DISCLOSURE OF GENETIC RISK: LESSONS FOR EARLY RECOGNITION

1
DISCLOSURE OF GENETIC RISK LESSONS FOR EARLY
RECOGNITION

• Deborah Blacker, MD, ScD
• Assistant Vice Chair for Research,
• Department of Psychiatry,
• Massachusetts General Hospital
• Associate Professor of Psychiatry,
• Harvard Medical School
• Associate Professor in Epidemiology,
• Harvard School of Public Health

2
REASONS FOR GENETIC TESTING

• Limit uncertainty and associated anxiety
• Obtain information about the future to guide
  reproductive, financial, and other life choices
• Select appropriate individuals for prevention or
  early intervention
• Select optimal prevention or early intervention
  strategy
• Shared with early recognition

3
TYPES OF GENETIC TESTING FOR ADEarly-onset
familial AD (EOFAD)

• Relevant to individuals with rare Mendelian
  family history
• Locus and allelic heterogeneity multiple
  mutations in three genes
• Deterministic virtually all are fully penetrant

4
(No Transcript)
5
Testing for early-onset AD genes

• Over 200 mutations among the three genes see
  online AD and FTD Mutation Database
• Testing requires sequencing and interpretation,
  so optimally done in a site with an active AD
  genetic research program
• Commercial testing for PSEN1, testing for other
  genes must be done in CLIA labs affiliated with
  research labs
• Offered with Huntingtons protocol as for HD,
  there has been limited uptake

6
Risks and Complications of Genetic Testing for
EOFAD
Testing is complex and expensive ideally
requires sequencing an affected family
member Pre- and post-test counseling strongly
recommended Genetic information affects the whole
family may be unwelcome for some Insufficient
information on when disease will strike to time
interventions, especially for PSEN2 and APP DIAN
project aims to understand early detection
strategies in setting of clear-cut genetic risk,
with benefits for gene carriers and more broadly
7
TYPES OF GENETIC TESTING FOR ADLate-onset AD

• Garden variety AD
• APOE is the only confirmed risk factor gene
  probabilistic, not deterministic
• Three alleles APOE-2 reduces risk, APOE-4
  increases risk
• APOE-4 has non-linear gene dose effect on risk
  and onset age

8
Testing for APOE

• Multiple blue ribbon panels have concluded that
  APOE testing has insufficient predictive value
  for clinical use
• However, likely to have a role in identifying
  high risk individuals for intervention
• REVEAL study (Robert Green, MD, MPH, PI),
  developed and tested methods to communicate
  meaning of APOE test results in the context of
  family history
• Showed that complex results can be communicated,
  even in a primary care setting, without undue
  psychological burden

9
Cupples et al., Genetics in Medicine, 2004 slide
courtesy of Dr. Robert Green
10
Risks and Complications of Genetic Testing for
Late-onset AD
Information remains of limited value for the
individual in the absence of early
intervention Despite successes of REVEAL study,
disclosure is time-consuming, and draws heavily
on limited understanding of probability among
physicians and their patients Large numbers at
risk put big demands on primary care Again,
genetic information affects the whole
family Little information on when the disease
will strike to guide timing of intervention
11
Late onset AD genes beyond APOE?

• AlzGene website documents multiple positive
  reports and multiple non-replications across the
  genome
• Meta-analyses suggest that some may have modest
  impact on risk
• Might contribute to risk profiling in
  combination with APOE and other factors (see
  AlzRisk website for non-genetic risk factors)

12
Meta analysis CHRNB2, rs4845378 (T allele vs. G
allele)
Alzheimer Research Forum Gene Database
www.alzgene.org
13
Alzheimer Research Forum Gene Database
www.alzrisk.org
14
LESSONS FOR EARLY DISCLOSURETwo continua of
uncertainty

• Given survival issues, risk and timing arent
  entirely separate, but its critical to
  distinguish them in assessing early recognition
  strategies
• Continuum of uncertainty in estimated risk For
  long-term prevention strategies, whether may be
  as important as when
• Continuum of uncertainty in estimated time to
  onset For short-term early intervention
  strategies, and for selection of subjects for
  clinical trials of such strategies, when is much
  more important

15
Progression of Alzheimers Disease
Presymptomatic
Preclinical
Clinical Dementia
CognitiveFunction
Prevention
Early intervention
Treatment
Disease Progression
16
Genetics and the continua of uncertainty

• Genetics can be very strong predictor of whether
• EOFAD genes gtgt APOE gt family history
• Genetics is relatively silent on when
• PSEN1 gt other EOFAD genes gtgt APOE gt family history

17
Multiple modalities in early recognition

• Focus of these efforts is to identify subjects
  for early interventionat a point where when is
  soon, and to know this with some precision
• Low tech questionnaires on memory complaints or
  subtle functional loss, standard neuropsych tests
  and diagnosis of MCI, NP challenge tests
• High tech structural MRI (subtle hippocampal
  atrophy, cortical thinning), fMRI (increased
  activation), PIB (amount, location)
• More work is needed to place these methods on the
  two continua of uncertainty

18
NP testing and clinical interviews both can
identify "preMCI" individuals more likely to
progress to dementia
Dickerson et al, Arch Gen Psychiatry, 2007
19
Structural MRI Pattern of cortical thinning may
provide an early cortical signature of AD,
visible in early MCI

• In some regions, MCI converters suffer as much
  thinning as mild ADs, but this is not true for
  all regions

Bakkour et al, Neurology, 2009 slide courtesy of
Dr. Brad Dickerson
20
fMRI Baseline hippocampal activation in early
MCI predicts degree of clinical decline over four
years
MCI subjects with 4 year clinical follow-up
(n25), greater hippocampal activation predicts
greater degree of cognitive decline (plt.02)
Miller S et al JNNP 2008) slide courtesy of Dr.
Reisa Sperling
21
PIB-PET Many clinically normal elders have
significant PiB retention longitudinal studies
in process
Slide courtesy of Drs. Keith Johnson and Reisa
Sperling
22
Risk vs. early recognition

• These strategies are sometimes said to identify
  those at increased risk of AD, but ideally
  identify those with underlying early disease
  pathology
• However, not all those identified go on to
  develop AD, at least not in several years of
  follow-up
• Critical to understand identify modifiers of
  underlying disease progression and the likelihood
  of symptoms for a given degree of pathology
• More data needed to understand which indicators
  or combinations thereof offer best prediction of
  whether and when disease will occur
• Such data are critical in designing clinical
  trials of early intervention strategies