Age-related Macular Degeneration Phenotype Consensus Meeting

One of the major research opportunities identified by the Retinal Diseases Panel during development of the National Eye Institute's National Plan for Eye and Vision Research was the standardization of the definitions and characteristics of retinal phenotypes in macular disease. The panel thought that more precise disease definitions based on genotype/phenotype/environment correlations would aid in the study of disease progression and the development of new therapies.

Key factors in successfully meeting this challenge are the need for common disease descriptors and a uniform fundus reading/grading system to identify new phenotypes. With such systems in place, it will be possible to identify new phenotypes and explore their underlying genotype. A decision was made to begin with age-related macular degeneration (AMD) because the identification of new AMD phenotypes could be accomplished through an existing network of reading/grading centers that review photographs of ocular pathology. Currently, these centers have established in-house methodologies and phenotypic definitions that are specific to an individual reading center and may not be broadly applicable to other systems.

In an effort to share information about grading methodologies and descriptors currently being used at major AMD grading centers, the NEI hosted an Age-related Macular Degeneration Phenotype Consensus Meeting on October 29, 2003. This meeting assembled investigators with the largest AMD reading/grading collections in the U.S. and abroad. Each of the grading centers has collected phenotype data from large cohorts for various epidemiology and treatment studies. The studies represented include the Beaver Dam Eye Study, the Age-Related Eye Disease Study, the Complications of Age-Related Macular Degeneration Trial, the Blue Mountains Eye Study, the Melbourne Visual Impairment Study, and the Rotterdam Eye Study.

The meeting participants, listed at the end of this document, were asked to consider several important questions in using phenotypical information to reach the overarching goal of identifying genes that influence the development of AMD. These questions and capsule summaries reflecting a sense of the discussions at the meeting are listed below.

How can AMD phenotypes enable the fields of epidemiology and molecular genetics to understand the complex interplay between environment and genetics in this disease?
AMD is a complex disorder with multiple phenotypes which are ultimately associated with vision loss in individuals with late stage disease. Epidemiologic studies describing AMD phenotypes in different populations have begun to inform research aimed at understanding the genetic and environmental factors involved in AMD because they allow researchers to more precisely define these different phenotypes, thereby increasing the likelihood of identifying the genes and environmental factors predisposing to a given phenotype.

NEI added an ocular component to the Age Gene/Environment Susceptibility Study (AGES), a large cohort of 9500 older-age Icelanders, to strategically investigate the environmental and genetic risk factors predisposing this genetically homogeneous population to develop specific AMD phenotypes. Similarly, the Age-related Macular Degeneration Amish Study is being funded by NEI to identify specific genetic and environmental factors underlying susceptibility to AMD phenotypes among Amish families. The genetic basis of early AMD phenotypes in different ethnic subgroups is under investigation by NEI in the Family component of the Multi-Ethnic Study of Atherosclerosis.

The meeting participants agreed that the identification of distinct phenotypes in homogenous populations creates opportunity to distinguish environmental and genetic factors in complex diseases like AMD. In more heterogeneous populations like the United States, researchers have not yet identified distinct phenotypes that would usefully direct genetic studies. The meeting participants suggested that ongoing investigations of early onset disease and familial and twin studies might identify useful phenotypes. Clinical findings such as retinal pigment epithelium pigmentation, and the size, distribution and symmetry of drusen deposits might also identify distinct phenotypes.

What have we learned from phenotype/genotype studies of classically inherited forms of macular degeneration such as Best and Stargardt disease? Do these lessons have relevancy to a multigenic disease like AMD?
The meeting participants described issues of phenotypic similarities among different genetic diseases. For example, a rare, genetic form of type 2 diabetes, maternally inherited diabetes and deafness (MIDD) also has associated retinal abnormalities that can resemble age-related macular degeneration. Choroideremia and retinitis pigmentosa can have very similar phenotypes and without genotype information, it is often difficult to correctly diagnose these patients. These examples highlight the potential pitfalls in selecting phenotypes to base future genetic studies.

In classic Mendelian eye disease, phenotypes have not been of great value for genetic research. Traditionally, geneticists have looked for large, multigenerational pedigrees and mapped the candidate chromosomal region as a first step to isolate the gene. However, for AMD genetic analysis is hampered by the late onset of the disease making it difficult to find surviving multigenerational families.

Do current AMD fundus reading center collections lend themselves to meta-analysis?
With some modification of these data sets, meta-analysis is possible. However, each of the reading centers has very large collections and therefore meta-analysis may not be necessary.

Can we reach consensus on an agreed classification system for phenotyping AMD?
As discussed previously, there are very few recognized phenotypes in AMD.
The ones in existence are currently well-known and agreed upon. Additional phenotypes will have to come from fundus reading/grading center collections. The meeting participants discussed and compared their classification schemes for reading AMD fundus photos. The eight reading centers represented at this meeting are the largest AMD fundus photo collections in existence. In reviewing classification schemes, it became clear that these eight centers had similar classifications systems that provide a cohesive system to read/grade AMD fundus photos and to describe new phenotypes. A consensus was reached to use these systems as a standard for the identification of future phenotypes. The two tables describing these classification systems were compiled by Dr. Tunde Peto of the Institute of Ophthalmology and are included as an appendix in this report.

What recommendations would the meeting participants make as a next step toward promoting more uniform grading classification systems?
The group suggested that because the classification systems used by each of the individual grading centers represented at the meeting were in close concordance, it would be helpful to make these grading protocols widely available as a suggested set of guidelines for other interested researchers.

The tables included in this report describe the grading protocols used by the various fundus reading centers represented at this meeting. Table 1 describes grading protocols for fundus photography. Table 2 describes grading protocols used for fluorescein angiography.

AMD Phenotype Consensus Meeting Participants

Alan C. Bird, MD
Dept of Clinical Ophthalmology
Moorfields Eye Hospital

Neil M. Bressler, MD
Wilmer Eye Institute
Johns Hopkins Medical School

Susan B. Bressler, MD
Wilmer Eye Institute
Johns Hopkins Medical School

Deborah Carper, PhD
National Eye Institute

Emily Y. Chew, MD
National Eye Institute

Karl G. Csaky, MD, PhD
National Eye Institute

Matthew D. Davis, MD
Ophthalmology & Visual Science
University of Wisconsin

Frederick L. Ferris, III, MD
National Eye Institute

Stuart L. Fine, MD
Scheie Eye Institute

Robyn Guymer, MD
Macular Research Unit
University of Melbourne

David Hinton, MD
USC School of Medicine

Larry D. Hubbard, MAT
Ophthalmology & Visual Science
University of Wisconsin

Paulus TVM de Jong, MD
Netherlands Ophthalmic Research Institute

Barbara E.K. Klein, MD
Ophthalmology
University of Wisconsin

Ronald Klein, MD
Ophthalmology
University of Wisconsin

Maureen G. Maguire, Ph.D
Ophthalmology
Scheie Eye Institute

Paul Mitchell, MD, PhD
Ophthalmology
University of Sydney
Eye Clinic Westmead Hosp

Tunde Peto
Institute of Ophthalmology
Moorfields Eye Hospital

Paul A. Sieving, MD, PhD
National Eye Institute

The following tables describe the highly technical metrics used to grade fundus photographs of patients with macular degeneration. The eight participating fundus reading centers surveyed in this document provide an exemplary manual for distinguishing new phenotypes of macular degeneration.