Yi Wang

Yi Wang

Auto-generated profile, based on publicly available information

Yi Wang

Department of Cancer Screening and Prevention Zhongshan Hospital, Fudan University Shanghai China

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
5 July 2025 in International Journal of Medical Informatics
Verified
Journal Article
Open access
YWYi Wang
Yi Wang
JLJing Lei
Jing Lei
ZJZhiping Jin
Zhiping Jin
YJYing Jiang
Ying Jiang
NZNingping Zhang
Ningping Zhang
TLTianshu Liu
Tianshu Liu
Immune checkpoint inhibitor (ICI)-related liver injury poses a considerable clinical challenge for cancer patients. This study aimed to develop and validate an interpretable predictive model employing machine learning (ML) algorithms to accurately identify patients at high risk of acute liver injury within one month of initiating ICI treatment. This longitudinal cohort study included pan-cancer patients who received their first ICI treatment between March 2019 and September 2022 at Zhongshan Hospital. Six ML algorithms, namely neural networks (NN), gradient boosting classifier (GBC), eXtreme gradient boosting (XGBoost), logistic regression (LR), categorical boosting classifier (CatBoost) and random forest (RF), were utilized to construct predictive models for acute ICI-related liver injury. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and brier score (BS). The SHapley Additive exPlanations (SHAP) method was applied to rank the feature importance and interpret the final model, providing insights into the contribution of each feature to liver injury prediction, thereby enhancing clinical interpretability. This study is registered with the Chinese Clinical Trial Registry (ChiCTR2300067470). A total of 863 patients were enrolled in the study, with 22.71% experiencing liver injury within one month of ICI initiation. Among the six preliminary models, the RF model exhibited the best performance and was selected for the development of the final model. The SHAP method was utilized to rank variables from the six pre-models, with 10 variables selected for the final model by identifying the intersection of the top 20 most important variables across these models. The final RF model exhibited robust performance, achieving an AUC of 0.81 (95% CI: 0.73-0.90) on the test set, and 0.79 (95% CI: 0.72-0.88) and 0.80 (95% CI: 0.72-0.89) in the 5-fold and 10-fold cross-validation, respectively. The Decision Curve Analysis (DCA) curve illustrated solid clinical benefit, and the calibration curve reflected good predictive consistency. An interpretable RF model was developed to predict acute liver injury occurring within one month after ICI treatment. This clinical-friendly model enables early identification of high-risk patients, facilitating optimized clinical management and ultimately improving treatment outcomes.

15 October 2026 in Cancer Medicine
Verified
Journal Article
Open access
JLJing Lei
Jing Lei
YYYiyi Yu
Yiyi Yu
QLQing Liu
Qing Liu
NZNingping Zhang
Ningping Zhang
YWYi Wang
Yi Wang
MLMinzhi Lv
Minzhi Lv
Background: Immune checkpoint inhibitors (ICIs) are increasingly used in patients with gastrointestinal cancer (GIC). However, ICI‐associated adverse event (AE) profiles in GIC remain poorly characterized, and whether AE incidence is associated with clinical outcomes is unclear. Methods: Randomized controlled trials (RCTs) were identified through February 2025. Primary outcomes included treatment‐related AEs (TRAEs) and immune‐related AEs (irAEs) (any grade/Grade 3+). Single‐arm random‐effects meta‐analysis pooled AE incidences; risk ratios (RRs) compared toxicity across ICI regimens. Digitized Kaplan–Meier curves reconstructed survival data for pooled analysis and Cox models. Weighted Spearman's tests assessed correlations between AE incidences and median overall survival (OS)/progression‐free survival (PFS). Results: Forty RCTs enrolling 10,233 patients investigated ICI monotherapy, dual‐ICI therapy, and ICI combination regimens. Pooled any‐grade TRAE incidences were 68.0% (monotherapy, 95% CI: 60%–76%, I 2 = 97.4%, p < 0.001), 87.0% (dual therapy, 95% CI: 73%–100%, I 2 = 98.0%, p < 0.001), and 90.0% (combination therapy, 95% CI: 83%–96%, I 2 = 94.6%, p < 0.001). IrAE incidences reached 24.0% (monotherapy, 95% CI: 18%–29%, I 2 = 67.2%, p < 0.05) and 37.0% (combination therapy, 95% CI: 25%–49%, I 2 = 98.8%, p < 0.001). Relative to monotherapy, dual‐ICI therapy elevated any‐grade (RR = 1.28, 95% CI: 1.05–1.55) and grade ≥ 3 TRAEs (RR = 2.71, 95% CI: 1.84–3.99); ICI combination therapy increased any‐grade (RR = 1.32, 95% CI: 1.15–1.51) and grade ≥ 3 TRAEs (RR = 2.98, 95% CI: 2.30–3.87). Any‐grade TRAEs moderately correlated with median OS ( R = 0.46, p < 0.001) and PFS ( R = 0.50, p < 0.001). Any‐grade irAEs strongly correlated with median OS ( R = 0.90, p < 0.001) and moderately with PFS ( R = 0.57, p < 0.001). Conclusion: While AE incidence is high among GIC patients receiving ICIs, most events are low‐grade. Dual‐ICI and ICI‐based combination therapies carry higher TRAE and irAE risks than monotherapy. At the trial level, higher non‐severe irAE incidence correlated with favorable survival outcomes; this exploratory ecological association does not apply to individual patients. Systematic Review Registration: This study was registered on International Platform of Registered Systematic Review and Meta‐analysis Protocols (INPLASY202230144).