CT-Based Body Composition Metrics for Prognosis in Advanced Pancreatobiliary Cancer

A retrospective study of 1,066 metastatic or recurrent pancreatobiliary cancer patients diagnosed between 2006 and 2017. The dataset includes baseline clinical, laboratory, and nutritional data, along with CT-derived body composition metrics. It was authored by Se Eung Oh and last updated on 2026-05-15.

Use Cases

• Build prognostic models for overall survival based on CT-measured body composition parameters.
• Analyze the relationship between sarcopenia, adiposity indices, and clinical factors like performance status.
• Validate a composite risk score integrating muscle and adipose tissue metrics for patient stratification.
• Study the interaction between systemic inflammatory markers (like CRP) and body composition in cancer outcomes.

Strengths

• Includes data from 1,066 patients, providing a substantial sample size for analysis.
• CT-derived metrics show excellent interobserver agreement (ICC range, 0.969–0.995).
• The composite risk score demonstrated significant prognostic discrimination with median OS ranging from 4.9 to 11.7 months.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• The data is stored in a DOCX file format, which is less structured for direct analysis than CSV or database formats.

Provenance

Source

figshare

Collection Method

Retrospective analysis of patient records and CT images.

Time Range

Patients diagnosed between 2006 and 2017.

Freshness

Last updated 2026-05-15 08:36:04; freshness should be verified.