A decisional intervention for older patients with Acute Myeloid Leukemia and their oncologist: A pilot randomized controlled trial (RCT)

Abstract Introduction: Treatment shared decision-making (SDM) is challenging for older adults with newly diagnosed acute myeloid leukemia (AML) due to acute disease onset, need for urgent high-risk treatment decisions, and limited patient prognostic understanding. Yet SDM is critical for these patients who often have various aging-related conditions that increase the risk of adverse outcomes.We assessed the preliminary efficacy of a decisional intervention (UR-GOAL) on SDM and prognostic understanding. Methods: In a pilot RCT, we enrolled adults aged ≥60 years with newly diagnosed AML and their oncologists. Patients were randomized to the UR-GOAL or usual care arms. In UR-GOAL, patients viewed an educational video about AML diagnosis, treatment, and prognosis; completed the Best-Worst Scaling (BWS) value clarification process; and reviewed a summary report of their values with tailored question prompts and resources for supporting SDM. Using BWS, older patients with AML ranked the relative importance of various attributes of their care. Specifically, patients were presented with choice tasks, each task consisting of four of eight pre-selected attributes (daily activities, quality of life, location of treatment, survival), from which they selected the most and least important attributes of AML treatment. Oncologists then reviewed a summary report of the patient's aging-related conditions, perception of prognosis, and values. Within a month of the treatment decision, we assessed patient-perceived SDM SDM-Q-9, 0-100; higher scores are better and patient-oncologist concordance in estimated chance of cure and life expectancy. We also assessed observed SDM from recorded treatment decision-making visits (OPTION-5). Two trained staff coders used the OPTION scale to code for 5 items related to treatment decision-making, and a consensus score was generated for each patient (individual item score 0-20, total 0-100, higher scores are better). We also conducted qualitative interviews with patients and oncologists to obtain feedback on the UR-GOAL intervention. Results: We enrolled 100 patients and 8 oncologists. Mean patient age was 74 years, 37% females, 93% White, 6% Black, and 61% completed college or higher education. Patient-perceived SDM was high (SDM-Q-9=85), with no difference between arms. Observed total SDM score was numerically greater in the UR-GOAL vs usual care arm (OPTION5: 77.4 vs. 74.0). Observed SDM for each item of OPTION5 in the UR-GOAL vs. usual care arm was as follows: 1) presenting options (17.1 vs. 17.0); 2) establishing partnership (16.6 vs. 15.4); 3) describing options (14.7 vs. 13.8); 4) exploring preferences (14.6 vs. 13.5); and 4) making the decision (14.5 vs. 14.6). At post-intervention, 32% of patients in the UR-GOAL arm estimated their chance of cure to be ≤50% vs. 43% in usual care; only 2% of patients in both arms had oncologist-reported estimates above this threshold. Regarding life expectancy, 43% of UR-GOAL patients estimated their survival to be 5 years vs. 53% in usual care, while no oncologist in either arm estimated life expectancy beyond 5 years. Numerically, patient-oncologist dyads in the UR-GOAL arm were more often concordant in estimates of curability (71% vs. 58%, p=0.37) and life expectancy (57% vs. 47%, p=0.56). The qualitative analyses resulted in 2 key findings: 1)UR-GOAL enhanced patients' understanding and preparedness for acute myeloid leukemia (AML) by clarifying what to expect and supporting more informed conversations with their care team; 2) UR-GOAL promoted self-reflection, helping patients identify and prioritize what matters most, such as survival or quality of life, when making treatment decisions. In qualitative interviews with oncologists, information on prognosis and patient values was helpful for framing treatment discussions, though it rarely altered the oncologists' treatment recommendations. Conclusions: UR-GOAL may enhance SDM and prognostic understanding in older patients with AML by supporting more patient-centered treatment conversations between patients and oncologists. A fully powered randomized controlled trial is currently underway.