Background: The use of artificial intelligence (AI) in medicine has surged. Given the sensitive nature of palliative care, it is crucial to apply AI tools in a patient-centered and ethical manner. Aim: We sought to understand the ethical implications of implementing AI-based tools in palliative care from various stakeholders. Design: Using a framework of ethics of care, we conducted thematic analysis to identify emerging themes. Setting/participants: We recruited and interviewed 22 participants: six patients, four care-partners, five clinicians, three bioethicists, and four clinician-bioethicists, recruited through a single hospital clinic for patients and care-partners and purposeful snowball sampling for clinicians and bioethicists based on their knowledge and interest in AI-based tools. The mean age of participants was 48 years old; 64% were male. Clinician participants had practiced for a mean of 11 years, and ethicists for 15 years. We aimed to recruit clinicians and bioethicists from multiple states and specialties. Results: From the interviews we extracted five main themes: (1) Primacy of the doctor-patient relationship over AI performance; (2) Humans have intuition and nuance that AI lacks; (3) Agreement about the importance of oversight of AI tools; (4) New AI technologies should include a process for patient education; and (5) AI increases efficiency, scalability, and a more unified approach to serious illness. Conclusions: When building and implementing AI-based tools, we recommend: establishing oversight committees; reflecting on the unique contributions of humans to care; proactively educating patients and contextualizing the tools; and ensuring data use is restricted to clinical care.

Background: Advances in large language models (LLMs) provide a means for scalable tracking of patient symptoms in clinical trials and post-marking surveillance using the electronic health record (EHR). Therefore, we sought to validate symptoms extracted from the EHR using a LLM to scale symptom extraction from the EHR. Methods: Across a dataset of 499 randomly chosen clinical notes from patients seen in a neuro-oncology clinic, GPT-4o annotated symptoms (headache, fatigue, nausea, anxiety, difficulties sleeping, numbness and tingling, rash, constipation, and diarrhea) with an average sensitivity and specificity of 0.97 relative to expert manual review. We then applied the LLM to an external dataset of 51,541 notes representing 1,642 patients to obtain real-world symptom prevalence for temozolomide, bevacizumab, lomustine, immune checkpoint inhibitors (ICI), and methotrexate. Results: In the external dataset, the average number of symptoms per note was 3.92, and the most common symptom was fatigue (83% of patients). Surprisingly, patients receiving ICIs suffered from the most symptoms (mean = 4.68) and those receiving methotrexate had the least (mean = 2.92). We also found that the prevalence of reported symptoms in this real-world cohort was often much greater than the prevalence of reported symptoms in clinical trials of similar treatment regimens. Conclusions: Large language models offer the ability to scale symptom extraction from health records, which is crucial to understand symptom burden and power symptom-related interventions and studies in real-world patient cohorts.

Background: Patients with glioblastoma experience high physical and psychosocial symptom burden. Poor social relationships have been shown to increase the risk of neurologic illnesses and decline, and conversely, strong personal social networks (PSN) have been shown to reduce the risk of mortality and improve quality of life. The aim of this pilot study is to determine the feasibility of measuring PSN in glioblastoma patients. Methods: We recruited 25 adult glioblastoma patients between the initial diagnosis and the first cycle of adjuvant chemotherapy from March to September 2023 in the outpatient neuro-oncology clinic and adapted PERSNET, a quantitative PSN assessment tool, to this population. We collected demographics, tumor measures, treatment courses, and the European Organization for Research and Treatment of Cancer QOL for Patients with Brain Tumors (EORTC QLQ-BN20) and conducted qualitative interviews. Results: The average age was 64.5 years old, 56% were female, and 84% had a Karnofsky Performance Status of 70 or higher. Patients had large network sizes (total size of patient’s PSN; mean = 8.8). Participants stressed the importance of social support and how different people filled different roles in their oncology care. Behavioral and/or cognitive changes resulted in delayed presentation, and children, especially daughters, were important in advocating for presentation to the hospital. Conclusion: This is the first proof-of-concept study showing that PSN can be measured in patients with glioblastoma. Qualitative data showed that patients felt social support was very important, and different people in their networks addressed different domains of care: physical, emotional, and logistical.

Background: The occurrence (incidence or prevalence) of delirium in brain tumors is unknown, yet delirium is associated with increased morbidity and mortality and worse quality of life. We conducted a systematic review and meta-analysis to determine the occurrence of delirium in hospitalized patients with brain tumors. Methods: PubMed, Scopus, and Web of Science were systematically searched for papers from 1 January 1999 to 12 July 2024, including references from texts. Cross-sectional, prospective, and other cohort study designs were included, and individual case reports, case series, editorials, and reviews were excluded. The included papers were scored using a validated sensitivity analysis tool and tested for quality and bias using funnel plots and Egger’s test. We used random effects models for the summary estimates. We performed subgroup analyses by tumor type, tumor location, delirium subtype, and length of stay. Results: Of the 452 studies screened, 27 were included, representing 35,958 patients. The overall occurrence of delirium was 0.17 (95% CI [0.11–0.24]). Delirium occurrence in patients with low-grade gliomas, high-grade gliomas, and brain metastases was 0.10 [0.06–0.16], 0.21 [0.10–0.40], and 0.31 [0.16–0.50], respectively. Compared to the occipital lobe, there was a higher occurrence of delirium for tumors in the frontal (RR 3.08 [1.35–8.22]) and temporal lobes (RR 2.88 [1.22–7.93]). The patients were more likely to have hypoactive (RR 1.61 [1.30; 1.98]) than hyperactive delirium. Delirium was associated with 4.62 additional hospitalized days compared to those without delirium (CI [3.23–6.01]). Discussion: We confirmed high occurrence rates of delirium in patients hospitalized with brain tumors. Patients with brain metastases had a higher occurrence of delirium compared to patients with gliomas, and delirium occurrence rates were higher in patients with frontotemporal tumors. Delirium occurrence rates in the literature are very heterogeneous and point toward a need for tailored assessments in patients with brain tumors.

Introduction: Patients with cognitive impairments, who may be unable to engage in traditional psychotherapies, may require support through non-cognitive interventions. This review synthesizes the existing literature to assess the evidence base for these interventions and to propose future directions for research. Methods: A comprehensive literature search was conducted across seven electronic databases on October 17, 2024. Eligible studies assessed psychological, emotional, social, or spiritual outcomes. Results: Seventy-three studies were included, conducted across 18 countries. Music therapy and art therapy had the strongest empirical support, with consistent effects on distress, emotional expression, social and spiritual connectedness, and finding meaning. Other approaches, including movement therapy, spiritual interventions, and pet therapy, showed promising but preliminary results. Aromatherapy and massage therapy yielded mixed findings, while energy-based methods lacked robust evidence. Most studies had a high risk of bias, primarily due to confounding. Few interventions were conducted with patients with cognitive impairments, despite their potential relevance. Conclusion: A variety of interventions compatible with cognitive impairment show promise in palliative care settings. However, existing empirical support is constrained by bias and a lack of research in cognitively-impaired populations.

PURPOSE: To understand from patients, care partners (P/CPs), and physicians the ideal timing of specialty palliative care (SPC) involvement. METHODS: Using a framework of phenomenology and care ethics, we analyzed interviews through a thematic analysis approach. Two investigators independently coded transcripts of recorded interviews. P/CPs were recruited from a single hospital clinic from multiple different oncologists. Physicians were recruited via purposeful snowball sampling based on their knowledge and expertise. RESULTS: Thirty-one participants were recruited: 14 (45%) P/CPs and 17 (55%) clinicians. The mean age of P/CP was 49 years, whereas clinicians' mean age was 44 years. Half of P/CPs were female, and a similar portion of clinicians (53%) were female. More than half of patients had gliomas (57%), and all clinicians were physicians who had practiced neuro-oncology for a mean of 12 years. We identified four themes: (1) Participants felt patients would have benefited from earlier SPC, but there were slight differences in timing based on brain tumor type and grade; (2) neuro-oncology patients have unique supportive care needs compared to other cancers; (3) specialty knowledge of neuro-oncology is important to P/CP for advance care planning, prognostication, and anticipatory guidance; and (4) the broadly intensive role played by care partners in neuro-oncology means they have equally heavy supportive needs. CONCLUSION: This is one of the first studies to explore timing of SPC for patients with CNS tumors. P/CPs and clinicians agreed that early, time-based SPC is beneficial for management of high-grade tumors. Increasing knowledge of neuro-oncology among SPC clinicians working with this population is an area for further study.