New frontiers in retinal medicine

Over the past decade retinal medicine has entered a new era in which neovascular age-related macular degeneration (AMD) – a previously untreatable chronic and progressive condition – cannot only be stabilised but can even be reversed. However, with this potential for improved vision comes a small but not insignificant risk of total vision loss in the treated eye due to endophthalmitis.

[caption id='attachment_583' align='alignright' width='400' caption='A retinal oximetry image showing a mild central retinal vein occlusion. The pseudocolour scale on the right side shows the oxygen saturation percentage. The arterioles are red indicating near full oxygen saturation of hemoglobin. The venules indicate low oxygen saturation, ie, hypoxia'][/caption]

As a procedure, intravitreal injections appear to be about as safe as a cataract extraction procedure in terms of the endophthalmitis rate. In prospective clinical trials the incidence of the complication per injection has ranged from 0.05 per cent, in the Marina trial, to 0.16 per cent, in the Vision trial. However, it is because patients must receive repeated injections, generally about one a month, the risk per patient is generally much higher, ranging from 0.7 per cent, in the ANCHOR study, to 2.7 per cent in the FOCUS study.

“It is important to realise that you can count the risk either per injection (typically around one per 1,000 as in cataract surgery), or per patient, in which case you would multiply the risk because the treatment is repeated for 24 times or more,†said Jan van Meurs MD, The Rotterdam Eye Hospital and Erasmus University, Rotterdam, The Netherlands.

On the other hand the risks involved in not treating patients are also very considerable, noted EURETINA president, Gisbert Richard MD, University Medical Centre, Hamburg, Germany. The principal diseases treated by anti-VEGF injections – the wet form of AMD, diabetic macular oedema and vein occlusion – are also the main reasons for legal blindness in developed countries, he told EuroTimes.
He noted that in a recent meta-analysis published in the journal Retina, the overall rate of endophthalmitis following intravitreal injections of ranibizumab and bevacizumab was 0.05 per cent per injection or less, and the rate of other complications like retinal detachment was even lower (0.01 per cent).

“Regarding the great efficacy of the intravitreal injection of anti-VEGF, the relevance of local and systemic side effects is remarkably low. Therefore, the intraocular injection justifies the small risk of postoperative complications, even considering that the drug requires repeated injection into the eye,†Dr Richard added.

He noted that in his experience, anti-VEGF agents have a cumulative and residual effect, such that each treatment results in a better and longer lasting effect and over time the need for retreatment becomes less and less. Moreover, a new anti-VEGF agent called VEGF trap (Regeneron) may reduce the need for repeated injections in the future.

Avastin vs Lucentis

Currently the most frequently administered intravitreal agents are the anti-VEGF agents, bevacizumab (Avastin, Genentech) and related compound ranibizumab (Lucentis). The main indication currently is for choroidal neovascularisation resulting from AMD, but the future is likely to see an increasing use of the agents for diabetic retinopathy.

Of the two agents only ranibizumab has received FDA approval for intravitreal use in the treatment of CNV. However, many patients cannot afford the 40-fold higher price of ranibizumab, and as a result must decide between off-label treatment with bevacizumab or gradually going blind. In fact, Avastin injections account for 60 per cent of patients receiving intravitreal anti-VEGF agents in the US.
Until recently, the arguments in favour of ranibizumab over bevacizumab have been largely theoretical. Proponents of Lucentis had originally argued that because Avastin is a much larger molecule it would not penetrate the retinal tissues as well as Lucentis and would therefore be less effective and it would also be more immunogenic.

Subsequent research has suggested that there may be only slight or negligible differences between the two agents in terms of safety or efficacy. For example, in the Comparison of AMD Treatments Trials (CATT) both agents produced virtually identical results in terms of visual acuity throughout at all time points during one year of follow-up.

There was a slightly higher incidence of serious adverse events in the bevacizumab group (24 per cent vs. 19 per cent, p=0.040), mainly hospitalisations. However, the authors point out that the bevacizumab group had a higher rate of risk factors for hospitalisation, such as smoking, diabetes and a history of cardiovascular disease (The CATT Research Group, N Engl J Med 2011; 364:1897-1908).

At the same time, the incidence of endophthalmitis per patient was fairly high in both treatment groups in the CATT study. Among 600 patients randomised to receive ranibizumab, the incidence was one in 300 patients, and among 600 patients randomised to receive bevacizumab the incidence was one in 150 patients.

The microorganisms identified included coagulase negative staphylococcus in two cases and alpha haemolytic streptococcus in two cases. Two cases were culture-negative. Visual acuity upon resolution of the infection ranged from 20/25 to 20/60 in four cases, in one case it was 20/125, and in another it was counting fingers.

Despite there being little differences between the two agents in the context of the trial, proponents of ranibizumab maintain that the situation could be very different in the context of vast numbers of compounding pharmacists using a range of different repackaging techniques.
In fact, there have been several recent reports of outbreaks of endophthalmitis following intravitreal injections of the medication has to be sterile as well,†Dr Richard said.

Dr Van Meurs told EuroTimes that useful additional measures might include making sure only the patient and injector are in the room; using a controlled air flow system; cap, mask and gloves for the injector, drape that separates nose, mouth and hair from the eye for the patient. He added that if research shows that outbreaks of endophthalmitis really are more common after Avastin (prepared in batches), injection through a bacterial filter at the price of €2.00 extra per injection would then be an inexpensive option of reducing the excess risk. 'This could be done either in the office, although you would need larger aliquots, because the filter has a considerable dead space, or in the pharmacy for each prepared vial, to further minimise batch contamination,†he said.

New delivery routes  

[caption id='attachment_590' align='alignright' width='400' caption='Vitrectomy for endophthalmitis post anti-VEGF injection - Courtesy of Peter Barry FRCS'][/caption]

Intravitreal steroids carry about the same risk of endophthalmitis as intravitreal anti-VEGF agents and also have some additional risks. However, steroids have also been the focus of the development of a number of other methods to deliver medicine to the retina, in the form of slow-release implants, such as Ozurdex, and most recently a dexamethasone preparation designed for topical administration.

The earliest steroid implant was Retisert (Bausch + Lomb), which used fluocinolone acetonide. The implant is sutured to the eye wall using a 3.5mm pars plana incision designed to last up to three years. In the FDA clinical trial all phakic patients receiving the implant have developed cataracts and 40 per cent of eyes needed filtration surgery to control the IOP.

The Iluvien (Alimera) implant also uses fluocinolone acetonide but appears to have a less severe effect on IOP. In the Fluocinolone for Macular Edema (FAME) study, 75 per cent of patients developed cataracts, but only five per cent requires filtering surgery.
There is also a biodegradable dexamethasone implant, Ozurdex (Allergan). Unlike the Retisert and the Iluvien implants, it does not leave an empty husk behind once the drug is used up. Instead, as the drug is released, the implant’s polymer, polylactic glycolic acid, biodegrades to lactic acid glycolic acid, water, and carbon dioxide.

In the PLACID trial, only 15 per cent of eyes required IOP-lowering medication. The only specific adverse event to occur significantly more often in the implant group was increased IOP, which occurred in 20 per cent, compared to 1.6 per cent of the laser alone group.
Another, completely different approach, now under development by a French Company called Novogali, is to inject dexamethasone in a pro-drug form. The pro-drug’s chemical structure, dexamethasone palmitate, is relatively inert in the aqueous and the vitreous and is therefore less likely to affect the lens or the trabecular meshwork. However, in the retina it breaks down and becomes active. A Phase I, Open-Label, Dose-Escalation Clinical Study is now under way.

Non–invasive options

However, even these approaches remain invasive procedures. A better alternative might be the development of agents that can
be applied topically, said Einar Stefánsson MD, PhD, University of Iceland,
Reykjavik, Iceland.

“Clearly from a standpoint of safety we need to look at ways of eliminating or at least reducing the need for intravitreal injections. The complications like endophthalmitis and intraocular inflammation seen with intravitreal injections show we need to focus on developing non-invasive ways of delivering drugs into the eye,†Dr Stefánsson told EuroTimes.

He noted that in a recently published phase II study a topical 1.5 per cent dexamethasone aqueous solution prepared by cyclodextrin nanoparticle technology produced significant reductions in mean macular thickness and significant improvements in mean visual acuity in 19 eyes of 19 patients with chronic diabetic macular oedema (Tanito et al, Invest Ophthalmol Vis Sci. 2011;52(11):7944-7948).

Dr Stefánsson said that so far patients have had a fairly low incidence of elevated IOP. Moreover, should a patient be a steroid responder and develop a high IOP, the treatment is very easily reduced simply by withdrawing the medication.

Gene therapy is another technology that is beginning to bear fruit in the treatment of retinal disease, notably in the treatment of Leber’s congenital amaurosis. Clinical studies are under way employing gene therapy in the treatment of retinitis pigmentosa.

Another approach is the use of encapsulated cell technology. This approach, in effect, places a drug factory in the eye, in the form of a cluster of encapsulated cells, which secrete a therapeutic agent. In a presentation at the 11th EURETINA Congress, Weng Tao MD, US, described how, in three dose-ranging studies involving a total of 120 patients with retinitis pigmentosa or geographic atrophy, an implant employing encapsulated cell technology to secrete a neurotrophic agent was able to increase the thickness of the photoreceptor layers of the retina in the eyes with implant.

In the meantime, practitioners of medical retina are to some extent flying blind in their attempts to reduce the rate of endophthalmitis in their patients, said Dr Barry. That is because as yet there is no comprehensive epidemiological evidence on which to base prophylactic measures.
“It might be a good idea to create an endophthalmitis registry like the Swedish registry, but covering all of Europe and encompassing both cataract procedures and intravitreal injections, through a combination of the efforts of EURETINA and the ESCRS,†he added.

contacts
Jan van Meurs – janvanmeurs@cs.com
Gisbert Richard – augenklinik@uke.uni-hamburg.de
Peter Barry – peterbarryfrcs@eircom.net
Einar Stefánsson – Einarste@landspitali.is