TRIALS SHOW EFFICACY

Photographs of a child with keratitis ichthyosis deafness (KID) syndrome, before and after limbal injection of bevacizumab, showing marked reduction in vessel calibre but ‘ghost’ vessels still present. Courtesy of Department of Ophthalmology, Great Ormond Street, London, UK

As is often the case with clinical decision-making in paediatric ophthalmology, the use of an anti-vascular endothelial growth factor (anti-VEGF) agent for treating corneal neovascularisation should be assessed on a case-by-case basis, said William Moore BSc, MBBS, FRCOphth, at the 3rd World Congress of Paediatric Ophthalmology and Strabismus in Barcelona, Spain.

Reviewing the published literature on this topic, Dr Moore concluded that there is evidence from animal and human trials showing efficacy of anti-VEGF treatment for reducing corneal neovascularisation. However, the clinical experience is limited, mostly involved adults, and represents off-label use.

“While anti-VEGF agents are giving us some hope for treating corneal neovascularisation, there is no established safe dosage or mode of administration. Therefore, further investigation is needed before anti-VEGF drugs can become the useful therapeutic agents for inhibiting corneal angiogenesis that we all wish them to be,” said Dr Moore, consultant ophthalmic surgeon, Great Ormond Street Hospital, London, UK.

“Nevertheless, there are children under our care whose corneas are failing, rejecting and clouding, and so we must continue to try to help them, using anti-VEGF agents based on our best judgment calls,” he added.

RESEARCH NEEDED

Most reports on anti-VEGF treatment for corneal neovascularisation in human eyes describe the use of bevacizumab, which is also the least expensive of the available agents. It has been administered topically and also as an intrastromal, subconjunctival or intraocular injection.

Topical use has been associated with epitheliopathy, with a 60 per cent incidence in one study evaluating 1.25 per cent bevacizumab applied three times daily. More research is needed to understand the safety of subconjunctival and intrastromal injections.

Of note, one case report described regression of extensive vessel growth in the stroma, with no recurrence during six months of follow-up after an intrastromal injection of bevacizumab.

“It makes sense that the intrastromal injection might be a more effective route of administration since the medication will not diffuse away as quickly,” said Dr Moore.

Compared with bevacizumab, ranibizumab binds with higher affinity to VEGF-A, and aflibercept has even higher affinity for VEGF-A while also binding to placental growth factors 1 and 2. Its broader binding profile may provide aflibercept with greater anti-angiogenic activity. Dr Moore noted that aflibercept also has the longest duration of anti-VEGF action (twofold longer than ranibizumab after intravitreal injection), which may be a drawback when treating children.

Although pegaptanib, which binds only to the VEGF165 isoform, seems to be less effective than the other anti-VEGF agents for reducing corneal neovascularisation, its good safety profile when used for retinal neovascularisation suggests the potential for it to have a role as maintenance therapy, Dr Moore said.

The idea that maintenance therapy may be necessary derives from observations that available anti-VEGF agents can stop new vessel growth if treatment is started early, but are less effective once the blood vessels become covered with pericytes, which occurs after about two weeks.

“If we watch the eyes and wait, it may be too late. However, if we also treat the underlying cause for the corneal neovascularisation aggressively, the eye may stay calm,” Dr Moore said.

Investigational anti-angiogenic agents with novel mechanisms may offer greater activity than the available anti-VEGF agents, but much more work is needed to define their efficacy and safety. Compounds in development include silencing RNA molecules (bevasiranib and Sirna-027, for example) that can block both the intracellular and extracellular effects of VEGF and its receptors.

In addition, there are tyrosine kinase inhibitors being developed that target a pathway initiated after VEGF binds to its receptors. Agents within the latter class include pazopanib, which improved graft survival in a mouse model of corneal transplantation, and sunitinib, which has been shown to reduce pericytes in established vessels. The effect of tyrosine kinase inhibition on normal vasculature needs to be determined, Dr Moore said.

He concluded by stating that development of anti-VEGF agents as a treatment for corneal neovascularisation will require large randomised, controlled trials.

In the meantime, Dr Moore encouraged ophthalmologists who see large numbers of patients with corneal neovascularisation to collaborate and publish their data.

William Moore: william.moore@gosh.nhs.uk