The Primary Open-Angle Glaucoma Genes and Environment (GLAUGEN) Study

Background Information: Primary open-angle glaucoma (POAG) is an age-related, intraocular pressure (IOP)-dependent progressive optic neuropathy that ultimately leads to irreversible blindness. Vision loss from POAG is a condition of public health significance. Current evidence suggests that POAG is a polygenetic disease modified by environmental influences. Despite the fact that a positive family history of disease is an important risk factor for POAG, conventional linkage and candidate gene approaches have revealed less than 5% of the genetic component of the disease. Furthermore, there is no consensus on environment risk factors for POAG. Elevated IOP is the only modifiable risk factor for POAG; yet, lowering IOP slows, but does not halt the disease process.

Study Objectives: The overall goal of our research is to elucidate the pathogenesis of POAG so that cost-effective disease detection and primary prevention strategies can be implemented. The primary aim of the Glaucoma Gene Environment Initiative (GLAUGEN), funded by the Human Genome Research Institute (NHGRI), is to discover genetic loci associated with POAG. The secondary aim of GLAUGEN, funded by NHGRI and the National Eye Institute, is to discover gene environment interactions in POAG.

Methods and study populations: For this study, we have assembled cases and controls from three studies: the Nurses' Health Study (NHS), the Health Professionals Follow-up Study (HPFS) and the Genetic Etiologies of POAG (GEP) project based at Massachusetts Eye and Ear Infirmary (MEEI). This case-control group includes 1057 unrelated cases and 1272 controls. Members of the NHS and HPFS also have repeated environmental exposure data collected prior to a diagnosis of POAG.

NHS - The Nurses Health Study started in 1976 under the direction of Dr. Frank E. Speizer. With funding from the NIH, registered nurses from 11 US states were invited to complete a detailed questionnaire regarding lifestyle and health biennially. Initially, 121,000 women responded to the baseline questionnaire. Currently, Dr. Susan Hankinson serves as the program director for the NHS.

HPFS - The Health Professionals Follow-up Study began in 1986 under the direction of Drs. Walter Willett and Meir Stampfer. Under the auspices of the NIH, they enlisted 51,529 male health professionals from throughout the US to complete similarly designed biennial questionnaires.

Beginning in 1990, questions regarding ocular health were added to biennial questionaires completed by health professionals participating in the NHS and HPFS. This allowed us to formulate (PI: S. Hankinson; NEI) and maintain (PI: L. Pasquale; NEI) a cohort at risk for POAG derived from the respective general cohorts who were under ophthalmic care. We then developed a definition of POAG that allowed us to identify cases from a population that was geographically dispersed. The centerpiece of this definition is the presence of reproducible visual field loss consistent with nerve fiber layer (NFL) dropout (the NFL contains the axons that comprise the optic nerve) on reliable tests. Reproducible visual field loss occurred in the context of anterior segment findings that did not suggest a secondary cause of elevated IOP and posterior segment findings that did not suggest a secondary cause of visual field loss. We selected controls from the cohort at risk for POAG on the basis of age, gender and time period when cases were identified.

GEP - The Genetic Etiologies of POAG was initiated in 1996 with funding from the National Eye Institute under the direction of Dr. Janey Wiggs. The purpose of this work was to discover novel genetic loci associated with POAG. In the GEP, cases were derived predominantly from the Glaucoma Service at MEEI. The majority of cases had an examination by a glaucoma specialist and met the definition for POAG used in NHS and HPFS. Cases with only one reliable visual field consistent with NFL dropout were included if there was a cup-disc ratio of 0.7 or more. The majority of controls were patients who presented to the MEEI comprehensive ophthalmology service for routine eye examination or from spouses of MEEI patients with secondary forms of glaucoma. Other controls were identified from regional glaucoma screenings held throughout Massachusetts. Members of GEP have detailed ocular phenotype data but limited information on environmental exposures.

This study is part of the Gene Environment Association Studies initiative (GENEVA, http://www.genevastudy.org) funded by the trans-NIH Genes, Environment, and Health Initiative (GEI). The overarching goal is to identify novel genetic factors that contribute to primary open-angle glaucoma through large-scale genome-wide association studies of three well-characterized cohorts of cases and controls, some in matched pairs. Genotyping was performed at the Broad Institute of MIT and Harvard, a GENEVA genotyping center. Data cleaning and harmonization were performed at the GEI-funded GENEVA Coordinating Center at the University of Washington.

• Study Design:
  • Case-Control
• Study Type:
  • Case-Control
• dbGaP estimated ancestry using GRAF-pop
• Subject Sample Telemetry Report (SSTR)

Case inclusion criteria:

1. Age ≥ 40 years old.
2. European-derived Caucasian or Hispanic Caucasian.
3. Slit lamp exam findings did not reveal secondary causes for elevated intraocular pressure (IOP). Specifically cases showed no signs of exfoliation syndrome, pigment dispersion syndrome, uveitis or trauma on slit lamp examination when the diagnosis of POAG was made. Subsequent discovery of these signs after a diagnosis of POAG was made does not exclude a subject.
4. For cases, the filtration apparatus was open to the filtering portion of the trabecular meshwork for at least 180 degrees. Alternatively, there was evidence that the pupils were pharmacologically dilated without an elevation of IOP.
5. There was visual field (VF) loss (on reliable VFs) consistent with nerve fiber layer loss. A reliable and abnormal VF consistent with glaucoma was defined according tothe following considerations:
1. A reliable visual field was defined by fixation loss ≤ 33%, false positive rate ≤ 20% and false negative rate ≤ 20%.
2. For automated perimetry there was no requirement for the type of perimeter used except that the perimeter must have an age-matched controlled database that allows for the equivalent of a total deviation and pattern deviation plot. Thus Dicon and Octopus perimetry data were acceptable.
3. Goldmann visual fields (GVFs) were acceptable unless the perimetrist regarded the test as unreliable.
4. The VF defect must be regarded as consistent with nerve fiber layer loss (loss not consistent with choroidal-retinal pathology, chiasmal disease or post-chiasmal lesions).
5. The VFs were graded in a standardized manner using the pattern deviation plot.
6. For automated VFs, the pattern deviation plot or its equivalent was stratified into superior and inferior paracentral, nasal step, Bjerrum area and temporal wedge regions. A cluster of 3 or more point that were -5dB reduced from normal constitute the minimal change consistent with nerve fiber layer pathology. In the superior Bjerrum area, the superior-most points were disregarded because they could be influenced by the lid position. Each VF was graded on a Glaucoma VF Review Sheet.
7. The VF loss was not explained by choroidal-retinal disease mimicking nerve fiber layer loss (such as retinitis pigmentosa) or other optic nerve disease (such as optic nerve drusen or ischemic optic neuropathy).
8. For GVFs, the investigator stratified the field into similar regions to assess whether the VF contained defects consistent with NFL loss.
9. If the patient had other reasons to have VF loss, such as a cerebrovascular accident (CVA), then only the segment of the pattern deviation plot that was not affected by the CVA was used to determine whether the participant met criterion for a case.
6. The VF loss was reproduced on a subsequent reliable VF in the same region. For example, if the 1st VF shows a superior nasal step, then the subsequent VF must also showed a superior nasal step. In the GEP, in lieu of a subsequent VF, the cup to dic ratio (CDR) could be used for confirmation if the CDR was greater than or equal to 0.7 in the eye showing loss. When multiple VFs were available and the latest available VF did not show minimal criteria for a NFL defect then the patient was excluded as a POAG case.
7. Cases in GEP all had less than 8 diopters of myopia. There was no refractive error inclusion/exclusion criterion for cases in NHS and HPFS.

Control inclusion criteria:

1. Age ≥ 40 years old.
2. European-derived Caucasian or Hispanic Caucasian.
3. Because the study also aimed for the discovery of gene environment interactions in POAG, controls from NHS and HPFS (where extensive environmental exposure data was available) were allowed to have a family history of glaucoma in 1st degree relatives (parent, sibling or child) or 2nd degree relatives (aunts, uncles, and cousins). Controls from GEP with a family history of glaucoma in 1st degree relatives were excluded, but they could have a 2nd degree relative with glaucoma.
4. There were no specific criteria for exclusion based on refractive status for participants in the NHS and HPFS. In the GEP, participants with more than 8 diopters of myopia were excluded.
5. For controls in GEP, it was required that slit lamp exam findings did not reveal a secondary cause for elevated intraocular pressure. Specifically, there were no signs of exfoliation syndrome, pigment dispersion syndrome, uveitis or ocular trauma on slit lamp examination. Controls in NHS and HPFS did not report a diagnosis of glaucoma or IOP > 25 mm Hg in either eye in the biennial questionnaires (but there is a remote possibility that some participants had slit lamp signs of pigment dispersion syndrome, exfoliation syndrome, uveitis or prior ocular trauma as these were not assessed in the biennial questionnaires).
6. For controls in GEP, slit lamp exam showed > 0.25 for the corneal thickness anterior chamber depth at the peripheral cornea with the van Herick technique. Subjects with shallower anterior chamber depth with the van Herick technique had gonioscopy showing the filtration apparatus to be open to the filtering portion of the trabecular meshwork for at least 180 degrees. For controls in NHS and HPFS, there was no exclusion criterion on the basis of anterior chamber depth.
7. For controls in GEP, participants with IOP > 21 mm Hg in either eye by applanation tonometry were excluded. In NHS and HPFS, only controls who denied a report that they were aware of an IOP > 25 mm Hg in either eye were included; otherwise, there was no exclusion criterion on the basis of IOP.
8. For controls in GEP, the CDR was less than 0.7 in both eyes and the CDR asymmetry was less than 0.2 on fundus examination of the optic nerve. In NHS and HPFS, there was no inclusion/exclusion criterion on the basis of optic nerve appearance.

Exclusions in cases and controls:

1. People of Asian or African decent were excluded.
2. People under the age of 40 years old were excluded.
3. Potential cases in whom the most recent available VF is reliable and normal in both eyes were excluded; these were ineligible regardless of prior VF findings and regardless of the disc appearance.
4. Patients with significant retinal vascular disease (such as retinal venous occlusive disease or proliferative diabetic retinopathy treated with pan-retinal photocoagulation) that could produce VF defects were excluded.

Note #1. There were no exclusions for IOP among POAG cases because POAG occurs across the entire spectrum of IOP (A World Glaucoma Society consensus statement, ARVO 2007). Cases of "normal tension glaucoma" will be analyzed together with "high tension" open-angle glaucoma as there is no epidemiologic evidence that clearly separates these two populations.

Note #2. There is a wide spectrum of CDRs associated glaucomatous optic neuropathy owing to the variation in disc size and shape that exists in the population. Thus, there were no a priori disc structural criteria for POAG if there were no supporting VF data.

The NHS and HPFS were initiated in 1976 and 1986 respectively, when 121,000 registered nurses participated and 51,529 male health professionals completed detailed questionnaires about lifestyle and general health. Questionnaires have been repeated every year and the content of these questionnaires can be found at the following websites: www.channing.harvard.edu/nhs and www.hsph.harvard.edu/hpfs/hpfs_qx.htm. Follow-up has been approximately 85% of the total person time of follow-up.

In 1990, questions regarding ocular health were added to the biennial questionnaires in both the NHS and HPFS cohort. In 1996, under the direction of Dr. Susan Hankinson, Dr. Jae Hee Kang formed the cohort at risk for POAG. In 2004, Dr. Pasquale, capitalizing on DNA samples collected from 1990 through 2004, studied the relation between candidate gene-environment interactions in POAG. This provided an opportunity to expand the cohort at risk for POAG and to form a case-control group nested within that cohort. This nested case-control group served as the basis for the NHS/HPFS portion of this case-control group.

In 1996, Dr. Janey Wiggs began her study of the genetic determinants of POAG. In 2000, she published a low-density genome wide scan for POAG, a study that demonstrated that POAG is a polygenetic disease. Subsequent funding allowed for the expansion of her clinic-based repository, which served as the basis for the GEP component of this case-control cohort.

• Primary Phenotype: Glaucoma, Open-Angle
• Low Tension Glaucoma

• BioProject
• PubMed
• BioSample
• MeSH
• PMC

• Principal Investigator
  • Louis R. Pasquale, MD. Harvard Medical School, Boston, MA, USA.
• Co-Principal Investigator
  • Janey L. Wiggs, MD, PhD. Harvard Medical School, Boston, MA, USA.
• Co-Investigator
  • Jae Hee Kang, ScD. Harvard Medical School, Boston, MA, USA.
• Collaborators
  • Teresa Chen, MD. Harvard Medical School, Boston, MA, USA.
  • Elizabeth DelBono, MPH. Harvard Medical School, Boston, MA, USA.
  • Jonathan L. Haines, PhD. Vanderbilt University, Nashville, TN, USA.
  • David Hunter, MBBS, DrPH. Harvard Medical School, Boston, MA, USA.
  • Peter Kraft, PhD. Harvard Medical School, Boston, MA, USA.
  • Stephanie Loomis, MPH. Harvard Medical School, Boston, MA, USA.
  • Lana M. Olson, MS. Vanderbilt University, Nashville, TN, USA.
  • Douglas Rhee, MD. Harvard Medical School, Boston, MA, USA.
  • Brian Yaspan, MPH, PhD. Vanderbilt University, Nashville, TN, USA.
• Funding Sources
  • R01 EY015473. National Eye Institute, National Institutes of Health, Bethesda, MD, USA.
  • R01 EY015872. National Eye Institute, National Institutes of Health, Bethesda, MD, USA.
  • RO1 EY09611. National Eye Institute, National Institutes of Health, Bethesda, MD, USA.
  • P01 CA87969. National Cancer Institute, National Institutes of Health, Bethesda, MD, USA.
  • U01 HG004728. National Human Genome Research Institute, National Institutes of Health, Bethesda, MD USA.
  • U54 RR020278. National Center for Research Resources, National Institutes of Health, Bethesda, MD USA.