Can genetic therapy solve the AMD puzzle?

How advances in pharmacogenetics may help to diagnose and manage the disease.

By Karen Appold, Contributing Editor

Pharmacogenetics holds great promise for unlocking some mysteries, but ophthalmologists shouldn’t let themselves become too optimistic: Experts doubt that it will result in a cure of AMD in the next 10 to 20 years.

“Pharmacogenetics will most likely improve our understanding of how patients with AMD respond to various medications, including both existing therapies and hopefully newer treatments that aren’t yet available,” says Stephen G. Schwartz, MD, MBA, associate professor at Bascom Palmer Eye Institute, University of Miami. “But because pharmacogenetics seeks to understand inter-patient differences in treatment outcomes, both in terms of efficacy and toxicity, it is unlikely that this type of research will identify a single cure that works for all AMD patients.”

Pharmacogenetics may help in identifying subgroups that respond more effectively to one drug vs. others, Dr. Schwartz says. Here’s a look at what pharmacogentics may teach clinicians about the pathophysiology of AMD.

TODAY’S KNOWLEDGE BASE

Sometimes conflicting data

Multiple smaller clinical series have reported statistically significant associations between various risk alleles (particularly complement factor H [CFH] and age-related maculopathy susceptibility 2 [ARMS2]) and certain clinical outcomes using photodynamic therapy and anti-VEGF agents.

“However,” Dr. Schwartz says, “two large prospective clinical trials — Comparison of AMD Treatments Trials (CATT) and a randomized controlled trial of alternative treatments to Inhibit VEGF in Age-Related Choroidal Neovascularisation (IVAN) — reported no statistically significant pharmacogenetic correlations with anti-VEGF drugs.”

He adds, “Interestingly, there may be stronger evidence for a link between genotype and outcomes with the Age-Related Eye Diseases Study (AREDS) on vitamin supplementation, but this relationship is still being explored.”

Testing costs decrease

Because a cure for cancer is still elusive after 50 years of intensive research, Carl C. Awh, MD, doubts a cure for AMD will emerge any time soon — especially in light of the complexity of the disease.

However, he does believe pharmacogenetics will lead to more effective treatments for AMD, both preventative and therapeutic. “The cost of genetic testing has dramatically decreased, making it feasible to integrate it into every major clinical trial,” says Dr. Awh, president of Tennessee Retina in Nashville. “This will rapidly increase our understanding of how different genetic make-ups correlate with differing responses to therapy.”

Where current investigations focus

Given the cost and logistical burden of repeated anti-VEGF injections, clinicians have expressed great interest in the potential of viral vectors to introduce genes that produce an anti-VEGF protein. “This is exciting work, but we are a long way from having this available for patients with AMD,” Dr. Awh says.

Lampalizumab targets gene implicated in GA

Lampalizumab is an antigen-binding fragment, known as Fab, of a humanized, monoclonal antibody directed against complement factor D, a rate-limiting enzyme involved in the activation of the alternative complement pathway (ACP). Investigators have implicated genetic polymorphisms as well as ACP hyperactivity in the development of AMD, including geographic atrophy.

Round or oval regions of hypopigmentation characterize geographic atrophy.

Switzerland-based Roche has been developing lampalizumab and is sponsoring the MAHALO phase 2 study in patients with GA.

Early phase 2 results presented at the Retina Subspecialty Day session of the American Academy of Ophthalmology last year reported on a sub-population of GA patients positive for the CFI biomarker who received lampalizumab monthly.¹ An exploratory analysis showed a 44% decrease in the rate of disease progression at 18 months.

In every-other-month administration of lampalizumab, the rate of disease progression decreased 18% in the biomarker defined sub-group of patients. The MAHALO study detected no unexpected or unmanageable adverse events.

CFI biomarker implicated

In the MAHALO study, 57% of genotype samples collected from 93 patients were positive for the CFI biomarker. Although the phase 2 study was relatively small, the results have suggested the CFI biomarker is both prognostic for GA area progression and predictive for lampalizumab treatment response. Only MAHALO patients positive for the CFI biomarker showed a treatment effect with lampalizumab.

The phase 2 trial was a multi-center, randomized, single-masked, controlled study of the safety, tolerability and evidence of activity of lampalizumab in patients with GA associated with AMD. Study participants received lampalizumab injections in one eye either monthly or every other month for 18 months. The primary endpoint was change of GA area from baseline to month 18 compared with control, as assessed with fundus autofluorescence (FAF). Four genetic biomarkers were examined in MAHALO: complement factor H (CFH), C3, C2/CFB and CFI. In the MAHALO study, most patients had a background of CFH and C2/CFB.

Lampalizumab showed a 20.4% reduction rate in the area of geographic atrophy at 18 months that was statistically significant per pre-specified protocol criteria in patients with this advanced form of dry AMD. The efficacy assessed by FAF was observed in those receiving monthly injections beginning at month six and maintained through month 18. Color fundus imaging was used to assess a secondary endpoint of change in GA area from baseline to month 18, these results of which confirmed the FAF primary endpoint outcome.

GA progression rate reduced

In a sub-population of GA patients treated monthly with lampalizumab that were positive for the CFI exploratory biomarker, the GA progression rate decreased 44% at 18 months. The study authors reported adverse event rates of intraocular inflammation rates and IOP elevation were consistent with ranibizumab (Lucentis, Genentech, South San Francisco, Calif.) rates in wet AMD.

The most frequently reported adverse events in the trial were associated with the injection procedure. The trial did not report any intraocular infections, unexpected or unmanageable serious AEs, death or serious ocular AEs thought to be associated with the drug.

The recent phase 2 lampalizumab study¹ has suggested that this intended treatment for geographic atrophy may be most effective in patients with the risk form of complement factor I (CFI). “One day we will classify AMD using genetic markers as routinely as we do with slit lamp examination, optical coherence tomography and photography,” Dr. Awh says.

Dr. Awh and colleagues recently reported an association between CFH and ARMS2 genotypes and differential responses to zinc and antioxidants, components of the AREDS formulation. This suggested that genotype-directed nutritional therapy could significantly decrease the number of patients who progress to advanced AMD.

“The public health benefits and cost savings associated with even a modest reduction in the incidence of advanced AMD could be substantial,” Dr. Awh says.

Dr. Schwartz and his collaborators are also focused on exploring genetic associations with AMD progression, as well as pharmacogenetic associations with anti-VEGF treatments.

RECENT RESEARCH ADVANCES

Gene-environment interactions

No doubt the knowledge acquired through the Human Genome Project and gene discovery work has opened up new approaches to therapy, even though a full understanding of the pathophysiology of AMD has still not been realized, says David J. Wilson, MD, director of the Casey Eye Institute at Oregon Health and Sciences University in Portland.

However, recent AMD-related gene associations are shedding light on the pathophysiology of AMD. Also, the progress in the use of gene therapy for other diseases, such as Leber congenital amaurosis and Stargardt disease, has driven the development of vectors and devices that can be used in AMD.

Two companies, Oxford Biomedica in the UK and Neurotech Pharmaceuticals in Cumberland, R.I., have ongoing clinical trials that employ gene therapy in treatment of choroidal neovascularization in AMD.

“We are increasing our understanding of gene-environment interactions, such as gene-smoking interactions,” Dr. Schwartz says. “This is important for complex genetic diseases such as AMD, in which both genetic and environmental factors are thought to contribute to disease pathogenesis.”

Seeking to identify rare variants

Researchers have described new genetic risk alleles associated with AMD. Because all of the common genetic variants have been discovered, increasingly large numbers of patients are needed to identify rare variants, such as the C3 variant that several groups of researchers reported in 2013.

“Understanding the role of rare variant genes may speed the development of new therapies,” Dr. Awh says. “The multiple risk alleles identified in the complement system suggest that drugs that can selectively suppress complement-mediated inflammation may benefit patients with AMD.”

The promising initial results of gene therapy for choroideremia — recently reported in The Lancet² — in which investigators used an adenovirus vector to introduce the missing CHM gene into the retina, echoes the ground-breaking work reported in 2008 for patients with Leber congenital amaurosis.

“Similar therapies could prove useful for patients with AMD,” Dr. Awh says.

Dr. Awh also believes the eyeGENE project, which is integrating clinical and genetic research information to better understand the pathogenesis of inherited eye disease, will lead to a better understanding of the genetics of AMD.

HOW THERAPIES WORK

No single gene to target

As a complex genetic disease, AMD probably arises from an unknown combination of one or more genetic predispositions and one or more environmental triggers. “This is not the same as a monogenic disease, also known as a Mendelian or single-gene disease, in which a mutation in one gene causes the disease,” Dr. Schwartz says. “In general, monogenic diseases such as Leber congenital amaurosis are more potentially amenable to genetic therapies than are complex genetic diseases like AMD.”

Nevertheless, if research determined a single gene defect is responsible for AMD in a subset of patients, the theory follows that using gene therapy to deliver the missing gene product may be beneficial in this subset, he continues. However, he adds, this is highly theoretical.

Stem-cell and small-molecule approaches

Clinical research has also focused on stem-cell and small-molecule therapies. Stem cell therapy aims to replace the damaged or missing cell population in hopes of restoring function, according to Dr. Wilson. “This type of therapy, if proven successful, is hoped to be applied to patients with the dry form of macular degeneration in whom areas of the retinal pigment epithelium are absent,” he says. Advanced Cell Technology, a company based in Marlborough, Mass., is currently evaluating stem-cell therapy in a clinical trial.

While Dr. Awh finds the idea of injecting stem cells or new genes into eyes with AMD compelling, genetic therapies with the greatest potential will involve treatments selected for patients on the basis of their genetic make-up. “This will dramatically alter treatments in all areas of medicine,” he says. “AMD is the most genetically influenced of all human multigenic disorders.”

Small-molecule therapy has the advantage of being the simplest approach. Multiple agents are undergoing investigation for efficacy in both dry and wet AMD — with many holding promise for AMD patients. “Of course, these forms of therapy are not mutually exclusive,” Dr. Wilson says. “One can easily envision the use of combinations.”

Targeting loci

Still other treatment options may involve loci. Nineteen loci have been statistically associated with AMD, but this does not necessarily mean these gene products are directly involved in disease pathogenesis. In fact, environmental factors, such as smoking, exogenous estrogens, nutrient intake, and others may very well modify the effects of these genes.

“Targeting these loci may or may not be beneficial for individual patients,” Dr. Schwartz explains. The first step is to study the relevant loci and to determine whether they are disease-causing, or at least disease-influencing. However, this is not a straightforward proposition.

For example, one of the major AMD risk alleles, ARMS2, codes for a gene product with an unknown function. ARMS2 is in strong linkage disequilibrium with another gene, HTRA1. Their effects are indistinguishable statistically, so investigators cannot even parse out whether ARMS2 or HTRA1 is more relevant for AMD patients.

The single nucleotide polymorphisms (SNPs) associated with AMD do not necessarily have a direct impact on disease development, Dr. Awh points out. Rather, they serve as markers for adjacent DNA, or even for relatively distant areas of DNA that are inherited in a linked manner.

“We don’t have a complete understanding of the causative genes, but there is certainly powerful evidence pointing to the role of the complement system in the pathogenesis of AMD,” he says.

A controversial proposition

Today, genetic testing is useful as a research tool in AMD, but the clinical utility of such testing remains controversial. “As clinicians, we should be very careful about offering genetic testing to our patients with AMD, because there is no consensus about what we should do with this information,” Dr. Schwartz says.

The American Academy of Ophthalmology has recommended avoiding genetic testing for AMD and complex diseases generally until more published clinical trials demonstrate a benefit for this approach, he notes. “In my opinion, the currently available peer-reviewed evidence is insufficient for this purpose,” Dr. Schwartz says. “Hopefully, as we continue to collect results from new clinical trials, the role of genetic testing for AMD patients will become clearer, both in terms of diagnosis and treatment.”

Dr. Awh offers a different viewpoint. He notes that the recent 3 Continents AMD Consortium³ reported a substantial (35%) improvement in the ability to predict advanced AMD if genetic risk information is added to a model including AMD status and non-genetic factors.

“The knowledge that an individual patient is at substantially greater risk of AMD progression is clinically useful and can lead to changes in patient behavior and patient management that result in earlier detection of advanced AMD,” he says. “The accuracy of predictive genetic tests for AMD is uncontested. That there is no consensus on what to do with the information is typical of many new technologies, but it should not prevent us from utilizing these valuable tools.” OM

REFERENCES