3357 - Failure Patterns by PET for Relapsed / Refractory Multiple Myeloma (r/r MM) after CAR T Cell Therapy: Potential Role for Radiotherapy (RT)?

A. Dreyfuss¹, B. Fregonese², H. G. Hubbeling³, S. Mailankody⁴, S. Usmani⁵, J. Yahalom², and B. S. Imber¹; ¹Department of Radiation Oncology, Memorial Sloan Kettering Cancer Center, New York, NY, ²Radiation Oncology, Memorial Sloan Kettering Cancer Center, New York, NY, ³Department of Radiation Oncology, University of Pennsylvania, Philadelphia, PA, ⁴Memorial Sloan Kettering Cancer Center, New York, NY, ⁵Hematology, Medical Oncology, Memorial Sloan Kettering Cancer Center, New York, NY

Purpose/Objective(s): Chimeric antigen T cell receptor therapy (CART) offers impressive response rates in MM. However, a significant proportion of patients still relapse or remain refractory. Here we characterize failure patterns after CART for r/r MM, using positron emission tomography computed tomography (PET) to identify high risk lesions that may benefit from pre-CART addition of local RT. Materials/

Methods: We analyzed 71 MM patients treated between 2017-2022 with BCMA-directed CART (38 experimental, 33 commercial products). All had pre- and post-CART PET. 529 FDG-avid lesions on pre-apheresis (if received bridging therapy) or pre-lymphodepleting chemotherapy (if no bridging) PET and 321 lesions on post-CART PET at time of first failure were categorized, with failure defined as radiologic and/or serologic progression according to Lugano and International Myeloma Working Group criteria, respectively. At the patient level we classified disease as either bony (B), extramedullary (EM) / paramedullary (PM), or mixed B and EM/PM. Individual lesions were classified as B, EM, or PM. Lesions were also classified as progressive, stable, or new at pre-CART PET according to SUV change from preceding PET. Failure of a lesion present on pre-CART PET was defined as progression involving the same anatomic structure within 3 cm of initial lesion. Pearson’s X² test was used to compare pre- and post-CART PET.
Results: Pre-CART PET occurred at a median of 1.4 months (range, 0.2-8.5) prior to CART, with 27 (39%) patients receiving systemic bridging, 10 (14%) RT +/- systemic, and 32 (46%) none. Median number of PET avid sites pre-CART was 6 (0-9) per patient. Of the 71 patients, 34 (53%) had mixed B and EM/PM, 24 (37%) B only, and 6 (10%) EM/PM only; 7 (10%) had no identifiable lesions. Of the 529 lesions identified on pre-CART PET, 351 (66%) were B, 131 (25%) EM, and 47 (9%) PM. 61 (86%) patients progressed post-CART, 28 (46%) of whom had first failure in new and pre-existing sites, 17 (20%) in new sites only, 6 (10%) in pre-existing sites only, and 10 (16%) serologically only. At the lesion level, 101/321 (31%) failures occurred in lesions identified on pre-CART PET. At time of failure, EM (44% vs 25%) and PM (12% vs 9%) lesions were present in increasing proportion compared to pre-CART (p<0.001). Of the initial 529 lesions, failure occurred in 44/131 (34%) EM sites, 16/47 (34%) PM sites, and 41/351 (12%) B sites (p<0.001). Lesions progressing at time of pre-CART PET failed more frequently (41/143, 29%) than those that were stable (20/130, 15%) or new (24/175, 14%) (p=0.002).
Conclusion: EM and PM sites progressed locally at higher rates than B sites, as did lesions progressing pre-CART. These data suggest higher risk lesion characteristics for which targeted RT ablation or cytoreduction may be beneficial. Further analysis could inform a novel approach of integrating selective site RT pre-CART as a bridging intervention.