LAUSR

TO THE EDITOR Acute myeloid leukemia (AML) primarily affects older people, and thus many patients are felt to be medically less-than-fit (“unfit”) at the time of initial diagnosis (e.g., because of comorbid illnesses). How best to treat such individuals remains under debate, with ongoing uncertainty about the optimal treatment intensity. Results from three randomized trials conducted in the 1980s highlight this uncertainty and point to the delicate balance between efficacy and toxicity [1–3]. In a trial that included 45 patients aged ≥70 years, lower doses of daunorubicin, cytarabine, and 6-thioguanine were associated with fewer deaths within 60 days of treatment initiation, superior survival, and better quality of life (QOL) than full drug doses, while remission rates were similar [1]. A second trial, assigning 60 patients aged ≥65 years to either intensive induction chemotherapy or supportive care with hydroxyurea or low-dose cytarabine as necessary, found significantly higher complete remission (CR) rates and better survival for intensely treated patients, whereas the proportion of days spent in hospital was similar [2]. Yet another conclusion was be reached in a third trial, in which 87 patients aged ≥65 years were randomized between low-dose cytarabine or intensive chemotherapy: here, the more intensive therapy yielded more remissions, but survival was similar and the less intensive therapy required less supportive care and shorter hospital stays [3]. The uncertainty regarding optimal treatment intensity in unfit adults with newly diagnosed AML is ideally addressed via randomized treatment allocation. However, while these older studies suggest randomized assignment to lowervs. higher-intensity chemotherapy is feasible, strong perceptions and pre-conceived preferences regarding treatment intensity may render randomization challenging or impractical. The latter is suggested by the experience in the NCRI/ MRC AML 14 trial. This trial offered randomization between higher and lower intensity treatments in some patients but only 8 of the 1485 participants were randomized between the 2 approaches [4]. We therefore conducted a trial to evaluate the feasibility of randomization between different therapy intensities in unfit adults with newly diagnosed AML or other high-grade myeloid neoplasm. As intensive therapy, we used CLAG-M (cladribine, high-dose cytarabine, G-CSF, escalated-dose mitoxantrone), our institutional standard AML chemotherapy for fit adults [5]. Rather than selecting different classes of drugs in the two treatment arms– which may influence decisionmaking and therefore confound the question of treatment intensity – we used the same regimen at reduced dose (RD CLAG-M) as the lessintensive comparator. Adults≥ 18 years with untreated non-APL AML or another highgrade myeloid neoplasm (≥10% blasts in blood or marrow) were eligible if they had a left ventricular ejection fraction ≥ 45% and a Treatment-related Mortality (TRM) score of ≥13.1, corresponding to a ≥10–15% probability of day 28 mortality with intensive AML chemotherapy [6]. Patients with CML in blast crisis, concomitant illness with life expectancy of <1 year, and uncontrolled infection were excluded. Prior low-intensity treatment for antecedent lowgrade myeloid neoplasm (<10% blasts) was permitted. Disease risk was categorized based on the Medical Research Council (MRC) [7] and, when available, European Leukemia Net (ELN) 2017 criteria [8]. Best responses [8] were assessed after up to 2 cycles of induction therapy. Measurable residual disease (MRD) was quantified by multiparametric flow cytometry [9]. The study (ClinicalTrials.gov: NCT03012672), largely conducted before venetoclax availability (enrolled 01/24/17-11/17/2020), was approved by Fred Hutchinson Cancer Research Center’s Institutional Review Board. All participants gave written informed consent in accordance with the Declaration of Helsinki. 1:1 randomization occurred if patient-physician pairs were amenable to randomization. If not agreeable to randomization, patients and/or physicians chose the preferred treatment intensity and completed surveys to document the decision-making process (physician surveys included after the third patient enrolled). CLAGM with escalated-dose mitoxantrone (18 mg/m) was given as described [5]. RD CLAG-M, which we rarely used before