2026 ASH®
Annual Meeting & Exposition December 12-15, 2026 ▪ New Orleans, Louisiana

Paper Details

Abstracts, authors, affiliations, and disclosures.

Sunday, December 13, 2026
06:00 p.m. - 08:00 p.m.
MCCNO - Hall F
Access, Representation, Population Health, Technology and Procedures, Imaging, Treatment Considerations, Supportive Care, Biological Therapies, Transplantation, Bispecific Antibody Therapy, (CAR)-T Cell Therapies, Workforce, Adverse Events, Artificial Intelligence (AI), Generative AI, Natural Language Processing, Biological Processes, Pathogenesis, Metabolism, Clinical Practice (Health Services and Quality), Clinical Guidelines, Health Economics, Patient Safety, Practice Models, Quality Improvement, Quality of Care, Value, Diseases, Immunology, Immunodeficiency, Lymphoid Malignancies, Lymphoid Leukemias, Lymphomas, Plasma Cell Disorders, Myeloid Malignancies, Acute Myeloid Malignancies, Chronic Myeloid Malignancies, CLL, Non-Hodgkin Lymphoma, B Cell Lymphoma, Indolent Lymphoma, Aggressive Lymphoma, Research, Translational Research, CML, Clinical Research, Clinical Trials, Epidemiology, Health Outcomes Research, Health Disparities Research, Patient-Reported Outcomes, Real-World Evidence, Registries, Survivorship, Study Population, Human, Adult, Elderly
PAPER

4412 Patient perspectives on myelo and artificial intelligence: Needs, hopes, and concerns within the multiple myeloma community

Miko Santos, Michelle Faber, BS, MS2, Peter Anton3, Heather Cooper Ortner4, S. Vincent Rajkumar, MD5, Richard C. Godby, MD6
1 International Myeloma Foundation, Studio City, California, United States
2 International Myeloma Foundation, Studio City, California, United States
3 International Myeloma Foundation, Studio City, California, United States
4 International Myeloma Foundation, Studio City, California, United States
5 Mayo Clinic, Rochester, Minnesota, United States
6 Mayo Clinic, Rochester, Minnesota, United States

Introduction: The integration of artificial intelligence (AI) into healthcare presents a novel paradigm for enhancing patient health literacy. Multiple myeloma (MM) is a complex hematologic malignancy requiring patients to navigate medical data, dynamic treatment regimens, and extensive laboratory monitoring. While general AI tools are increasingly accessible, patient attitudes toward clinical integration and predictors of disease-specific AI adoption remain underexplored. This project aimed to evaluate baseline AI familiarity, characterize predictors of adoption for a disease-specific AI tool (Myelo) through the International Myeloma Foundation (IMF), and understand the primary hopes, fears, and desires of MM patients regarding AI.

Methods: A voluntary, anonymous, cross-sectional survey was made available to patients with MM and their care partners through the IMF for approximately two months (April-May 2026). It was made available through IMF social media platforms and email listservs. Domains assessed included baseline demographic characteristics, educational attainment, clinical treatment status, and baseline health literacy. Descriptive statistical analysis was performed to establish baseline cohort characteristics; Chi-squared and regression analyses were performed where appropriate. Open-ended responses regarding patients' greatest hopes and fears were qualitatively analyzed to capture perspectives.

Results: The cohort (n=177) consisted primarily of patients (90%) and reflected an older, highly educated demographic. Specifically, 72% of respondents were aged 65 or older, and 77% completed college. General AI familiarity was high, with 45% reporting frequent use and 32% reporting occasional use of general AI tools. Chi-square analysis indicated a significant negative association between age and general AI familiarity (p=0.0178).

Adoption of disease-specific AI (Myelo) was driven primarily by timing and active community engagement. Lasso regression revealed a "readiness window" for adoption: patients newly diagnosed (<1 year) were 71% less likely to use Myelo, and long-term survivors (>5 years) were 61% less likely to use the tool, compared to those in the intermediate phase. Active engagement in IMF programs strongly predicted adoption; Support Group Leader Summit attendance increased odds of Myelo use 417% while attendance at other programs each increased the likelihood of Myelo use by 45%. Respondents who reported not being confident in conducting their own research were 23% less likely to adopt the tool, whereas those who felt their healthcare team educated them well were 22% more likely to be users. There was a “trust gap” as 76% of Myelo users trusted it more than general AI, compared to 79% of non-users who remained neutral. Furthermore, 59% of users were comfortable with their clinical team utilizing Myelo, versus only 29% of non-users.

Qualitative analysis of open-ended responses revealed uniformity in patient anxieties and desires. When asked about their biggest fear regarding AI in healthcare, 68% of respondents specifically cited "hallucinations" or misinformation. Secondary fears included privacy (8%) and AI replacing human clinicians (7%). The primary hope for AI was to accelerate research toward a cure (42%), followed by achieving better access to information (28%). When asked to design their ideal health technology feature, 38% requested a tool capable of tracking laboratory results and providing plain-language explanations. Qualitative feedback underscored a preference for full disclosure, requesting that AI translate complex diagnostic data without "dumbing it down."

Conclusions: The MM community reported high baseline familiarity with general AI but feared "hallucinations." Adoption of curated, disease-specific AI is heavily influenced by the time since diagnosis and structured educational engagement. The trust gap between users and non-users, coupled with the overwhelming qualitative demand for AI-driven laboratory interpretation, highlights opportunities for increased awareness and engagement with Myelo. Integrating disease-specific AI into standard clinical workflows can help to address complex health literacy needs while preserving the physician-patient relationship. Future developmental roadmaps should prioritize longitudinal memory to retain patient history and the integration of lab result tracking with trend visualization and transparent, plain-language interpretations.

Miko Santos: No relevant conflicts of interest to declare.