Decomplicating Complications: Researchers Propose Standardizing List of Complications From Anti-VEGF Agents

Anti-vascular endothelia growth factor agents (VEGF) have transformed the treatment of wet, or neovascular, age-related macular degeneration (AMD), diabetic macular edema and other conditions related proliferating, leaky blood vessels in the choroid that impinge on the retina. But along with successes, such as Lucentis (ranibizumab) and Eylea (aflibercept), have come some setbacks, such as AbbVie’s abandonment of abicipar pegol, even though the phase 3 trial results for the agent were promising.

Rajeev H.Muni, M.D., M.Sc.

Rajeev H. Muni, M.D., M.Sc., of St. Michael’s Hospital in Toronto and an assistant professor at the University of Toronto, and his colleagues argue in the journal Ophthalmology this month that clearer delineation of complications from anti-VEGF would make them more apparent during late-stage clinical trials and avoid confusion and worry about the safety of anti-VEGF agents. They used the Delphi method, a standardized approach for researching consensus among experts, to arrive at list of 67 ocular complications and 47 unique “specifiers” of those complications.

Click here to see the list of 67 complications.

If researchers conducting clinical trials of anti-VEGF agent used their list of complications, they would improve consistency and the quality of adverse event reporting and help make risk-benefit analyses more accurate, Muni and the authors concluded.

Medical Dictionary for Regulatory Activities (MedDRA) terms are considered the gold standard for adverse events in clinical trials, and the latest version of MedDRA includes many of the ophthalmic terms on the list that they arrived at with their Delphi method , Muni and his colleagues note. But they said the unique specifiers, such as unilateral versus bilateral manifestation and more precise labels of anatomical location, are absent from MedDRA terms. They also note that MedDRA’s complexity and redundancy can “lead to inconsistent reporting of adverse events specific to anti -VEGF therapy as investigators may select terms for the same complication based on their initial encounter in the MEDAR hierarchy.”

The imprecision of the MedDRA terms for ophthalmic adverse events made it difficult to determine whether the intraocular inflammation events associated with Beovu (brolucizumab) included retinal vasculitis, according to Muni and his colleagues. “Indeed, it was only learned that severe retinal vascular events had occurred in the trials upon an independent post-marketing reanalysis of the clinical trial data.”The complications caused permanent loss of vision in some patients and reduced confidence in the treatment they wrote in the Ophthalmology article that was posted earlier this month.

To conduct their Delphi method study, Muni and his colleagues recruited a group of 25 retina specialist from the American Society of Retina Specialists, the Asia Pacific Vitreo-retina Society, the British and Eire Association of Vitreoretinal Surgeons, the Canadian Retina Society, the European Society of Retina Specialists, the Macular Society and the Retina Society. They searched the Ovid Medline, Embase and Cochrane Central databases to identify reports of randomized clinical trials of anti-VEGF agents that mentioned ocular adverse events. A list of 18,229 articles was eventually winnowed down to 145 studies and 244 ocular complications of intravitreal (into the eyeball) administration of anti-VEGF agents. After some repetition was removed, Muni and his colleagues had a preliminary list of 127 complications to put before experts and through the Delphi process.

The Delphi process involves rounds of surveys to arrive at a consensus. Muni and his colleagues said they used commonly used thresholds with consensus defined as 70% or greater agreement on 90% or greater of agreement of items. An item — in this case, a complication — was rephrased or excluded if less than 30% of the experts chose to include it.

After the Delphi method yielded a list of complications, a smaller steering committee came up with specifiers. Muni and his co-authors gave uveitis as an example: uveitis is the complication but bilaterial (both eyes) and acute (sudden) are specifiers. The specifiers were also put through the Delphi process.

The researchers said that they are not alone in using the Delphi process to identify and standard the reporting of complications, pointing to publication of results that used the method for complications from glaucoma surgery and rhegmatogenous retinal detachment. Muni and his colleagues said they also took the step of harmonizing their terms with the Standardization of Uveitis Nomenclature (SUN) criteria.

As for the limitations of their work, Muni and his colleagues noted that countries that the experts were from tended to those with developed economies and that Delphi method inherently relies on the opinion of the participants and can introduce possibly misleading subjectivity. They also note that although patient-reported outcomes can provide valuable insights into patient experiences, they are not included. Ocular complications, they said, “require a specialized understanding of the underlying clinical assessment and multimodal imaging interpretation that can only be offered by retinal specialists.”