Pharmacodynamics Of Visual Cycle Modulation In the Treatment of GA

Emerging data from clinical studies demonstrate the biological activity of emixustat in the retina.


Pharmacodynamics Of Visual Cycle Modulation In the Treatment of GA

Emerging data from clinical studies demonstrate the biological activity of emixustat in the retina.


Following the recent successes of new therapies for exudative age-related macular degeneration, a significant challenge in AMD still remains: therapeutic development to address geographic atrophy (Figure 1).


Figure 1. Geographic atrophy is characterized by round or oval regions of hypopigmentation within which choroidal vessels are more visible than adjacent regions.


Complicating this effort is our limited understanding of the etiology and pathophysiological mechanisms underlying the development of GA lesions in AMD patients. The complexity of the photoreceptor/retinal pigment epithelium/choroid system and the strong interdependence of these tissues make it exceedingly difficult to isolate a single causative factor that triggers the disease process.

Nevertheless, general agreement exists among researchers and clinicians that RPE dysfunction is an early component of the pathogenesis and that an inherited susceptibility and/or inflammatory event likely precipitates it.1,2

Fortunately, scientists have identified the key pathways involved in the development of GA in AMD and are currently investigating pharmacologic targeting of various components of these pathways.

Nathan L. Mata, PhD, is the principal clinical research scientist at Acucela, Inc., Seattle. Ryo Kubota, MD, PhD, is chairman, president, and CEO of Acucela. Susan Schneider, MD, is vice president at Acucela. Jennifer Kissner, PhD is a director at Acucela. David G. Birch, PhD, is the chief scientific and executive officer and director of the Rose-Silverthorne Retinal Degeneration Laboratory at Retina Foundation of the Southwest in Dallas. Pravin U. Dugel, MD, is managing partner of Retinal Consultants of Arizona in Phoenix and a founding member of Spectra Eye Institute in Sun City, AZ. Drs. Mata, Kubota, Schneider, and Kissner report significant financial interest, and Drs. Birch and Dugel report minimal financial interest in Acucela. Dr. Dugel’s e-mail is


A large, well-controlled, interventional phase 2b/3 clinical trial (the SEATTLE study, NCT01802866) is currently recruiting to evaluate the safety and efficacy of an oral small molecule, emixustat hydrochloride, in GA.

Emixustat modulates activity of the visual cycle by inhibiting the retinoid isomerase, a retinal pigment epithelium–specific 65-kDa protein (RPE65), and it is the first compound with this mechanism of action under evaluation for the treatment of GA.

The RPE65 protein produces a light-sensitive retinoid (11-cis retinol) from stores of dietary vitamin A (all-trans retinyl esters) in the RPE.

Because cis retinoids are needed for the production and regeneration of rhodopsin, inhibition of 11-cis retinol biosynthesis with emixustat is expected to reduce rhodopsin levels in the retina.

Acucela (Seattle, WA) designed emixustat specifically to inhibit the catalytic activity of RPE65, an activity that is unique to retinal tissue. With the exception of one report, which described the identification of RPE65 in human keratinocytes,3 expression of the RPE65 protein is restricted to retinal tissue. As a result, off-target effects of emixustat should be minimal.

Data from a recently completed phase 2a clinical study evaluating the safety and tolerability of emixustat in GA subjects (NCT01002950) appear to confirm this expectation, because no significant systemic adverse events occurred in subjects treated with emixustat.4


The aberrant accumulation of lipofuscin in the RPE is an important pathological characteristic of GA.5 A major molecular component of lipofuscin is N-retinylidene-N-retinylethanolamine (A2E), a cytotoxin generated from retinoid by-products of the visual cycle.

The numerous cytotoxicities associated with A2E have led to the hypothesis that A2E contributes to lesion progression in GA.6 Importantly, reduction of rhodopsin levels, through systemic retinol reduction7 or by inhibition of key visual cycle enzymes,8,9 has led to significant reductions of A2E in animal models.

Rhodopsin Biosynthesis

These findings indicate that modulation of rhodopsin biosynthesis would be a tractable approach for determining whether A2E plays a role in the progression of GA lesions.

Pharmacologic modulation of rhodopsin biosynthesis may also prove an effective method to reduce light-induced retinal pathology. We have known for many years that light accelerates retinal degeneration and that rhodopsin is the molecular switch that triggers phototoxic mechanisms in photoreceptor cells (reviewed by Organisciak and Vaughan10).

The rate of rhodopsin regeneration, which requires the transport of vitamin A compounds to and from the RPE, sets the photon-catch capacity of the retina and is a major determinant in susceptibility to light damage. Protection from light damage in rodents deficient in either rhodopsin 11,12 or RPE65 protein,13 and observations that regional increases in rhodopsin within the retina correlate with areas of greater photoreceptor cell damage14,15 attest to the importance of rhodopsin as a primary mediator of light-induced cell death in the retina.

These various lines of evidence suggest that limiting rhodopsin levels through visual cycle modulation may be a useful therapeutic strategy for retinal and macular degenerations such as GA.


A key feature of visual cycle modulation is the ability to monitor drug effects in the retina noninvasively, using electroretinography (ERG). The electrophysiological activity of rod photoreceptors depends upon rhodopsin. As a result, a proportional relationship exists between rod photoreceptor-derived ERG signals and rhodopsin levels.16

Photobleaching of rhodopsin elicits distinct ERG waveforms, a- and b-waves, which describe the integrity of signal processing from photoreceptors and bipolar cells, respectively (Figure 2).


Figure 2. Components of the dark-adapted ERG. An illustration of the retina (left) and a representative ERG tracing (right) are shown. In the dark-adapted retina, a light stimulus (1) elicits a presynaptic response from photoreceptor cells, represented by the downward-deflecting a-wave (2). The subsequent postsynaptic response, mediated largely by bipolar cells, produces the b-wave (3). The a-wave amplitude (measured from the baseline to the trough of the a-wave) depends on the intensity of the light stimulus and the integrity of the photoreceptors. The b-wave amplitude (measured from the trough of the a-wave to the peak of the b-wave) depends on the a-wave and the integrity of signal transmission within the retina.

Following exposure to a bright bleaching light in the ERG dome, ERG a-wave and b-wave amplitudes decrease severely, but they recover over ~30 minutes. The degree of recovery reflects rhodopsin levels and/or regeneration rate. A visual cycle modulator such as emixustat limits the rate of rhodopsin regeneration, resulting in suppressed ERG amplitudes following the photobleach.

Evidence From Clinical Trials

In the phase 2a study of emixustat in GA subjects, the investigators implemented a dose-ranging approach with the intent of obtaining a wide range of suppression of rod photoreceptor amplitudes after the photobleach.

The standardized ERG protocol for the measurement of a pure rod response requires a light intensity that produces a prominent b-wave but very little a-wave.17 As a result, investigators routinely utilize measurement of the rod b-wave amplitude as a measurement of rod photoreceptor activity; the emixustat phase 2a study used it as the index for drug effect on rhodopsin regeneration.

Measurements taken following 14 days of treatment (during the ERG steady-state period) showed a linear, dose-dependent suppression of the recovery of rod b-wave amplitude, with ~35% and ~90% reductions at the 2-mg and 10-mg doses, respectively (Figure 3A). This drug effect was reversible following one to two weeks of drug cessation (Figure 3B).


Figure 3. Dose-dependent reduction of rod photoreceptor b-wave amplitude and reversibility with emixustat. Electroretinographic data are shown from GA subjects participating in the phase 2a study of emixustat. Measurements of rod photoreceptor b-wave amplitudes (30 minutes postbleach) were obtained at baseline, on day 14 of treatment, and at study exit (one to two weeks following drug cessation). For each subject, measurements from right and left eyes were averaged and are shown as a single data point (symbol) in each panel. The numbers of subjects in each treatment arm are as follows: placebo = 17; 2 mg = 12; 5 mg = 11; 7 mg = 10; 10 mg = 5. In panel A, the data show the percentage change of rod photoreceptor activity on day 14, relative to baseline, for subjects in each treatment group. The mean percentage change of b-wave amplitude (± SD) from baseline to day 14 for the treatment groups are as follows: placebo, -2.35 ± 35.22; 2 mg, -35.42 ± 42.12; 5 mg, -53.98 ± 40.26; 7 mg, -63.41 ± 36.53; and 10 mg, -85-53 ± 14.45. A similar analysis, performed at study exit, showed a return of rod b-wave amplitudes toward baseline values (panel B). The mean percentage change of b-wave amplitude (± SD) from baseline to study exit for the treatment groups are as follows: placebo, 5.99 ± 35.39; 2 mg, 16.96 ± 50.69; 5 mg, -0.55 ± 46.13; 7 mg, 9.46 ± 44.85; and 10 mg, 43.79 ± 112.30. The solid line in each panel is a linear regression fit of all of the data points in the panel.

The relatively high variability in rod suppression among GA subjects, which we can attribute to varied degrees of retinal pathology and the test-retest variability of ERG measurements, precluded the demonstration of a statistically significant correlation between rod suppression and emixustat dose. Nevertheless, the ability to vary the degree of suppression of rod function with emixustat is ideal for evaluating the potential therapeutic benefit of visual cycle modulation in GA patients.

These ERG data helped to guide dose selection in the current phase 2b/3 GA clinical study, and they constitute a benefit of visual cycle modulation not available with other therapeutic approaches in GA to date.


When signal transmission from photoreceptors to bipolar cells fails, or deficits occur in signaling within bipolar cells, the ERG a-wave is present while the b-wave is absent or significantly reduced.

This “negative” ERG phenotype has occurred in patients affected with congenital stationary night blindness and X-linked retinoschisis.18 Alternatively, a proportional relationship between the a- and b-wave across a range of amplitudes implies normal signal processing within the retina.19 In the phase 2a study of emixustat discussed above, the investigators interpreted the reduction of the rod b-wave as the drug’s effect on rhodopsin regeneration in rod photoreceptors.

This approach is common because accurate measurement of the a-wave following a photobleach requires high-intensity stimuli, which could interfere with the time course of dark adaptation, particularly in GA patients.

Subanalysis Results

To confirm that emixustat actually modulates photoreceptor activity and that no additional modulation occurs at the postsynaptic (bipolar cell) level, analysis of prebleach ERG amplitudes was performed, with investigators measuring a- and b-wave amplitudes from fully dark-adapted GA subjects immediately following a brief flash of dim white light.

The analyzed ERG response is a mixed rod-cone response but mainly reflects the activity of rod photoreceptors because only a minor contribution comes from cone photoreceptors. The analysis included measurements from GA subjects in the 5-mg dose group (26 eyes), taken at baseline and on day 14 during treatment.

A scatter plot with linear regression of the a-wave amplitude as a function of the b-wave amplitude for all of the eyes showed no change in slope values during emixustat treatment (Figure 4).


Figure 4. Relationship between a-wave and b-wave amplitudes during emixustat treatment. Prebleach ERG measurements obtained from GA subjects participating in the phase 2a study of emixustat (5-mg dose group) are shown. Following a period of dark adaptation, the subjects were exposed to a brief dim flash of white light, and the resulting a- and b-wave amplitudes were recorded. Measurements were obtained for each subject at baseline (open symbols) and on day 14 (filled symbols) of treatment (n = 26 eyes at each time point). Linear regression lines for values at baseline (solid line) and day 14 (dashed line) were fit for the data at each time point.

The downward shift in the slope for eyes at day 14 shows that emixustat reduces the a-wave and b-wave amplitudes in a proportional manner. This finding suggests that no further postsynaptic modulation occurs.

In the phase 2a clinical study, the majority of ocular adverse events reported, such as delayed dark adaptation, were attributable to the mechanism of emixustat action on the target (RPE65) and the ensuing reduction in rhodopsin levels.

These subjective reports are consistent with the ERG data, which indicate a specific effect of emixustat on rod photoreceptors with no collateral effect that would compromise the functional integrity of the retina.


Visual cycle modulation with emixustat represents a novel therapeutic approach for the potential treatment of GA due to AMD. The ability to use noninvasive ERG techniques to monitor pharmacodynamic effects of emixustat could represent a significant advantage of this therapeutic approach.

The measurement of ERG parameters in GA subjects treated with emixustat has revealed a dose-dependent response, which is consistent with the proposed mechanism of action.

This key feature of the therapeutic strategy with emixustat has been utilized to guide dose selection during clinical development and may be useful to monitor response to treatment. RP


1. Donoso LA, Kim D, Frost A, et al. The role of Inflammation In the pathogenesis of age-related macular degeneration. Surv Ophthalmol. 2006;51:137-152.

2. Kanda A, Abecasis G, Swaroop A. Inflammation in the pathogenesis of age-related macular degeneration. Br J Ophthalmol. 2008;92:448-450.

3. Hinterhuber G, Cauza K, Brugger K, et al. RPE65 of retinal pigment epithelium, a putative receptor molecule for plasma retinol-binding protein, is expressed in human keratinocytes. J Invest Dermatol. 2004;122:406-413.

4. Dugel PU, Novack RL, Csaky KG, et al. A Phase 2 Double-masked, placebo-controlled, dose ranging study of emixustat hydrochloride (ACU-4429) in subjects with GA associated with dry AMD. Invest Ophthalmol Vis Sci. 2013;54:ARVO E-Abstract 4506.

5. Choudhry N, Giani A, Miller JW. Fundus autofluorescence in geographic atrophy: a review. Semin Ophthalmol. 2010;25:206-213.

6. Nowak JZ. Age-related macular degeneration (AMD): pathogenesis and therapy. Pharmacol Rep. 2006;58:353-363.

7. Radu RA, Han Y, Bui TV, et al. Reductions in serum vitamin A arrest accumulation of toxic retinal fluorophores: a potential therapy for treatment of lipofuscin-based retinal diseases. Invest Ophthalmol Vis Sci. 2005;46:4393-4401.

8. Radu RA, Mata NL, Nusinowitz S, et al. Treatment with isotretinoin inhibits lipofuscin accumulation in a mouse model of recessive Stargardt’s macular degeneration. Proc Natl Acad Sci U S A 2003;100:4742-4727.

9. Maeda A, Maeda T, Golczak M, et al. Effects of potent inhibitors of the retinoid cycle on visual function and photoreceptor protection from light damage in mice. Mol Pharmacol. 2006;70:1220-1229.

10. Organisciak DT, Vaughan DK. Retinal light damage: mechanisms and protection. Prog Retin Eye Res. 2010;29:113-134.

11. Humphries MM, Rancourt D, Farrar GJ, et al. Retinopathy induced in mice by targeted disruption of the rhodopsin gene. Nat Genet. 1997;15:216-219.

12. Grimm C, Wenzel A, Williams T, et al. Rhodopsin-mediated blue-light damage to the rat retina: effect of photoreversal of bleaching. Invest Ophthalmol Vis Sci. 2001;42:497-505.

13. Grimm C, Wenzel A, Hafezi F, et al. Protection of Rpe65-deficient mice identifies rhodopsin as a mediator of light-induced retinal degeneration. Nat Genet. 2000;25:63-66.

14. Rapp LM, Naash MI, Wiegand RD, et al. Morphological and biochemical comparisons between retinal regions having differing susceptibility to photoreceptor degeneration. In: La Vail MM, ed. Retinal Degeneration: Experimental and Clinical Studies. New York, NY; Alan R Liss; 1985:421-437.

15. Rapp LM, Williams TP. A parametric study of retinal light damage in albino and pigmented rats. In: Williams TP, Baker BN., eds. The Effects of Constant Light on Visual Processes. New York, NY; Plenum Press; 1980:133-159.

16. Bonting SL, Caravaggio LL, Gouras P. The rhodopsin cycle in the developing vertebrate retina. I. Relation of rhodopsin content, electroretinogram and rod structure in the rat. Exp Eye Res. 1961;1:14-24.

17. Perlman I. Relationship between the amplitudes of the b wave and the a wave as a useful index for evaluating the electroretinogram. Br J Ophthalmol. 1983;67:443-448.

18. Raghuram A, Hansen RM, Moskowitz A, et al. Photoreceptor and postreceptor responses in congenital stationary night blindness. Invest Ophthalmol Vis Sci. 2013;54:4648-4658.

19. Brown KT, Wiesel TN. Localization of origins of electroretinogram components by intraretinal recording in the intact cat eye. J Physiol. 1961;158:257-280.