Outcomes and Toxicities in Patients with Diffuse Large BCell Lymphoma of the Gastrointestinal Tract

Diffuse large B-cell lymphoma (DLBCL) involving the gastrointestinal (GI) tract is rare and long-term outcomes are not well defined. Combined modality therapy (CMT) with radiotherapy (RT) in addition to systemic therapy in this setting is not commonly pursued. We aim to characterize outcomes in patients with GI DLBCL treated with systemic therapy, with or without RT.Patients diagnosed with DLBCL of the GI tract (with or without mesenteric involvement) treated at a single institution from 1988-2022 were retrospectively reviewed on an IRB-approved protocol. Clinical and treatment data were collected including adverse events (AE; acute vs late defined as before or 4 weeks after therapy end). Kaplan-Meier and Cox regression models were used to estimate survival.Of 207 patients, 62% were male and median age at diagnosis was 63 (IQR 52-73). Gastric involvement was most common (n = 130, 63%), followed by small intestines (n = 48, 23%) and colon/rectum (n = 24, 12%). Most presented with early-stage disease (n = 124, 60%), with a median IPI score of 1. All patients received chemotherapy. Of 182 treated with CHOP/EPOCH, 36 (20%) were treated in the pre-rituximab era while 146 (80%) received rituximab. 66 patients (32%) were treated with RT, 89% as part of first line CMT. 50 cases (76%) received consolidative RT, while 10 (15%) targeted residual gross disease and 4 (6%) targeted distant sites. Median dose and fractionation were 36Gy (IQR 30.6-39.6) in 18 fractions (IQR 17-22). Over half (n = 132, 64%) developed grade 3+ acute chemotherapy AEs, and the most common were anemia (n = 64), febrile neutropenia (n = 40), and neutropenia (n = 20). Grade 3+ late chemotherapy AEs occurred in 14 patients (7%). Acute grade 3+ radiation AEs were uncommon (n = 2, 3%; colitis, emesis). No grade 3+ late radiation AEs were noted. Median follow-up was 46 months (IQR 16-97). 169 (81.6%) had a complete response (CR), with 154 (91%) after first line chemotherapy, 9 (5%) after second line, and 6 (4%) after RT. CR was defined by PET (62%), endoscopy (22%), CT (9%), or other methods (7%). The 5-year progression-free survival for those treated with one line of chemotherapy with or without RT was 95%. Median overall survival (OS) was not reached. Improved OS was associated with early-stage disease (p = 0.003), low IPI (p = 0.001), fewer chemotherapy lines (p<0.001), and CR (p<0.001). OS did not differ by gender, age, immunophenotype, GI site, SUVmax, or RT. Patients with early stage DLBCL treated with RT in the post-rituximab era received fewer chemotherapy cycles compared to those treated without RT (p = 0.02; median of 4 (IQR 3-6) vs 6 cycles (IQR 4-6)), with no OS difference.GI DLBCL patients have favorable outcomes after CMT with minimal late toxicity. CMT with RT to the GI tract is well tolerated with no OS difference compared to chemotherapy alone, and may mitigate risks from additional chemotherapy cycles for selected early-stage patients.