Patient Survival and Body Composition Data from Early-Phase ADC Trials, 2015-2023

From March 2015 to March 2023, a retrospective analysis of 136 cancer patients treated with antibody-drug conjugates (ADCs) in early-phase clinical trials. The dataset includes AI-quantified 3D body composition metrics from pre-treatment CT scans and their association with survival outcomes and toxicity. It was created by Matthieu Delaye and shared under a CC-BY-4.0 license.

Use Cases

• Predicting progression-free survival based on AI-quantified subcutaneous adipose tissue (SAT) and total adipose tissue (TAT) metrics.
• Modeling overall survival using visceral adipose tissue (VAT) and SAT measurements derived from CT scans.
• Analyzing the association between body composition parameters and all-grade toxicity events in ADC-treated patients.
• Investigating the prognostic value of the Royal Marsden Hospital (RMH) score in conjunction with anthropometric data.

Strengths

• Includes data from 136 patients with detailed clinical and AI-derived body composition metrics.
• Provides specific hazard ratios and p-values for survival associations, such as HR=0.67 for SAT and PFS.
• Covers a defined time range of eight years (March 2015 to March 2023) for patient inclusion.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• The dataset is very small at 12.6 KB, indicating limited scope.

Provenance

Source

figshare

Collection Method

Retrospective analysis of patients treated at a single institution, using a deep learning software (Anthropometer3DNet) to quantify body composition from CT scans.

Time Range

March 2015 to March 2023

Freshness

Last updated 2026-05-13 04:56:26; freshness should be verified.