Ong and Larson provide an excellent review of acute lymphoblastic leukemia (ALL) in adults. They thoroughly discuss such basic issues as the diagnosis and classification of ALL, prognostic factors, and the principles of treatment. They also discuss specific problems that arise, such as the treatment of ALL in the elderly and in those with Philadelphia chromosome-positive ALL. In addition, the authors comment on areas that do not yet have fully defined roles in treatment, such as the detection of minimal residual disease and various methods of admin-\nistering high-dose chemotherapy supported by allogeneic or autologous progenitor cells obtained from blood or marrow. Their views, as expressed in this paper, are reasonable and supported by appropriate references. This review will therefore expand on and underline comments made by the authors in several areas.\n\nEvaluation of Newly Diagnosed Patients\n\nAll patients with acute leukemia should undergo a series of studies, preferably on cells obtained from bone marrow. Peripheral blood may contain sufficient blasts to be helpful, and may be the only source of information if the marrow cannot be aspirated. These studies should include the careful review, by an experienced hematologist or hematopathologist, of slides of the bone marrow aspirate and peripheral blood stained with a polychrome stain as well as Sudan black B or peroxidase. In most patients, this will suggest the diagnosis of either ALL or acute myeloid leukemia (AML), but such stains should not be considered sufficient. All patients should have immunophenotyping performed to define the presence or absence of surface markers indicative of B-lymphocytic, T-lymphocytic, or myeloid differentiation.\n\nAs pointed out by Ong and Larson, myeloid markers can occur on the cells of patients with ALL, but markers of T- or B-cell differentiation are also found on the cells of patients with AML. The combination of morphology/cytochemistry and immunophenotyping provides the clinician with the greatest degree of confidence that the correct diagnosis is being made and that the choice of treatment is appropriate.\n\nDifferent treatments are required for ALL, the variant of ALL called L3 (or Burkitt's leukemia), and AML. All are potentially curable. Molecular studies, such as rearrangement of the immunoglobulin genes or T-cell receptors, may also define the presence of clonality, but their significance is unclear. The finding of a clonal rearrangement provides no greater assurance of a correct diagnosis than defining lymphoid surface antigens, and discrepant molecular findings can occur with any variety of acute leukemia.\n\nChromosome Analysis\n\nIn addition, all patients with leukemia of any description should have chromosomal analyses performed, preferably on cells obtained from bone marrow. Ong and Larson comment on various prognostic factors for achieving complete remission or long-term disease-free survival. The most important finding, in a specific patient, is one that indicates no chance of cure with "standard" approaches. The presence of the Philadelphia chromosome, which occurs in as many as 30% of adults with ALL, defines patients who will clearly relapse with treatment that is effective and curative in the majority of adults with ALL. Such patients should undergo allogeneic bone marrow transplantation if a suitable donor can be identified. No other piece of information obtainable from adults with ALL so clearly defines outcome. Although other chromosomal abnormalities, such as the t(4;11) karyotype, may suggest a similar prognosis, the frequency with which these abnormalities occur is far less than that of the Philadelphia chromosome.\n\nThe use of aggressive, multiagent induction chemotherapy has resulted in high complete remission rates of 80% to 90%. Such approaches are more toxic than the use of vincristine, prednisone, and an anthracycline, but the results justify the increase in toxicity. Aggressive induction therapy followed by further aggressive cycles of treatment after remission (consolidation, intensification) is associated with median survival durations of 2 to 3 years, which are considerably longer than survivals attained in studies conducted more than a decade ago. Clinicians are well advised to select the treatment recipe that seems optimal to them, unless a suitable clinical trial is available. A number of suitable alternatives are described in the review.\n\nAs to the value of specific elements of treatment after remission, regardless of whether the label is "consolidation," "intensification," or "maintenance," prolonged therapy seems appropriate since the results are quite favorable. However, compelling evidence suggesting that a given type of treatment or duration of treatment is required for best results has not been published.\n\nBurkitt's (L3) Leukemia\n\nPatients with L3 ALL have leukemic cells with a characteristic morphologic appearance (vacuoles in the cytoplasm of large agranular, otherwise undifferentiated blasts), typical immunophenotypic characteristics (the presence of monoclonal surface immunoglobulin and CD10 [Calla] and the absence of terminal deoxynucleotidyl transferase), and characteristic cytogenetic abnormalities (in- volving translocation of the myc-oncogene on chromosome 8 to the heavy-chain gene locus on chromosome 14, the kappa gene locus on chromosome 2, or the lambda gene locus on chromosome 22). Such patients represent only 1% to 2% of all adults with ALL.\n\nIn most ALL treatment studies that have included patients with L3, the results of approaches that have been successful in L1 and L2 ALL have met with dismal results in L3. However, over the past 5 years, a series of publications have documented the fact that patients with L3 can be cured with short-duration, high-intensity chemotherapy, usually involving high doses of alkylating agents, methotrexate, and cytarabine. Such patients should therefore receive a chemotherapeutic regimen tailored specifically to the treatment of this entity.\n\nTransplantation\n\nThe role of allogeneic transplantation in patients with unfavorable cytogenetic abnormalities has been discussed, and such treatment is also appropriate for patients in whom relapse occurs. There does not appear to be a role for early transplantation of any sort in patients who do not have adverse cytogenetic findings.\n\nAutologous transplantation has been used primarily in patients whose ALL has relapsed and who do not have a suitable allogeneic donor. Given the improved results of treatment of standard-risk adults with ALL, the role of autologous transplantation during first remission remains undefined. Perhaps high-risk patients, defined by chromosomal abnormalities, will benefit from this approach.\n\nSummary\n\nThe outcome has improved considerably for patients with ALL. Studies have demonstrated that aggressive, prolonged treatment combined with central nervous system prophylaxis produces improved results. Ong and Larson provide a blueprint for achieving optimal results with currently available treatments, and also discuss areas of current interest for improving on these results. It is to be hoped that the next decade will see changes in treatments that will have as much impact on prognosis as has been seen during the last decade.