Abstract
Introduction
Historically, the management of relapsed or refractory diffuse large B-cell lymphoma (r/r-DLBCL) involved chemotherapy and autologous stem cell transplant, though outcomes were often suboptimal. Chimeric antigen receptor T-cell (CAR-T) therapy has transformed the therapeutic landscape for r/r-DLBCL, achieving high response rates and improving progression-free and overall survival. However, a significant proportion of patients relapse after CAR-T, and optimal treatment strategies for post-CAR-T relapse remain unclear. Radiotherapy (RT), a highly effective treatment for lymphoma, is increasingly recognized for its potential role as both a bridging therapy and a salvage option following CAR-T relapse.
Methods
A comprehensive literature review was conducted using databases including PubMed, Scopus, EMBASE, and Cochrane Library, with search terms combining “radiotherapy,” “radiation therapy,” “lymphoma,” and “CAR T-cell.” A total of 690 records were screened, and 14 studies were included in the analysis after applying inclusion and exclusion criteria.
Results
RT demonstrates high response rates in CAR-T relapsed DLBCL, with overall response rates (ORR) ranging from 35% to 82.4% and complete response rates (CRR) from 17% to 59%. One-year local control rates ranged between 62% and 84%. Salvage RT showed comparable or superior outcomes to systemic therapies in multiple studies, particularly in patients with localized relapses. The toxicity profile of RT was favorable, particularly when modern techniques such as IMRT were employed. Case reports and retrospective series highlighted its effectiveness in achieving durable responses and controlling localized disease progression.
Conclusions
Radiotherapy is a safe and effective treatment option for patients with DLBCL relapsed or refractory after CAR-T therapy. It achieves high local control rates and favorable outcomes, particularly in patients with localized relapse. Incorporating RT into the therapeutic workflow may enhance the management of this challenging population. Further prospective studies are needed to define its role and optimize treatment sequencing.
Introduction
Historically, the only therapeutic option for relapsing or refractory diffuse large B-cell lymphoma (r/r-DLBCL) was chemotherapy, followed by autologous stem cell transplant for eligible patients, though this often resulted in suboptimal outcomes. 1
The chimeric antigen receptor T-cell (CAR-T) therapy has revolutionized the treatment for r/r-DLBCL, providing high overall response rates (ORR) and complete response (CR) rates and improving progression free survival (PFS) and overall survival (OS).2,3 In CAR-T, autologous T-cells are extracted from the patient's blood, engineered by adding a chimeric antigen receptor and then re-infused in the patient: this procedure requires 3–6 weeks and, due to the potential of rapid disease progression, often a “bridging” therapy is required to contain the disease. 4 Radiotherapy (RT) has been increasingly adopted as a bridging strategy for patients undergoing CAR-T therapy, demonstrating optimal results in terms of disease control and the potential to reduce treatment-related toxicity compared with systemic bridging options.4-7 Despite high initial ORR, the majority of patients experience disease relapse, with only 20–50% maintaining a durable CR after CAR-T therapy at 5 years.2,3,8 Although several agents are approved for treating r/r-LBCL, the optimal sequencing of treatments before and after CAR-T remains unclear. Literature on outcomes following salvage therapy after CAR-T relapse is limited, generally reporting poor outcomes.9,10
Radiotherapy could help address this unmet clinical need, as it provides high rates of local response and disease control, even in chemoresistant lymphomas. 11 Additionally, it might exert a potential synergy with CAR-T by enhancing the antitumoral immune response.4,12,13 Moreover, the relatively low radiation dose required for lymphoma treatment results in optimal toxicity, which can be further mitigated by the adoption of modern techniques like Intensity Modulated Radiation Therapy (IMRT). 14 Nonetheless, indications for incorporating RT in treatment workflow for DLBCL that relapse after CAR-T are not yet well defined. In this paper we present a comprehensive review of the current literature on the use of RT in this challenging setting.
Methods
A review of the current literature was performed according to the recommendations of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement. The keywords “radiotherapy” or “radiation therapy” in combination with “lymphoma” and “CAR T-cell” were entered in multiple databases including PubMed, Scopus, EMBASE and Cochrane Library. Some articles were as well identified from the references of selected papers. Research was performed with no publication time restriction on October 11, 2024 by two independent reviewers; no conflicts or disagreements were identified between the two reviewers. A total of 690 records was retrieved and the first selection was based on the title and abstract, including only articles in English and excluding duplicates. The full text of the resulting 28 articles was analyzed and reviews with no original data and papers not specifying details of radiation therapy were discarded. Finally, 14 papers were selected and included in the review (Figure 1). As this is a review paper with no original data, approval by an Ethics/Review board was not required.

Workflow Adopted for Selecting the Records to Include in the Final Literature Review.
Literature Review
Current literature regarding the adoption of RT to treat DLBCL refractory or relapsing after CAR-T encompasses several retrospective series and case reports. The main studies on the topic are summarized in Table 1.
Main Studies Assessing Salvage RT After Post CAR-T Cell Therapy Relapse. DLBCL = Diffuse Large B-Cell Lymphoma; CAR-T = Chimeric Antigen Receptor T-Cell Therapy; CR = Complete Response; PR = Partial Response; SD = Stable Disease; PD = Progressive Disease; 1 yr LCR = one Year Local Control Rate.
Ababneh et al. 15 retrospectively analyzed a cohort of 120 patients with DLBCL relapsed post-CAR-T: 54 patients received RT to 93 sites as a salvage treatment, median dose/fractionation was 30 Gy in 10 fractions (range 4-50.4 Gy in 1-28 fractions). For the 81 assessable lesions, in-field response was as follows: CR 59%, PR 23%, SD 4%, PD 14%. One-year local control rate was 84% and median OS from RT was significantly higher among patients irradiated to all the sites of relapse versus focal RT (19.1 months vs 3.0 months; p < 0.001).
In the retrospective international multicentric study by Iacoboni et al., 16 237 DLBCL patients who progressed after CAR-T received salvage treatment, including 37 subjects that underwent RT as first subsequent therapy. In the RT group, the ORR was 45%, with a CR rate of 35%. PFS and OS at 12 months were 30.8% and 37.5%, respectively. These outcomes were superior to those achieved by patients who received lenalidomide, immune check points inhibitors (ICIs) and chemotherapy as first salvage regimen, remarkably among those relapsing 2–6 months after CAR-T infusion. On the other hand, patients who received polatuzumab-bendamustine-rituximab or bispecific antibodies as their first salvage treatment had better outcomes. It should be noted that a relatively large proportion of patients receiving RT had more localized relapse.
In the cohort presented by Fan et al., 17 60 patients progressed after CAR-T infusion, of whom 40 received salvage treatment, with 20 undergoing salvage RT. Radiotherapy was well tolerated, grade 3 toxicities were registered for 2 patients and no grade >3 toxicity was observed. Local response was documented for 17 patients, with 82.4% achieving either a partial or complete response. Patients who received salvage RT for localized relapses had significantly better OS rates from CAR-T relapse compared to those with localized relapse treated with systemic therapy (p = 0.03) and those with advanced relapses treated with RT (p = 0.02), with a 1-year OS of 88.9%, 37.5%, and 35%, respectively.
The multicentric retrospective study by Saifi et al. 18 compared DLBCL patients who underwent bridging RT (BRT) before CAR-T with those who received salvage RT (SRT) on sites that relapsed after CAR-T therapy, with 35 patients and 59 irradiated sites in the BRT group versus 48 patients and 65 irradiated sites in the SRT group. Considering the treated lesions, the in-field relapse rate was 13% in the BRT group compared to 32% in the SRT group, resulting in 1-year local control rates of 84% and 62%, respectively (p = 0.009). In patients receiving SRT for limited relapse, those who underwent comprehensive RT on all sites (n = 26) had better OS (51% vs 12%; p = 0.028) and freedom from subsequent progression (31% vs 0%; p < .001) at one year, compared to those who received focal RT (n = 11).
In another paper by Saifi et al., 19 61 patients with NHL who received CAR-T therapy and achieved PR or SD on a PET-CT scan performed 30 days after infusion were retrospectively analyzed: 45 of them were addressed to exclusive follow up (eFU) and 16 received consolidative RT (cRT). In the first group, 33% of patients experienced spontaneous CR and 60% progressed, whereas in the cRT group, 63% achieved CR and 25% developed PD, with no relapses observed in the irradiated sites. There was a statistically significant difference between cRT and eFU in terms of 2-year local relapse-free survival (100% vs 31%, p < 0.001) and 2-year PFS (73% vs 37%, p = 0.025). There was a non-significant difference in 2-year OS (78% vs 43%, p = 0.12). Patients receiving comprehensive cRT (n = 13) had significantly better 2-year OS (86% vs 43%; p = 0.047) compared to those underwent observation or receiving focal cRT.
In a retrospective analysis by Alarcon Tomas et al., 20 among 135 patients with NHL receiving salvage treatment for relapse after CAR-T therapy, subjects treated with involved site RT (n = 15) obtained CR and OR rates of 27% and 54%, respectively. These results were comparable with those achieved with polatuzumab-based salvage (34% CR and 48% OR) and lenalidomide-based salvage (33% CR and 43% OR), and were superior to those obtained with other regimens. Nonetheless, it is important to consider that the group receiving RT had the highest proportion of stage 1 disease (60%).
Imber et al. 21 analyzed a cohort of 14 DLBCL patients that received salvage RT for disease progression after CAR-T therapy. In 71% of cases sRT was administered at first relapse and ISRT was prescribed with heterogeneous doses (20-46 Gy, 1.5-4 Gy/fraction). Among the 6 patients with localized progression, ORR was 100% (three CR and three PR). In contrast, patients with advanced progression had poor outcomes, with 71% showing in-field PR and 71% experiencing out-of-field PD. No unexpected toxicities were reported. With a median follow-up of 10.2 months post-SRT, median OS was 10.4 months (not reached for localized relapses vs 2.6 months for advanced relapses). The localized relapse group had significantly better freedom from subsequent relapse and OS.
Di Blasi et al. 22 identified 238 DLBCL patients who experienced progression or relapse after CAR-T therapy from the DESCAR-T registry. Among there, 154 received salvage treatment, including 12 patients with localized relapse who were treated with salvage RT. Compared to patients receiving systemic salvage therapy, salvage RT resulted in a higher ORR of 35% versus 8–14.3%, with comparable or better PFS.
The multicenter retrospective study by Zurko et al. 23 evaluated DLBCL patients receiving allogenic transplant after CAR-T failure. Among the most common treatment regimens given between CAR-T and transplant, 18 courses of radiotherapy were provided, resulting in an ORR of 56% and a CR of 17%.
In another paper, Zurko et al. 9 identified 24 patients receiving sRT, among a cohort of 167 subjects who received salvage treatment for progression after CAR-T therapy. The ORR and complete response rates (CRR) were 25% and 13%, respectively, which were lower compared to those achieved with salvage systemic treatments. However, the PFS was similar (2.2 months). On the other hand, RT administered as a second-line salvage treatment (5 patients) achieved the highest ORR (60%), with all responses being PR.
A meta-analysis published in 2024 24 including 41 studies assessed CR and ORR of first-line treatments for CAR-T-relapsed/refractory LBCL. Non-CD19 CAR-T cells yielded the best CR (56%), significantly higher than other interventions, except for CD19 CAR-T (CR = 30%). Bi-specific monoclonal antibodies, radiotherapy, lenalidomide and polatuzumab-based regimens resulted in similar CR rates (28%, 26%, 19%, and 24%, respectively). Immune check-point inhibitors and Bruton's tyrosine kinase inhibitors showed the lowest CR rates (12% and 8%, respectively). Nonetheless, given the variability among studies and the heterogeneity of the estimates, the results should be interpreted cautiously. Moreover, only a limited number of the most recent and largest studies on sRT were included in this meta-analysis.
A few case reports have been already published regarding the adoption of salvage RT for relapse after CAR-T theapy.
Abou-Samra et al. 25 retrieved the case report of a patient with DLBCL presenting with papillitis and vitritis upon completion of CAR-T therapy, with a diagnostic vitreous tap that revealed the presence of intraocular lymphoma. Intraocular infiltration, edema and vitritis rapidly improved after RT start (30 Gy in 15 fractions), but the patient deceased after a few days due to cerebral disease progression.
Saifi et al. 26 presented a series of 14 patients who received bridging RT before CAR-T infusion. The one-year local control was 86%, with in-field relapse occurring in only 2 of 15 radiated sites. One of these patients subsequently received salvage RT, with a dose of 36 Gy in 12 fractions, resulting in a partial response.
Another paper 27 presented a case report of a 60-year-old female with mediastinal persistence of metabolically active lymph nodes after CAR-T therapy for DLBCL, after multiple lines of systemic treatment. She received ISRT at a dose of 50 Gy in 25 fractions with IMRT technique, developing only mild reversible odynophagia and radiodermitis. PET-CT scans conducted 2 and 8 months after radiotherapy revealed a complete metabolic response.
Discussion
CAR-T revolutionized the landscape of relapsing/refractory DLBCL, offering high response rates and improving survival.2,3,8 Nonetheless, a significant proportion of patients (40-60%) are either refractory to treatment or develop disease relapse after CAR-T administration.10,28 This relevant subset of patients is characterized by dismal prognosis, with an overall survival rate of about 30% at 12 months from relapse.10,28 Resistance to treatment and relapse are driven by multiple mechanisms, including variability in target antigens expression, heterogeneity in immune memory and CAR-T cells expansion, antigen loss and modulation (often due to clonal treatment-induced selection), escape to immune response and immunosuppressive tumor microenvironment.29-31
Radiotherapy might potentially mitigate these factors by enhancing antigenic expression, stimulating the release of chemokines and cytokines that promote the recruitment and migration of CAR-T cells to the tumor site and facilitating T-cell-mediated killing, thus hindering escape to immune response.30,32,33
The combination of RT and CAR-T cell therapy encompass different strategies according to timing and disease presentation. Bridging therapy is a treatment typically delivered between autologous T-cells extraction and CAR-T cell re-infusion with the aim to contain the disease during the 3–6 weeks required for manufacturing. Multiple options can be adopted for bridging therapy, including chemotherapy and other systemic agents. Radiotherapy emerged as an optimal bridging modality and most of the literature regarding the combination of RT and CAR-T cell therapy is represented by adoption of RT as a bridging strategy before CAR-T cells administration. In this setting, the role of RT is better established, as it demonstrated several advantages, and preliminary clinical guidelines with a practical framework for clinical implementation have been published. 34 Bridging RT resulted in fair toxicity profile and favorable clinical outcomes, including higher response rates, improved relapse free survival (RFS) and event-free survival (EFS)35,36 when compared to patients that did not receive bridging RT and historical controls. Comprehensive bridging RT was associated with superior PFS, RFS and OS, remarkably in patients with limited disease.35-37 In patients with high burden and bulky disease RT allows significant cytoreduction.38,39 Moreover, compared with chemotherapy, bridging RT is associated with a reduction in severe cytokine release syndrome (CRS) 38 and severe neurologic complications. 4
In the context of r/r DLBCL after CAR-T cell therapy, the main scenarios in which RT is indicated are consolidative treatment in case of partial response and salvage radiotherapy for primary refractory disease or for relapsing disease after initial response to CAR-T therapy.
Currently, best treatment option and therapeutic sequence are not well codified for DLBCL refractory/relapsing after CAR-T therapy. Lymphomas are generally radiosensitive neoplasms, requiring low prescription doses of radiotherapy.10,40 This results in optimal toxicity profiles, which can be further improved by using modern conformal techniques, 14 currently widely available and extensively utilized in most centers. 41 As RT has already demonstrated high response rates in heavily pre-treated lymphomas before the CAR-T era, 11 it should be considered a safe and potentially effective treatment option for patients with r/r DLBCL following CAR-T therapy.
Most studies included in this review reported high, though heterogeneous, OR rates (35-82.4%) and CR rates (17-59%). These results also translated into high local control rates at one year (62-82%).15-18,23 In the large majority of the studies comparing different salvage treatment modalities, RT achieved similar or better ORR and CRR compared to systemic therapies.20,22 Conversely, in a study by Zurko et al. 9 identifying 24 patients treated with first-line sRT, ORR and CRR were lower and worse compared to systemic therapies, while RT provided the best ORR when used as a second-line salvage treatment.
Since about 80% of post-CAR-T failures occur in previously avid sites, 15 RT should also be considered as a consolidation treatment. In a study by Ababneh et al., 15 consolidative RT improved both ORR and CRR in patients with incomplete response on a PET-CT scan performed 30 days after CAR-T therapy. Nonetheless, compared to bridging RT, salvage RT was associated with higher in-field recurrence rates. 18 This finding suggests that, when feasible and clinically indicated, RT should be anticipated and considered as an optimal bridging therapy.
Data regarding OS and PFS were extremely variable across different studies, highlighting the heterogeneity of enrolled cohorts. Overall survival was consistently better for patients with localized relapse17,21 and for patients receiving comprehensive RT on all sites of disease compared to patients that received only focal RT.15,17,18 In confront with salvage systemic therapy, sRT resulted in similar OS 16 and comparable or better PFS.9,22 In a study by Fan et al., 17 sRT demonstrated significantly better OS for patients with limited relapse compared to systemic salvage therapy (1-year OS 88.9% vs 37.5%).
Consolidative RT resulted in a significant improvement in PFS and a non-significant improvement in OS compared to exclusive observation following an incomplete PET response 30 days after CAR-T therapy. 19
One of the advantages of RT is that it can be administered concurrently with most systemic salvage treatments. Although data on the combination of systemic therapies and RT for lymphoma are limited, preliminary experiences reporting concurrent administration of agents approved for r/r DLBCL are reassuring.4,13,42 Radiotherapy demonstrated high response rates also in heavily pre-treated relapsing/refractory hematologic neoplasms, therefore could be the optimal option for salvage treatment, remarkably in case of localized relapse.43-46 Moreover, RT could synergize with salvage systemic therapy through multiple mechanisms, such as generation of new antigens that foster the immune response or direct cell death induction via DNA damage.4,13,47 It has also been reported that RT does not negatively affect CAR-T cell expansion48-50 but, conversely, it can induce a CAR-T re-expansion in cases of resistant disease. 51
Although preliminary, the encouraging data reported in this review, supported by a solid biologic rationale, suggest that radiotherapy should be considered as a crucial element in the treatment of DLBCL r/r after CAR-T therapy, highlighting the need to clarify its role in this setting. Nonetheless, the limits of the studies included in this review must as well be taken into account. The papers published to date encompass heterogeneous patients cohorts, all the analyses were retrospective and most of the studies had rather limited sample sizes. The included studies show a remarkable variability in RT doses, fractionation schemes, and patient selection criteria and those features were not consistently reported, as summarized in Table 1. Moreover, radiation-induced toxicity was reported only in a few instances, although generally with a fair profile. This heterogeneity could be mostly due to the retrospective nature of the studies, but might hinder the interpretation and conclusions of this review.
Several questions remain unanswered regarding the optimal treatment schedule and timing for salvage RT in r/r DLBCL post CAR-T. Results of this review suggest that, when feasible and remarkably in case of limited disease, all lymphoma sites should be irradiated. The best timing is yet to be defined, however RT should ideally be delivered as soon as possible, as local control seems to be superior for bridging therapy compared with salvage treatment, follow up resulted in worse control rates compared with consolidative RT and residual CAR-T cell number and activity are likely to decline over time. 52 Radiotherapy schedules were not consistently reported and variable across different studies, therefore no definitive conclusions can be drawn from the current literature. Previous studies suggest that regimens with higher EQD2 might improve disease control, especially for bulky tumors. 45 Nonetheless, treatment should be personalized on the basis of the presentation of disease. Indeed, r/r DLBCL after CAR-T presents with variable burden and sites of progression and encompasses different categories, including incomplete response, primary resistance, early and late relapse. Biomarkers, such as inflammatory signatures, might help to identify earlier patients with a higher likelihood of disease relapse and mortality. 53 Larger cohorts and prospective studies with well defined objectives and rigorous methodology are eagerly awaited to validate the promising results of RT for DLBCL R/R after CAR-T therapy summarized in this review.
Conclusion
The therapeutic workflow for refractory/relapsing DLBCL after CAR-T therapy is not well defined. While the available literature on the adoption of radiotherapy in this scenario is limited by retrospective design, small sample sizes, and heterogeneity in treatment protocols, preliminary results are promising. Radiotherapy has generally shown high overall response and complete response rates (also compared to systemic salvage treatment) and has offered promising results in terms of local control and progression-free survival. Evidence suggests that early use of RT, possibly encompassing all the sites of disease persistence, may offer superior local control compared to delayed intervention. Optimal dose and fractionation or radiotherapy have yet to be defined. Considered the extremely variable clinical presentation of r/r DLBCL following CAR-T cell therapy, a personalized approach is essential. Data from larger, prospective studies are needed to further consolidate the role of radiotherapy in this setting.
Footnotes
Abbreviations
Data Availability Statement
Data and material are stored according to our Institutional protocols and are available upon request.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
