Development and internal validation of prognostic models for event-free survival in newly diagnosed Ewing sarcoma patients based on routinely collected clinical characteristics from the European randomised controlled trial, EE2012.

INTRODUCTION

Ewing sarcoma is a rare paediatric cancer. Currently, there is no way of accurately predicting these patients' survival at diagnosis. Disease type (ie, localised disease, lung/pleuropulmonary metastases and other metastases) is used to guide treatment decisions, with metastatic patients generally having worse outcomes than localised disease patients. However, not all patients fit this trend. An accurate prognostic model could be used to guide treatment decisions in clinical practice to avoid patients being incorrectly under or overtreated.

METHODS AND ANALYSIS: This study aims to develop and internally validate prognostic models in newly diagnosed Ewing sarcoma patients, using the EE2012 clinical trial data set. The models will incorporate prognostic factors, identified from a literature review, to predict patients' probability of event-free survival at clinically important time points. Three models will be developed, for comparison of their performance and stability, using different methods of model selection and penalisation for overfitting (full model or backwards selection applying uniform shrinkage; and lasso variable selection). The models will be internally validated using bootstrapping to give optimism-adjusted performance statistics (calibration and discrimination) and model stability results. Patient and clinical user groups will be asked to determine risk thresholds to guide treatment decisions in clinical practice based on the model. Decision curve analyses will examine clinical utility at these thresholds.

ETHICS AND DISSEMINATION: This study is a secondary analysis of EE2012 clinical trial data. The EE2012 trial received ethical approval from the competent authorities (UK ethics reference approval number 12/NW/0827). This study is covered by the trial ethics in place. The results from this study will be published in peer-reviewed journals to act as a benchmark for future studies.

TRIAL REGISTRATION NUMBER: EudraCT number 2012-002107-17. ISRCTN number 92192408.