Clark RA, Mostoufi-Moab S, Yasui Y, et al. - Given gonadal impairment can result from cancer treatment, and the definition of acute ovarian failure is the permanent loss of ovarian function within 5 years of cancer diagnosis, researchers undertook this cohort study to construct as well as validate risk prediction tools to afford proper clinical guidance for paediatric patients with cancer. They used eligible female US and Canadian participants in the Childhood Cancer Survivor Study (CCSS) cohort to construct models to predict acute ovarian failure risk. Validation of the models was obtained in the St Jude Lifetime Cohort (SJLIFE) Study. To construct the risk prediction models in the CCSS cohort, they employed logistic regression, random forest, and support vector machines as candidate methods. Similar results were generated by these three candidate methods, and experts selected a prescribed dose model with abdominal and pelvic radiation doses and an ovarian dose model with ovarian radiation dosimetry applying logistic regression. History of haematopoietic stem-cell transplantation, cumulative alkylating drug dose, and an interaction between age at cancer diagnosis and haematopoietic stem-cell transplant were documented as common predictors in both models. Findings revealed the good performance of both acute ovarian failure risk prediction models. Preference was given to the ovarian dose model if ovarian radiation dosimetry was available. An online risk calculator meant for clinical use was developed using these models. The clinical discussions about the requirement for fertility preservation interventions in girls and young women newly diagnosed with cancer could be aided by these models, along with the online risk calculator.