Reporting Surgical Resection Margin Status for Osteosarcoma: Comparison of the AJCC, MSTS, and Margin Distance Methods.

Cates JM
Am J Surg Pathol. 2017 41 (5): 633-642

PMID: 28125453 · DOI:10.1097/PAS.0000000000000815

Multiple different schemes are used to assess surgical resection margins in orthopedic pathology, but which is optimal for reporting resection margin status of osteosarcoma is uncertain. Moreover, the minimum tumor clearance (metric width of resection margin) necessary for local control is not well defined. In this investigation, the American Joint Committee on Cancer (AJCC) R system, Musculoskeletal Tumor Society (MSTS) system, and margin distance method for reporting resection margin status were compared in a series of 186 high-grade osteosarcomas. Hazard ratios for local recurrence for each resection margin category were compared with other categories within each margin assessment scheme to assess discriminatory ability. Cross-model comparisons of regression coefficients from parametric survival and logistic regression models were also performed. Predictive accuracy of each margin assessment scheme for determining 2-year local recurrence-free survival was evaluated by comparing the areas under receiver-operating characteristic curves generated from logistic regression analyses. Concordance with clinical outcomes was also calculated. Both the MSTS and margin distance schemes showed significantly greater predictive accuracy and concordance with observed outcomes than the AJCC R system. A margin distance of ≥2 mm significantly decreased the risk of local recurrence. Results were similar after adjustment for confounding prognostic factors (anatomic site, macroscopic lymphovascular invasion, and chemotherapy status). Therefore, surgical resection margins for osteosarcoma should be reported using either the MSTS or margin distance method instead of the AJCC R system.

MeSH Terms (26)

Adolescent Adult Area Under Curve Bone Neoplasms Decision Support Techniques Disease-Free Survival Female Humans Kaplan-Meier Estimate Logistic Models Male Margins of Excision Multivariate Analysis Neoplasm, Residual Neoplasm Grading Neoplasm Recurrence, Local Osteosarcoma Osteotomy Predictive Value of Tests Proportional Hazards Models Risk Assessment Risk Factors ROC Curve Time Factors Treatment Outcome Young Adult

Connections (1)

This publication is referenced by other Labnodes entities: