Background Gastric cancer liver metastasis (GCLM) patients usually accompany by abnormal serum liver function tests (LFTs) more or less; however, the prognostic value of LFTs is not fully understood. This study aimed to develop a liver chemistry score (LCS) based on LFTs and incorporate it into prognosis determination for GCLM patients who received palliative chemotherapy. Methods Data were derived from hospitalized GCLM patients in two general hospitals in China. LCS was generated based on the results of LFTs by LASSO regression. Cutoff value of the score was determined by restricted cubic spline. The score was then incorporated into Cox regression analysis to construct a predictive nomogram; the model was then evaluated internally and externally by AUC of time‐dependent receiver operating characteristic curves (ROC) and calibration curves. Results Three hundred and thirty‐six and 72 patients were included in development and validation cohort, respectively. LASSO regression analysis in development cohort finally reached a two‐parametric LCS calculated on AST and ALP levels as 0.03343515 × ln (AST, U/L) + 0.02687997 × ln (ALP, U/L), and 0.232 was set as optimal cutoff value. Patients in low (LCS < 0.232) or high (LCS ≥ 0.232) score group experienced different survival times; median OS was 13.54 (95% CI: 11.1–15.6) months in the low LCS group and 7.3 (6.6–9.3) months in the high LCS group ( p < 0.001). A nomogram including LCS and other clinical parameters was constructed and showed superior performance than model not including LCS. AUC of 6‐month ROC improved from 0.647 (95% CI: 0.584–0.711) to 0.699 (0.638–0.759) in internal validation, and 0.837 (0.734–0.940) to 0.875 (0.784–0.966) in external validation. Conclusions Liver chemistry score is useful in determining the prognosis of gastric cancer patients with liver metastasis and may be helpful to clinicians in decision‐making.

Purpose This study aimed to identify prognostic factors for patients with distant lymph node-involved gastric cancer (GC) using a machine learning algorithm, a method that offers considerable advantages and new prospects for high-dimensional biomedical data exploration. Materials and Methods This study employed 79 features of clinical pathology, laboratory tests, and therapeutic details from 289 GC patients whose distant lymphadenopathy was presented as the first episode of recurrence or metastasis. Outcomes were measured as any-cause death events and survival months after distant lymph node metastasis. A prediction model was built based on possible outcome predictors using a random survival forest algorithm and confirmed by 5×5 nested cross-validation. The effects of single variables were interpreted using partial dependence plots. A contour plot was used to visually represent survival prediction based on 2 predictive features. Results The median survival time of patients with GC with distant nodal metastasis was 9.2 months. The optimal model incorporated the prealbumin level and the prothrombin time (PT), and yielded a prediction error of 0.353. The inclusion of other variables resulted in poorer model performance. Patients with higher serum prealbumin levels or shorter PTs had a significantly better prognosis. The predicted one-year survival rate was stratified and illustrated as a contour plot based on the combined effect the prealbumin level and the PT. Conclusions Machine learning is useful for identifying the important determinants of cancer survival using high-dimensional datasets. The prealbumin level and the PT on distant lymph node metastasis are the 2 most crucial factors in predicting the subsequent survival time of advanced GC. Trial Registration ChiCTR Identifier: ChiCTR1800019978