The Population-Scale Analysis of DLBCL Patients Receiving Standard-Dose RCHOP Identified Impactful Prognosticators in the Real World

Background

Real-world clinical data analysis is critical for identifying prognostic factors among diffuse large B-cell lymphoma (DLBCL) patients who are outside clinical trials of therapeutics. The heterogeneity of this disease and treatment as well as the wide range of comorbidities require a large-scale harmonized database to tap into potential impacts from various clinical parameters. This study employed Taiwan National Health Insurance Research Database (NHIRD), coupled with Taiwan Cancer Registry (TCR) and the nationwide Death Registration to circumvent aforementioned difficulties.

Patients and Methods

We collected clinical data of 8,190 DLBCL adult patients, aged between 20 and 80 years old, newly diagnosed between 2015 and 2020 from TCR and NHIRD and follow up their survival until the end of 2021. Screened for receiving the standard-dose RCHOP, 2,924 patients were qualified for the downstream analyses. To identify prognosticators for overall survival (OS), the adjusted hazard ratios (aHRs) and 95% confidence intervals (CIs) were used to compare between 36 clinical parameters. The definition of acute hepatitis was the occurrence of serum transaminase levels higher than five times of the normal upper limit. To analyze relapsed/refractory DLBCL, the event of progression (EP) was defined as having one of the following situations: the use of rituximab three months after the last cycle of RCHOP, stem cell transplantation (SCT), or death.

Results

The median age of these 2924 patients was 60 years old. Male patients consisted of 54.2%. Approximately half of this cohort (49.0%) was Ann Arbor stage III and IV. The distribution of international prognostic index scores of 0-1, 2, 3, 4-5 were 44.3%, 23.9%, 20.6%, 11.2%, respectively. Twenty-three percent were hepatitis B virus carrier. Overall, 691 patients (23.6%) died in this cohort treated with the standard-dose RCHOP (the median follow-up period for OS was 4.1 years). We found negative prognosticators for OS including age at diagnosis (60-69 vs. 20-59, aHR: 1.48; 95% CI: 1.21-1.82; 70-79 vs. 20-59, aHR: 2.24; 95% CI: 1.77-2.82), sex (male vs. female, aHR: 1.19; 95% CI: 1.01-1.39), Ann Arbor stage IV vs. stage I (aHR: 1.44; 95% CI: 1.02-2.03), IPI score (2 vs. 0-1, aHR: 1.78, 95% CI: 1.37-2.31; 3 vs. 0-1. aHR: 1.78, 95% CI: 1.31-2.43; 4-5 vs. 0-1, aHR: 2.55; 95% CI: 1.79-3.64), active hepatitis (aHR: 2.16; 95% CI: 1.76-2.64), dementia (2.86; 95% CI: 1.47-5.56), chronic pulmonary disease (aHR: 1.49; 95% CI: 1.10-2.01), peptic ulcer disease (aHR: 1.45; 95% CI: 1.10-1.91) and renal disease (aHR: 1.56; 95% CI: 1.04-2.36). One thousand and eighteen patients (34.8%) developed EP (the median follow-up period for EFS was 3.4 years). To focus on 512 patients (50.3%) who was salvaged with platinum-containing therapy and/or SCT, 306 patients (59.8%) developed EP within 1 year, 116 patients (22.7%) within 1-2 years, and 90 patients (17.6%) after 2 years. Importantly, both the absence of SCT (aHR:1.85; 95% CI:1.39-2.47) and lacking platinum drugs in salvage therapy (2.58;1.03-6.43) were associated with the increased risk of death. The longer interval bewteen the initiation of RCHOP and EP was associated with a lower risk of death (aHR: 0.81 per year; 95% CI: 0.71-0.93).

Conclusion

This is one of the largest population-based studies about DLBCL patients treated with standard-dose RCHOP in the real world. We identifies significant OS prognosticators not only corroborating with prior reports but also revealing the negative impact by acute hepatitis.