Assessing Patient Preferences for the Benefits and Risks of Treating for Acute Myeloid Leukemia (AML): A Pilot Study

Introduction:Acute Myeloid Leukemia (AML) is a blood cancer which progresses rapidly in the absence of treatment. There have been few advances in the treatment of AML over the last three decades. In 2015 the Leukemia and Lymphoma Society (LLS) initiated a research program to assess patient preferences for AML treatments. The aim of this research was to promote patient-focused drug development and inform future regulatory decisions. We sought to develop and pilot a patient-centered survey instrument to assess patient preferences for the benefits and risks of AML treatments.

Methods: Development was informed by a targeted literature review and engagement with an expert stakeholder committee (n=12) to guide the clinical accuracy and relevance of the survey instrument. A community stakeholder committee, consisting of patients with AML and caregivers (n=15), provided information about their experiences with AML and various treatments. They also engaged in pretest interviews to test comprehension and ensure it captured the patient experience. A discrete-choice experiment (DCE) was developed spanning 5 benefits and risks, including event-free survival (EFS), complete remission (CR), time in hospital, short-term side effects, and long-term side effects. This DCE consisted of 16 pairs of hypothetical treatments, with participants being asked to identify which treatment they would prefer in each pair. Results of a pilot study with AML patients and caregivers were assessed by Z-score that were derived from a conditional logistic model regressing each attribute upon their choices.

Results: The pilot included 18 patients and 8 caregivers with a mean age of 50 years (range=24-81). Most participants were college educated (n=22), Caucasian (n=19), privately insured (n=21), and employed (n=13). Participants valued CR the most (Z-score=7.95, p<0.001), followed by EFS (5.32, p<0.001). They were most averse to time in hospital (-3.41, p=0.001), followed by long-term side effects (-3.03, p=0.002) and short-term side effects (-1.99, p=0.047), which was marginally significant.

Conclusions: This study demonstrates the value of rigorous community engagement in developing survey instruments to measure patient preferences. The results of this pilot study demonstrate the ability of our DCE to measure treatment preferences of AML patients and caregivers. Given this success, we are currently engaged in a nationally study where we will recruit a larger and more diverse sample. The results of this national study will inform drug developers and regulatory decision makers.