Combining CAR-T therapy with radiotherapy or not in refractory/relapsed diffuse large B-cell lymphoma: A comparative study

Highlights

• •BRT group showed obviously longer PFS and OS than no BRT group.
• •Comprehensive BRT subgroup improved prognosis in PFS and OS when compared with focal BRT subgroup.
• •High-EQD2 subgroup did not improve PFS or OS, except for local control in patients with high tumor burden.

Abstract

Background

Radiotherapy and Chimeric Antigen Receptor(CAR)-T therapy may exhibit a synergistic effect, suggesting that incorporating radiotherapy into CAR-T could improve the prognosis for patients with refractory/relapsed diffuse large B-cell lymphoma (R/R DCBCL). A lack of standardized treatment protocols and relevant guidelines in bridging radiotherapy(BRT) prior to CAR-T therapy still exists. Consequently, we retrospectively analyzed the outcomes of R/R DLBCL patients treated with BRT prior to CAR-T therapy or not, aiming to evaluate the efficacy and satety of BRT as well as the impact of radiotherapy dose on prognosis.

Methods

Between December 2017 and January 2025, 80 patients diagnosed with R/R DLBCL were treated with CAR-T. Thirty-five of them received BRT during leukapheresis and lymphodepletion. The primary endpoint of this study was progression-free survival(PFS), and secondary endpoints included overall survival(OS), disease-specific survival(DSS), in-field PFS, best objective response rate(ORR), and complete response rate(CRR). PFS and OS of CAR-T were compared between BRT group and no BRT group. In the subgroup of radiotherapy patients, PFS, OS and in-field PFS were compared between the low-Equivalent dose to 2 Gy per fraction(EQD₂) subgroup and the high-EQD₂ subgroup.

Results

BRT group showed obviously longer PFS and OS than no BRT group(p = 0.001, p = 0.043). In addition, BRT did not increase the incidence of CAR-T toxicities during follow-up (median:35.27 months). Comprehensive BRT subgroup improved prognosis in PFS(p = 0.015) and OS(p = 0.029) when compared with focal BRT subgroup, no significant effect on DSS was noted(p = 0.109). High-EQD2 subgroup did not significantly improve PFS(p = 0.181) and OS(p = 0.665) except for local control(p = 0.079) especially in patients with high tumor burden(p = 0.005). There is no impact on prognosis between early salvage radiotherapy(SRT) and salvage chemotherapy(SCT) cohorts in patients with PR response to CAR-T therapy.

Conclusions

Our analysis demonstrated that BRT is an effective and safe approach for patients with R/R DLBCL preparing for CAR T-cell therapy, which indicated that BRT may enhance the anti-tumor effect of CAR T-cells.

Introduction

DLBCL is the most common non-Hodgkin lymphoma (NHL) subtype. Although most patients are sensitive to first-line immunotherapy and chemotherapy treatment, about 30–40 % of patients are ineffective with this regimen or experience recurrence in the later stages [1]. Most of R/R DBCLC after chemotherapy face a dismal outcome, with their median overall survival of 6.3 months, a 2-year OS rate of 20 %, and most of them die due to disease progression [2]. With the boom in CAR-T in recent years, the treatment paradigm for R/R DLBCL patients has changed significantly. Three CD19 CAR T-cell therapies—axicabtagene ciloleucel, tisagenlecleucel, lisocabtagene maraleucel have been approved for use in adult patients who have failed at least two lines of therapy. CAR T-cell therapy delivers encouraging results with ORR of 52–82 % in R/R DCBCL patients [[3], [4], [5]]. However, only about 40 % of them achieve long-term remission from CAR-T. Patients who relapse or progress after CAR-T face a poor prognosis [6]. One of the attempts to improve the prognosis is to bridge multiple treatments such as radiotherapy, chemotherapy, targeted therapy and hormones between T-cell collection and lymphodeletion [[7], [8], [9]]. However, there is no clear evidence on which bridging therapy is more beneficial to improve CAR T-cell therapy efficacy due to the paucity of relevant studies.

Radiotherapy has been shown to enhance CAR-T efficacy by mediating cytotoxic effects via the TRAIL pathway, inducing target antigen expression and releasing cytokines to enhance CAR-T efficacy. It breaks down physical barriers, improves CAR-T cell migration and infiltration, and remodels the tumor microenvironment to promote adaptive immunity [[10], [11], [12]]. Previous studies have indicated that patients who underwent bridging radiotherapy(BRT) had longer local lesion control times and lower local relapse rates [13]. Zhu [14] et al. and Qu [15] et al. found that DLBCL patients with high tumor burden treated with BRT had a higher ORR, CRR and a lower incidence of high grade cytokine release syndrome (CRS) and immune effector cell-associated neurotoxicity syndrome (ICANS) than those treated with bridging chemotherapy. Based on the above small cohort analyses, BRT has a synergistic effect on CAR-T therapy and has the potential to improve patient prognosis and overall survival, which has attracted increasing attention to BRT in CAR-T treatment. Hypofractionated radiotherapy(5*5 Gy) triggers immune cell recruitment and reshapes the tumor immune microenvironment, facilitating CAR-T cell infiltration into tumors [16]. Owing to the absence of prospective analyses and limited retrospective data about BRT, there is disagreement among institutions regarding the effect of focal versus comprehensive BRT on prognosis of patients with R/R DCBCL [17,18]. BRT makes reducing the total radiation dose a logical consideration as it serves solely as an immune primer in CAR-T therapy. Recent limited cohort studies indicate that reduced radiation doses correlate with an increased risk of intra-field recurrence [18]. Additionally, radiotherapy is applied as early salvage therapy following CAR-T therapy. Early SRT has been shown to achieve CR in R/R DLBCL patients with residual disease post CAR-T therapy, linked to elevated CAR-T cell copies in vivo [15,19].

Prior studies on combining BRT and CAR-T were limited to small retrospective cohorts, we retrospectively analyzed a cohort of R/R DLBCL patients who received BRT prior to CAR T-cell therapy or not in our institution, aimed to provide additional data on this approach and examine whether early SRT improves outcomes in R/R DCBCL patients with PR responses to CAR-T.

Patients and methods

A total of 80 patients diagnosed with R/R DLBCL were treated with CD19 CAR T-cell infusion between December 2017 and January 2025 at Shanghai Tongji Hospital. We subdivided the cohort into 2 groups: BRT group and no BRT group. All procedures were performed in accordance with relevant guidelines and regulations, and the study was approved by our institutional review board. All participants gave informed consent. The study was also approved by the Ethics Committee of Shanghai Tongji Hospital. Inclusion criteria: age ≥ 16 years; diagnosis of DCBCL by histopathology; relapse or progression in all patients after receiving at least two lines of treatment in complete or partial remission; no other synchronous regimen during radiotherapy; exclusion criteria: R/R DCBCL combined with other malignant tumors;≥Ⅲ myelosuppression; be in a coma of consciousness.

BRT, response evaluation, toxicity and follow-up

BRT was defined as radiotherapy between leukapheresis and lymphodepletion. We divided the cohort into BRT and No-BRT groups. The BRT group was further stratified into high-EQD₂ (>20 Gy) and low-EQD₂ (≤20 Gy) subgroups. Comprehensive BRT was defined as radiation therapy that includes all active lesions identified by Positron Emission Tomography − Computed Tomography(PET-CT) prior to leukapheresis. Focal radiotherapy was defined as the partial active lesions requiring radiation therapy, as determined through multidisciplinary discussion with a hematologist and an oncologist. The efficacy of the lesions in the radiation field was evaluated by Computed Tomography(CT) 4 weeks after the end of the BRT. PET-CT examination was performed 1 month after CAR-T cell infusion to evaluate the therapeutic response, and the efficacy was evaluated using Lugano 2014 criteria. SRT/SCT was defined as patients assessed as PR through PET-CT one month after CAR-T cell infusion treated with radiotherapy/chemotherapy. Progression-free survival(PFS) was defined as the time from the completion of CAR-T infusion to disease progression, recurrence, death, or the last follow-up. Overall survival(OS) was defined as the time from the completion of CAR-T infusion to death or the final follow-up. Disease-specific survival(DSS) was defined as the time from the completion of CAR-T infusion to death solely due to R/R DCBCL. In-field-PFS was defined as the time from the patient's end of radiotherapy to the time of in-field lesion progression or follow-up cut-off. The definitions of complete response(CR), partial response(PR), stable disease(SD), progressive disease(PD) are in reference to the Lugano 2014 criteria. ORR was defined as sum of the proportions of CR and PR. Complete Response Rate(CRR) was defined as the proportion of CR. The primary endpoint of this study was PFS and secondary endpoints included OS, in-field-PFS, best ORR and CRR. Adverse effects of radiotherapy were determined by the CTCAE 5.0 grading criteria [20]. CAR-T therapy cytotoxicity was graded according to the consensus grading system of the American Society for Transplantation and Cellular Therapy (ASTCT) [21]. Large masses were localized as masses with a maximum diameter of > 7 cm. Age, tumor stage, general condition (ECOG score), lactic dehydrogenase (LDH), international prognostic index(IPI) score, bulky disease status and were documented before leukapheresis when patients were confirmed to receive CAR-T therapy.

Statistical analysis

Statistical analysis was performed using SPSS version 26.0 (IBM Corporation, Armonk, NY, USA). The estimation of OS, PFS and in-field-PFS was performed via Kaplan-Meier survival curves, with group differences analyzed through log-rank tests and chi-square tests. Potential factors associated with these indicators were identified through univariate and multivariate Cox proportional hazard regression models. If the indicator in the univariate analysis is less than or equal to 0.1, it is included in the multivariate analysis. A p-value of less than 0.05 was considered statistically significant.

Results

Our retrospective analysis included a total of 80 patients diagnosed with R/R DLBCL. A total of 35 patients received BRT before lymphodepletion, and the remaining 45 individuals did not undergo BRT. The demographic details of patients diagnosed with R/R DCBCL were categorized into two groups and presented in Table 1. For the entire cohort, the median follow-up period was 32.27 months(range: 0.6–68.27 months). Thirty patients were documented with disease-specific mortality, one patient died of cerebral hemorrhage, one died of acute heart failure, and one died of pulmonary embolism. The estimated median PFS was 6.53 months (95 % CI: 0–15.243 months), estimated median OS was 22.07 months (95 % CI: 7.483–36.657 months), estimated median DSS was 29.50 months (95 % CI: 13.395–45.065 months). The BRT group showed superior PFS, OS and DSS outcomes when contrasted with non-bridging treatment strategies(Fig. 1).

Table 1.

Baseline characteristics of patients in BRT and no BRT group.

	ALL	BRT	No BRT	P-value
	N = 80	N = 35	N = 45
Gender
Male	48(60.0 %)	20(57.1 %)	28(62.2 %)	0.645
Female	32(40.0 %)	15(42.9 %)	17(37.8 %)
Age
＞60	27(33.8 %)	12(34.3 %)	15(33.3 %)	0.929
≤60	53(66.3 %)	23(65.7 %)	30(66.7 %)
Stage
I-II	7(8.8 %)	4(11.4 %)	3(6.7 %)	0.727
III-IV	73(91.3 %)	31(88.6 %)	42(93.3 %)
Extranodal lesions
Yes	75(93.8 %)	33(94.3 %)	42(93.3 %)	1
No	5(6.3 %)	2(5.7 %)	3(6.7 %)
Elevated LDH
Yes	42(52.5 %)	15(42.9 %)	27(60.0 %)	0.194
No	38(47.5 %)	20(57.1 %)	18(40.0 %)
ECOG
0–1	74(92.5 %)	30(85.7 %)	44(97.8 %)	0.109
02-Apr	6(7.5 %)	5(14.3 %)	1(2.2 %)
IPI score
<3	28(35.0 %)	15(42.9 %)	13(28.9 %)	0.194
≥3	52(65.0 %)	20(57.1 %)	32(71.1 %)
Subtype
GCB	27(33.8 %)	12(34.3 %)	15(33.3 %)	0.929
Non-GCB	53(66.3 %)	23(65.7 %)	30(66.7 %)
Prior lines of therapy
≤3	5(6.3 %)	3(8.6 %)	2(4.4 %)	0.771
>3	75(93.8 %)	32(91.4 %)	43(95.6 %)
Bulky disease
≥7cm	41(51.2 %)	22(62.9 %)	19(42.2 %)	0.067
＜7cm	39(48.8 %)	13(37.1 %)	26(57.8 %)

BRT: bridging radiotherapy; LDH: lactic dehydrogenase; IPI: International Prognostic Index; GCB: Germinal Center B-cell.

Fig. 1.

Patient outcomes in BRT and No BRT group.

The median EQD2 of the BRT cohort was 30 Gy (range: 10–50.05 Gy), with a median fraction dose of 2 Gy (range: 1.75–5 Gy). Of the 35 patients in the BRT group, 13 were treated with an EQD₂ dose less than 20 Gy, 20 patients received radiotherapy for multiple active sites. Comprehensive radiotherapy was administered to seventeen patients, and focal radiation therapy to another eighteen patients. The main focus radiotherapy site is the abdomen (n = 22), followed by the head and neck region (n = 16), bone (n = 9), pelvic cavity (n = 5), chest (n = 3), and soft tissues (n = 2). The incidence of acute side effects in BRT is low and slight, with the following rates: fatigue(n = 2), oral mucosal reaction(n = 4), radiation Pneumonitis(n = 2), neutropenia(n = 5), edema(n = 1). CT was re-examined at the end of BRT, no progression was observed before CAR T-cell infusion, indicating that radiotherapy has significant dominance in controlling local lesions.

The occurrence of high grade (>grade 3) toxicities related to CAR-T therapy showed comparable rates between BRT and no BRT cohort, which means radiation did not worsen CAR-T toxicities. It is worth noting that 2 patients who received brain BRT both developed high-grade ICANS. After successful CAR T-cell infusion, best response was recorded, including n = 23 CR, n = 29 PR, n = 14 SD, n = 4 PD, with a CRR of 28.75 % and an ORR of 65 %. There was a p-value between the two cohorts in the response to CAR-T therapy(p = 0.047)(Table 2).

Table 2.

Toxicities and therapy responses of CAR T in BRT and no BRT group.

Toxicity	ALL	BRT	No BRT	p value
	N = 80	N = 35	N = 45
CRS
≤Grade 2	71(88.8 %)	31(88.6 %)	40(88.9 %)	1
>Grade 2	9(11.3 %)	4(11.4 %)	5(11.1 %)
ICANS
≤Grade 2	76(95.0 %)	33(94.3 %)	43(95.6 %)	1
>Grade 2	4(5.0 %)	2(5.7 %)	2(4.4 %)
Best responses of CAR T
CR	23(28.75 %)	14(40 %)	9(20 %)	0.047
PR	29(36.25 %)	12(34.29 %)	17(37.78 %)
SD	14(17.5 %)	7(20 %)	7(15.56 %)
PD	14(17.5 %)	2(5.71 %)	12(26.67 %)

BRT: bridging radiotherapy; CRS:cytokine release syndrome; ICANS:immune effector cell-associated neurotoxicity syndrome;

In an effort to determine suitable radiotherapy doses for combination therapies, dose-volume analysis was implemented in the BRT subgroup, despite its limited cohort size (Fig. 2). Among the 35 R/R DLBCL patients receiving BRT, 22 patients received a high EQD₂ > 20 Gy, while the remaining patients received ≤ 20 Gy, allowing us to categorize them into two cohorts. The finding showed that increasing the intensity of BRT in synergistic treatment strategies might not result in improved clinical outcomes(PFS: p = 0.181; OS: p = 0.665; DSS: p = 0.971). As expected, the high-dose group may exhibited better local control rates within the in-field PFS(p = 0.079), the difference was not significant. Among the subgroup receiving BRT, 22 patients had bulky disease sites irradiated. In this population, higher EDQ2 was associated with better local control(p = 0.005). Another key issue we focused on was the selection of target area in BRT. Seventeen patients who received comprehensive BRT demonstrated improved prognosis in PFS (p = 0.015) and OS (p = 0.029), no significant effect on DSS was noted (p = 0.109) (Fig. 3).

Fig. 2.

Comparison of high and low doses subgroup in terms of prognosis.

Fig. 3.

Comparison of comprehensive and focal target area in terms of prognosis.

Of the 29 patients achieving PR response to CAR-T therapy, 9 underwent early SRT for residual disease, 13 chose SCT, 2 died prior to treatment initiation, and 5 had no further treatment records. Unfortunately, early SRT has not demonstrated a more significant advantage compared to early SCT (Fig. 4). Among patients with SD or PD response to CAR-T therapy, 12 also received SRT.

Fig. 4.

SRT and SCT treatment of patients with PR response to CAR-T therapy.

In univariate analysis, the prognostic factors for poorer PFS in all patients were elevated LDH (p = 0.001), ECOG > 1(p = 0.001), IPI score ≥ 3 (p = 0.003), No BRT (p = 0.001), SD/PD response to CAR-T(p = 0.000). Moreover, ECOG > 1(p = 0.000), elevated LDH (p = 0.000), IPI score ≥ 3(p = 0.000), bulky disease(p = 0.005), No BRT(p = 0.043) and SD/PD response to CAR-T(p = 0.000) were connected with a worse OS. In the BRT group, univariate analysis revealed a correlation between poorer PFS and IPI score ≥ 3(p = 0.008), ECOG > 1(p = 0.009), elevated LDH (p = 0.009), SD/PD response to CAR-T(p = 0.000), focal BRT(p = 0.015). For patients in the No-BRT group, univariate analysis indicated that poorer PFS was related to elevated LDH(p = 0.003), ECOG > 1(p = 0.003), IPI score ≥ 3(p = 0.010), SD/PD response to CAR-T(p = 0.001). Multivariable analysis was shown in Table 3.

Table 3.

Multivariable analysis of factors associated with poorer PFS and OS.

	All	BRT group	No BRT group
Factors affecting PFS
Elevated LDH	0.011	0.256	0.048
IPI score ≥ 3	0.388	0.119	0.800
Bulky disease	0.236	0.851	0.516
No BRT	0.076	−	−
SD/PD to CAR-T	0.000	0.000	0.023
Focal BRT	−	0.027	−
Factors affecting OS
Elevated LDH	0.005	0.053	0.056
IPI score ≥ 3	0.585	0.691	0.099
Bulky disease	0.033	0.078	0.201
No BRT	0.462	−	−
SD/PD to CAR-T	0.017	0.045	0.004
Focal BRT	−	0.550	−

BRT: bridging radiotherapy; LDH: lactic dehydrogenase; IPI: International Prognostic Index; SD: stable disease; PD: progressive disease;

Discussion

Our retrospective analysis on radiotherapy pre-CAR-T demonstrates that BRT, particularly comprehensive BRT can improve the prognosis of R/R DCBCL patients by prolonging their PFS and DSS without compromising CAR-T raleted cytotoxicity or therapeutic efficacy. BRT plays a crucial role in allowing patients with high tumor burden, vital organ involvement, or symptoms requiring urgent relief to successfully transition to CAR T-cell infusion [22]. Additionally, radiotherapy is a well-tolerated and safe bridging treatment that alleviates symptoms and promotes lymphodepletion without affecting the in vivo expansion of CAR-T cells [23]. BRT does not exacerbate CAR-T-specific toxicities and is associated with a lower incidence of CRS [[24], [25], [26], [27], [28]]. Even in a retrospective central nervous system BRT by Cederquist et al, only 1/12 patients experienced grade > 2 CRS, and 3/12 patients experienced Grade > 2 ICANS [29]. However, some studies have pointed out the opposite view that radiotherapy can cause damage to normal tissue within the irradiated area, especially in the case of brain tumors, where radiotherapy may exacerbate CAR-T-related neurotoxicity [30]. The enhanced local control provided by radiotherapy, or the increased tumor-killing efficacy of CAR-T cells during the synergistic process, might be responsible for this [12].

There is currently no established guideline for determining the radiation dose for bridging radiation therapy when used in conjunction with CAR-T therapy. Higher doses of BRT often mean longer radiotherapy intervals, which may delay CAR-T cell infusion. However, lower doses of BRT allow some lesions to continue progressing shortly after CAR-T therapy, presenting a challenging dilemma. In the management of R/R DCBCL lymphoma patients, radiotherapy should serve as an adjunctive rather than the main treatment approach. As our multivariate analysis indicated, for R/R DCBCL patients, the response to CAR-T therapy was the most significant factor affecting prognosis. The focus should be on improving the success rate of CAR-T therapy in these patients. Radiotherapy can potentially enhance CAR-T efficacy by modifying the tumor immune microenvironment in various direct or indirect ways, an effect that bridging chemotherapy cannot replicate.

Swift BRT not only saves economic costs for patients and facilitates a rapid transition to CAR-T therapy, even Ruan [31] et al. attempted 1.5 Gy twice daily for 10 days before CAR-T cell infusion, but also minimizes acute toxic side effects, such as gastrointestinal perforation which was reported as a relatively common and fatal complication in CAR-T [32]. DeSelm [12] et al. noted that exposing tumors to low-dose radiation can sensitize tumor cells, even those that do not express the relevant target antigen, to CAR-T-cytotoxicity, thereby reducing the recurrence of antigen-negative tumors. A fractionated dose ≥ 3 Gy is the only parameter confirmed to be associated with the abscopal effect, with a single high-dose exposure also facilitating dendritic cell homing and T-cell activation [33]. Our research indicated that there was no significant statistical difference in PFS and OS between patients in different dose groups of BRT (EQD2 > 20 Gy v.s EQD2 ≤ 20 Gy). Recent studies have also indicated that the comparison of OS between the high-dose BRT group (BED₁₀ > 30 Gy) and the low-dose group (BED₁₀ ≤ 30 Gy) reveals no significant differences, although there might be a beneficial effect on PFS [18,34]. However, in the subgroup of patients with bulky disease, higher radiation doses were associated with PFS in patients, as noted in Saifi's [35] retrospective analysis, patients with bulky disease, high SUV max, elevated LDH, and extranodal involvement were more likely to fail locally. Numerous studies have shown a positive correlation between high tumor burden and the incidence of high grade CRS and ICANS following CAR T-cell therapy [[36], [37], [38]]. This finding requires further data to support, and we plan to conduct more in-depth research on the total dose of bridging radiotherapy for different sites.

Some researchers have noted that including all active lesions in the bridging radiotherapy target area is associated with a longer PFS trend [17,18]. This might be due to the fact that patients who can receive comprehensive radiotherapy typically have better baseline conditions than those who receive focal radiotherapy. However, it is reasonable to assume that comprehensive radiotherapy can enhance the infiltration of CAR-T cells into each active lesion. Therefore, some scholars [39] have tried a BRT approach, which involves administering Dt 20–30 Gy/5-15F to areas requiring long-term control, while all other active lesions receive Dt 4 Gy/2F. It may be applicable to R/R DCBCL patients with extensive lesions. Multiple prospective clinical trials are currently being conducted to investigate different total radiation doses, fraction doses, and target areas(Supplemental Table 1).

Patients experiencing progression after CAR-T therapy face a critical gap in standard treatment options, contributing to poor long-term results [40]. We placed greater emphasis on the subsequent treatment of patients who achieved a PR to CAR-T therapy, as for this group of patients, follow-up consolidation treatment might lead to better outcomes. Although some prior studies suggested that SRT following CAR-T therapy improves survival outcomes, our retrospective analysis failed to replicate these findings, potentially due to inconsistencies in baseline clinical characteristics among patients and the small sample size of those studies [41,42]. SRT is more suitable as a salvage treatment for these patients as it has fewer side effects on hematopoietic function when SRT and SCT have no impact on patient prognosis.

The primary value of this article is to clarify the feasibility of combined radiotherapy and CAR-T therapy in R/R DLBCL patients and to summarize the current clinical application outcomes. Based on current evidence, we propose that a comprehensive BRT target area characterized by high fraction dose, low total-dose administration may represent the most appropriate BRT strategy. We plan to explore optimal bridging parameters through expanded radiation cohort studies.

The single-center retrospective study to evaluate whether R/R DCBCL patients received CAR-T treatment could benefit from BRT faced multiple challenges, such as a small sample size, lack of prospective data, absence of long-term follow-up data, inconsistent follow-up time, inability to randomly assign patients, challenges in controlling confounding factors, patient selection bias, etc. This will indeed have a negative impact on the reliability and generalability of our findings. We have strictly screened during the process of sample inclusion, paid much attention to the diversity of baseline characteristics of R/R DCBCL patients and conducted multivariate analysis to partially counterbalance single-center/small-sample limitations. Over the past few months, the number of patients receiving BRT in our center has steadily increased. In the future, we plan to conduct multi-center prospective and randomized studies to identify treatment parameters in BRT. We believe that combination therapy will ultimately demonstrate superior tumor suppression and survival benefits compared to monotherapy.

Conclusion

BRT prior to CAR T-cell therapy is a safe and effective treatment strategy for R/R DLBCL patients. BRT group improved prognosis in PFS, OS and DSS when compared with No BRT group. Comprehensive BRT demonstrated improved prognosis in PFS and OS, but increasing the dose of BRT can not. For PR patients post CAR-T therapy, it may be necessary to find the best combination therapy plan.

Consent for publication

Not applicable.

CRediT authorship contribution statement

Yun Yang: Conceptualization, Data curation, Formal analysis, Project administration, Software, Writing – original draft. Bichun Xu: Data curation, Formal analysis, Software, Supervision, Validation, Writing – review & editing. Hong Zhu: Supervision, Validation, Writing – review & editing. Weikai Sun: Supervision, Validation. Aibin Liang: Writing – review & editing. Judong Luo: Writing – review & editing.

Ethics approval and consent to participate

Prior to the enrollment and treatment, all patients signed an informed consent form and received thorough information from their oncologists regarding to the potential toxicity and benefits of radiotherapy and CAR T-cell therapy. This retrospective study was reviewed and approved by the Ethics Committees of Shanghai Tongji Hospital, it was conducted according to the Declaration of Helsinki.

Funding

This work is supported by Research Start-up Fund of Tongji Hospital(RCQD2407).

Availability of data and materials.

The data that support the findings of this study are not openly available due to reasons of sensitivity and are available from the corresponding author upon reasonable request. Data are located in controlled access data storage at the Shanghai Tongji Hospital.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following are the Supplementary data to this article: