An important factor in the diagnosis of acute lymphoblastic leukemia (ALL) is that karyotype is an independent prognostic indicator, with an impact on the choice of treatment. Outcome is related to the number of chromosomes. For example, high hyperdiploidy (51–65 chromosomes) is associated with a good prognosis, whereas patients with near haploidy (23–29 chromosomes) have a poor outcome. The discovery of recurring chromosomal abnormalities in the leukemic blasts of patients with ALL has identified a large number of genes involved in leukemogenesis. Certain specific genetic changes are related to prognosis. The ETV6/AML1 fusion arising from the translocation (t12;21) (p13;q22) has been associated with a good outcome; the BCR/ABL fusion of (t9;22)(q34;q11), rearrangements of the MLL gene, and abnormalities of the short arm of chromosomes 9 involving the tumor suppressor genes p16^INK4A have a poor prognosis. Unfortunately, the classification of patients into prognostic groups based on cytogenetics is not always as predicted. Even when other clinically based risk factors are taken into account, some patients with good‐risk cytogenetic features will relapse. In the search for new measures of prognosis, it has recently emerged that the level of minimal residual disease following induction therapy can be a reliable predictor of outcome in ALL.