Development and Validation of Prognostic Nomograms for Elderly Patients with Osteosarcoma.

Area of Science:

• Oncology
• Biostatistics
• Geriatric Medicine

Background:

• Osteosarcoma disproportionately affects elderly patients, a population often underrepresented in clinical trials.
• Accurate prognostic tools are crucial for managing treatment and predicting outcomes in this demographic.

Purpose of the Study:

• To develop and validate prognostic nomograms for predicting cancer-specific survival (CSS) and overall survival (OS) in elderly osteosarcoma patients.
• To identify key clinical predictors influencing survival in this patient group.

Main Methods:

• Utilized data from 816 elderly patients (≥40 years) with osteosarcoma from the SEER database (2004-2016).
• Employed Least Absolute Shrinkage and Selection Operator (Lasso) Cox regression to identify significant predictors.
• Constructed nomograms for 1-, 3-, and 5-year CSS and OS, validated internally.

Main Results:

• Identified five key predictors: age, chemotherapy, surgery, AJCC stage, and summary stage.
• Nomograms demonstrated satisfactory performance, validated by C-index, calibration, and decision curve analysis.
• Grade and M stage showed differential associations with OS and CSS, respectively.

Conclusions:

• The developed nomograms serve as effective tools for precise prognosis prediction in elderly osteosarcoma patients.
• These nomograms can aid clinicians in optimizing patient management and improving clinical decision-making.