ADNI Clinical Core Paul Aisen Ron Petersen Michael Donohue Jennifer Salazar ADNI Steering Committee Meeting Washington, DC April 20, 2015

©2012 MFMER | 3188678-1

ADNI 2 Enrollment by Cohort

Total at initial entry (includes 276 ADNI 1 + 120 ADNI GO rollovers):

1180

Current total (minus reported withdrawals):

917

350 Enrolled 313 311 298 300 Minus Full Withdrawals 280 251 250 220 200 150 150 108 102 100 64 50 0 CN SMC EMCI

Cohort

LMCI AD ©2012 MFMER | 3188678-2

Age (yrs) Female Education CDR-SB ADAS 13 MMSE Part. ECog Study Part. Ecog CN n=184

73.4 (6.3) 94 (51%) 16.5 (2.5) 0.0 (0.1) 9.2 (4.5) 29.0 (1.3) 1.3 (0.3) 1.2 (0.3)

ADNI GO + 2 Baseline

SMC n=103

72.2 (5.6) 61 (59%) 16.7 (2.6) 0.01 (0.2) 8.9 (4.3) 29.0 (1.2) 1.6 (0.3) 1.3 (0.3)

EMCI n=301

71.3 (7.4) 132 (44%) 16.0 (2.7) 1.3 (0.8) 12.7 (5.4) 28.3 (1.6) 1.8 (0.5) 1.6 (0.5)

LMCI n=160

72.2 (7.5) 74 (46%) 16.5 (2.6) 1.7 (1.0) 18.7 ( 7.1) 27.6 (1.8) 1.8 (0.5) 1.9 (0.7)

AD n=145

74.6 (8.1) 59 (41%) 15.8 (2.7) 4.5 (1.7) 31.0 (8.4) 23.1 (2.1) 1.9 (0.6) 2.7 (0.7)

Combined n=893

72.5 (7.3)

P

<0.001

420 (47%) 0.027

16.3 (2.6) 0.009

1.5 (1.7) 15.5 (9.6) 27.6 (2.6) 1.7 (0.5) 1.7 (0.7) <0.001

<0.001

<0.001

<0.001

<0.001

©2012 MFMER | 3188678-3

100

Dropout Rate

Drop out (ADNI 1+GO+2) p < 0.0001

75 50 25 0

AD LMCI EMCI SMC CN

333 550 301 103 410 0 2 165 446 232 3 335 4 Years 0 199 49 0 155 Number active 6 0 147 0 0 132 dropout = reported withdrawals 8 0 64 0 0 49 ©2012 MFMER | 3188678-4

100

Dropout Rate

Drop out (ADNI 1+GO+2) p < 0.0001

75 50 25 0

AD LMCI EMCI SMC CN

333 550 301 103 410 0 2 155 421 232 3 323 4 Years 0 191 49 0 155 Number active 6 0 139 0 0 131 8 0 62 0 0 49 dropout = reported withdrawals or no new data in last 18 months ©2012 MFMER | 3188678-5

Instrument sensitivity to APOE related change

• • • The following slides summarize MMRM estimates of the APOE-ε4 group difference in change from baseline at 24 months.

Estimated differences are reported on a common scale (mean:SD).

CN, EMCI, LMCI, and AD are modeled separately.

NC APOE group diff. in 2-yr change

ADAS13 RAVLT Digit Ecog Pt MOCA ADAS DWR Log Mem Del Trails B MMSE Trails A FAQ Log Mem Imm CDRSB Vegetable Fl Animal Fl Ecog SP Boston Naming −0.5

0.0

0.5

1.0

Mean:SD

EMCI APOE group diff. in 2-yr change

FAQ CDRSB Ecog SP MOCA Animal Fl MMSE Log Mem Del RAVLT Log Mem Imm Trails B Boston Naming ADAS13 ADAS DWR Trails A Ecog Pt −0.5

0.0

Mean:SD 0.5

1.0

LMCI APOE group diff. in 2-yr change

Log Mem Del Log Mem Imm FAQ MMSE CDRSB MOCA Digit Trails B ADAS13 Vegetable Fl ADAS DWR Animal Fl Boston Naming Ecog SP RAVLT Trails A Ecog Pt −0.5

0.0

Mean:SD 0.5

1.0

AD APOE group diff. in 2-yr change

FAQ Log Mem Del ADAS DWR Digit Boston Naming ADAS13 Vegetable Fl RAVLT MOCA Log Mem Imm Trails A CDRSB MMSE Trails B Animal Fl Ecog SP Ecog Pt −0.5

0.0

0.5

1.0

Mean:SD

NC Amyloid group diff. in 2-yr change MMSE Digit Trails B Animal Fl RAVLT Ecog Pt Trails A Log Mem Del Log Mem Imm CDRSB Ecog SP FAQ MOCA Boston Naming ADAS13 ADAS DWR Clock Vegetable Fl −0.5

0.0

Mean:SD 0.5

1.0

EMCI Amyloid group diff. in 2-yr change MOCA CDRSB Ecog SP Log Mem Del RAVLT Log Mem Imm Boston Naming Trails B Trails A Ecog Pt MMSE FAQ Animal Fl Clock ADAS DWR ADAS13 −0.5

0.0

0.5

1.0

Mean:SD

LMCI Amyloid group diff. in 2-yr change Log Mem Imm ADAS13 CDRSB Log Mem Del MMSE ADAS DWR MOCA FAQ Vegetable Fl Ecog SP Digit Trails B Boston Naming Clock Trails A RAVLT Ecog Pt −0.5

0.0

0.5

1.0

Mean:SD

AD Amyloid group diff. in 2-yr change Log Mem Del Log Mem Imm FAQ ADAS DWR ADAS13 MMSE MOCA Vegetable Fl CDRSB Boston Naming Clock Animal Fl Digit Trails A Ecog SP Trails B Ecog Pt RAVLT −0.5

0.0

Mean:SD 0.5

1.0

1.00

0.75

Transitions from NL

Transitions from NL

Transition

NL to Dementia NL to MCI 0.50

0.25

0.00

0.0

2.5

5.0

Year 7.5

1.00

0.75

Transitions from MCI

Transitions from MCI

Transition

MCI to Dementia MCI to NL 0.50

0.25

0.00

0.0

2.5

5.0

Year 7.5

Transitions from “de novo” MCI

Transitions from MCI 1.00

Transition

MCI to Dementia 0.75

MCI to NL 0.50

0.25

0.00

1 2 Year 3 4

ADNI 3 CLINICAL CORE PLANS

Paul Aisen Ron Petersen Mike Donohue Mike Weiner

The aims of the ADNI3 Clinical Core will include:

     Oversight of ADNI3 clinical activities, data management, tracking and quality control, recruitment and retention of participants, regulatory oversight and financial management.

Characterization of the cross-sectional features and longitudinal trajectories of cognitively normal older individuals and mild cognitive impairment.

Study of the relationships among clinical/demographic, cognitive, genetic, biochemical and neuroimaging features of AD from the preclinical through dementia stages.

Assessment of genetic, biomarker and clinical predictors of decline.

Refinement of clinical trial designs, including secondary prevention, slowing of progression in symptomatic disease, and cognitive/behavioral management.

Key hypotheses of ADNI3 Clinical Core

 All or almost all normal participants with brain amyloidosis will show cognitive decline compared to those without amyloidosis, and will progress to MCI.  Confirmation of this hypothesis is critical to early stage trial design and regulatory support.

 MCI participants who are biomarker positive (amyloid and tau) will progress more rapidly than those who are negative

Other hypotheses

 Amyloid-related cognitive decline involves episodic memory, executive function and orientation across the spectrum of AD  AD-related cognitive decline can be captured by unsupervised web-based testing  Early stage AD cognitive decline predicts later functional and clinical decline  Web-based registries will facilitate recruitment for ADNI (and therapeutic trials)

ADNI3 cohorts

 ADNI3 will carry forward roughly 300 normals (w/wo subjective concerns) and 300 MCI (EMCI+LMCI)  ADNI3 will enroll modest numbers of new normal and MCI participants  ADNI3 will follow MCI participants who progress to AD dementia

Possible adjustments to assessments

 Drop RAVLT, add FCSRT.

 Drop Boston Naming.

 Drop Clock Drawing.

 Add web-based cognitive testing.

 CFI instead of eCOG?

 Other subjective concerns measures?

 Reaching a consensus will be challenging, but we need to begin the discussion even as we work on additional analyses.