Age of onset can show severity of glaucoma

December 1, 2007
Nancy Groves

New research in a large family affected with autosomal dominant juvenile-and adult-onset primary open-angle glaucoma has shown that age of onset can predict severity of disease. Researchers also have demonstrated that at least one modifier gene or locus alters the severity of glaucoma caused by a particular myocilin mutation.

Key Points

Québec City-New analyses of a large French-Canadian family in which one myocilin mutation is the main cause for glaucoma have shown that age of onset is a reliable predictor of disease severity in this pedigree.

The investigative team also has found that myocilin mutation carriers cluster in large groups with extreme age at onset for the glaucoma phenotype and that there is at least one modifier gene that alters glaucoma severity when it is caused by the myocilin mutation.

Several potential modifiers exist, and they also may be interacting with other myocilin mutations, said Vincent Raymond, MD, PhD, professor, Laval University Hospital, Québec City, Quebec, Canada. He discussed recent findings from an ongoing study of this family, which was first reported in the literature in 1998.

In 2003 and 2004, the researchers re-contacted family members, hoping to recruit the youngest children for molecular diagnostic purposes but also to look for modifier genes or loci that could be interacting with the myocilin mutation to cause the wide phenotypic variability, Dr. Raymond said.

The goal at that time was first to find whether a subphenotype was clustering within a particular branch of that family that might influence a characteristic such as age of onset or cup-to-disc ratio. If such a subphenotype were found, then the long-term goal would be to identify the modifier genes associated with this characteristic.

For this analysis, individuals classified as affected were those with diagnosed POAG of varying severity and those with suspected glaucoma who had undergone treatment. Individuals with suspected glaucoma who had not received treatment were classified as asymptomatic.

The investigators compiled all of their information into a database, which allowed them to conduct analyses quickly for different parameters. They investigated 312 members of the family; DNA for 375 individuals had been sequenced or genotyped to screen for presence of the Lys423Glu myocilin mutation. In all, the pedigree as of last spring included 749 individuals, dating back more than 200 years. The branch of the family under closest scrutiny began with one of the three daughters of the first identified common ancestor. She was born in 1834 and had 10 children, six of whom developed glaucoma; most were blind when they died.

This branch included 156 myocilin mutation carriers; 152 were heterozygotes and of them 97 were affected, 19 had ocular hypertension, and 36 were asymptomatic. Another 149 individuals did not carry the mutation and were not affected.

The distribution of age at onset among the affected individuals varied over 5 decades; although most had glaucoma that was diagnosed between 20 and 39 years of age, 19 had developed juvenile-onset open-angle glaucoma by the age of 19, and a few were aged more than 60 years when POAG initiated.

The penetrance rate was almost 80% for people who were carriers and aged more than 40 years, but as with other traits in this family, variability existed, Dr. Raymond said.

To look at subphenotypes and test for carriers of modifier genes, Dr. Raymond used the concept of the neighborhood, a grouping in which the members have a significant probability of sharing an allele by descent. In such a neighborhood, the kinship coefficient between the carrier and the others was defined as equal to or greater than 0.0625, a value that reflects the number of genes shared between cousins.

Dr. Raymond and his colleagues calculated the median age at onset for the neighborhood of each one of the 152 heterozygotic carriers. They then looked to see whether these carriers may be grouped together based on three categories for the median age at onset of their neighborhoods: equal to or less than 25 years, between 26 and 33 years, and 34 years or more. Results clustered into five distinct groups. Almost all members of clusters 1, 2, and 3 had a neighborhood median age at onset of 34 years or more. In clusters 4 and 5, the median age at onset was 25 or fewer years of age or less. The search for genetic modifiers was then concentrated in the clusters that showed extremes in the age at onset.