Interpretable Machine Learning Models for Prognosis in Gastric Cancer Patients

Development and Validation of Interpretable Machine Learning Models for Prognosis in Gastric Cancer Patients: a Multicenter Retrospective Study

Status

Completed

Phases

Unknown

Study type

Observational

Source

ClinicalTrials.gov

Registry ID

NCT06548464

Enrollment

18000

Registered

2024-08-12

Start date

2024-06-01

Completion date

2024-08-06

Last updated

2024-08-12

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Gastrectomy, Machine Learning, Stomach Neoplasms

Keywords

Machine Learning, Prognosis, Stomach Neoplasms, Gastrectomy

Brief summary

This multicenter, retrospective cohort study aimed to develop and validate an explainable prediction model for prognosis after gastrectomy in patients with gastric cancer.

Detailed description

This multicenter, retrospective cohort study aimed to develop and validate an explainable prediction model for prognosis after gastrectomy in patients with gastric cancer. The study included patients who underwent radical gastrectomy for primary gastric or gastroesophageal junction cancer across multiple institutions in China. The primary objective was to create a machine learning-based model to predict postoperative outcomes following gastrectomy, using readily available clinical and pathological parameters. The main outcome of interest was early recurrence within 2 years after surgery, which significantly impacts overall prognosis. The study employed various machine learning algorithms to develop prediction models, which were then compared and validated. Model performance was assessed through measures such as area under the receiver operating characteristic curve (AUC), calibration, and Brier score. The SHapley Additive exPlanations (SHAP) method was used to interpret the model and rank feature importance. This research aims to provide clinicians with a tool for identifying patients at higher risk of poor postoperative outcomes who may benefit from more intensive post-operative monitoring and early intervention strategies, potentially improving prognosis for gastric cancer patients.

Interventions

None listed

Study design

Observational model

COHORT

Time perspective

RETROSPECTIVE

Eligibility

Sex/Gender

ALL

Healthy volunteers

Inclusion criteria

* Patients diagnosed with primary gastric or gastroesophageal junction cancer * Underwent radical gastrectomy * Complete clinical and pathological data available

Exclusion criteria

* Presence of distant metastases before surgery * Non-adenocarcinoma histology * Incomplete follow-up data

Design outcomes

Primary

Measure	Time frame	Description
Survival	Up to 5 years after surgery	Assessment of overall survival outcomes in gastric cancer patients after gastrectomy.

Secondary

Measure	Time frame	Description
Early Recurrence	Within 2 years after surgery	Incidence of cancer recurrence within 2 years after gastrectomy.
Late Recurrence	From 2 years up to 5 years after surgery	Incidence of cancer recurrence occurring more than 2 years after gastrectomy.
Postoperative Complications	Within 30 days after surgery	Incidence and severity of complications following gastrectomy.
Neoadjuvant Treatment Efficacy	From initiation of neoadjuvant therapy to surgery (typically 2-3 months)	Assessment of tumor response to neoadjuvant therapy before gastrectomy.
5-Year Survival Rate	5 years after surgery	Percentage of patients alive 5 years after gastrectomy.

Countries

China

Outcome results