The prognostic value of clinical and inflammatory biomarkers in diffuse large B-cell lymphoma (DLBCL) overall survival (OS) is becoming increasingly recognized. Traditional analyses often use categorical variables, leading to data loss and limiting quantitative assessments. Evaluating dose-response relationships using restricted cubic spline (RCS) functions can reveal information lost in categorization, providing a flexible method for modeling non-linear relationships and enhancing prognostic tools.
To evaluate the dose-response relationship between inflammatory biomarkers-including the neutrophil/lymphocyte ratio (NLR), lymphocyte/monocyte ratio (LMR), systemic immune-inflammation index (SIII), serum albumin, hemoglobin levels, and platelet counts-and OS in patients with DLBCL in Latin America (LATAM).
We conducted a retrospective cohort study using a database of patients diagnosed with DLBCL and treated at various specialized centers in LATAM, members of the Grupo de Estudio de Latinoamericano de Linfoproliferativos (GELL) between 2012 and 2022. Patients with incomplete data (n=323) or could not be contacted (n=10) were excluded. The primary outcome evaluated was OS defined as time from diagnosis until death from any cause or censoring. NLR was calculated by dividing the number of neutrophils by the number of lymphocytes, LMR by dividing the absolute lymphocyte count by the absolute monocyte count, and SIII by multiplying the neutrophil count by the platelet count and dividing by the lymphocyte count at diagnosis. Serum albumin, hemoglobin level, and platelet count were also measured at diagnosis. To determine the dose-response relationship, RCS with 3 knots was used in a univariate Cox regression model, reporting hazard ratios (HR) and 95% confidence intervals (CI) to determine the nonlinear association between inflammatory biomarker levels and OS, confirmed by the Wald test. Harrel's concordance index (C-Statistics) was calculated. Analyses were performed using R (version 4.1.2; R Project for Statistical Computing) and relevant R packages (‘rms’, ‘survival’, ‘survminer’, and 'ggplot2'). Missing data were evaluated and imputed using the 'mice' package.
A total of 1,137 patients were included, with a median age of 64 years (range 18-96), and 50% were female. Seventy percent received R-CHOP-based treatments. The median follow-up was 32 months (range 0.1-152). In the RCS analysis, only serum albumin (p=0.286) and hemoglobin (p=0.194) exhibited linear prognostic relationships with OS. The remaining biomarkers showed non-linear relationships (p<0.001 for all factors). Serum albumin demonstrated the highest predictive power with an effect of -0.78 and a range between the max and min value of 0.85 (HR: 0.46; 95% CI: 0.36-0.58), with a C-Statistics of 0.65, indicating that an increase of 0.85 units in albumin reduces the risk of death by 54.3%. LMR showed an effect of -0.37 with a range between the max and min value of of 2.5 (HR: 0.68; 95% CI: 0.57-0.81) and a C-Statistics of 0.60, indicating that an increase of 2.5 units in LMR reduces the risk of death by 31.3%.
Inflammatory biomarkers demonstrated both linear and non-linear dose-response relationships with OS in patients with DLBCL in LATAM. Serum albumin at diagnosis emerged as the best predictor of OS and should be considered in clinical practice and future research. Continuing to evaluate the utility of RCS in the clinical research of inflammatory biomarkers in lymphoma is crucial for improving our understanding and management of the disease.
Perini:BMS, Roche, Abbvie, AsstaZeneca, Beigene: Speakers Bureau. Gomez-Almaguer:BMS: Consultancy, Other: Advisory board, Speakers Bureau; Kartos Therapeutics: Research Funding; Gilead/Forty Seven: Research Funding; Tevas: Speakers Bureau; AbbVie: Research Funding, Speakers Bureau; Roche: Speakers Bureau; Seattle Genetics: Research Funding; Incyte: Research Funding; Amgen: Consultancy, Other: Advisory board, Research Funding, Speakers Bureau; Blueprint Medicines: Research Funding; ConstellationPharmaceuticals: Research Funding; Astex Pharmaceuticals: Research Funding; Takeda: Consultancy, Other: Advisory board, Research Funding, Speakers Bureau; Sanofi: Speakers Bureau; Novartis: Consultancy, Other: Advisory board, Speakers Bureau; Janssen: Consultancy, Other: Advisory board, Speakers Bureau. Castillo:AbbVie: Consultancy, Research Funding; LOXO: Consultancy, Research Funding; AstraZeneca: Consultancy, Research Funding; Mustang Bio: Consultancy; Kite Pharmaceuticals: Consultancy; Pharmacyclics: Consultancy, Research Funding; Cellectar Biosciences: Consultancy, Research Funding; Janssen: Consultancy; BeiGene: Consultancy, Research Funding.
This feature is available to Subscribers Only
Sign In or Create an Account Close Modal