Predictive Value and Modeling of Peripheral Blood Lymphocyte Subsets for the Treatment of Diffuse Large B-Cell Lymphoma

Objective:

This study aims to investigate the distribution of lymphocyte subpopulations in patients diagnosed with diffuse large B-cell lymphoma (DLBCL), analyzing their impact and predictive value on treatment efficacy, and constructing a multifactorial nomogram model for predicting patient outcomes.

Methods:

Lymphocyte subpopulation percentages, other biochemical indexes, and clinical characteristics were collected from 125 patients initially diagnosed with DLBCL at the Second Affiliated Hospital of Nanchang University and 33 healthy controls with similar age and gender ratios who underwent physical examinations at our hospital during the same period. A total of 86 patients were included in the study based on specified inclusion and exclusion criteria. For group comparisons, the Mann-Whitney U test was employed for independent samples while the Wilcoxon test was used for paired samples. The chi-square test was utilized to compare qualitative data. Variables showing statistical significance were identified through univariate logistic binary regression analysis and included as potential predictors. Independent risk factors were identified using multifactorial logistic regression analysis, followed by the development of a predictive nomogram for CR. The performance of the nomogram was evaluated through receiver operating characteristic (ROC) curves, calibration plots, and decision curve analysis (DCA). Statistical analysis was performed using SPSS 26.0, and graphs were generated using GraphPad Prism 9.0. Nomogram model construction and performance tests were performed utilizing the R programming language (version 4.2.2) along with the “rms” program package.

Results:

The proportions of peripheral blood CD3+T cells, B lymphocytes, and CD4+T/CD8+T ratio were significantly lower in patients with DLBCL compared to the control group. Specifically, the levels of CD4+T/CD8+T and B cells showed statistically significant differences between the two groups(P=0.005,P=0.04). Furthermore, the percentage of CD8+T cells was significantly higher than that of the controls(P<0.05).The percentages of CD3+T and CD8+T cells exhibited a significant increase in DLBCL patients following chemotherapy compared to pre-treatment levels (P<0.05). Conversely, the percentages of CD4+T cells, B lymphocytes, and the ratio of CD4+T/CD8+T demonstrated a significant decrease after chemotherapy (P<0.05).Comparison of lymphocyte subsets in relation to efficacy between the effective group (CR+PR) and the ineffective group (SD+PD), as well as within the complete response group (CR) versus the non-CR group (PR+SD+PD), revealed a consistent trend: patients with elevated the percentages of CD4+T cells and CD4+T/CD8+T ratio exhibited superior treatment response.The CD4+T/CD8+T ratio, ALB levels, and BCL2/MYC expression were identified as independent predictors, and a nomogram model was constructed to predict CR, exhibiting a C-exponent of 0.903. Calibration curves and clinical decision curves (DCA) further validated the excellent clinical predictive performance of this model.

Conclusion:

This study identified disparities in the peripheral blood immune profile between patients with DLBCL and healthy controls, while the distribution of lymphocyte subsets at initial diagnosis demonstrated predictive value for treatment efficacy.We developed a nomogram model to predict CR following treatment for DLBCL and verified the model's good performance in prediction.

No relevant conflicts of interest to declare.