NIA AD Genetics Initiative: Multiplex Family Study
Meeting Summary: Friday October 1 st , 2004
Toronto, Ontario
Attendees: The following Centers/Institutions were represented
Boston University
Columbia University
Case Western Reserve University
Duke University
Indiana University
Johns Hopkins University
Massachusetts General Hospital
Mayo Clinic (Rochester and Jacksonville)
Mount Sinai
NCRAD
Oregon Health and Science University
Rush University
University of Alabama
University of California, Los Angeles
University of California, San Diego
University of Kentucky
University of Pennsylvania
University of Pittsburgh
University of Southern California
University of Texas Southwestern
NIA/NIH including ADEAR
NACC
NINDS
University of Toronto
University of Washington
Washington University, St. Louis
Alzheimer’s Association
Northwestern University
A brief summary is below with a more detailed account of the meeting following the overview.
Slide presentations are available in the “What’s New” section of the online manual at www.NIAgeneticsinitiative.org
MEETING SUMMARY
 As of 9/24/04
­ 658 approved pedigrees
­ 1,885 samples from 422 families at NCRAD
­ 1,298 have a complete Minimum Data Set (MDS)

Current focus on providing a complete MDS for every subject sampled and ensuring data is clean

Online LOAD manual has new domain name: www.NIAgeneticsinitiative.org
 Control Series: Goal is 1,000 controls, 50-60 per center. Unresolved issues re: inclusion/exclusion criteria will be addressed through an open forum.

Phenotype committee updates
­ Age at onset (AAO): Guidelines on how to more effectively elicit AAO are now in the Data Element Dictionary in the LOAD manual
­ Neuropathological: 11 new variables added to the MDS: criteria used, stage assigned, etc.
­ Cognitive testing: A battery of 7 cognitive tests administered to all LOAD subjects
Scores are included in the MDS. Scores will be stored longitudinally in the database.
­ Biomarker: Future goal is to collect plasma and serum to make quantitative trait markers available

Sampled pedigrees: When a family meets minimum criteria and samples/MDS for the three essential family members are at NCRAD, a “sampled pedigree” is faxed to Columbia.
­ Included in pedigree: who is sampled/has MDS at NCRAD, who is evaluated/had cognitive testing

­ Crucial for QA/QC
­ A sampled pedigree report is generated for each center each time a pedigree is faxed: report contains all people sampled with MDS
 MDS: IT’S CRUCIAL WE RECEIVE A MDS FOR EACH FAMILY MEMBER
SAMPLED. As of 9/24/04, 205 families have at least one MDS at NCRAD, but only 90 families meet criteria according to the MDS.

NINDS Repository and Controls: 3,000 samples of affecteds, controls, and families for PD, stroke, epilepsy, ALS
­ Approximately 1,000 controls have been collected
­ Need to gather enough clinical data to allow investigators to include subjects that are ideal for their study
­ See http://locus.umdnj.edu/ninds for more information on repository and data collected

Continuation of LOAD
­ Project was initially hampered by paperwork such as MTAs and consent forms
­ Centers have made tremendous progress in year 2 and 3, but funds are limited
­ Project will continue through a cooperative agreement (U24) between five Centers to collect remaining subjects to meet goal of 1,000 families and controls

Indiana University, Columbia University, University of Washington,
Washington University, and Mayo Center

Subcontracts will be generated for all other sites to follow families and controls previously enrolled and possibly enroll heavily loaded families
­ Under U24, we will apply to CIDR to genotype the first 500 LOAD families and make the raw genotype data available through a secure website
­ We will issue subcontracts for genetic analysis—OPEN TO ANYONE
­ Any publications that result from analyzing the genotype data will be required to identify the NIA Genetics Initiative.
COMPLETE MEETING MINUTES
NIA welcome : Creighton Phelps, PhD and Richard Mayeux, MD, MSc
LOAD Progress Report as of 9/24/04 : Susan La Russe, MS
Slides were included in the meeting folder. Copies of the slides can also be available at www.NIA.geneticsinitiative.org
.

Consent forms and MTAs: All 18 funded Centers have an approved consent form. All funded
Centers have an MTA in place for sending samples to NCRAD.

Centers: Several non-funded Centers have a consent form approved/under review
-Northwestern University
-New York University
-Vanderbilt University
-University of California, San Diego
-University of Michigan
-University of California, Davis

Pedigrees approved: There were 658 approved pedigrees. Not all of the pedigrees will be qualifying LOAD families.

Pedigree characterization: 46% of the pedigrees approved have 3 or more affected individuals per family.


Samples at NCRAD: There are a total of 1885 samples from 422 families at NCRAD. 1298 of these samples have minimum data sets. 89% of the sampled individuals are white.

Collaboration between Centers: Centers have worked together on 25 family members. These centers include University of Pennsylvania, University of Texas Southwestern, University of
Pittsburgh, University of California Los Angeles, and Columbia University.

ADEAR efforts: A third printing of 10,000 Genetics Study brochures has taken place. Most of the brochure requests have been from ADCs. ADEAR was present, along with NCRAD, at the
Alzheimer’s Association Education/International Research Conference in Philadelphia. The
North American Precis Syndicate (NAPS) has a potential of reaching 15-20 million readers.

LOAD media coverage: The LOAD Genetics study has been featured in the following:
Time Magazine
NBC Nightly News with Tom Brokaw
Indianapolis Star
Seattle Times
Philadelphia Fox News
AARP
227 families have been referred by NCRAD. Centers also need to advertise on their own.

Minimum data set and data cleaning: There are 9 “required for all” and 17 “required for sampled” variables in the minimum data set. Data cannot be imported unless the “required for all” variables are coded. Family and individual Ids must be unique. “Required for sampled” and other conditional variables must be coded.
Procedures Manual, Control Series & Phenotype Committee Update: Stephanie Doan, MPH

Procedures manual: The website has a new domain name, www.NIAgeneticsinitiative.org
.
The minimum data set has been updated to include:
1. Control variable (specifies whether or not the subject is a control)
2. 11 additional autopsy-specific variables
3. 16 new cognitive testing variables
4. GENRSCH variable (specifies that the subject consents to researchers studying the genetics of any human disease having access to his or her clinical data and DNA; this is in accordance with NIA’s data sharing policy).

Control series: Each of the 18 Centers has been asked to ascertain 50-60 unrelated, nondemented control subjects. The goal is collect 1,000 controls. Controls are defined as individuals without subjective complaints of memory loss or cognitive decline, and whose neuropsychological test performance falls within the range of norms for age and education.*
Frequency matching based on the demographic profiles of probands already in the database will be used. Matching will be done at NCRAD.
* There is much debate about the definition and inclusion/exclusion criteria for controls. There will be an open forum via www.NIAgeneticsinitiative.org
, where questions, concerns, and suggestions can be sent. More details will be provided in the near future.

Phenotype committees: Working groups were appointed to standardize methods for querying age at onset, neuropathology, cognitive testing, and biomarkers. The subcommittees met regularly via conference calls during the last year. The results of these calls are in the following recommendations.

Age at Onset (AAO) Committee: Committee recommends that the AAO should be identified through the clinical history, and the first choice of informant should be a knowledgeable caregiver or family member. Memory decline should be accompanied by symptoms that reflect significant functional change in the individual’s abilities i.e. in judgment, home activities, and

orientation. Questions that probe for age at onset have been added to the Data Element
Dictionary of the Procedures Manual.

Neuropath/Autopsy Committee: The committee elected to have the following 11 questions added to the minimum dataset regarding autopsy:
1. Were CERAD criteria used?
2. If CERAD were used, what category was assigned?
3. Were NIA-Reagan criteria used?
4.
If NIA-Reagan were used, what category was assigned?
5. Were Braak and Braak criteria used?
6. If Braak and Braak were used, what category was assigned?
7. Was another neuropathological criteria used?
8. Describe the type of neuropathological criteria and assigned category.
9. If Lewy bodies were present, in what part of the brain were they found?
10. What stain was used to look for Lewy bodies?
11. Describe the type of stain used to look for Lewy bodies.

Cognitive Testing Committee: The new battery consists of 7 cognitive tests:
1. Logical Memory IA
2. Digit Span Forward
3. Digit Span Backward
4. Category Fluency: Animals
5. Category Fluency: Fruits and Vegetables
6. Digit Ordering
7.
Logical Memory IIA
The new battery can be administered in 15 to 20 minutes in person or via telephone to everyone that is sampled. This battery can be administered by telephone since participants in the study are often widely dispersed geographically. It would be compatible with the Alzheimer’s
Disease Centers’ unified data set, and would minimize burden on all programs and sites. A training session for cognitive testing was held in Indiana University. A certification process has begun so that coordinators would become proficient. Over half the coordinators from the 18
Centers have been certified to administer the new cognitive testing battery.

Biomarkers Committee: The goal is to collect and store plasma and serum from all participating family members. Investigators will be able to use these samples as quantitative trait biomarkers in families with AD. We are not ready to start collecting.
NCRAD Update: Tatiana Foroud, PhD

Appropriate families for Genetics Initiative study:
1.
At least 2 siblings with LOAD (onset > 60 years) who have biological samples
2.
At least 1 other living related family member who either has AD (onset > 50 years)
OR is unaffected (> 60 years)

Who to collect: Genetic analysis seeks to study the transmission of marker alleles throughout the family. When we can determine the inheritance of all marker alleles unambiguously, we have the greatest power to find genes for the disease. Unaffected individuals may be very important for collection.

I.
Who to collect
­ Collect the parents in generation I, if available
­ Collect all offspring in generation II (affected and unaffected)
­ Collect any individuals in generation III with memory loss
­ Collect any individuals in generation III > 60 years
II.
­ Query for any other affected cousins, half siblings, aunts, uncles
­ Younger people (< 60) may be important for
III. reconstruction, but need follow-up

Who are the best families: Families with the largest number of affected individuals are the best families. A strong family history indicates more genetic effects. Unaffected individuals in families with many affected individuals are also important, especially if they are examined clinically.

Preliminary pedigree: Is sent to the Coordinating Center at Columbia and should include diagnosis, current age, age at onset, and who will likely be sampled. It is highly recommended to include the family and individual numbers.

Minimum dataset: A minimum dataset must be completed for all individuals from whom a biological sample has been sent to NCRAD. The minimum dataset is transferred each quarter to
Indiana University.

Sampled pedigree: Is sent to the Coordinating Center at Columbia after minimum criteria are met and should include all individuals identified by their family and individual numbers, who were sampled, who had a minimum dataset collected, and who had neuropsychological testing.
A family cannot be considered “sampled” until the minimum dataset is transferred to NCRAD for all biologically sampled individuals. This is very important, since without the minimum dataset, we cannot determine whether the family meets the study criteria.

Pedigree sample reports: NCRAD sends out a pedigree sample report after receiving a copy of the sampled pedigree. This report highlights missing data that is required for a family to be considered sampled or complete. It is important for sites to respond and submit missing information.

Qualified families based on submitted data: 205 families have a minimum dataset submitted to NCRAD for at least one family member. Only 90 families consist of at least 2 siblings with a minimum dataset supporting AD and at least one other biologically related family member meeting the age requirements and with a minimum dataset. The majority of the families do not qualify because a minimum dataset has not been submitted for the critical individuals or have missing variables in the minimum dataset. There is small number of families who will never qualify.

Samples to Researchers: Researchers can request LOAD samples, but at this time there is limited minimum dataset data. An MTA Out will be required. Requests for samples should include IRB approval for proposed analyses, investigator CV, and a short description of proposed studies. See Section 14 of the manual for details.
What data should be collected on control subjects? : Katrina Gwinn-Hardy, MD


NINDS Repository:
Samples and clinical data on Parkinson’s, stroke, epilepsy, ALS are available at the repository. There are over 3,000 samples on affected, controls, and families with accompanying data. The website is: http://locus.umdnj.edu/ninds .

Goal for control samples: The goal for the control samples is to have enough clinical details to allow investigators to include those ideal for his/her study. This can be useful across multiple collections. Controls tend to be harder to ascertain than cases.

Control Collections: Most control collections have minimal, if any, data. They are usually population based, and age and gender matched. Some data are cases; this may not allow broad use for other disease entities.

Data collected: Some of the data collected include:
1.
Demographics (age, gender, race)
2.
Type of control (unaffected spouse, population control, convenience
3.
Medical history
4.
Family history (same disorders, relationship)
5.
Any other samples from family
6.
Smoking history
7.
Who collected data (neurologist, other MD, non-MD)?
8.
Was the subject examined?
9.
Were results of neuro-exam normal?
10.
MMSE

Current Control Data: Current control data include:
1.
No neurological disease
2.
Detailed medical history
3.
No family history of neurological disease
4.
Smoking
5.
Normal neurological examination

Currently Collecting: Approximately 1,000 controls have been collected. About 50 are available now, and nearly 300 controls will be available soon at a repository near you.
Continuation of LOAD: Marcelle Morrison-Bogorad, Ph.D & Richard Mayeux MD, Msc

Future of the NIA-LOAD Study: The five Centers, Columbia, Indiana, Mayo, University of
Washington, and Washington University, will complete the collection of the 1,000 families. A standardized follow-up procedure will be implemented, and subcontracts for follow up will be available for the remaining 13 Centers.

Specific Aims:
Aim 1: Complete the current collection of 1,000 families. Recruitment will primarily be at the five centers. The 5 Centers will be expected to travel for assessment. Recruitment of ethnic minorities is highly encouraged.
Aim 2: Identify newly affected members and evaluate appropriate unaffected individuals in participating families. All pedigrees will be expanded. Eligible family members not previously studied will be examined and family members that were unavailable or postponed will be re-contacted.
Aim 3: Develop and implement standardized follow-up procedures. Follow-up data forms will be created and validated. In person assessments using the regional travel team will be continued. Procedures for autopsy will be developed and implemented.
Aim 4: Improve statistical power for the subsequent genetic linkage analysis by implementing procedures to ascertain a series of quantitative traits for:
-Age at onset
-Cognitive performance, particularly memory

-Biological markers of disease risk and progression
Aim 5: Complete the recruitment and begin follow-up of controls. Similar assessments for controls will be used as in families with Alzheimer’s disease. Similar follow-up methods will be used.
Aim 6: Apply to CIDR for genotyping of first 500 families. All raw genotyping data will be provided in a secure, web-based format to perform genetic analyses. Proposals to complete basic preliminary linkage analyses, which will be posted on a website available to all qualified investigators, will be solicited. Genotype-phenotype information for qualified investigators will be provided.

What we will have produced: 500 families with familial Alzheimer’s disease and an additional
500 families with LOAD will be produced. 1,000 controls will be recruited and assessed. All data, including family structure and genotypes, will be available to qualified investigators.

Continued Role of ADC/ADRC: This is an ADC/ADRC related program. The consortium will report directly to NIA and ADC leadership. The original 18 centers will continue to be involved through
1.
Subcontracts for follow-up
2. Identification of new families
3. Subcontracts for genetic analysis in later years.
Publications will be required to identify all the NIA-LOAD Study investigators.
Scientific Symposium
 Genetic influences on Alzheimer’s disease: where we stand, where we are going by Peter St.
George-Hyslop, MD, DSc

Phenotype definition and analytic strategies for late-onset AD by Deborah Blacker, MD