Abstract The identification of reliable tumor prognostic markers can help clinicians and researchers predict tumor development and patient survival outcomes more accurately, which plays a vital role in clinical diagnosis, treatment effectiveness assessment, and prognostic evaluation. Existing web tools supporting online survival analysis are gradually failing to meet the increasing demands of researchers in terms of the dataset size, richness of survival analysis methods, and diversity of customization features. Therefore, there is an urgent need for a large-scale, one-stop pan-cancer survival analysis web server. We developed PanCanSurvPlot ( https://smuonco.shinyapps.io/PanCanSurvPlot/ ), a Shiny web tool that has incorporated a total of 215 cancer-related datasets from the GEO and TCGA databases, covering nearly 100,000 genes (mRNAs, miRNAs, and lncRNAs), approximately 45,000 samples, 51 different cancer types, and 13 different survival outcomes. The website also provides two cutoff methods based on median and optimal cutpoints. All survival analysis results from the log-rank test and univariate Cox regression are presented in a clear and straightforward summary table. Finally, users can customize color schemes and cutpoint levels to quickly obtain high-quality Kaplan-Meier survival plots that meet publication requirements.

PanCanSurvPlot: A Large-scale Pan-cancer Survival Analysis Web Application