Pathological estimation of tumor necrosis after chemotherapy is essential for patients with osteosarcoma. This study reports the first fully automated tool to assess viable and necrotic tumor in osteosarcoma, employing advances in histopathology digitization and automated learning. We selected 40 digitized whole slide images representing the heterogeneity of osteosarcoma and chemotherapy response. With the goal of labeling the diverse regions of the digitized tissue into viable tumor, necrotic tumor, and non-tumor, we trained 13 machine-learning models and selected the top performing one (a Support Vector Machine) based on reported accuracy. We also developed a deep-learning architecture and trained it on the same data set. We computed the receiver-operator characteristic for discrimination of non-tumor from tumor followed by conditional discrimination of necrotic from viable tumor and found our models performing exceptionally well. We then used the trained models to identify regions of interest on image-tiles generated from test whole slide images. The classification output is visualized as a tumor-prediction map, displaying the extent of viable and necrotic tumor in the slide image. Thus, we lay the foundation for a complete tumor assessment pipeline from original histology images to tumor-prediction map generation. The proposed pipeline can also be adopted for other types of tumor.

PURPOSE: The primary aim of this phase III randomized trial was to test whether the addition of vincristine, topotecan, and cyclophosphamide (VTC) to interval compressed chemotherapy improved survival outcomes for patients with previously untreated nonmetastatic Ewing sarcoma. METHODS: Patients were randomly assigned to receive standard five-drug interval compressed chemotherapy (regimen A) for 17 cycles or experimental therapy with five cycles of VTC within the 17 cycles (regimen B). Patients were stratified by age at diagnosis (< 18 years and ≥18 years) and tumor site (pelvic bone, nonpelvic bone, and extraosseous). Tumor volume at diagnosis was categorized as < 200 mL or ≥ 200 mL. Local control occurred following six cycles. Histologic response was categorized as no viable or any viable tumor. Event-free survival (EFS) and overall survival (OS) were compared between randomized groups with stratified log-rank tests. RESULTS: Of 642 enrolled patients, 309 eligible patients received standard and 320 received experimental therapy. The 5-year EFS and OS were 78% and 87%, respectively. There was no difference in survival outcomes between randomized groups (5-year EFS regimen A v regimen B, 78% v 79%; P = .192; 5-year OS 86% v 88%; P = .159). Age and primary site did not affect the risk of an EFS event. However, age ≥ 18 years was associated with an increased risk of death at 5 years (hazard ratio 1.84; 95% CI, 1.15 to 2.96; P = .009). The 5-year EFS rates for patients with pelvic, nonpelvic bone, and extraosseous primary tumors were 75%, 78%, and 85%, respectively. Tumor volume ≥ 200 mL was significantly associated with lower EFS. CONCLUSION: While VTC added to five-drug interval compressed chemotherapy did not improve survival, these outcomes represent the best survival estimates to date for patients with previously untreated nonmetastatic Ewing sarcoma.

Ewing sarcoma (EWS) is a bone tumor occurring primarily in adolescence and young adulthood. Multi-institutional clinical trials have improved the outcome for patients with nonmetastatic EWS, but not with metastatic EWS. Furthermore, although 30% of EWS recur, multi-institutional studies have not been completed for this high-risk group. Planning such studies has been hampered by both the lack of novel therapies and the inability to incorporate the biology of EWS. While the importance and detail of the EWS-FLI-1 translocation between chromosomes 11 and 22 are described, these have not yet led to new drug development for this orphan tumor. However, recent evidence supporting novel cytotoxic therapy, antiangiogenic therapy, and receptor-targeted therapy provides reason for optimism for patients with high-risk disease.