Abstract Background. Patients with acute leukemia (AL) or myelodysplastic syndrome (MDS) are less likely to receive palliative care (PC) services than those with solid malignancies, and more likely to receive aggressive end-of-life care and die in the hospital. Whether early PC referral might affect outcomes such as place of death for patients with AL or MDS remains unclear. This Phase II trial aimed to evaluate whether early and sustained PC intervention increases the proportion of patients with newly diagnosed high-risk or relapsed-refractory AL or MDS who die at home rather than in the intensive care unit (ICU). Methods. A single-site, randomized Phase II trial was conducted at an academic medical center from October 2020 to May 2025. Eligible patients included those age ≥65 with newly diagnosed AML (excluding APML), ALL, CMML, or high-risk MDS as well as adult patients of any age with relapsed or refractory disease. Participants were randomized 3:2 to either the early PC arm (initial PC visit within three days while inpatient or 14 days while outpatient, follow by regular follow-up visits) or standard of care (PC consultation ordered at the discretion of the treatment team). The primary endpoint was the place of death (ICU vs. elsewhere). Secondary endpoints included one-year overall survival, hospice use, and patient-reported quality of life. The chi-square statistic was used to assess differences by arm for place of death and hospice use, and summary measures were used to describe differences in other secondary endpoints. The analysis detailed in this abstract is limited to the subset of patients who died. Results. The trial enrolled 101 participants (63 to the early PC arm and 38 controls) with a median age of 69 years. After excluding one participant who withdrew consent and one with incomplete data, 78% (77/99) experienced death prior to the end of the trial including 80.9% (51/63) of early PC participants and 68.4% (26/38) of controls. Among the 77 decedents, 40.3% were female, 5.2% identified as non-white, 76.6% had AML, and 49.3% had relapsed or refractory disease; these proportions were similar between arms. One-year survival estimates were 25.5% and 26.9% in the early PC and controls arms, respectively. ICU deaths occurred in 25.5% (13/51) of patients in the early PC arm compared to 15.4% (4/26) in the control arm, but this difference was not statistically significant (p = 0.3). A similar proportion of patients died at home (51.0% among early PC vs. 57.7% among controls), at an outpatient hospice facility (7.84% vs. 3.85%), and on a hospital floor (15.7% vs. 23.1%) between the two treatment groups. Hospice services were used by 54.9% of participants in the early PC arm compared to 69.2% of controls (p = 0.2). Differences in patient-reported quality of life scores between arms are currently under analysis. Notably, 53.8% (14/26) of decedents in the control arm received PC consultation “on demand” (at the request of the treating oncologist) during the trial at a median time of 30 days (range 0 – 599 days) after randomization. Conclusions. 1.Early and sustained PC intervention did not decrease the risk of dying in the ICU for patients with AL or high-risk MDS compared to “on-demand” PC consultation. 2. One-year overall survival rates were not significantly different between the two treatment arms. 3. Hospice utilization rates were lower in the early PC arm, but this difference did not reach statistical significance. We hypothesize that these surprising findings may be explained by the fact that a significant proportion of patients in the control arm (over 50%) were referred to PC providers upon request of the treatment team and thus were able to experience some of the possible benefits of palliative care, including advance care planning and better symptom control. Results from this trial challenge the assumption that early PC intervention necessarily reduces ICU deaths and highlight the complex end-of-life trajectories in patients with aggressive hematologic malignancies. These findings also suggest that training oncologists to recognize when PC referral is warranted could potentially obviate the need for structured PC intervention for all patients with AL and high-risk MDS. A larger Phase III trial will be undertaken to further explore the role of early and sustained PC in management of this patient population.
Blackhall et al. (Mon,) studied this question.