Chemotherapy of childhood lymphoblastic leukaemia: the first 50 years.

It is 50 years since the first effective drug for childhood lymphoblastic leukaemia (ALL) was described. At that time the outlook for such children was certain death. Now patients have an odds-on chance of normal health and life expectancy. Although the greatest gains have been made in recent years, the classes of drug that have achieved this have all been available for over 20 years. It is their better deployment and the greater understanding of their pharmacology that have allowed both more effective protocols to be devised and long term adverse effects to be recognised and avoided. Supportive treatment has also improved in parallel. Three major problems remain: (i) how to recognise children in whom conventional therapy will fail; (ii) how to prevent failure; and (iii) how to treat it if it occurs. Therapy will fail in some children for pharmacological reasons--noncompliance or constitutional (genetic) drug resistance. For such children in vitro drug sensitivity testing and greater pharmacological vigilance may help by identifying those at risk and allowing intervention. In others, treatment will fail because of intrinsically resistant disease that either develops despite therapy or regrows from a minimal residue. Despite wider application of sophisticated immunological and genetic studies both at diagnosis or later, recognising poor-prognosis children prospectively is hampered by the lack of a biological classification system that is sufficiently sensitive and specific to categorise all patients reliably. In those where there is no doubt about high-risk status, treatment failure rates are still unacceptably high whatever therapy is given, and salvage therapy in any child who relapses is a continuing challenge.