Outcomes in primary gastrointestinal (GI) follicular lymphoma (FL): results from a multicenter analysis

Abstract

Introduction: GI FL is a rare disease, accounting for less than 5% of GI non-Hodgkin’s lymphomas (NHL). We formed a consortium to clarify baseline features, common practices and survival outcomes in primary GI FL. Methods: A retrospective analysis of 182 patients (pts) with primary GI FL from 8 institutions was conducted. Overall survival (OS) and progression-free survival (PFS) were assessed. Cox regression models were used to estimate hazard ratios (HR) with 95% confidence intervals (CI). Results: Median age at diagnosis (dx) was 60 years (yrs) (IQR: 51–68). 89% (n = 141) of pts had low-grade FL by histology, and most presented with a low-risk score by FLIPI (68%, n = 111) or FLIPI-2 (76%, n = 103). At presentation, 18% (n = 32) of pts were asymptomatic, and 71% (n = 124) had at least one GI symptom, most commonly abdominal pain (40%, n = 70). 5% (n = 9) of pts presented with overt GI bleeding, and 9% (n = 15) with bowel obstruction. The most common first-line strategy was observation (33%, n = 59), of which 36% (n = 21) underwent surgical resection at dx. Other regimens included single-agent rituximab (R) (22%, n = 39), chemoimmunotherapy (CI) with R-bendamustine (18%, n = 32) or R-CHOP (11%, n = 19), and radiation therapy (5%, n = 9). The remainder received other CI combinations or R-lenalidomide. For those observed, median time to first treatment (tx) was 4.6 yrs. For those treated, overall response rate was 83% (95% CI, 76%–89%), and median time to second-line tx was 3.7 yrs. With median follow-up of 5.7 yrs, median OS was not reached. 5-yr PFS and OS were 71% (95% CI, 64%–79%) and 93% (95% CI, 89%–98%), respectively. Relevant factors that did not show a significant effect on PFS/OS included advanced disease by Paris and Ann Arbor staging systems, high-risk disease by FLIPI or FLIPI-2, B symptoms, GI symptoms, location of disease within the GI tract, and bone marrow involvement. Patients who were initially observed had similar OS to those who were treated up-front; PFS/OS did not improve significantly with use of maintenance rituximab. Factors that significantly impacted PFS and/or OS are outlined in Table 1. Conclusion: This multi-institutional analysis describes characteristics of the largest cohort of pts with GI FL to date. Observation or tx with rituximab ± chemotherapy were common initial approaches. Patients in this cohort had a good overall prognosis. While some factors including age, grade and BCL-2 expression by immunohistochemistry (IHC) predicted survival, traditional prognostic indices and staging systems did not. The research was funded by: No funding to disclose. Keyword: Extranodal non-Hodgkin lymphoma Conflicts of interests pertinent to the abstract. T. Alrifai Consultant or advisory role: Previous independent consultancy for MJH Life Sciences; completed in June of 2022. J. Kline Employment or leadership position: Speaker for Kite Pharma, Inc./Gilead. B.A.D Consultant or advisory role: Advisory boards for Merck, Kite Pharma, Inc./Gilead, Seattle Genetics, and Verastem Oncology Honoraria: Has received advisory board fees from Janssen, Sanofi, and consulting fees from COTA Healthcare. Research funding: Research support from Merck, iTeos, and Verastem Oncology R. Karmali Employment or leadership position: Speakers Bureau: AstraZeneca, BeiGene, Morphosys Consultant or advisory role: Advisory Board: BMS, Gilead Sciences/Kite Pharma, Janssen, Pharmacyclics, Morphosys, Epizyme, Genentech/Roche, Calithera, Miltenyi, Lilly Oncology Research funding: Grants/Research Support: BMS, Takeda, BeiGene, Gilead Sciences/Kite, Calithera