Soumerai JD, et al. - Researchers sought to develop as well as validate a prognostic model to identify high-risk chronic lymphocytic leukaemia (CLL) patients who do not achieve a good outcome with available targeted therapies. They used a training dataset (n=727), and an internal-validation dataset (n=242) and three external-validation datasets for validation. Four factors that were included in the derived model were: one point each; serum β₂-microglobulin ≥5 mg/dL, lactate dehydrogenase >upper limit of normal, haemoglobin <110 g/L for women or <120 g/L for men, and time from initiation of last therapy <24 months; these enabled to classify patients into low (score 0–1), intermediate (score 2–3), and high risk (score 4) groups. In the training and all validation cohorts, the risk score was prognostic for overall survival. This model represents the first validated risk score that allows overall survival prediction in cases with relapsed or refractory CLL treated with targeted therapy, and has applicability for cases treated with all currently approved targeted therapies (ibrutinib, idelalisib, and venetoclax) and chemoimmunotherapy. A well defined cohort of previously treated patients with CLL who are at high risk of death can be detected with this model.