ML Decision Model for G-NEC Adjuvant Therapy (G-NEC)

Machine Learning-Based Decision Model for Optimal Adjuvant Therapy in Primary Gastric Neuroendocrine Carcinoma: a National Real-World Evidence Study

Gastric neuroendocrine carcinoma (G-NEC) is a rare and aggressive tumor originating from neuroendocrine cells in the stomach lining. It is characterized by a high propensity for recurrence and a generally poor prognosis. Due to its rarity, there is limited data and no established consensus on the optimal postoperative adjuvant therapy, making treatment decisions challenging for healthcare providers.

This study is a retrospective analysis focusing on evaluating survival rates, identifying prognostic factors, and formulating treatment recommendations for patients with G-NEC. By analyzing real-world clinical data, we aim to better understand the factors that influence patient outcomes and to develop evidence-based strategies for improving survival. Our goal is to provide clinicians with valuable insights and tools to make more informed treatment decisions, ultimately enhancing the quality of care and outcomes for patients with this challenging disease.

Study Overview

• Status: Completed
• Conditions: Machine Learning; Survival Outcomes; Gastric Neuroendocrine Carcinoma (G-NEC); Postoperative Adjuvant Therapy for G-NEC

• Study Type: Observational
• Enrollment (Actual): 1505

Contacts and Locations

Study Locations

• China
  • Fujian
    • Fuzhou, Fujian, China, 350001
      • Fujian Medical University

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Child; Adult; Older Adult
• Accepts Healthy Volunteers: No

• Sampling Method: Non-Probability Sample

Study Population

This study includes adult patients diagnosed with gastric neuroendocrine carcinoma (G-NEC) or mixed adenoneuroendocrine carcinoma (MANEC) who underwent radical surgery at 38 tertiary hospitals in China between January 2006 and December 2020. The study population consists of patients who received either no adjuvant chemotherapy, etoposide and platinum derivatives-based chemotherapy, or fluorouracil-based chemotherapy following surgery.

Description

Inclusion Criteria:

• (1) patients who underwent radical surgery without any neoadjuvant therapy;
• (2) pathology confirmed NEC or mixed adenoneuroendocrine carcinoma (MANEC).

Exclusion Criteria:

• (1) history of other malignant neoplasms;
• (2) treatment with endoscopic submucosal dissection or endoscopic mucosal resection or thoracotomy;
• (3) incomplete clinical data (including pathological, adjuvant chemotherapy, and follow-up information);
• (4) receipt of alternative adjuvant treatment regimens;
• (5) death within 30 days postoperatively.

Study Plan

How is the study designed?

Number of groups / cohorts

1

Cohorts and Interventions

Group / Cohort

Gastric Neuroendocrine Carcinoma (G-NEC) Patients; This study focuses on patients diagnosed with gastric neuroendocrine carcinoma (G-NEC) who have undergone radical surgery. The cohort includes adult patients (≥18 years) treated at 38 tertiary hospitals in China between January 2006 and December 2020. Patients are divided into three groups based on their postoperative adjuvant treatment: no adjuvant chemotherapy, etoposide and platinum derivatives-based chemotherapy, and fluorouracil-based chemotherapy. The study aims to develop and validate a machine learning-based decision support model to optimize individualized adjuvant therapy strategies for G-NEC patients, with the primary outcome being disease-free survival (DFS).

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Disease-Free Survival (DFS)	Disease-free survival is defined as the time from the date of surgery to disease recurrence, death from any cause, or last follow-up, whichever occurs first. The machine learning model's performance in predicting DFS and recommending optimal adjuvant therapy will be evaluated.	From date of surgery up to 5 years

Collaborators and Investigators

• Sponsor: Chang-Ming Huang, Prof.

Study record dates

Study Major Dates

• Study Start (Actual): January 1, 2024
• Primary Completion (Actual): June 1, 2024
• Study Completion (Actual): June 30, 2024

Study Registration Dates

• First Submitted: October 26, 2024
• First Submitted That Met QC Criteria: October 26, 2024
• First Posted (Actual): October 29, 2024

Study Record Updates

• Last Update Posted (Estimated): November 27, 2024
• Last Update Submitted That Met QC Criteria: November 25, 2024
• Last Verified: October 1, 2024

More Information

Terms related to this study

• Keywords: Machine Learning; Adjuvant Chemotherapy; Gastric Neuroendocrine Carcinoma (G-NEC); Decision Support Model; Random Survival Forest
• Additional Relevant MeSH Terms: Neoplasms; Neoplasms by Histologic Type; Neoplasms, Glandular and Epithelial; Adenocarcinoma; Neuroectodermal Tumors; Neoplasms, Germ Cell and Embryonal; Neoplasms, Nerve Tissue; Neuroendocrine Tumors; Carcinoma; Carcinoma, Neuroendocrine
• Other Study ID Numbers: 2024GNEC

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: UNDECIDED

IPD Plan Description

Patient Privacy: As this study involves sensitive medical information, we must ensure that any data sharing plan complies with all relevant privacy laws and regulations.

Ethical Considerations: We are consulting with our ethics committee to determine the most appropriate approach to data sharing that respects patient consent and maintains the integrity of the research.

Data Standardization: Given the multi-center nature of this study, we are working on standardizing our data collection and storage processes across all 38 participating hospitals to ensure consistency and quality of any potentially shared data.

Collaborative Potential: We recognize the value of data sharing in advancing gastric neuroendocrine carcinoma research and are exploring potential collaborations with other research groups.

Resource Allocation: We are assessing the resources required to prepare the data for sharing, including de-identification processes and creation of data dictionaries.

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No
• product manufactured in and exported from the U.S.: No