Automated 3D body composition analysis on chest CT scans for survival prediction in high-grade extremity soft tissue sarcomas.

Higher IMAT volume was associated with shorter overall survival, disease-free survival, and local recurrence-free survival in patients with high-grade extremity soft tissue sarcomas.

• IMAT was assessed using dedicated AI-based automated 3D quantitative analysis software on staging chest CT scans
• The study was a single-center retrospective study at the Jules Bordet Institute including patients diagnosed between January 2010 and January 2023
• IMAT volume showed statistically significant associations with OS, DFS, and local recurrence-free survival
• IMAT did not show a significant association with metastatic recurrence-free survival

Higher PAT volume was associated with shorter overall survival, disease-free survival, and local recurrence-free survival.

• PAT (pericardial adipose tissue) was among the anatomical compartments quantified by the automated 3D volumetric analysis
• PAT volume was significantly associated with OS, DFS, and local recurrence-free survival
• PAT did not show a significant association with metastatic recurrence-free survival
• PAT was measured from staging chest CT scans as part of the initial evaluation workup

Increased EAT volume was associated with reduced overall survival.

• EAT (epicardial adipose tissue) was one of the anatomical compartments quantified using the dedicated AI software
• EAT showed a statistically significant association specifically with OS
• EAT did not show significant associations with DFS, local recurrence-free survival, or metastatic recurrence-free survival

Higher VAT volume was associated with worse overall survival and disease-free survival.

• VAT (visceral adipose tissue) was quantified using automated 3D analysis on chest CT scans
• VAT volume showed significant associations with both OS and DFS
• VAT did not show significant associations with local recurrence-free survival or metastatic recurrence-free survival

The study used an AI-based imaging software to perform automated 3D quantitative analysis of multiple body composition compartments from routine staging chest CT scans.

• Anatomical compartments analyzed included IMAT, PAT, EAT, and VAT
• The software performed rapid volumetric (3D) assessment rather than conventional single-slice 2D measurements
• CT scans were those already indicated by current guidelines for systematic chest imaging during initial assessment of extremity high-grade soft tissue sarcomas
• The study was retrospective and single-center, covering patients diagnosed between January 2010 and January 2023

The study evaluated body composition metrics against four survival endpoints: overall survival, disease-free survival, local recurrence-free survival, and metastatic recurrence-free survival.

• Four distinct survival outcomes were analyzed to capture different aspects of disease progression
• Local recurrence-free survival was significantly associated with IMAT and PAT but not EAT or VAT
• Metastatic recurrence-free survival was not significantly associated with any of the body composition metrics evaluated
• The study population comprised patients with high-grade soft tissue sarcomas of the extremities, described as rare cancers with poor prognosis