TO THE EDITOR:
Treatment options remain limited for patients with relapsed/refractory (R/R) multiple myeloma (MM) who progress on T-cell–engaging bispecific antibodies (BsAbs). 1 In the context of other malignancies, for patients with focal resistance or progression, 2 targeted radiation therapy (RT) has been used to sustain or salvage response to systemic therapy. 3 , 4 , 5 However, the use and outcomes of RT and BsAbs in R/RMM remain understudied. Therefore, we performed a single-center, retrospective study of patients with R/RMM who received RT with teclistamab or talquetamab from June 2018 to April 2024. Data collection concluded on 13 February 2025, with a median follow-up time of 12.7 months from BsAb initiation (exclusion criteria outlined in supplemental Figure 1 ; approved by the University of Pennsylvania Institutional Review Board). This study was conducted in accordance with the Declaration of Helsinki.
RT was categorized as follows: 1. Concurrent RT (CRT) was administered within the first 2 bsAb cycles to preexisting or evolving disease, acknowledging that BsAb responses generally occur within the first month of treatment. 2. Salvage RT (SRT) was administered after an initial response to BsAb for radiographic sites of progression. We noted whether SRT targeted all sites of progression (comprehensive SRT) or only a subset (noncomprehensive SRT).
An RT course was defined as treatment to ≥1 site in parallel, except for 1 case with sequential irradiation to multiple lesions in close succession. Responses, including radiologic responses of extramedullary disease (EMD) and paramedullary disease (PMD), were assessed per the International Myeloma Working Group criteria. 6 RT response for skeletal lesions was determined by symptomatic resolution.
Forty-one of the 208 consecutive patients received RT during BsAb therapy: 34 patients received RT with teclistamab (22 concurrent, 16 salvage, and 4 both), and 7 were treated on a clinical trial. Twelve received RT with talquetamab (7 concurrent, 7 salvage, and 2 both; none on a clinical trial; supplemental Figure 1 ). Seventeen patients received both teclistamab and talquetamab; 5 received RT during both therapies.
The median age of patients was 67.9 years ( supplemental Table 1 ). Nearly half (48.8%) of the patients had high-risk MM per the 2025 International Myeloma Working Group classification. 7 Patients who received RT with talquetamab had received more prior lines of therapy and exposure to anti-BCMA therapies than patients who received RT with teclistamab ( supplemental Table 2 ). Patients in the CRT group more frequently exhibited features at baseline that predict poor response to BsAb, 8 including elevated ferritin (56% vs 36.4%) and high-risk MM (58.6% vs 39.1%), than those receiving SRT ( Table 1 ). Table 1. Patient characteristics by bsAb and RT groups Characteristics Tec + CRT (n = 22) Tal + CRT (n = 7) Tec + SRT (n = 16) Tal + SRT (n = 7) Age, y 67.8 (37.4-81.5) ∗ 59.9 (43.9-78.8) 68 (56-87) 67.9 (55-77.7) Time to treatment since MM diagnosis, y 4.6 (1.9-13.3) 3.8 (1.5-10.1) 3.6 (2.2-14.5) 8.2 (3.8-13.5) Course duration, mo 2.1 (0.6-34.2) 4.8 (0.8-7.3) 11.1 (3.3-53.1) 5.3 (2.8-12.1) Met criteria for HRMM 12 (54.5) 5 (71.4) 6 (37.5) 3 (42.9) Baseline Anemia (hemoglobin <10 g/dL) 14 (63.6) ∗ 5 (71.4) 9 (56.3) 4 (57.1) Thrombocytopenia (<75 × 10 3 /μL) 6 (27.3) 3 (42.9) 3 (18.8) 1 (14.3) Ferritin >400 ng/mL † 10 (52.6) 4 (66.7) 6 (40) 2 (28.6) Bone marrow plasma cells ≥60% 0/4 ∗ 0/0 1/7 0/1 Presence of EMD 11 (50) 6 (85.7) 9 (56.3) 4 (57.1) Nonsecretory/oligo-secretory disease at progression ‡ 15 (71.4) 2 (22.2) Radiation characteristics Total number of RT courses 25 7 21 9 Time to RT since bispecific initiation, mo 0.1 (−0.1 to 2) 0.1 (−0.1 to 1.1) 7 (2.7-33.3) 4.5 (1.9-11.2) RT course duration, d 9 (1-19) 8 (1-43) 10 (1-16) 9 (1-16) Total RT dose per course, Gy 25 (4-90) 30 (16-40) 20 (8-60) 20 (8-40) Radiation site Total number of sites irradiated 44 12 29 11 Number of sites irradiated per course 1 (1-4) 2 (1-3) 1 (1-3) 1 (1-2) RT dose to each site, Gy 20 (4-30) 20 (8-30) 20 (8-30) 20 (8-30) Lesion type Extramedullary 17 8 12 6 Paramedullary 14 3 3 4 Skeletal 13 1 14 1 Comprehensive SRT Yes 9 (42.9) 6 (66.7) No 12 (57.1) 3 (33.3) Best IMWG response criteria § PD 13 (59.1) 0 (0) 0 (0) 0 (0) SD 0 (0) 2 (28.6) 0 (0) 3 (42.9) MR 0 (0) 0 (0) 1 (6.3) 2 (28.6) PR 3 (13.6) 1 (14.3) 4 (25) 2 (28.6) VGPR 6 (27.3) 3 (42.9) 7 (43.8) 0 (0) CR 0 (0) 1 (14.3) 4 (25) 0 (0) CR, complete response; CRS, cytokine release syndrome; HRMM, high-risk MM; ICANS, immune effector cell-associated neurotoxicity syndrome; IMWG, International Myeloma Working Group; max, maximum; min, minimum; MR, minimal response; PD, progressive disease; SD, stable disease; SFLC, serum free light chains; tal, talquetamab; tec, teclistamab; VGPR, very good PR. ∗ Median (minimum, maximum); n (%); number of bone marrow biopsies that meet criteria/total number of bone marrow biopsies performed. † Not available for 3 patients in tec + CRT, 1 patient in tal + CRT, and 1 patient in tec + SRT groups. ‡ Nonsecretory or oligo-secretory disease (defined as M-spike <0.5 and SFLC <100 mg/L) at the time of disease progression before SRT. § Best response per IMWG criteria, irrespective of whether achieved before or after radiation.
Table 1.
Patient characteristics by bsAb and RT groups
CR, complete response; CRS, cytokine release syndrome; HRMM, high-risk MM; ICANS, immune effector cell-associated neurotoxicity syndrome; IMWG, International Myeloma Working Group; max, maximum; min, minimum; MR, minimal response; PD, progressive disease; SD, stable disease; SFLC, serum free light chains; tal, talquetamab; tec, teclistamab; VGPR, very good PR.
∗ Median (minimum, maximum); n (%); number of bone marrow biopsies that meet criteria/total number of bone marrow biopsies performed.
† Not available for 3 patients in tec + CRT, 1 patient in tal + CRT, and 1 patient in tec + SRT groups.
‡ Nonsecretory or oligo-secretory disease (defined as M-spike <0.5 and SFLC <100 mg/L) at the time of disease progression before SRT.
§ Best response per IMWG criteria, irrespective of whether achieved before or after radiation.
| Characteristics | Tec + CRT (n = 22) | Tal + CRT (n = 7) | Tec + SRT (n = 16) | Tal + SRT (n = 7) |
| Age, y | 67.8 (37.4-81.5) ∗ | 59.9 (43.9-78.8) | 68 (56-87) | 67.9 (55-77.7) |
| Time to treatment since MM diagnosis, y | 4.6 (1.9-13.3) | 3.8 (1.5-10.1) | 3.6 (2.2-14.5) | 8.2 (3.8-13.5) |
| Course duration, mo | 2.1 (0.6-34.2) | 4.8 (0.8-7.3) | 11.1 (3.3-53.1) | 5.3 (2.8-12.1) |
| Met criteria for HRMM | 12 (54.5) | 5 (71.4) | 6 (37.5) | 3 (42.9) |
| Baseline | ||||
| Anemia (hemoglobin <10 g/dL) | 14 (63.6) ∗ | 5 (71.4) | 9 (56.3) | 4 (57.1) |
| Thrombocytopenia (<75 × 10 3 /μL) | 6 (27.3) | 3 (42.9) | 3 (18.8) | 1 (14.3) |
| Ferritin >400 ng/mL † | 10 (52.6) | 4 (66.7) | 6 (40) | 2 (28.6) |
| Bone marrow plasma cells ≥60% | 0/4 ∗ | 0/0 | 1/7 | 0/1 |
| Presence of EMD | 11 (50) | 6 (85.7) | 9 (56.3) | 4 (57.1) |
| Nonsecretory/oligo-secretory disease at progression ‡ | 15 (71.4) | 2 (22.2) | ||
| Radiation characteristics | ||||
| Total number of RT courses | 25 | 7 | 21 | 9 |
| Time to RT since bispecific initiation, mo | 0.1 (−0.1 to 2) | 0.1 (−0.1 to 1.1) | 7 (2.7-33.3) | 4.5 (1.9-11.2) |
| RT course duration, d | 9 (1-19) | 8 (1-43) | 10 (1-16) | 9 (1-16) |
| Total RT dose per course, Gy | 25 (4-90) | 30 (16-40) | 20 (8-60) | 20 (8-40) |
| Radiation site | ||||
| Total number of sites irradiated | 44 | 12 | 29 | 11 |
| Number of sites irradiated per course | 1 (1-4) | 2 (1-3) | 1 (1-3) | 1 (1-2) |
| RT dose to each site, Gy | 20 (4-30) | 20 (8-30) | 20 (8-30) | 20 (8-30) |
| Lesion type | ||||
| Extramedullary | 17 | 8 | 12 | 6 |
| Paramedullary | 14 | 3 | 3 | 4 |
| Skeletal | 13 | 1 | 14 | 1 |
| Comprehensive SRT | ||||
| Yes | 9 (42.9) | 6 (66.7) | ||
| No | 12 (57.1) | 3 (33.3) | ||
| Best IMWG response criteria § | ||||
| PD | 13 (59.1) | 0 (0) | 0 (0) | 0 (0) |
| SD | 0 (0) | 2 (28.6) | 0 (0) | 3 (42.9) |
| MR | 0 (0) | 0 (0) | 1 (6.3) | 2 (28.6) |
| PR | 3 (13.6) | 1 (14.3) | 4 (25) | 2 (28.6) |
| VGPR | 6 (27.3) | 3 (42.9) | 7 (43.8) | 0 (0) |
| CR | 0 (0) | 1 (14.3) | 4 (25) | 0 (0) |
Figure 1 A-B presents each individual bsAb course with the timing of RT and outcomes. Figure 1. RT outcomes in patients treated with bsAbs for MM. (A) Swimmer plot for patients with a BsAb course duration <6 months. (B) Swimmer plot for patients with a BsAb course duration >6 months. (C) PFS from the time of BsAb initiation among patients who received CRT. (D-E) PFS (D) and TNT-D (E) from the time of SRT in groups receiving comprehensive SRT vs noncomprehensive SRT. Of note, 2 radiation courses were combined for 2 patients, given the close timing of administration. PFS was calculated from the initiation of the first radiation course until subsequent progression. For all patients who received multiple courses of rescue radiation (n = 4), we calculated TNT-D from the start of their first SRT course. P values were calculated using the log-rank test. CR, complete response; MR, minimal response; SD, stable disease; VGPR, very good PR.
In the CRT group (n = 29), there were 56 anatomic sites irradiated (25 EMD, 17 PMD, and 14 skeletal) across 32 RT courses ( supplemental Tables 3 and 4 ). The median time to RT from BsAb initiation was 0.1 months ( Table 1 ). After RT, 16 of the 17 EMD lesions (94.1%), 6 of the 10 PMD lesions (60%), and 13 of the 14 skeletal lesions (92.9%) with imaging available had a partial response (PR) or greater ( supplemental Table 5 ). Among the 17 patients receiving RT during BsAb step-up dosing, 4 developed cytokine release syndrome (all grade 1-2), and 5 developed immune effector cell–associated neurotoxicity syndrome (grade 1-2, n = 4; grade 3, n = 1). None experienced “RT-related” cytopenias, although 1 patient experienced BsAb hold due to neutropenia. Cytopenias were considered possibly “RT related” if systemic therapy did not appear causal, they were worse than baseline, and they met Common Terminology Criteria for Adverse Events version 5.0.
Best overall response rate, encompassing the effects of BsAb and CRT, was 48.3% (14/29) with a very good PR or better rate of 34.5% (10/29; Table 1 ). CRT and BsAb had a median progression-free survival (PFS) of 1.4 months, although 12-month PFS probability was 0.23 (95% confidence interval [CI], 0.14-0.54), driven by a subset of patients who received RT with teclistamab with long-term response ( Figure 1 C).
In the SRT group (n = 23), 30 RT courses were administered, targeting a total of 40 anatomic sites (18 EMD, 7 PMD, and 15 skeletal; Table 1 ). The median time to RT was 6.9 months. Half (15/30) of the patients received comprehensive SRT. Patients who received comprehensive SRT had fewer known focal sites of disease than those who received noncomprehensive SRT (1 vs 4; P < .0001; supplemental Table 6 ). PR or better was achieved in 10 of the 15 EMD (66.7%), 2 of the 5 PMD (40%), and 14 of the 15 skeletal lesions (93.3%) with post-RT imaging available ( supplemental Table 5 ). No patients developed cytokine release syndrome, immune effector cell–associated neurotoxicity syndrome, or “RT-related” cytopenias during SRT.
Best response at any time during BsAb therapy in the SRT group, compared to pre-BsAb baseline, was PR or better in 17 of the 23 patients (73.9%) and very good PR or better in 11 of the 23 patients (47.8%; Table 1 ). The estimated median PFS after SRT was 2.2 months (95% CI, 0.92-3.75); the median time to next treatment or death (TNT-D) was 3.2 months (95% CI, 1.5-5.8; Figure 1 D-E). Patients with EMD as the target lesion had shorter PFS than patients with non-EMD targets (0.99 vs 3.19 months; P = .027), and those who received SRT to 1 site vs multiple sites had longer PFS (3.19 vs 0.92 months; P = .037; supplemental Figures 2-3 ). Compared to noncomprehensive SRT, patients who received comprehensive SRT had longer post-RT PFS (2.8 vs 0.9 months; P = .0015) and TNT-D (5.4 vs 1.2 months; P = .003). The 6-month PFS and TNT-D probability in comprehensive SRT courses was 0.25 (95% CI, 0.094-0.65) and 0.38 (95% CI, 0.16-0.87), respectively.
The median OS was 24.8 months (teclistamab) and 8.7 months (talquetamab; supplemental Figure 4 ). Most patients progressed through BsAb and were either transitioned to another therapy (50%) or hospice (30.4%); 9 were continued on BsAb (n = 7) or observed off therapy (n = 2) by the end of data collection ( supplemental Tables 2,7 ).
In this retrospective cohort study, we observed that RT administered with BsAb was feasible and, in select cases, augmented the clinical benefit of BsAb therapy. This cohort included a high proportion of patients with adverse prognostic features, including extensive previous therapy, elevated ferritin, EMD, and need for RT alongside bispecific treatment. 8 Nonetheless, a subset of patients in both the CRT (n = 12) and SRT groups (n = 7) achieved PFS >12 months.
The discrete contribution of RT is most clearly illustrated in patients who received SRT, in whom the benefit was more pronounced with comprehensive SRT; such patients demonstrated a 6-month TNT-D probability of 38% (n = 6), with 2 still progression free and continuing on BsAb after 12 months. The patients who received comprehensive SRT had fewer sites of active disease at progression, which likely contributed to improved outcomes independently of SRT. Conversely, patients who received RT to multiple sites or EMD had poorer PFS. In the comprehensive SRT subset, post-SRT TNT-D was substantially longer than PFS (5.4 vs 2.8 months), suggesting that clinical features of progression after SRT may be less likely to necessitate urgent transition to subsequent systemic therapy.
The use of RT with BsAbs may offer particular value for EMD, our cohort’s most common radiation target. EMD is associated with aggressive disease biology and resistance to immunotherapies, including chimeric antigen receptor T-cell therapy and bsAb. 8 , 9 RT may overcome local immune resistance by enhancing T-cell infiltration, antigen presentation, and immune costimulatory pathways. 10 Preclinical data support the immunomodulatory effects of RT, and its use in combination with T-cell–engaging therapies warrants further prospective evaluation.
Our single-center retrospective study has limitations, including heterogeneity in imaging practices and response assessments. RT was used at the clinician’s discretion, with a mix of palliative and cytoreductive intent. In addition, small sample sizes, especially among patients who received RT with talquetamab, limited subgroup analyses. Nonetheless, this study provides insight into real-world outcomes of RT in patients receiving BsAbs and identifies clinical scenarios, such as progression in a limited number of sites amenable to comprehensive salvage RT, where RT may yield clinically meaningful delays in progression and need for subsequent systemic therapy.
Conflict-of-interest disclosure: A.D.C. reports research funding (paid to the institution) from Novartis, Johnson & Johnson, Genentech/Roche, Poseida, and GSK; and consulting or data safety and monitoring board (DSMB) service fees from Novartis, Johnson & Johnson, Genentech/Roche, GSK, Kite, Bristol Myers Squibb (BMS), Legend, AstraZeneca, Regeneron, iTeos, Sanofi, Caribou, Moderna, AbbVie, Ichnos, Prothena, and Pfizer. D.T.V. reports research funding from Takeda and Active Biotech; and consulting fees from Takeda, BMS, AbbVie, Genentech, Janssen, and GSK. A.L.G. reports research funding (paid to the institution) from CRISPR Therapeutics and Johnson & Johnson; consulting relationships with Johnson & Johnson, Novartis, BMS, AbbVie, Regeneron, and Smart Immune; and DSMB service for Johnson & Johnson. S.S.-A. reports research funding (paid to the institution) from Johnson & Johnson; and consulting fees from Sanofi. The remaining authors declare no competing financial interests.
References
1 Yu K Ho M Paruzzo L Clinical outcomes after teclistamb failure in relapsed/refractory multiple myeloma Clin Lymphoma Myeloma Leuk 25 suppl 2 2025 S64 S65
2 Ravella R Feliciano EJG Dee EC Gomez DR Iyengar P Management of oligoprogressive and oligopersistent disease in advanced NSCLC Clin Adv Hematol Oncol 23 1 2025 40 50 39907576
3 Saifi O Lester SC Breen WG Incorporating radiation with anti-CD19 chimeric antigen receptor T-cell therapy for relapsed/refractory non-Hodgkin lymphoma: a multicenter consensus approach Am J Hematol 99 1 2024 124 134 37950857 10.1002/ajh.27155
4 Srinivas S Zhu LL Raje N Patel CG Radiation therapy to manage isolated relapse after Chimeric antigen receptor T-cell therapy in multiple myeloma Haematologica 110 10 2025 2543 2547 40468959 10.3324/haematol.2025.287358 PMC12485353
5 Fregonese B Imber BS Dreyfuss A Hajj C Yahalom J Upfront radiation therapy (RT) for post-stem cell transplant multiple myeloma (MM) patients with oligo-relapse/progressive disease: factors associated with favorable outcomes Blood 142 suppl 1 2023 4719
6 Durie BGM Harousseau JL Miguel JS International uniform response criteria for multiple myeloma Leukemia 20 9 2006 1467 1473 16855634 10.1038/sj.leu.2404284
7 Avet-Loiseau H Davies FE Samur MK International Myeloma Society/International Myeloma Working Group consensus recommendations on the definition of high-risk multiple myeloma J Clin Oncol 43 24 2025 2739 2751 40489728 10.1200/JCO-24-01893 PMC13445576
8 Razzo BM Midha S Portuguese A Real-world experience with teclistamab for relapsed/refractory multiple myeloma from the U.S. Myeloma Immunotherapy Consortium Blood Cancer Discov 6 6 2025 561 571 40629516 10.1158/2643-3230.BCD-24-0354
9 Gagelmann N Eikema DJ Verdonck LF Impact of extramedullary disease on outcomes after BCMA-directed CAR T-cell therapy in multiple myeloma: a European retrospective analysis Blood Adv 7 4 2023 702 711
10 Demaria S Formenti SC Golden EB The intersection of radiotherapy and immunotherapy: mechanisms and clinical implications Nat Rev Cancer 20 9 2020 516 529
Supplementary Material
Supplemental Tables and Figures
Acknowledgment: A.L.G. acknowledges support from a Scholar in Clinical Research award from the 10.13039/100005189 Leukemia and Lymphoma Society .
Contribution: H.P. performed the research, collected, analyzed, and interpreted data, performed statistical analysis, and wrote the manuscript; H.H. performed the research and collected, analyzed, and interpreted data; K.Y. and M.C. performed the research and collected data; M.J.L., J.P.P., A.D.C., S.K., D.T.V., A.J.W., and E.A.S. performed the research and analyzed and interpreted data; A.L.G. designed and performed the research, analyzed and interpreted data, performed statistical analysis, and wrote the manuscript; and S.S.-A. designed and performed the research, collected, analyzed, and interpreted data, performed statistical analysis, and wrote the manuscript.
Data are available from the corresponding author, Sandra Susanibar-Adaniya ( sandra.susanibaradaniya@pennmedicine.upenn.edu ), on request.
The full-text version of this article contains a data supplement.