Investigation into the prognostic factors of early recurrence and progression in previously untreated diffuse large B-cell lymphoma and a statistical prediction model for POD12.

Objective: The objective of this study is to evaluate the incidence, prognostic value, and risk factors of progression of disease within 12 months (POD12) in patients with diffuse large B-cell lymphoma (DLBCL).

Methods: A retrospective analysis of the clinical, pathological, and follow-up data was carried out on 69 DLBCL cases in Shanxi Bethune Hospital from January 2016 to June 2020. One-way ANOVA and multivariate Cox regression analysis were used to explore the correlation between POD12 and prognosis, and logistic regression analysis was used to explore the risk factors of POD12, accompanied by prediction models based on convolutional neural networks and long short-term memory (CNN-LSTM), as well as particle swarm optimization and general regression neural network (PSO-GRNN) models.

Results: (1) POD12 is significantly correlated with PFS ( 0.001) and OS ( = 0.008). (2) From the univariate logistic regression analysis corrected by the first-line chemotherapy regimen, LDH, β-MG, stage, ECOG, NLR, and SII are identified as risk factors for POD12 ( 0.1), while β-MG and ECOG are identified as independent risk factors from the multivariate logistic regression analysis (< 0.05). (3) A prediction model for POD12 is established based on LDH, β-MG, stage, ECOG, NLR, and SII. The AUC is 0.846 (95% CI: 0.749~0.944, 0.001), suggesting that the model is reasonable. A prediction method for the characteristic variables of POD12 risk is proposed using the CNN-LSTM deep learning model based on chaotic time series. Comparatively, the CNN-LSTM and PSO-GRNN models are the most suitable to predict the risk level of the POD12 in the future.