A Prognostic Model Based on Nutritional Indexes for Patients With Pan-Cancer: A Real-World Cohort Study

Area of Science:

• Oncology
• Nutritional Science
• Biostatistics

Background:

• Pan-cancer research necessitates accurate prognostic tools.
• Nutritional status is increasingly recognized as a critical factor in cancer outcomes.
• Predicting individual survival probability is essential for treatment planning.

Purpose of the Study:

• To identify key nutritional indexes for pan-cancer prognosis.
• To construct and validate a predictive model and nomogram for survival probability.
• To stratify pan-cancer patients into distinct survival risk groups.

Main Methods:

• Collected nutritional indicators, clinicopathological data, and treatment details from 2020 patients.
• Utilized Least Absolute Shrinkage and Selection Operator (Lasso) regression for variable selection.
• Developed a Cox proportional hazards model and visualized it as a nomogram, validated in separate cohorts.

Main Results:

• A prognostic model incorporating 13 characteristics was developed.
• The model demonstrated strong discrimination with C-indexes ranging from 0.799 to 0.851 across different time points and cohorts.
• The model accurately stratified patients into different survival risk groups, with significant differences in overall survival (OS).

Conclusions:

• A validated prognostic model based on nutritional indexes can effectively predict survival in pan-cancer patients.
• The nomogram provides a practical tool for assessing individual survival risk.
• This model aids in stratifying patients and potentially guiding personalized therapeutic decisions.