.333 vs 1, p=0.006). Notably, hyper-fractionated radiotherapy significantly reduced both the absolute count (135×106/L vs. 39×106/L, p=0.030) and proportion (15.8% vs. 10.0%, p=0.015) of CD8?PD-1? T cells. No significant differences were observed in other immune cell populations. Transcriptome of the tumor tissues were compared after HFRT and regular RT in humanized mice. Gene expressions related to T-cell exhaustion including TIGIT, CTLA1, PDCD1(PD1), LAG3 and LGALS9 were significantly decreased in mice received HFRT.
Conclusion:
HFRT is a promising bridging strategy to CAR-T therapy in DLBCL, especially in patients with CNS involvement. It may act synergistically with CAR-T therapy by remodeling the exhausted microenvironment, thereby facilitating durable long-term remission in patients with R/R DLBCL.
DISCLOSURES
Jing Ruan, DO: No relevant conflicts of interest to declare.