CHARACTERIZING REAL-WORLD DATA AVAILABILITY FOR MONITORING ANTI-AMYLOID THERAPY IN ALZHEIMER'S DISEASE
Author(s)
John Schwarz, PhD1, John Foley, MD2, Victoria Shada, MD3, Brooke Rhead, PhD4, Rebekah Foster, MBA4, Hailey Lefkofsky, BS4.
1Bristol Myers Squibb, Princeton, NJ, USA, 2Rocky Mountain MS Clinic, Salt Lake City, UT, USA, 3Atrium Health, Winston-Salem, NC, USA, 4Tempus AI, Inc, Chicago, IL, USA.
1Bristol Myers Squibb, Princeton, NJ, USA, 2Rocky Mountain MS Clinic, Salt Lake City, UT, USA, 3Atrium Health, Winston-Salem, NC, USA, 4Tempus AI, Inc, Chicago, IL, USA.
OBJECTIVES: Anti-amyloid monoclonal antibodies have introduced new diagnostic and monitoring workflows in Alzheimer’s disease (AD), increasing the need for real-world data that reflect current clinical practice. This study characterized completeness and longitudinal availability of a curated real-world AD dataset from the Tempus Database for evaluating treatment monitoring and disease progression in routine care.
METHODS: This descriptive study used a curated dataset of 500 de-identified patients in the Tempus Database with AD-spectrum diagnoses treated with lecanemab and/or donanemab. Data included treatment, imaging, cognitive assessments, functional measures, symptoms, fluid biomarkers, and APOE status. We evaluated completeness across domains, quantified analysis-ready cohorts using pre-specified data-availability definitions, and assessed repeated measurements relative to treatment initiation at baseline, 0-6 months, 6-12 months, and 12-18 months.
RESULTS: Treatment exposure was available for all 500 patients; 62% received lecanemab, 31% donanemab, and 7% more than one anti-amyloid drug. At curation, 86% remained on treatment. Imaging was available for 93% of patients, including MRI (91%) and amyloid PET (57%). Cognitive assessments were commonly captured (MMSE 66%, MoCA 70%), as were FAQ (59%), symptom assessments (85%-90%), and ADL functional status (89%). Fluid biomarker data were available for 68% of patients and APOE genotype for 71%. A core clinical cohort requiring cognition, MRI, and function measures included 424 patients (85%); adding APOE and fluid biomarkers reduced this to 273 (55%). A practical longitudinal threshold of at least 2 brain images, at least 2 standardized assessments, and at least 1 fluid biomarker was met by 280 patients (56%). Among 398 patients with sufficient follow-up, imaging was captured at baseline and repeated within 6 months in 89%.
CONCLUSIONS: This real-world AD dataset from the Tempus Database includes substantial multimodal and longitudinal data for studying anti-amyloid treatment in routine practice. In addition to imaging and cognitive testing, functional measures may expand follow-up available for monitoring disease progression.
METHODS: This descriptive study used a curated dataset of 500 de-identified patients in the Tempus Database with AD-spectrum diagnoses treated with lecanemab and/or donanemab. Data included treatment, imaging, cognitive assessments, functional measures, symptoms, fluid biomarkers, and APOE status. We evaluated completeness across domains, quantified analysis-ready cohorts using pre-specified data-availability definitions, and assessed repeated measurements relative to treatment initiation at baseline, 0-6 months, 6-12 months, and 12-18 months.
RESULTS: Treatment exposure was available for all 500 patients; 62% received lecanemab, 31% donanemab, and 7% more than one anti-amyloid drug. At curation, 86% remained on treatment. Imaging was available for 93% of patients, including MRI (91%) and amyloid PET (57%). Cognitive assessments were commonly captured (MMSE 66%, MoCA 70%), as were FAQ (59%), symptom assessments (85%-90%), and ADL functional status (89%). Fluid biomarker data were available for 68% of patients and APOE genotype for 71%. A core clinical cohort requiring cognition, MRI, and function measures included 424 patients (85%); adding APOE and fluid biomarkers reduced this to 273 (55%). A practical longitudinal threshold of at least 2 brain images, at least 2 standardized assessments, and at least 1 fluid biomarker was met by 280 patients (56%). Among 398 patients with sufficient follow-up, imaging was captured at baseline and repeated within 6 months in 89%.
CONCLUSIONS: This real-world AD dataset from the Tempus Database includes substantial multimodal and longitudinal data for studying anti-amyloid treatment in routine practice. In addition to imaging and cognitive testing, functional measures may expand follow-up available for monitoring disease progression.
Conference/Value in Health Info
2026-11, ISPOR Europe 2026, Vienna, Austria
Value in Health, Volume 29, Issue 12S
Code
RWD51
Topic
Clinical Outcomes, Health Service Delivery & Process of Care, Real World Data & Information Systems
Topic Subcategory
Distributed Data & Research Networks
Disease
Neurological Disorders, No Additional Disease & Conditions/Specialized Treatment Areas