Impact of Radiotherapy and Rituximab on Outcome of Stage I(E) Diffuse Large B-Cell Lymphoma

Purpose:
For patients with limited stage diffuse large B-cell lymphoma (DLBCL) the overall survival (OS) in the pre-rituximab era was initially reported to be superior for abbreviated CHOP followed by radiotherapy compared to full course CHOP chemotherapy. At long term follow-up this benefit was lost. Rituximab(R) has increased the outcome of advanced stage patients, but the impact on limited stage DLBCL is less clear. Whether abbreviated R-CHOP followed by radiotherapy or full course R-CHOP is the treatment of choice remains controversial.

Patients and methods:
We performed an observational population-based cohort study of patients with stage I (E) DLBCL identified from 2 databases from the northern part of the Netherlands. Patients had been treated with abbreviated CHOP (3 cycles) followed by radiotherapy or full course CHOP (6-8 cycles). R was given to all patients after the drug became available in the Netherlands from 2004 onward. Primary endpoints were time to tumor progression (TTP) and OS

Results:
From 1984 until 2014 196 patients with stage I(E) were identified. Median follow-up was 53 and 110 months for patients treated with R and without R respectively Characteristics at diagnosis: median age 63 years (range 15-90), 20% ≥ 75 years; IPI-score 0-1: 88%; extranodal disease: 58%.
Treatment: 190 patients completed at least 3 cycles of (R-)CHOP, 63% R and 37% without R . Full course (R-)CHOP was administered in 35% of patients. Patients with extranodal disease more often received full course (R-)CHOP (44%) compared to nodal disease (22%).
Outcome: There were 32 relapses; 10 / 66 (15%) in full course (R-)CHOP and 22 / 124 (18%) in abbreviated (R-)CHOP followed by radiotherapy. Of the latter 19% were infield relapses. 5-year TTP and OS for the entire cohort were 83% (95%CI 78-89%) and 77% (95%CI 71-83%) respectively. In univariate analysis only IPI and age had significant impact on OS. Patients with an IPI-score of 0 had a hazard ratio (HR) of 0.50 compared to patients with an IPI-score ≥1 (95%CI 0.28-0.89, p 0.013). Patients ≤60 years and 60-75 years had a HR of 0.13 (95%CI 0.06-0.27, p < 0.01) and 0.34 (95%CI 0.18-0.63, p< 0.01) respectively compared to patients ≥75 years for OS. No difference was found in OS between patients treated with abbreviated chemotherapy followed by radiotherapy or full course chemotherapy (HR 1.09, 95%CI 0.62-1.91, p 0.77); nor for treatment with or without R (HR 1.03, 95%CI 0.56-1.89, p 0.93). In univariate analysis only age ≤75 years was prognostic (HR 0.37 95%CI 0.17-0.84, p 0.017) for TTP. Of the relapses 25% occurred ≥ 3 years after treatment irrespective of rituximab exposure.

Conclusions: 
In this large population-based analysis of patients with stage I(E) DLBCL no differences in OS and TTP were found between abbreviated R-CHOP followed by radiotherapy and full course R-CHOP. No clear benefit of R on OS, TTP and late relapses was found. With similar outcomes, toxicity should primarily determine the chose of treatment regimen.