Combining lymph node ratio to develop prognostic models for postoperative gastric neuroendocrine neoplasm patients

Area of Science:

• Oncology
• Surgical Pathology
• Biostatistics

Background:

• The lymph node ratio (LNR) is a significant prognostic factor in various cancers.
• Limited research exists on the prognostic value of LNR in gastric neuroendocrine neoplasms (NENs).

Purpose of the Study:

• To evaluate the prognostic significance of LNR in patients with resected gastric NENs.
• To develop and validate predictive models for disease-specific survival (DSS) using LNR.

Main Methods:

• Utilized data from the SEER database (training/validation sets) and a single-institution cohort (test set).
• Employed Cox regression, Random Survival Forest (RSF), and Cox proportional hazards (CoxPH) models.
• Compared model performance against the 8th edition AJCC TNM staging system.

Main Results:

• LNR was identified as an independent prognostic factor for DSS in gastric NEN patients.
• The RSF model demonstrated superior predictive performance (C-index 0.769) compared to CoxPH (0.744) and AJCC TNM (0.723) in the test set.
• RSF model showed good calibration, clinical utility, and effective risk stratification.

Conclusions:

• Elevated LNR correlates with decreased DSS in postoperative gastric NEN patients.
• The RSF model offers improved prognostic prediction for gastric NENs over existing methods, with potential clinical applicability.