April 2014
Volume 55, Issue 13
Free
ARVO Annual Meeting Abstract  |   April 2014
A model for predicting the likelihood of identifying ABCA4 mutations for Stargardt disease
Author Affiliations & Notes
  • Jillian Huang
    Ophthalmology and Visual Sciences, Kellogg Eye Ctr, Univ of Michigan, Ann Arbor, MI
  • Sarwar Zahid
    Ophthalmology and Visual Sciences, Kellogg Eye Ctr, Univ of Michigan, Ann Arbor, MI
  • Kari E Branham
    Ophthalmology and Visual Sciences, Kellogg Eye Ctr, Univ of Michigan, Ann Arbor, MI
  • John R Heckenlively
    Ophthalmology and Visual Sciences, Kellogg Eye Ctr, Univ of Michigan, Ann Arbor, MI
  • Thiran Jayasundera
    Ophthalmology and Visual Sciences, Kellogg Eye Ctr, Univ of Michigan, Ann Arbor, MI
  • Footnotes
    Commercial Relationships Jillian Huang, None; Sarwar Zahid, None; Kari Branham, None; John Heckenlively, None; Thiran Jayasundera, None
  • Footnotes
    Support None
Investigative Ophthalmology & Visual Science April 2014, Vol.55, 3255. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Jillian Huang, Sarwar Zahid, Kari E Branham, John R Heckenlively, Thiran Jayasundera; A model for predicting the likelihood of identifying ABCA4 mutations for Stargardt disease. Invest. Ophthalmol. Vis. Sci. 2014;55(13):3255.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract
 
Purpose
 

Stargardt disease is the most common autosomal recessive macular dystrophy, and is caused by mutations in ABCA4. Nearly 50% of patients, however, have only one or no mutations identified in ABCA4. Our aim is to develop a model that allows for a prediction of the probability of identifying two mutations in ABCA4 for Stargardt patients..

 
Methods
 

Clinical and demographic variables were collected from 70 Stargardt disease patients seen at the Kellogg Eye Center, who had full sequencing of all 50 exons of ABCA4 that resulted in either two pathogenic mutations (n=31) or no mutations (n=39) found. A bivariate analysis was performed to identify the potential association of demographic (gender, family history of Stargardt disease or age-related macular dystrophy, age at visit, and age at symptom onset by decade) or clinical testing results (visual acuity, electroretinogram b-wave scotopic and photopic amplitudes as a percentage of normal, Ishihara plate color vision testing, and the presence or absence of a central scotomas on Goldmann Visual Field testing, fundus flecks, geographic atrophy, a dark peripapillary ring, or bull’s-eye maculopathy) with a positive genotype. The variables that resulted in a p-value <.10 were submitted to logistic regression. Regression analysis results were used to construct integer risk weights for each independent predictor variable. An overall genotype score was calculated by summing the risk weights for each variable for each patient. Tertiles of the genotype score were used to construct 3 relative risk groups: low (genotype+ probability 19%, score range -2 to 0), intermediate (genotype+ probability 42%, score range 1-3), and high (genotype+ probability 81%, score range 4-6).

 
Results
 

The independent predictors that showed positive association with mutation identification were age of onset, scotopic b-wave amplitude, presence of fundus flecks, and presence of a bull’s-eye maculopathy. The model had a receiver operator curve of 0.844 and Hosmer-Lemeshow goodness-of-fit P=0.22.

 
Conclusions
 

The model is a reasonably accurate tool to predict the likelihood of identifying two mutations in ABCA4 in a Stargardt disease cohort. The genotype score may be a useful tool to increase the diagnostic yield of genetic testing in Stargardt disease, improving the cost effectiveness of genetic testing.

 
Keywords: 539 genetics • 464 clinical (human) or epidemiologic studies: risk factor assessment • 696 retinal degenerations: hereditary  
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×