Uploaded on

 

  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Be the first to comment
    Be the first to like this
No Downloads

Views

Total Views
232
On Slideshare
0
From Embeds
0
Number of Embeds
0

Actions

Shares
Downloads
0
Comments
0
Likes
0

Embeds 0

No embeds

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
    No notes for slide
  • Clinical sensitivity (or clinical detection rate) Clinical specificity (1- clinical false positive rate) Prevalence of clinical disorder Positive and negative predictive values Penetrance (including gene and environmental modifiers)

Transcript

  • 1. Clinical Validity: Prenatal Screening for Cystic Fibrosis Sue Richards, PhD Professor, Molecular & Medical Genetics Director, DNA Diagnostic Laboratory Oregon Health & Science University
  • 2.
    • Defines the ability of a test to detect or predict the phenotype or particular clinical outcome
    • Elements build upon analysis of analytic validity for a specific disorder in a defined setting
    Clinical Validity Effective Intervention (Benefit) Natural History Economic Evaluation Quality Assurance Education Facilities Pilot Trials Monitoring & Evaluation Ethical, Legal, & Social Implications (safeguards& impediments) Health Risks Clinical Specificity Clinical Sensitivity Prevalence PPV NPV Penetrance Assay Robustness Quality Control Analytic Specificity Analytic Sensitivity Disorder & Setting
  • 3. Test Result Disease Phenotype Yes No Pos A B Neg C D Clinical Sensitivity & Specificity Sensitivity: Proportion of persons with the disease who have a positive test = A / (A+C) Specificity: Proportion of persons who do not have the disease who have a negative test = D / (B+D)
  • 4. Issues in evaluating Clinical Validity
    • Study design
      • Case definitions may vary
      • Small numbers and potential biases
      • Populations may need to be stratified by variables such as gender, age, race/ethnicity
      • Testing protocols may not be comparable (e.g., numbers of mutations tested)
      • Comparability of case and control populations
    • Need to evaluate the impact of genetic and environmental modifiers
  • 5. Clinical Sensitivity of CFTR Testing: Prenatal Screening for Cystic Fibrosis
    • Clinical sensitivity is dependent on the panel of CFTR mutations selected
      • This review will focus on the ACMG recommended panel of 25 mutations
    • Clinical sensitivity is dependent of the race/ethnicity of the population screened
      • This review will provide estimates for self-reported race/ethnic groups (i.e., Hispanic and non-Hispanic Caucasians, Ashkenazi Jewish, African Americans and Asian Americans)
    • Clinical sensitivity is not dependent on the screening model used
      • Sequential (two-step), couple (one-step) and concurrent
  • 6. American College of Medical Genetics (ACMG) Recommended Panel of 25 CFTR Mutations
    • 7 mutations often tested in 7-13 pilot trials
      • delF508, G551D, G542X, 621+1G>T, W1282X, N1303K, and R553X
    • 7 mutations sometimes tested (2 to 6 trials)
      • delI507, 1717+1G>T, R117H, 3849+10kbC>T, R560T, A455E, R334W
    • 11 mutations rarely or never tested (0 or 1)
      • R347P, 711+1G>T, R1162X, I148T, 2789+5G>A, 3569delC, G85E, 2184delA, 1898+1G>A, 3120+1G>T, 1078delT
  • 7. To Reliably Estimate Clinical Sensitivity Requires
    • An unbiased set of DNA samples from individuals with clinically defined ‘classic’ cystic fibrosis
    • A set test panel of CFTR mutations
    • Two large datasets are available:
      • The Cystic Fibrosis Consortium (diagnostic laboratories)
      • The Cystic Fibrosis Foundation (clinics providing patient care)
  • 8. An ACCE Review Involves a Critical Evaluation of the Literature
    • The CF Consortium (Kazazian et al., 1994) reported mutation frequencies in North Americans (Caucasians). However, those numbers could not be used directly for two reasons:
      • This group includes reports that focus mainly on Hispanic Caucasians, Ashkenazi Jewish or African Americans
      • There was a computational error that underestimated mutation frequencies, especially for those less common
  • 9. Cystic Fibrosis Consortium Reanalysis Mutation Frequency (%) Mutation Original Revised* delF508 66.1 68.94 G542X 2.2 2.17 G551D 2.0 2.06 621+1G>T 1.5 1.91 ?? Others 7.4 8.20 A455E 0.25 0.54 711+1G>T 0.20 0.77 R347P 0.25 0.46 All 79.9 85.05 *After removing 15 reports and using a new denominator
  • 10. Mutation Frequencies in Non-Hispanic Caucasians Mutation CF Consort CF Found Average Cumulative delF508 68.94 75.90 72.42 72.42 G542X 2.17 2.39 2.28 74.70 G551D 2.05 2.44 2.25 76.95 621+1G>T 1.92 1.22 1.57 78.52 W1282X 1.42 1.57 1.50 80.02 20 others 8.55 8.12 8.33 88.35 Total 85.05 91.64 88.35 CF Consortium based on between 2,197 and 9,792 chromosomes CF Foundation based on 3,938 chromosomes
  • 11. Which Set of Mutation Frequencies Should be Used?
    • The Cystic Fibrosis Consortium data may under-represent ‘common’ mutations such as delF508 because of the ‘reference’ nature of participating laboratories
    • The Cystic Fibrosis Foundation data may over-estimate the more ‘severe’ mutations such as delF508, because they are more likely to be enrolled for services
    • The average of the two may be reasonable to use
  • 12. Relationship Between Carriers Identified & Detection of Carrier Couples (or Affected Fetuses) Carrier Individuals Carrier Couples Increase * 20 4 30 9 5 40 16 7 50 25 9 60 36 11 70 49 13 80 64 15 90 81 17 95 90 9 * Increase in detection for 10% increase carriers identified Proportion Identified (%)
  • 13. Carrier Couples Detected (Clinical Sensitivity) Among Non-Hispanic Caucasians 1 5 10 15 20 25 20 30 40 50 60 70 80 90 Number of CFTR Mutations Clinical Sensitivity (%) for Carrier Couples or Affected Fetuses 78% 75%
  • 14. Clinical Sensitivity in Other Racial/Ethnic Groups
    • Similar analyses are performed for individuals reported to be
      • Hispanic Caucasians
      • Ashkenazi Jewish
      • African American
      • Asian American
  • 15. Clinical Sensitivity of Prenatal Cystic Fibrosis Screening by Race/Ethnicity 1 5 10 15 20 25 20 30 40 50 60 70 80 90 100 Number of CFTR Mutations Clinical Sensitivity (%) Ashkenazi Jewish Non-Hispanic Caucasian Asian American Hispanic Caucasian African American
  • 16. Clinical Specificity of Prenatal Screening for Cystic Fibrosis
    • Analytic false positive results
      • Pre analytic
        • Sample mix-up prior to laboratory receipt
        • Degraded or mishandled sample
        • Non-paternity
      • Analytic
        • Sample mix-up after laboratory receipt
        • Benign polymorphism mistaken for a mutation
      • Post analytic
        • Data entry error
        • Incorrect laboratory interpretation
        • Report mix-up at the laboratory or provider site
  • 17. Clinical Specificity of Prenatal Screening for Cystic Fibrosis
    • Clinical false positive results
      • Incomplete penetrance
        • Can be minimized, but not eliminated, by reflexive testing associated with R117H and I148T (D1152H is not on the panel but is likely to be of low penetrance as well)
        • A small proportion (up to 5%?) of individuals with two mutations will not have the ‘classic’ CF phenotype
      • Mutation is not disease-causing
        • Possible that I148T is a polymorphism that is tightly linked to a ‘real’ mutation
  • 18. The Birth Prevalence of Cystic Fibrosis by Race/Ethnicity
    • Sources of data include:
      • prenatal (pre) screening trials
      • newborn (new) screening trials
      • population (pop) registries
    • Identified biases include:
      • number of mutations tested (pre, new)
      • race/ethnicity not accounted for (pre, new, pop)
      • prenatal screening and termination (new, pop)
      • length of follow-up (new, pop)
      • other study-specific biases (pre, new, pop)
  • 19. ACCE Methodology Attempts to Account for Biases Rather than Excluding Studies
    • Number of mutations tested, and proportion of cases detected by race/ethnicity can be taken into account by u sing the race/ethnic specific mutation rates described earlier
    • Several studies provide the relative rates of prenatal screening and termination that occurs in concert with newborn screening (Europe / Australia)
    • The proportion of cases detected during reported follow-up can be adjusted using age at diagnosis information from the CF Foundation
  • 20. Adjusted Prevalence of CF in non-Hispanic Caucasians Using Prenatal Screening Trials 9B 6 10 4 13 1 9A 12 5 7 3 11 8 Consensus 1000 2000 3000 5000 7000 1:2488 Prenatal Study Number Birth Prevalence (1:n)
  • 21. Prevalence of Cystic Fibrosis in non-Hispanic Caucasians Using Newborn Screening Trials 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Newborn Study Number Birth Prevalence (1:n) Observed Adjusted 500 1000 2000 3000 4000 6000 1:2509 UK Australia US Europe
  • 22. Prevalence of Cystic Fibrosis in non-Hispanic Caucasians Using Population Registries UK Canada US Consensus 1000 2000 3000 5000 7000 1 : 2499 Population-Based Registry Number Birth Prevalence (1:n)
  • 23. Summary: Birth Prevalence of Cystic Fibrosis in Selected Racial/Ethnic Groups Prevalence Racial/Ethnic Group Studies (1:n) Ashkenazi Jewish 4 2,271 Non-Hispanic Caucasians 33 2,500 Hispanic Caucasians 3 13,500 African Americans 3 15,100 Asian Americans 1 31,000
  • 24. Genotype/Phenotype Relationships
    • ‘ Classic’ cystic fibrosis phenotype occurs about 98% of the time when two mutations in the recommended mutation panel are identified.
    • Most mutations associated with pulmonary disease, but cannot accurately predict timing of onset & progression.
    • Pancreatic insufficiency in most cystic fibrosis patients, but 5 to15% have pancreatic function. Certain less common mutations are associated with pancreatic sufficiency.
    • Screening is expected to generate more information.
  • 25. Acknowledgements
    • ACCE Review: Population-based prenatal screening for cystic fibrosis via carrier testing (available on the CDC website)
    • ACCE Core Group
    • Slides provided by:
      • Glenn Palomaki & Linda Bradley