Artificial Intelligence-Guided Identification of IGFBP7 as a Critical Indicator in Lactic Metabolism Determines Immunotherapy Response in Stomach Adenocarcinoma

Abstract

Due to considerable tumour heterogeneity, stomach adenocarcinoma (STAD) has a poor prognosis and varies in response to treatment, making it one of the main causes of cancer-related mortality globally. Recent data point to a significant role for metabolic reprogramming, namely dysregulated lactic acid metabolism, in the evolution of STAD and treatment resistance. This study used a series of artificial intelligence-related approaches to identify IGFBP7, a Schlafen family member, as a critical factor in determining the response to immunotherapy and lactic acid metabolism in STAD patients. Computational analyses revealed that a high lactic metabolism (LM) state was associated with poor survival in STAD patients. Further biological network-based investigations identified a key subnetwork closely linked to LM. Machine learning techniques, such as random forest and least absolute shrinkage and selection operator, highlighted IGFBP7 as a crucial indicator in STAD. Functional annotations showed that IGFBP7 expression was linked to important immune and inflammatory pathways. In vitro experiments demonstrated that silencing IGFBP7 suppressed cell proliferation and migration. Furthermore, heightened susceptibility to several chemotherapeutic drugs was linked to elevated IGFBP7 levels. In conclusion, this work sheds light on the mechanisms by which the lactate metabolism-related indicator IGFBP7 affects the tumour immune milieu and the response to immunotherapy in STAD. The results point to IGFBP7 as a possible therapeutic target and predictive biomarker for the treatment of STAD.

Keywords: IGFBP7; STAD; immunotherapy; lactic metabolism; machine learning.