Real World Epidemiology and Treatment Strategies of Limited Stage Large B Cell Lymphoma: Spanish Multicenter Geltamo Study

Introduction

Large B cell lymphoma (LBCL) presents as limited stage (LS) in 25-30% of patients. This subgroup has been defined as I-II Ann Arbor (AA) stage with non-bulky disease (commonly <10 cm). Prognosis is favorable with overall survival (OS) at 10 years at least of 70-80%. The IPI has limited utility as most patients by definition have a favorable risk profile, so the stage modified IPI (smIPI) was developed including age >60 years, II AA stage, elevated LDH and ECOG >1 as poor-risk features. Improvements in the knowledge of disease biology, the availability of positron-emission tomography (PET) and clinical trials including only this population have changed the treatment paradigm. However, there has been still a heterogeneity in the optimal approach. The aim of this study is to analyze the real-world epidemiology of these patients including biological markers, extranodal sites and different treatment strategies for patients with LS-LBCL in Spain.

Patients and methods

We performed a retrospective multicenter study including patients from GELTAMO centers with LS-LBCL from January 2013 to December 2022. Diffuse LBCL NOS, high-grade B-cell lymphoma (HGBCL) double hit and NOS, transformed follicular lymphoma and other less frequent LBCL subtypes were included. Primary mediastinal, central nervous system, cutaneous lymphoma and patients with bulky mass >10cm were excluded. Disease status was assessed by PET/CT. The primary endpoints were response rate, progression-free survival (PFS) and overall survival (OS) in the overall series and depending on the different treatment strategies such as combined modality treatment using abbreviated R-CHOP (x 3-4 cycles) plus involved-site radiation therapy (RT), abbreviated R-CHOP alone with 4 cycles and standard R-CHOP consisting of 6 cycles.

Results

Four-hundred and thirty-five patients fulfilled the inclusion criteria; median follow-up 62 months (95%CI 56-69); median age 68 years (range 22-96); 91% diffuse LBCL NOS subtype, 5% HGBCL (4% double hit and 1% NOS); 75%, 85% and 46% had BCL2, BCL6 and MYC expression, respectively; 43% were non-germinal centre by Hans algorithms; 48% II AA stage; 31% with high LDH; 56% with extranodal disease (34% gastrointestinal, 15% head/neck and 51% other different sites) and 48% with smIPI ≥2. Regarding therapeutic approaches, 12% received R-CHOP x 3-4 plus RT, 19% R-CHOP x 4, 41% R-CHOP x 6, 20% other chemoimmunotherapy strategies and 7% palliative treatment. Overall and complete response rate at end of induction were 390 (93%) and 375 (89%), respectively.

Focusing on the cohort of patients that exclusively received R-CHOP x 3-4 plus RT, R-CHOP x 4 and R-CHOP x 6 (318/435, 73%), PFS and OS at 5 years were 80% (95%CI 68-83) and 88% (95%CI 81-90), respectively. The frequency of these strategies was maintained in different time periods. These 3 subgroups were comparable depending on biological and clinical characteristics except for BCL2 rearrangement, II AA stage, high LDH, extranodal disease and smIPI >1 which were more frequent in R-CHOP x 6 subgroup. These variables were included as potential confounders in the multivariate analysis for PFS and OS. PFS was influenced by age, ECOG, BCL6 expression, BCL2 and MYC rearrangement and smIPI (p<0.001). In the multivariate analysis, ECOG >1 [HR 2.6 (95%IC 1.1-6.3), p=0.031] was the only independent variable for PFS. OS was influenced by age, ECOG, BCL2 and MYC rearrangement and smIPI (p<0.001). In the multivariate analysis, age >60 years [HR 3.1 (95%IC 1.5-6.4), p=0.002] and ECOG>1 [HR 4.4 (95%IC 2.1-9.2), p<0.001] were the only variables associated with worse OS.

Conclusions

To our knowledge, this is the largest series analyzing patients with LS-LBCL in which 56% of patients had extranodal sites and 5% were diagnosed with HGBCL. Our real-world results do not suggest differences between R-CHOP x 3-4 plus RT, R-CHOP x 4 and R-CHOP x 6 strategies in terms of both PFS and OS as had already been shown in clinical trials. However, in our series R-CHOP x 6 was the most used strategy. It is planned to analyze the impact of PET on the therapy decision in this cohort.