Methods

Following institutional review board approval, a comprehensive chart review of patients treated for osteosarcoma at Monroe Carell Junior Children’s Hospital at Vanderbilt and Vanderbilt University Medical Center between 2000 and 2022 was performed. Follow up data was obtained until December 2023. Patients who were treated medically at outside institutions were included and data was obtained through review of outside records. Patients with unresectable disease, unclear neoadjuvant regimen, or inadequate information for analysis were excluded from the study. Inclusion criteria included the diagnosis of osteosarcoma as confirmed by pathologic evaluation and surgical resection performed at Vanderbilt. Additional inclusion criteria included completion of two cycles of neoadjuvant MAP regimen (defined by two courses of cisplatin and doxorubicin and at least two but no more than six doses of methotrexate), and age less than 40 years. Demographic and clinical data collected include age at diagnosis, sex, histologic classification, grade, site of tumor, location in long bone (if applicable), stage, year of chemotherapy initiation, number of neoadjuvant methotrexate doses given, margin status at time of surgery, type of event (if applicable) after the completion of chemotherapy, event-free-survival (EFS), and duration of follow-up. Metastatic disease at the time of diagnosis was defined as a patient having one pulmonary nodule greater than one centimeter, greater than two pulmonary nodules equal to or larger than five millimeters, or distant metastasis identified on initial staging workup. The interval from completion of adjuvant therapy to surgery was defined as the duration between the date of completion of the last methotrexate dose to the date of surgical resection. EFS was calculated as the duration from the date of completion of initial therapy until the date of an event, including recurrent localized disease, new or relapsed metastatic disease, or death. Patients with refractory disease who did not achieve remission were defined as having an EFS of zero days. In the cases of disease relapse, the date of the event was defined as the date when the relapse was confirmed via biopsy and pathologic review. If a biopsy was not obtained, the event was defined as the date of radiologic confirmation.

Statistical Analysis

All statistical analysis was conducted in R (R Core Team) and an alpha value of 0.05 was set as a threshold for statistical significance. A Spearman’s correlation analysis was utilized to analyze the correlation between neoadjuvant therapy completion to surgery interval length and tumor necrosis at the time of surgery. Kaplan Meier analyses were performed on those with an interval greater than versus less than or equal to 14, 15, 16, and 17 days. A multivariable Cox hazards regression for EFS and a linear regression for tumor necrosis were additionally performed, treating interval as a continuous variable with a linear effect, and controlling for age at diagnosis, margin status, tumor site, histologic subtype, and stage at diagnosis as potential confounders. Collapsed categories were used for tumor site and histologic subtype. A post-hoc multivariable Cox hazards regression was performed on patients with localized disease at the time of diagnosis, controlling for age at diagnosis, margin status, tumor site, and histologic subtype. A forest plot of hazard ratios (HR) was created for both multivariable Cox model analyses.