A web-based prognostic tool for extremity and trunk wall soft tissue sarcomas and its external validation

Area of Science:

• Oncology
• Medical Statistics
• Surgical Oncology

Background:

• Soft tissue sarcomas (STS) are rare malignancies requiring accurate prognostic tools.
• Predicting long-term survival is crucial for patient management and treatment planning.
• Existing prognostic models may not fully capture the complexity of STS behavior.

Purpose of the Study:

• To develop and validate an improved prognostic tool for extremity and trunk wall soft tissue sarcomas.
• To predict 10-year sarcoma-specific survival with enhanced accuracy.
• To make the prognostic tool accessible via a web-based calculator.

Main Methods:

• Development of two Cox proportional hazards models (SIN and SAM) using data from Helsinki University Central Hospital (1987-2002).
• The SIN model included size, necrosis, and vascular invasion.
• The SAM model extended SIN by adding grade, depth, location, and continuous tumor size. External validation was performed using data from Lund University Hospital.

Main Results:

• The extended SAM model demonstrated superior accuracy in predicting 10-year sarcoma-specific survival.
• Area Under the ROC Curve (AUC) for SAM was 0.81 (Helsinki) vs. 0.74 for SIN (P=0.0007).
• External validation showed AUCs of 0.77 for SAM and 0.73 for SIN (P=0.03), confirming improved predictive performance.

Conclusions:

• The addition of grade, depth, location, and continuous tumor size significantly enhances the accuracy of STS prognostic models.
• The SAM model offers a more precise prediction of 10-year sarcoma-specific survival compared to models relying solely on size, necrosis, and vascular invasion.
• A web-based calculator for the SAM model is available, facilitating clinical application.