Clinical Scorecard: A decision tree analysis reveals significant interaction between tumor size and close surgical margins affecting outcomes in limb salvage surgery for high-grade osteosarcoma

At a Glance

Key Highlights

• Close peri-neurovascular margin (PNM) <2.4 mm is a risk factor for local recurrence and survival in small-sized tumors.
• Tumor size greater than 31.6% of the limb predicts poor prognosis regardless of surgical margin status.
• Significant interaction effect between tumor size and PNM on local recurrence, metastasis, and overall survival demonstrated by decision tree and survival analyses.

Guideline-Based Recommendations

Diagnosis

• Use MRI to assess tumor size and proximity to neurovascular structures preoperatively.
• Classify surgical margins based on peri-neurovascular margin width, considering <15 mm as close margin.

Management

• Perform limb salvage surgery aiming for wide margins where possible, balancing preservation of vital structures.
• Administer standardized pre- and postoperative chemotherapy (doxorubicin, cisplatin, methotrexate, ifosfamide) per IOR/OS-N4 protocol.

Monitoring & Follow-up

• Monitor for local recurrence especially in patients with close PNM and small tumor size.
• Surveillance for metastasis and overall survival outcomes should consider tumor size and margin status interaction.

Risks

• Close peri-neurovascular margins increase local recurrence risk in small tumors.
• Large tumor size (>31.6% limb involvement) portends poor prognosis regardless of margin status.

Patient & Prescribing Data

High-grade extremity osteosarcoma patients undergoing limb salvage surgery with close surgical margins

Standardized chemotherapy regimen combined with surgical margin assessment informs prognosis; close margins require careful consideration in small tumors.

Clinical Best Practices

• Evaluate tumor size quantitatively relative to limb size to stratify risk.
• Aim for peri-neurovascular margins >2.4 mm when feasible to reduce local recurrence in small tumors.
• In large tumors, recognize that prognosis is poor regardless of margin width, guiding patient counseling and management.
• Use machine learning models such as decision trees to integrate multiple prognostic factors for personalized risk assessment.

Related Resources & Content

• Institutional Research Review Board of Ruijin Hospital, Shanghai Jiaotong University School of Medicine
• IOR/OS-N4 chemotherapy protocol