Hematological Dynamic Scores for Predicting Survival and Treatment Response for Advanced Gastric Cancer After Neoadjuvant Therapy

Gastric Cancer

Gastric cancer, Neoadjuvant chemotherapy, inflammation, Oxidative stress, Treatment response, Risk stratification

HMDLS, based on hematological markers, could effectively distinguish the long-term efficacy of AGC patients after NAT. The predictive performance of nomogram-HMDLS was better than ypTNM stage, achieving better prognostic stratification and tumor treatment response prediction.

In this research, we incorporated a total of 320 patients from the Union Hospital of Fujian Medical University to form the training cohort (TC). Additionally, we included 122 patients from four distinct medical centers to serve as the external validation cohort (EVC). The Hematological Marker Dynamic Load (ΔHMDL) was determined using the following formula: ΔHMDL = (HMDL pre-surgery - HMDL pre-NAT) / HMDL pre-NAT, where HMDL represents the hematological marker levels before surgery and before the initiation of Neoadjuvant Therapy (NAT), respectively. Employing LASSO regression analysis, we identified the most influential and statistically significant ΔHMDL indicators. These were then utilized to compute the Hematological Marker Dynamic Load Score (HMDLS), defined as: HMDLS = Σ(LASSO coefficient \* ΔHMDL), where the summation encompasses the products of the LASSO-estimated coefficients and the corresponding ΔHMDL values. Further, leveraging the outcomes of a multivariate COX regression analysis, we integrated clinical parameters with the HMDLS to formulate a predictive model, termed the Nomogram-HMDLS model. The efficacy of this model in terms of predictive accuracy, clinical utility, and calibration was meticulously assessed and confirmed through several metrics, including the concordance index (C-index), Receiver Operating Characteristic (ROC) curve analysis, decision curve analysis (DCA), and calibration curves.

No related papers found yet.

None listed

China