Construction of a Prognostic Predicting Model for Advanced Gastric Cancer Patients Treated with PD-1 Inhibitors in Combination with Taxane-Based Chemotherapy

Abstract

Background: Advanced gastric cancer (AGC) patients may benefit from programmed cell death protein 1 (PD-1) inhibitors, but this is not the case for everyone. Some patients need a predicting model to assess immunotherapy efficacy. Methods: This is a retrospective study that analyzes 12 biomarkers of 70 AGC patients receiving PD-1 inhibitors in combination with taxane-based chemotherapy to construct a prognostic index (PI) equation to assess the associations between biomarkers and outcomes alongside with the value of the prognostic risk prediction model in patient prognostication. The biomarkers are carcinoembryonic antigen (CEA), carbohydrate antigen 125 (CA125), CA153, CA199, high-sensitivity C-reactive protein (hs-CRP), eosinophils, neutrophil-to-lymphocyte ratio (NLR), body mass index (BMI), albumin/globulin (A/G) ratio, prognostic nutritional index (PNI), lactate dehydrogenase (LDH), creatine kinase (CK). Results: Baseline low CA153 and NLR as well as high BMI and CK, together with critical eosinophilic rise of 0.01 × 10⁹/L and an early expansion in BMI after 2 rounds of therapy might be valuable prognostic biomarkers of PFS in AGC patients. The PI equation is ∑βixi = 0.778 X1 + 0.726 X2 – 0.717 X3 – 0.682 X4, where X1 represents baseline CA153, and the value is 1 for CA153 ≥13.35 U/mL and 0 for CA153 <13.35 U/mL;X2 represents baseline NLR and is 1 for NLR ≥3.08 and 0 for NLR <3.08;X3 represents baseline BMI and is 1 for BMI ≥21.3 kg/m² and 0 for BMI <21.3 kg/m²;X4 represents baseline CK and is 1 for CK ≥60.3 U/L and 0 for CK <60.3 U/L. The difference in survival rate between two prognostic risk groups based on the PI was significant (p < 0.001). Conclusions: The PI shown in this study can forecast the prognosis of AGC patients using immunotherapy. Additionally, the PI can be used to guide the follow-up treatment and provide reference for the formulation of more acceptable therapeutic regimes.