[Correspondence] Primary ovarian insufficiency prediction in adult survivors of childhood cancer: model concerns – Authors’ reply

We appreciate the interest in our Article1 presenting age-specific risk prediction models for primary ovarian insufficiency among female survivors of childhood cancer. In brief, we developed these models using cohort study data from the Childhood Cancer Survivor Study (CCSS; n=7891), with external validation from the St Jude Lifetime Cohort (n=1349) where ovarian status was clinically or biochemically ascertained. The model that used prescribed pelvic or abdominal radiation dose as a predictor showed excellent external validation performance (AUROCs: 0·88–0·95 across ages 21–40 years).