Since 2017, therapeutic options for acute myeloid leukemia (AML) have expanded rapidly with the FDA approval of 8 new agents. This has provided new therapeutic options for both patients and healthcare providers (HCPs), but it has also introduced new challenges, including the selection of the optimal agent(s) and sequencing strategies that optimize patient benefit and minimize toxicity. To help address these challenges, we developed an online AML decision support tool designed to provide HCPs with expert guidance on optimal treatment (Tx) for personalized patient scenarios.


In February 2019, a panel of 5 experts in AML provided Tx recommendations for 330 distinct patient scenarios in the settings of newly diagnosed (ND; n = 150) and relapsed/refractory (RR; n = 180) disease. These scenarios were defined by patient and disease characteristics identified by the experts as important when making Tx choices and include disease setting, patient age and fitness, previous MDS or MPN, cytogenetic data, presence of specific biomarkers, and previous therapy. To use the tool, HCPs are prompted to select defined patient/disease characteristics from provided options and then are asked to provide their Tx choice for that specific patient scenario. After this, they are shown the recommendations of the 5 experts based on those same characteristics and are asked to indicate the impact of the expert recommendations on their planned Tx approach.


From June 2019 to July 2020, a total of 417 unique HCPs used the AML tool, 85% specializing in hematology and/or oncology and 59% being physicians. Two thirds of the HCPs reported treating ≤ 10 patients with AML each month. Tool users entered 934 patient case scenarios-748 (80%) ND and 186 (20%) RR cases; 51% of cases that were identified for patient type (actual or hypothetical) represented actual patients in their practices.

Overall, the Tx selections by HCPs agreed with expert choices only 36% of the time for ND fit patients and 33% of the time for ND unfit patients. Most of the entered ND cases focused on fit patients ≤ 75 years of age (n = 535; 72%). For this patient group, the experts recommended primarily 3 Tx options: CPX-351 (29%), 7+3 plus gemtuzumab ozogamicin (22%), and venetoclax + HMA (20%) (Table). Conversely, HCPs selected these top 3 Txs in only 20% of their cases, instead choosing the 7+3 regimen for 33% of their cases, with another 22% uncertain of Tx choice. For the entered ND unfit or > 75 years of age patient scenarios, the disparity in Tx choices continued, with experts selecting a venetoclax regimen in 74% of cases (64% combined with HMA and 10% with LDAC) vs in 33% of HCP cases (Table). Experts also recommended FLT3 and IDH1/2 inhibitor therapies as monotherapy in the upfront setting in unfit patients far more frequently than HCPs (18% vs < 1% of cases).

In RR AML patient scenarios, HCP Tx selections again differed considerably from the experts in both the fit population (n = 134) and older, less fit population (n = 52). For example, experts selected targeted therapy for 87% of patients with FLT3 or IDH1/2 mutations in RR AML, yet HCPs only selected targeted agents for 47% of these patient scenarios.

In an optional question, among HCPs who initially differed from the expert recommendations (50% of total), 56% planned to change their Tx approach, 28% reported that barriers prevented them from changing their Tx, and 16% remained uncertain about choice of therapy.


The analysis of data from this online tool found several areas of expert consensus regarding Tx strategies for patients with AML, particularly regarding the use of venetoclax + HMAs for older, less fit patients with ND AML, intensive induction strategies in younger/fit AML patients, and targeted therapies for RR AML patients with FLT3 or IDH1/2 mutations. Overall, practice patterns of community HCPs differed from the experts for the majority of entered cases. The disparity in Tx choices between community HCPs and experts suggests a continued need for education through Tx decision tools such as this as well as other innovative approaches to increase HCP awareness of expert recommendations and best practices in AML.

A detailed comparison of expert and user data from the online tool will be presented.


Ravandi:Macrogenics: Research Funding; Abbvie: Consultancy, Honoraria, Research Funding; Jazz Pharmaceuticals: Consultancy, Honoraria, Research Funding; Xencor: Consultancy, Honoraria, Research Funding; Orsenix: Consultancy, Honoraria, Research Funding; Astellas: Consultancy, Honoraria, Research Funding; Amgen: Consultancy, Honoraria, Research Funding; Celgene: Consultancy, Honoraria; BMS: Consultancy, Honoraria, Research Funding; AstraZeneca: Consultancy, Honoraria. Smith:Jazz: Consultancy, Membership on an entity's Board of Directors or advisory committees; Novartis: Consultancy, Membership on an entity's Board of Directors or advisory committees; Pfizer: Consultancy, Membership on an entity's Board of Directors or advisory committees; Celgene: Consultancy, Membership on an entity's Board of Directors or advisory committees; Agios: Consultancy, Membership on an entity's Board of Directors or advisory committees. Walter:Selvita: Research Funding; Race Oncology: Consultancy; Seattle Genetics: Research Funding; Pfizer: Consultancy, Research Funding; New Link Genetics: Consultancy; Macrogenics: Research Funding; Agios: Consultancy, Research Funding; Amgen: Consultancy, Research Funding; Amphivena: Current equity holder in publicly-traded company; Aptevo: Consultancy, Research Funding; Argenx: Consultancy; Arog: Research Funding; Astellas: Consultancy; BioLineRx: Consultancy, Research Funding; BiVictriX: Consultancy; Boston Biomedical: Consultancy; Celgene: Consultancy, Research Funding; Daiichi: Consultancy; Genentech: Consultancy; ImmunoGen: Research Funding; Jazz: Consultancy, Research Funding; Kite: Consultancy; StemLine: Research Funding. Wang:Astellas: Consultancy; Macrogenics: Consultancy; PTC Therapeutics: Consultancy; Stemline: Speakers Bureau; Genentech: Consultancy; Pfizer: Speakers Bureau; Abbvie: Consultancy; Jazz Pharmaceuticals: Consultancy; Bristol Meyers Squibb (Celgene): Consultancy. Lancet:Abbvie: Consultancy; Agios Pharmaceuticals: Consultancy, Honoraria; Astellas Pharma: Consultancy; Celgene: Consultancy, Research Funding; Daiichi Sankyo: Consultancy; ElevateBio Management: Consultancy; Jazz Pharmaceuticals: Consultancy; Pfizer: Consultancy.

OffLabel Disclosure:

In this report of results from a treatment decision support tool for AML, some of the drugs selected by the experts are recommended in off-label applications (eg, IDH1/2 inhibitors ivosidenib and enasidenib in frontline AML therapy).

Author notes


Asterisk with author names denotes non-ASH members.