Despite impressive progress in childhood acute lymphoblastic leukemia, an urgent clinical need remains. Patients still relapse, and outcomes after relapse have changed little between 1996-2006 and 2004-2014 [1]. Despite progress with late marrow relapse (first remission [CR1] >36 months), treatment of early relapse (CR1 <36 months) remains unsatisfactory [2], especially in patients deemed higher risk at diagnosis, such as adolescents and young adults (AYAs) [3].Progress to date derives from improved primary therapy. Further improvements in primary therapy have an ever-increasing price. Going from an event-free survival of 50% to 70%, a 40% reduction in relapses, five patients need to be treated to prevent one relapse. Similarly, going from 75% to 85%, again a 40% reduction relapses, 10 patients need to be treated to prevent one relapse. With improving outcomes with primary therapy, we are facing an increasing number needed to treat for further improvement [4]. All patients need to bear the burden of a novel intervention to benefit an ever smaller percentage. Not all interventions are successful.Recent experience suggests that we are reaching the limits of "intensification" of therapy, despite improvements in supportive care [5]. For some patients such as AYAs, we may have surpassed tolerable limits. Querying the Pediatric Health Information System database, Gupta et al. found a higher incidence of intensive care unit stays and increased toxicities in almost every organ system for AYAs [6]. Serious complications may prevent delivery of best care, resulting in relapse.