Absolute Risk of Progression to Cognitive Impairment Based on Plasma Phosphorylated Tau 217, Age, Sex, and APOE ε4
Importance: Alzheimer disease (AD) pathology can be detected years before symptom onset, but some individuals who are biomarker positive and cognitively unimpaired (CU) remain asymptomatic. Objective: To estimate 5- and 10-year absolute risks of mild cognitive impairment (MCI) and dementia using continuous plasma phosphorylated tau 217 (p-tau217) and p-tau217/amyloid β 42 (Aβ42). Design, Setting, and Participants: This cohort study included participants from the Mayo Clinic Study of Aging (MCSA) and the Alzheimer's Disease Neuroimaging Initiative (ADNI). Participants were CU individuals aged 60 to 90 years with baseline Fujirebio Lumipulse plasma p-tau217 measurements, APOE genotyping, and longitudinal clinical follow-up. The MCSA is a population-based longitudinal cohort of residents in Olmsted County, Minnesota; ADNI is a multicenter observational cohort. MCSA data were ascertained from October 2008 to September 2025; ADNI data from September 2005 to January 2026. Exposures: Age, sex, APOE ε4, p-tau217, and p-tau217/Aβ42. Main Outcomes and Measures: Transition from CU to MCI and dementia estimated using continuous-time multistate models. Results: In the MCSA cohort (n = 1644), the median (IQR) age was 72 (66-79) years; 804 participants (49%) were female and 840 (51%) were male. In the ADNI cohort (n = 534), the median (IQR) age was 72 (68-77) years; 297 participants (56%) were female and 237 (44%) were male. A p-tau217 of 0.324 pg/mL vs 0.072 pg/mL (90th vs 10th percentile) was associated with 2.66 (95% CI, 2.02-3.51), 3.02 (95% CI, 2.21-4.12), 2.62 (95% CI, 1.99-3.46), and 2.79 (95% CI, 2.04-3.81) greater hazard of clinical progression in MCSA female participants who did not carry the APOE ε4 variant (ε4-), females who did carry the APOE ε4 variant (ε4+), males who were ε4-, and males who were ε4+, respectively. In ADNI, hazards were 2.67 (95% CI, 1.86-3.85), 2.83 (95% CI, 1.89-4.22), 2.08 (95% CI, 1.48-2.92), and 3.99 (95% CI, 2.59-6.13), respectively. In MCSA, higher baseline p-tau217 and older age were associated with greater absolute risk of progression. At age 75 years and assuming sex and APOE are at the MCSA means of 0.511 and 0.268, 5-year absolute risk of MCI increased from 7% (95% CI, 5%-8%) to 17% (95% CI, 13%-20%) for a p-tau217 concentration of 0.072 vs 0.324 pg/mL; 10-year absolute risk of MCI increased from 15% (95% CI, 12%-18%) to 35% (95% CI, 29%-41%; P < .001 for both). At 0.324 pg/mL, 5-year MCI risk increased from 6% (95% CI, 4%-8%) to 37% (95% CI, 32%-43%) for ages 65 to 85 years; 10-year risks were 17% (95% CI, 12%-21%) and 66% (95% CI, 57%-74%; P < .001 for both). Results were similar for p-tau217/Aβ42. Conclusions and Relevance: This study found that higher plasma p-tau217 and older age were associated with higher 5- and 10-year risk of cognitive impairment. Plasma p-tau217 concentrations, combined with age, sex, and APOE ε4 status, can enable individualized estimation of short-term absolute risk of cognitive impairment in CU individuals, with applications in prognosis and trial enrichment in preclinical AD.
Authors
- Robin Bean
- Clifford R. Jack (ORCID: https://orcid.org/0000-0001-7916-622X)
- Alireza Atri (ORCID: https://orcid.org/0000-0003-4405-6973)
- Jonathan Graff‐Radford (ORCID: https://orcid.org/0000-0003-2770-0691)
- Michael Asbach
- Miriam Ashford
- Ruchi Aggarwal
- Terry M. Therneau (ORCID: https://orcid.org/0000-0003-1490-6711)
- Stephen D. Weigand (ORCID: https://orcid.org/0000-0002-4541-9011)
- Allison Ardolino
- Liana G. Apostolova (ORCID: https://orcid.org/0000-0003-1823-2451)
- Martin Ayres
- Alicia Algeciras‐Schimnich (ORCID: https://orcid.org/0000-0002-4455-6217)
- Howard Bergman (ORCID: https://orcid.org/0000-0002-7410-1707)
- Fatima S. Amtashar
- Haley Bernhardt
- Karen Blank
- Joshua A. Bornhorst (ORCID: https://orcid.org/0009-0003-9297-3271)
- Olusegun Adegoke
- Laura Barlow
- Maison Abu Raya (ORCID: https://orcid.org/0000-0002-7272-2104)
- Mara Acevedo
- Daniel J. Figdore (ORCID: https://orcid.org/0000-0002-4776-3203)
- Joseph Boonsiri
- Michael E Griswold (ORCID: https://orcid.org/0000-0001-8106-5516)
- Alisa Adhikari
- Zach Bonacorsi
- Tolulope Alo
- Virginia Boatwright
- Heather Bauerle
- Sarah Bohlman
- Marie Amoyaw
- Skieff Acothley
- Yair Avila
- Deborah Aloszka
- Hannatu Amaza
- Omobolanle Ayo
- Jahmila Al-Amin
- Sarah Basadre
- Kailey Basham
- Daniel Arcuri
- Alzheimer’s Disease Neuroimaging Initiative
- Paul Aisen
- Brandi Barger
- Kristi Ayers
- Sarah Binder
- Tuba Ahmed
- Allison Acree
- Karen Anderson
- Lindsey Booth Robbins
- Morgan Blackburn
- Heather J. Wiste (ORCID: https://orcid.org/0009-0002-8833-2693)
- Mauricio Becerra
- Michele Assaly
- Abbey Arnold
- Katherine Berner
- Kedir Adem Hussen
- Randall Barker
- Neetha Basheer
- Margaret Bassett
- Charna Bates
- Wendy Baker
- Anne Arthur
- Sarah Ash
- Aparna Bhan
- Jeffrey Bishop
- Laura Ashe
- Ala Abusalim
- Dr Rob Bartha
- Ronald Bailey
- Warren Barker
- Sandra Black
- Colleen Albers
- Marie Bernard
- Brandon Anderson
- Sarah Berman
- Vernice Bates
- Daniel Bandy
- Prashanthi Vemuri
- Bela Ajtai
- James Bateman
- Sanjay Asthana
- Adeyinka Ajayi
- Bret Borowski
- Amy Boegel
- Mingzhao Hu
- Laurel Beckett
- Ronald C. Petersen
- Nicole Barcelos
- Charles Bernick
- Maryam Beigi
- Bradley Boeve
- Traci Aladeen
- Michael Alosco
- Marilyn Albert
- David S. Knopman
- David Bissig
- Lisa Barnes
- Kimberly Blanchard
- Robin Bonomi
Institutions
- Mayo Clinic (US)
- University of Mississippi Medical Center (US)
- Alzheimer’s Disease Neuroimaging Initiative (US)
Publication Details
- Journal
- JAMA Neurology
- Published
- 2026-11-16
- DOI
- https://doi.org/10.1001/jamaneurol.2026.4292
- Primary Topic
- Dementia and Cognitive Impairment Research
- Type
- article
- Field-Weighted Citation Impact
- 0.00