The cure rate of childhood acute lymphoblastic leukemia (ALL) has exceeded 90% in some contemporary clinical trials. However, the dose intensity of conventional chemotherapy has been pushed to its limit. Further improvement in outcome will need to rely more heavily on molecular therapeutic as well as immuno- and cellular-therapy approaches together with precise risk stratification. Children with ETV6-RUNX1 or hyperdiploid>50 ALL who achieve negative minimal residual disease during early remission induction are suitable candidates for reduction in treatment. Patients with Philadelphia chromosome (Ph)-positive or Ph-like ALL with ABL-class fusion should be treated with dasatinib. BH3 profiling and other preclinical methods have identified several high-risk subtypes, such as hypodiplod, early T-cell precursor, immature T-cell, KMT2A-rearranged, Ph-positive and TCF-HLF-positive ALL, that may respond to BCL-2 inhibitor venetoclax. There are other fusions or mutations that may serve as putative targets, but effective targeted therapy has yet to be established. For other high-risk patients or poor early treatment responders who do not have targetable genetic lesions, current approaches that offer hope include blinatumomab, inotuzumab and CAR-T cell therapy for B-ALL, and daratumumab and nelarabine for T-ALL. With the expanding therapeutic armamentarium, we should start focus on rational combinations of targeted therapy with non-overlapping toxicities.