The Role of Estrogen in Glaucoma

Louis R. Pasquale, Boston, MA, USA

Primary open-angle glaucoma (POAG) is a strongly age-related optic nerve degeneration and declining sex hormones represent an important aspect of the aging process. At first pass, it may seem that gender biology plays little to no role in POAG pathogenesis. By the eighth decade, sex differences in POAG disease burden are not apparent¹ and the widely-held clinical impression that the normal tension variant of POAG is a female-predominant disorder has not been confirmed by population based studies. In fact, among the elderly, more women suffer visual loss from open-angle glaucoma (OAG), perhaps merely because they live longer than men.² Nonetheless, this does not mean that estrogen levels are unimportant in POAG pathogenesis. In fact, a sex-stratified incidence study indicated that in the fourth through seventh decade, there is a male-predominance for this condition, suggesting that estrogen may protect women from developing POAG during this time period.¹ Important traits related to the POAG phenotype, such as the intraocular pressure (IOP) level, optic nerve structure and visual field performance, are also influenced by estrogen levels. Furthermore, there are several reproductive attributes, genetic biomarkers and gene-environment interactions related to estrogen levels that are associated with POAG. Finally, the development of optic nerve degeneration in animal models of glaucoma is altered when estrogen levels are experimentally manipulated.

Intraocular pressure
IOP decreases during pregnancy, even though there is a concomitant increase in central corneal thickness3 and this effect is mediated by an increase in outflow facility.⁴ There is a report that the menopause transition is associated with a 2-3 mmHg IOP increase⁵ although more careful longitudinal study adjusting for age, body mass index and blood pressure are needed. A prospective randomized trial involving postmenopausal women demonstrated a 1-2 mmHg IOP reduction after a three-month course with transdermal 17beta-estradiol (E2) and medroxyprogesterone acetate for 12 days per menstrual cycle.⁶

Optic nerve structure and function
Retinal ganglion cells (RGCs) express beta-type estrogen receptors in both men and women.⁷ The exact role of these receptors is not known. During the luteal phase of the menstrual cycle, when serum estradiol levels are their nadir, estimates of neuroretinal rim area are lowest among healthy women.⁸ These data suggest that there are subtle but definitive dynamic optic nerve changes during the menstrual cycle. Furthermore, compared to the luteal phase, mean retinal sensitivity on short wavelength automated perimetry (SWAP) was higher during the follicular phase of the menstrual cycle when serum estradiol levels are higher.⁹ It is unclear if these differences reflect real changes in optic nerve or retinal function or whether they are related to non-ocular factors. Nonetheless, among women with migraine, there are regional retinal sensitivity differences on SWAP during different phases of the menstrual cycle. Specifically, normal women with migraine had lower nasal retinal sensitivity on SWAP during the luteal versus the follicular phase of the menstrual cycle.⁹

Attributes of female reproductive health and primary open-angle glaucoma
Several aspects of female reproductive health are related to POAG. Five years of oral contraceptive use (which interferes with the normal physiological ebb and flow of estrogen) was associated with a 25% increased risk of incident POAG.¹⁰ Bilateral oophorectomy before the age of 43 was associated with a 60% increased risk of glaucoma.¹¹ Early age of menopause before age 45 years was associated with a 2.6-fold increased risk of OAG in the Rotterdam study.¹² In a prospective cohort study representative of the United States, among nurses aged 65 and older, a later age of menopause (≥ 54 years of age) was associated with a 47% reduced risk of POAG.¹³ Stein et al. found that estrogen-only use was associated with 18% reduced risk of OAG after four years of follow-up in a large United States healthcare network.¹⁴

Sex-specific genetic loci associated with POAG
Nitric oxide synthase 3 (NOS3) - which is expressed in the optic nerve, the outflow pathway and in blood vessels throughout the body - is an enzyme that is strongly influenced by estrogen levels. Interestingly, common polymorphisms in the NOS3 gene are more strongly associated with POAG in women than in men.^15,16 Furthermore, the relation between NOS3 single nucleotide polymorphisms (SNPs) and POAG is modified by postmenopausal hormone (PMH) use. For example, the inverse relation between PMH use and POAG was enhanced by an intron 13 NOS3 SNP.¹⁶ Interestingly, NOS3 is biophysically juxtaposed with caveolin in biological membranes. Polymorphisms in the genomic region between CAV1 and CAV2 are associated with POAG and this association is also stronger in women than in men.¹⁷ These findings suggest that nitric oxide signaling is important in POAG overall and that women may be particularly vulnerable when this biochemical process is impaired. The notion that nitric oxide signaling is important in POAG is highlighted by a murine model¹⁸ where the alpha subunit of soluble guanylate cyclase (which serves as the intracellular receptor for nitric oxide) is knocked out. These mice exhibited modest age related elevation of IOP and optic nerve degeneration. Furthermore, the genomic region between the alpha and beta subunit of soluble guanylate cyclase (the GUCY1A3/GUCY1B3 intergenic region), harbors a SNP (rs11722059) that is associated with the POAG endophenotype categorized by early paracentral visual field loss among women only. Researchers in the NEIGHBORHOOD (the National Eye Institute Glaucoma Human Genetics Collaboration Overall Operational Database) consortium assessed whether the collective effect of polymorphisms related to estrogen metabolism are associated with POAG. Using a collection of 3108 cases and 3430 controls, a panel of polymorphisms related to estrogen metabolism was associated with POAG in women but not in men.¹⁹ Interestingly, polymorphisms in the 17-beta hydroxysteroid dehydrogenase 1 and catechol-O-methyltransferase (COMT) genes were important loci in this panel that drove the association with POAG among women.

Animal models supporting a role for estrogen in optic nerve degeneration
There is robust evidence that estrogen is protective in various animal model systems of optic nerve injury. Topically applied E2 protects against optic nerve injury from elevated IOP induced by hypertonic saline injection into the episcleral vasculature.²⁰ In the DBA2J mouse model where there are spontaneous anterior segment changes that result in angle closure and elevated IOP, E2 ameliorates optic nerve damage.²¹ E2 also protects against retinal ischemia-induced by acute IOP elevation that cuts off the blood supply to the posterior segment,²² optic nerve axotomy²³ and glutamate-induced RGC toxicity.²⁴

Conclusions
The existing evidence is compelling that deleterious optic nerve aging in women is related to declining estrogen levels which occurs via a complex array of environmental and genetic influences. While this subject requires further study, it is reasonable to ask how this body of work can be leveraged into novel interventions to treat POAG but broaching the subject raises more questions than answers. Can topical estrogen delivery be tolerated and can appropriate vitreous levels of estrogen that benefit the optic nerve be achieved in POAG patients? Could people of both sexes benefit from such treatment? When in the course of disease should such therapy be introduced? Are there other drug targets within the estrogen pathway that could be used to favorably alter the course of POAG? What is the role of the other sex steroids (specifically progesterone and testosterone) in POAG pathogenesis? Widespread PMH use probably should not be advocated given real concerns regarding increased risk of breast cancer and other gynecological malignancies. Furthermore, one cannot dismiss the possibility that local ocular estrogen therapy could also produce these side effects. Once an ideal target and intervention related to estrogen levels can be determined, a neuroprotective trial designed along the lines of the Low-Pressure Glaucoma Treatment Study (LoPGTS) could be useful. A study that includes patients who develop OAG on the basis of low estrogen levels (such as women with recurrent migraines, early oophorectomy and frequent disc hemorrhages) along with an intervention with promising therapeutic effect may not require a large sample size or prolonged serial assessments of optic nerve structural / function to realize a therapeutic effect. A judicious strategy to correct ocular estrogen levels could result in reduced vision loss from OAG.