Genomic Correlates of Radiosensitivity in Diffuse Large B Cell Lymphoma

Introduction: Radiotherapy (RT) is beneficial for select patients with localized diffuse large B cell lymphoma (DLBCL), in both early stage and in relapsed/refractory settings. Genomic predictors of local response to RT have yet to be adequately explored. We sought to identify genomic signatures of radiosensitivity in localized DLBCL. Methods: We analyzed our database of DLBCL patients with localized measurable disease treated with RT between 2007 and 2022 with pre-existing targeted exome sequencing. This included both Ann Arbor stage I/II patients and those with isolated sites of relapsed/refractory disease in order to increase cohort size. The public NCI LymphGen algorithm was used to further classify genomic subgroups. Using Fisher’s exact test and logistic regression we tested for univariable correlation between specific gene alterations and LymphGen subgroups with initial radiation response, defined by Lugano criteria and decrease in maximum SUV. Results: We identified 38 patients with 52 irradiated tumors. The median (range) RT dose was 36 Gy (4–54), with 35 (67%) tumors receiving at least 30 Gy. 21 (40%) tumors were treated in the early stage setting (10% definitively, 90% as consolidation), while 31 (60%) were treated in the localized relapsed/refractory setting. 54% of tumors were GCB subtype, 31% were ABC subtype, and 15% were not classified. The median (range) of prior lines of systemic therapy were 1 (0–2) and 1 (1–7) in the early stage and localized relapsed/refractory settings respectively. The median (range) lesion SUV at the time of RT was 7.7 (1.1–46.0). 48% of tumors were in complete response (CR) at time of first imaging after RT (median 3.12 months) and the mean (SD) percent decrease in SUV was 60% (43%). The most commonly altered individual genes were KMT2D (37%), MYD88 (33%), and TP53 (31%). LymphGen classified 21% of cases as EZB subtype, 14% as MCD, 2% as BN2, 2% as ST2; 62% were unclassified. Alteration in ETV6 (13% of tumors) was associated with a significantly higher CR rate in response to RT (86% with alteration, 42% without, p = 0.046). After restricting each patient to their first course of RT, ETV6 alteration trended towards significantly higher CR (83% with alteration, 34% without, p = 0.06). Lesions with vs. without ETV6 alteration had similar SUV (p = 0.88) and size (p = 0.8) at time of RT. Alteration in ETV6 (p < 0.0001), PIM1 (p < 0.0001), MYD88 (p < 0.0001), and TNFAIP3 (p = 0.01) were associated with greater reductions in SUV. The LympGen MCD subgroup trended towards increased odds of CR post-RT (log odds 1.9, p = 0.078) and greater reduction of SUV (p = 0.12) compared to the EZB subgroup. Keywords: Aggressive B-cell non-Hodgkin lymphoma, Genomics, Epigenomics, and Other -Omics, Radiation Therapy Conflicts of interests pertinent to the abstract. G. Salles Consultant or advisory role: AbbVie, Aptitude Health, Bayer, BeiGene, Ltd., Bio Ascend, Bristol-Myers Squibb, Celgene, Epizyme, Everest Clinical Research Corporation, GenMab, Genentech, Gilead Pharmaceutical, Incyte, Ipsen, Janssen Pharmaceuticals, Inc., Loxo Oncology, Miltenyi Biotec Incorporated, MorphoSys AG, Nordic Nanovector ASA, Novartis, Physicians' Education Resource, RAPT Therapeutics, Regeneron Pharmaceuticals, Inc., Roche, Scientific Education Support Ltd., Takeda Millennium Stock ownership: Owkin, Inc.