Correspondence
Post-Remission Chemotherapy for Acute Myeloid Leukemia
N Engl J Med 1995; 332:334-335February 2, 1995
- Article
To the Editor:
Mayer et al. (Oct. 6 issue)1 conclude that adults with acute myeloid leukemia (AML) in remission who are 60 years of age or younger benefit from high-dose cytarabine therapy. It should be noted that the population these authors treated was selected to exclude patients with prior hematologic disorders and those who had received cytotoxic chemotherapy or radiation therapy. At the M.D. Anderson Cancer Center, patients with these characteristics made up 23 percent (107 of 472) of the adults 60 years of age or younger who entered complete remission after receiving chemotherapy for previously untreated AML between 1980 and 1993. The probability of remaining alive and in remission from the onset of complete remission was less for the patients with prior hematologic disorders or who had received cytotoxic chemotherapy or radiation therapy than for the patients without these characteristics (P<0.001, by two-tailed log-rank test). The same was true regardless of whether the patients received high-dose cytarabine and regardless of whether the 29 (of 472) patients who received allogeneic or autologous bone marrow transplants were included. The poorer results in the patients who had prior hematologic disorders or had received cytotoxic chemotherapy or radiation therapy may reflect the association between these characteristics and the leukemia-cell karyotype. Thus, among the 107 adults 60 years of age or younger who had prior hematologic disorders or had received cytotoxic chemotherapy or radiation therapy, 38 of the 98 (39 percent) in whom karyotyping was successful had abnormalities of chromosome 5, 7, or both (-5/-7); trisomy 8 (+8); deletion of part of the long arm of chromosome 11 (11q-); or other abnormalities, excluding inv(16), t(8;21), and t(15;17). In contrast, these karyotypes were found in 94 of 340 adults 60 years of age or younger who underwent successful karyotyping and did not have a history of hematologic disorders, cytotoxic therapy, or radiation therapy (P = 0.035, by the chi-square test). Ghaddar et al.2 have reported the poor results of high-dose cytarabine therapy in patients with -5/-7, +8, 11q-, or other abnormalities, excluding inv(16), t(8;21), and t(15;17). Four years after entering complete remission, adults 60 years of age or younger had a mean (±SE) probability of remaining alive and in complete remission of 10±7 percent.
In conclusion, we believe physicians need to take into account a patient's history of hematologic disorders, cytotoxic therapy, or radiation therapy, as well as cytogenetic findings, when considering the use of high-dose cytarabine as postremission therapy in adults 60 years of age or younger with previously untreated AML. Bone marrow transplantation, investigational chemotherapy, or immunotherapy may be more fruitful in patients with such histories or with the cytogenetic findings described above.
2 ReferencesElihu H. Estey, M.D.
Michael J. Keating, M.D.
University of Texas M.D. Anderson Cancer Center, Houston, TX 770301
Mayer RJ ,Davis RB ,Schiffer CA , et al. Intensive postremission chemotherapy in adults with acute myeloid leukemia. N Engl J Med 1994;331:896-903
Full Text | Web of Science | Medline2
Ghaddar HM ,Plunkett W ,Kantarjian HM , et al. Long-term results following treatment of newly diagnosed acute myelogenous leukemia with continuous-infusion high-dose cytosine arabinoside. Leukemia 1994;8:1269-1274
Web of Science | Medline
To the Editor:
In their report on post-remission chemotherapy for acute myeloid leukemia, Mayer et al. note that of 1104 patients registered, “Sixteen patients were considered ineligible or unable to be evaluated.” They also note that “Of the 693 patients in complete remission, 97 (14 percent) did not continue in this trial for a variety of reasons.” The reasons for ineligibility, inability to be evaluated, and failure to continue the trial should be enumerated so that readers can evaluate their validity. There should also be an explicit statement to indicate whether patients were excluded before or after randomization.
In addition, the authors fail to provide details of the distribution of base-line characteristics among the three treatment groups. They also fail to indicate the treatment assignments of the 14 patients who underwent bone marrow transplantation and the 6 patients who were lost to follow-up.
Finally, the study methods were changed in April 1989, when patients over 60 years of age were excluded from the high-dose treatment group, creating an imbalance with regard to age among the treatment groups. The authors attempt to adjust for this imbalance by using the proportional-hazards model to obtain age-adjusted hazard ratios. This is not optimal, because that model assumes independence of variables.1 It would seem better either to restrict the survival analysis to patients 60 or younger or to analyze only patients randomized before April 1989.
The size of the trial was such that these issues are not likely to alter the authors' finding of the value of high-dose cytarabine in patients 60 years of age or younger. However, this likelihood does not obviate the need for clear reporting of methods and results.2
2 ReferencesCarl D. Atkins, M.D.
242 Merrick Rd., Rockville Centre, NY 115701
Lagakos SW. Statistical analysis of survival data. In: Bailar JC III, Mosteller F, eds. Medical uses of statistics. 2nd ed. Boston: NEJM Books, 1992:290.
2
Guidelines for statistical reporting in articles for medical journals: amplifications and explanations. In: Bailar JC III, Mosteller F, eds. Medical uses of statistics. 2nd ed. Boston: NEJM Books, 1992:313-31.
Author/Editor Response
The authors reply:
To the Editor: Dr. Atkins requests more details about the patients in our study. Of the 1104 registered patients, 16 (1.4 percent) could not be evaluated or were ineligible (6 because of a history of myelodysplasia, 4 because further review of the bone marrow smears led to a diagnosis of acute lymphoblastic leukemia, 2 because of major irregularities in the induction regimen, and 1 each because of treatment before registration, a myocardial infarction before the start of therapy, the patient's refusal to participate, and a decision not to participate by the attending physician). Of the 693 patients who entered complete remission, 97 (14 percent) did not continue in the trial and were not randomly assigned to receive post-remission intensive chemotherapy (44 because of severe toxic effects of induction therapy, 27 because of the patient's refusal to participate, 10 because the patient underwent allogeneic bone marrow transplantation, 4 because of persistent hepatitis, 4 because of a decision not to participate by the attending physician, 2 because of insufficient finances, and 6 for miscellaneous reasons).
Dr. Atkins desires further information about the base-line characteristics of the three treatment groups. As noted in the manuscript, the groups were well balanced with regard to leukocyte count at the time of diagnosis and histologic subtype and were stratified at the time of randomization according to the number of courses of induction therapy (one or two) and the patient's age (less than 40, 40 to 60, and more than 60 years). Of the 14 patients withdrawn from postremission treatment to undergo bone marrow transplantation, 7 were in the group assigned to receive 100 mg of cytarabine per square meter of body-surface area, 5 were in the group assigned to receive 400 mg per square meter, and 2 were in the high-dose cytarabine group. Of the six patients who were lost to follow-up, one was assigned to the 100-mg group, three to the 400-mg group, and two to the high-dose group. Finally, the primary comparison of treatment groups in our analysis was based not on a proportional-hazards model, as suggested by Dr. Atkins, but rather on a stratified log-rank test. The exclusion of patients over the age of 60 would have yielded even stronger support for the dose–response effect of cytarabine that we observed, but it would have been statistically inappropriate given the initial design of the study.
Drs. Estey and Keating focus on the poorer prognosis associated with patients with AML (including those 60 years of age or younger) who have prior hematologic disorders or previous exposures to cytotoxic chemotherapy or radiation therapy. The purpose of our clinical trial was to assess objectively the concept of a dose–response effect of cytarabine on a population of patients with AML that was as homogeneous as possible. Consequently, we restricted our study to patients with primary AML. The influence of cytogenetics on prognosis, to which Drs. Estey and Keating refer, was addressed in a preliminary analysis. High-dose cytarabine has a positive effect on the post-remission outcome of patients with favorable leukemic karyotypes, such as t(8;21)(q22;q22) and inv(16)(p13q22), or intermediate karyotypes, such as t(15;17)(q22;q12), or normal karyotypes, but it does not influence the outcome in people with unfavorable cytogenetic patterns involving abnormal karyotypes, such as those that are frequent in secondary AML.1
1 ReferencesRobert J. Mayer, M.D.
Dana–Farber Cancer InstituteRoger B. Davis, Sc.D.
Beth Israel HospitalDeborah T. Berg, R.N., B.S.N.
Dana–Farber Cancer Institute, Boston, MA 021151
Bloomfield CD ,Lawrence C ,Arthur DC ,Berg DT ,Schiffer CA ,Mayer RJ . Curative impact of intensification with high-dose cytarabine (HiDAC) in acute myeloid leukemia (AML) varies by cytogenetic group. Blood 1994;84:Suppl 1:111a-111a abstract.
Web of Science
