gms | German Medical Science

VI. International Symposium on AMD – Age-Related Macular Degeneration – Emerging Concepts – Exploring known and Identifying new Pathways

11. - 12.09.2015, Baden-Baden

The majority of familial AMD can be explained by clustering of common risk factors

Meeting Abstract

  • Eveline Kersten - Nijmegen
  • Y.T.E. Lechanteur - Nijmegen
  • N.T.M. Saksens - Nijmegen
  • M.J. Geerlings - Nijmegen
  • T. Schick - Cologne
  • S. Fauser - Cologne
  • C.J.F. Boon - Leiden
  • A.I. den Hollander - Nijmegen
  • C.B. Hoyng - Nijmegen

VI. International Symposium on AMD – Age-Related Macular Degeneration – Emerging Concepts – Exploring known and Identifying new Pathways. Baden-Baden, 11.-12.09.2015. Düsseldorf: German Medical Science GMS Publishing House; 2015. Doc15amd03

doi: 10.3205/15amd03, urn:nbn:de:0183-15amd031

Veröffentlicht: 1. Oktober 2015

© 2015 Kersten et al.
Dieser Artikel ist ein Open-Access-Artikel und steht unter den Lizenzbedingungen der Creative Commons Attribution 4.0 License (Namensnennung). Lizenz-Angaben siehe



Background: Age-related macular degeneration (AMD) is known to cluster in families. Prediction models have been established for AMD based on common non-genetic and genetic risk factors. In AMD families these risk factors may be distributed differently, and rare genetic variants have been identified. This raises the question whether these models can be used in families. This study aims to evaluate the validity of a prediction model based on case-control data in AMD families.

Methods: Two datasets were extracted from the European Genetic Database (EUGENDA). The first dataset contained 1037 advanced AMD patients and 1290 controls; these were randomly subdivided into a training and validation set. The second dataset, a family dataset, contained 119 advanced AMD cases and 103 controls. A prediction model for the development of advanced AMD was created in the training set using logistic regression analyses. Subsequently, this model was validated in the validation and family set using receiver operating characteristics curves and calculation of the area under the curve (AUC).

Results: The final model included five non-genetic and seven genetic factors and showed an AUC of 0.873 (95%CI 0.851-0.895) in the training set. The AUCs in the validation and family set were 0.854 (95%CI 0.808-0.900) and 0.842 (95%CI 0.787-0.897). Further analyses of individual families showed clustering of risk factors in most families. We also identified two densely affected families despite low predicted values for all affected family members.

Conclusion: This study demonstrates that a prediction model based on common genetic and non-genetic risk factors can be applied to AMD families due to clustering of these factors in most families. In a subset of families the risk of advanced AMD was not explained by these factors. This suggests that other factors, such as rare genetic variants, play a role in the development of advanced AMD in these families.