Development and Validation of a Prediction Model of Overall Survival in High-Risk Neuroblastoma Using Mechanistic Modeling of Metastasis

Prognosis of high-risk neuroblastoma (HRNB) remains poor despite multimodal therapies. Better prediction survival could help to refine patient stratification and better tailor treatments. We established a mechanistic model metastasis in HRNB relying on two processes: growth dissemination patient-specific parameters: the rate μ minimal visible lesion size S vis . This was calibrated using diagnosis values primary tumor size, lactate dehydrogenase circulating levels, meta-iodobenzylguanidine International Society for Paediatric Oncology European (SIOPEN) score from nuclear imaging, data 49 metastatic patients. It able describe total mass (lactate dehydrogenase, R 2 > 0.99) number metastases (SIOPEN, = 0.96). A overall (OS) then developed Cox regression. Clinical variables alone were not generate with sufficient OS prognosis ability ( P .507). The parameter found be independent clinical positively associated .0739 multivariable analysis). Critically, addition this computational biomarker significantly improved concordance index increasing 0.675 (95% CI, 0.663 0.688) 0.733 0.722 0.744, < .0001), resulting significant .0422).