Patients with grade 3b follicular lymphoma experience superior outcomes compared with diffuse large cell lymphoma, but show a continuous pattern of relapse.

Abstract

Background: Grade 3B Follicular lymphoma (G3BFL) comprises approximately 15% of FL and controversy exists over whether it follows the relapsing/remitting course of low-grade FL, or more closely resembles diffuse large B cell lymphoma (DLBCL), which is potentially curable. Thus it is largely excluded from clinical trials for either group, hence robust data and management consensus are lacking. Aim(s): To establish survival outcomes of G3BFL. Method(s): We performed an international retrospective study of patients aged > / = 18 years, with G3BFL from tertiary institutions in Australia, Canada and England. Pure G3B, composite Grade 3A/G3BFL and G3BFL/ DLBCL treated with rituximab (R) or obinutuzumab (O) + chemotherapy +/- radiotherapy (RT) or RT alone between 2002-2019 were included. FL grading was according to WHO criteria. A comparator DLBCL group, treated with R-CHOP +/- radiotherapy from 2012-2018 was identified using the same criteria from 2 Australian sites. OS was defined as the time from the date of diagnosis until death from any cause; PFS was defined from diagnosis until relapse/progression (to any B cell lymphoma subtype) or death; both were calculated according to the Kaplan-Meier method. Associations between prognostic factors and outcomes were analysed with log-rank tests and Cox proportional hazards models. Result(s): 161 G3BFL cases were evaluable. Histology was G3BFL in 84, G3A/G3B FL in 25, and G3B/DLBCL in 52 ("G3B" group). Median age was 62 (18-86) and 55% were male. 71% of patients had stage III/ IV disease. 58% had a high FLIPI score. Treatment was R or O-CHOPlike chemotherapy +/- RT in 96% of cases with RT alone or BR in the remainder (2% each). 90% received anthracyclines. 171 DLBCL comparator cases were included with a median age of 68 years (25-91) and a high R-IPI in 54%. Compared to the G3B group, there was a statistically significant difference in age > 60 years (G3B: 53%, DLBCL 71%; P = 0.001), ECOG performance status > 2 (G3B 5%, DLBCL 23%; P = < 0.001), elevated LDH (G3B: 40%, DLBCL 56%; P = 0.002) and median ki67 (G3B: 70%, DLBCL 80%; P = < 0.001). As survival outcomes of G3B, G3A/3B and G3B/DLBCL were similar (PFS: P = 0.23; OS: P = 0.27), these grades were analysed together. Median follow-up for all G3B was 4 years (0.21-16.1) with 4-year PFS 72% (95% CI 64-79%) & OS 84% (95% CI 77-90%). For the DLBCL group, median follow up was 2.3 years (0.42-6.3); with a 2-year PFS 61% (95% CI 51-70%) & OS 75% (95% CI 66-81%). G3B PFS & OS were superior to DLBCL (4yr PFS 72% vs 61%; P = 0.02; & OS 84% vs 75%; P = 0.05 respectively) with no plateau in the G3B group survival curves (Image 1). 41 patients (16%) had relapse/ progression of FL or transformation to DLBCL with biopsy confirmation in 29. Histology at relapse was G1/2FL in 2pts (7%), G3A in 3pts (10%), G3B in 4pts (14%) & DLBCL in 19pts (66%). Multivariate analysis, adjusting for baseline prognostic factors (raised LDH, bulk, stage 3-4, ECOG > 2), showed ECOG > 2 (PFS: P = 0.04; OS: P = 0.005) was adversely prognostic for OS and PFS in G3BFL. Summary/Conclusion: In our cohort, G3BFL had superior outcomes to DLBCL but also a more favourable prognostic profile. Despite better outcomes than DLBCL, G3BFL has an ongoing pattern of relapse with most biopsied cases demonstrating DLBCL. Approximately 50% of G3BFL have composite histology.