Comparison of Ranibizumab With or Without Verteporfin Photodynamic Therapy for Polypoidal Choroidal Vasculopathy: The EVEREST II Randomized Clinical Trial

November 2020

Lim TH, Lai TYY, Takahashi K et al. Comparison of Ranibizumab With or Without Verteporfin Photodynamic Therapy for Polypoidal Choroidal Vasculopathy: The EVEREST II Randomized Clinical Trial. JAMA Ophthalmol. Sept 2020; 138(9): 935-942.

Polypoidal choroidal vasculopathy (PCV) is a subtype of neovascular age-related macular degeneration characterized by aneurysmal polypoidal lesions best imaged with indocyanine green angiography (ICGA).  A lack of consensus still exists in the ophthalmic community regarding the optimal regimen for treatment of PCV.  The two predominant treatments of choice, taken separately or in combination, involve anti-vascular endothelial growth factor (anti-VEGF) injections or verteporfin photodynamic therapy (vPDT).  The current paper outlines the 24-month treatment outcomes of the phase IV EVEREST II study, which compares ranibizumab and vPDT combination therapy with ranibizumab monotherapy.

The ranibizumab monotherapy arm received ranibizumab 0.5mg intravitreal injections monthly for 3 months, followed by a pro re nata (PRN) regimen.  The combination therapy arm followed the same injection protocol but also received vPDT treatment covering the entire lesion during the initial baseline treatment visit, with subsequent treatments as needed.  Ranibizumab monotherapy patients were eligible to be switched after month 12 if indicated, however only 9% (14 of 154) required rescue laser.

At 24 months, the best-corrected visual acuity improvement was greater in the combination group (9.6 letters) vs the monotherapy group (5.5 letters), demonstrating superiority of combination therapy (P=0.005).  The combination therapy group also demonstrated higher rates of complete polypoidal lesion regression (56.6%) at 24 months compared to monotherapy group (26.7%, P<.001), as determined by followup ICGA.  There was also noted to be a greater reduction of central subfield thickness (-152.9 vs -109.3 um), higher proportion of patients with absence of leakage on fluorescein angiography (57.5% vs 32%), and lower proportion of patients with disease activity (27% vs 54.3%) in the combination group versus the monotherapy group.  On average, the number of ranibizumab injections received was lower in the combination group (8.1) than the monotherapy group (12.5).  The mean number of vPDT treatments was 2.2, with 44.6% of patients in the combination group needing only one vPDT treatment during the 24 months.

The results of the EVEREST II study reaffirm the results of the original EVEREST trial, namely that ranibizumab is an effective treatment for PCV, but that combination therapy with vPDT results in superior outcomes compared to ranibizumab alone.  However, EVEREST II had a couple advantages over its predecessor.  Firstly, the use of a central reading center and predefined criteria for diagnosis improved the quality of data collection, especially since PCV can sometimes be difficult to diagnose.  Secondly, the 24-month results allowed participants to switch groups after 12 months if the disease was still active.  These switched group participants demonstrated better complete polypoidal lesion regression compared to those who were not switched (47.5% vs 26.7%), suggesting for the first time that vPDT applied later in the treatment course can still be beneficial.

As a retina specialist still struggling to obtain the very best results for my PCV patients, this well-designed phase IV study reaffirms that the judicious use of vPDT in combination with anti-VEGF can lead to good visual outcomes for patients.  The data shows good maintenance of visual acuity gains despite less aggressive and less frequent therapy in year 2.  However, the most intriguing finding about the benefits of delayed vPDT was based off a very small sample size of 14 patients, which is only a small fraction of the patients who were eligible for delayed laser.  Therefore, a larger study would be required to assess the true effect of delayed vPDT on outcomes in PCV.