Related Experiment Videos
Integrated metabolomics and machine learning identify predictive biomarkers via SHAP analysis for sintilimab-induced
Wenxiu Tian1, Hehe Tang1, Xiaoyuan Liu1
1Department of Clinical Pharmacy, Affiliated Hospital of Jiangnan University, Wuxi, Jiangsu Province, China.
Background:
Sintilimab-induced rash is a significant clinical challenge in lung cancer treatment, often necessitating therapy interruption or discontinuation and thereby compromising patient outcomes. The underlying mechanisms of this adverse event remain poorly understood. This study aimed to investigate potential predictive biomarkers and mechanisms of sintilimab-induced rash through metabolomic profiling.
Methods:
A total of 55 patients with lung cancer who received sintilimab were enrolled, including 32 who developed rash and 23 matched controls without rash. Blood samples were collected before sintilimab infusion and at rash onset. Comprehensive clinical data were recorded. Untargeted metabolomic analysis of plasma was performed using ultra-high-performance liquid chromatography-tandem mass spectrometry (UHPLC-MS/MS). Differential metabolites were identified and analyzed using pathway enrichment, univariate analysis (AUC ≥0.800), and SHAP analysis.
Results:
No significant differences were observed between groups in demographic characteristics or most clinical parameters. However, the rash group exhibited significantly elevated total bile acids glucose (GLU), and basophil percentage (BAS%), along with reduced AST/ALT ratio, alkaline phosphatase lactate dehydrogenase phosphorus (P), neutrophil count (NEU), and high-sensitivity C-reactive protein (hsCRP) (P < 0.05). Metabolomic analysis identified 92 differentially expressed metabolites. Pathway enrichment revealed alterations in oxytocin signaling, GnRH signaling, platelet activation, FcγR-mediated phagocytosis, retrograde endocannabinoid signaling, pantothenate and CoA biosynthesis, FcεRI signaling, and aldosterone synthesis and secretion. Univariate analysis identified 25 metabolites with high predictive value (AUC ≥0.800), and SHAP analysis highlighted 20 metabolites. Cross-comparison identified five overlapping metabolites: N,N,N-trimethyl-L-histidine, laurolactam, 2-naphthalenesulfonic acid, limonenecarboxylic acid, and N-lauroylsarcosine.
Conclusion:
Distinct clinical and metabolomic alterations are associated with sintilimab-induced rash in lung cancer patients. The identified differential metabolites may serve as predictive biomarkers and potential therapeutic targets, providing new insights for clinical management and mechanistic research into immune-related adverse events.