Gastric cancer (GC) is one of the most common gastrointestinal malignancies worldwide, characterized by a high incidence rate, low screening rate, and subtle early symptoms. As a result, the majority of patients are diagnosed at an advanced or metastatic stage, contributing to poor overall prognosis. In recent years, the emergence and continuous advancement of immunotherapy have revolutionized the traditional treatment landscape for GC, offering new hope for precision medicine. Immunotherapy exerts its antitumor effects primarily by modulating the immunosuppressive tumor microenvironment and includes modalities such as immune checkpoint inhibitors (ICIs), adoptive cell therapies (ACTs), and cancer vaccines. Among these, ICIs and ACT have garnered significant attention. This review summarizes the underlying mechanisms, current applications, and major challenges of immunotherapy in GC. In addition, we discuss emerging biomarkers with potential utility for predicting immunotherapeutic efficacy in GC patients.

Exercise intervention demonstrates unique potential in treating emotional dysregulation, yet the ambiguity of its neuromodulation targets hinders the development of precise exercise prescriptions. This study investigates trait anxiety as a representative emotional disorder in 40 high-trait-anxiety university students, who were randomly assigned to either an exercise intervention group (40 min moderate-intensity aerobic exercise, n = 20) or a non-exercise control group (40 min quiet reading, n = 20), followed by resting EEG data collection. By integrating resting-state electroencephalography (EEG) after exercise with deep learning algorithms, we developed an alpha-band time-frequency predictive model to systematically decode the neural oscillatory reprogramming mechanisms in the prefrontal cortex induced by exercise. The deep learning model exhibited superior classification efficacy (accuracy 83.33%, F1 score 0.83, Kappa coefficient 0.67) in identifying exercise-induced alpha-band power spectral entropy alterations. This study pioneers the identification of prefrontal Alpha excitatory rebalancing through neural oscillation remodeling as the core mechanism underlying exercise-mediated anxiety mitigation.

Exercise intervention demonstrates unique potential in treating emotional dysregulation, yet the ambiguity of its neuromodulation targets hinders the development of precise exercise prescriptions. This study investigates trait anxiety as a representative emotional disorder in 40 high-trait-anxiety university students, who were randomly assigned to either an exercise intervention group (40 min moderate-intensity aerobic exercise, n = 20) or a non-exercise control group (40 min quiet reading, n = 20), followed by resting EEG data collection. By integrating resting-state electroencephalography (EEG) after exercise with deep learning algorithms, we developed an alpha-band time-frequency predictive model to systematically decode the neural oscillatory reprogramming mechanisms in the prefrontal cortex induced by exercise. The deep learning model exhibited superior classification efficacy (accuracy 83.33%, F1 score 0.83, Kappa coefficient 0.67) in identifying exercise-induced alpha-band power spectral entropy alterations. This study pioneers the identification of prefrontal Alpha excitatory rebalancing through neural oscillation remodeling as the core mechanism underlying exercise-mediated anxiety mitigation.