Arthur Brant, MD
Darius Bordbar, MD
Jimmy Chen, MD
Massachusetts Eye and Ear Infirmary
Thursday’s AMD I session at the Retina Society meeting in Los Angeles was moderated by Carl Regillo, MD, and Jennifer Kang-Mieler, PhD. The presentations examined sustained drug delivery, treatment intervals in routine practice, and a new approach to VEGF and TIE-2 signaling.
Dr. Ehsan Rahimy opened with the preclinical rationale for EYP-1901, a bioerodible insert that releases the tyrosine kinase inhibitor (TKI) vorolanib for at least six months. Along with VEGF and PDGF receptor inhibition, vorolanib inhibits JAK1, a component of IL-6 signaling. In a laboratory endothelial barrier assay, its effect resembled combined VEGF and IL-6 inhibition.
Dr. Carl Regillo then presented visual and anatomic results from LUGANO, which randomized 432 patients with neovascular AMD (nAMD) to EYP-1901 every six months or aflibercept 2 mg every eight weeks after aflibercept loading. The trial missed its primary endpoint of noninferiority in best-corrected visual acuity (BCVA), averaged over weeks 52 and 56. Mean gains were approximately two and six letters, respectively, despite similar control of central subfield thickness (CST). Regillo reviewed nine EYP-1901 patients whose substantial vision loss was attributed to causes other than wet AMD. Excluding these patients in a post hoc analysis narrowed the difference to 2.4 letters. That analysis did not change the primary result.
Dr. Charles Wykoff addressed LUGANO’s treatment burden and safety findings. Fifty-four percent of EYP-1901 patients required no supplemental aflibercept through week 56, and 79% received no more than one supplement. After loading, the total injection burden was 42% lower with EYP-1901, counting both scheduled and supplemental injections. Two cases of endophthalmitis occurred in the EYP-1901 arm. Each arm had one retinal detachment and one patient with intraocular inflammation. Floaters were reported more often with EYP-1901, at 9.0% versus 3.6%. During discussion, an audience member asked whether eyes that became dry after two loading injections might be better candidates for TKI treatment. The presenters said that subgroup had not yet been analyzed.
Dr. Dilsher Dhoot presented SOL-1, the phase 3 trial of intravitreal axitinib hydrogel (OTX-TKI). After two aflibercept loading injections, participants received a single dose of OTX-TKI or aflibercept, with subsequent treatment governed by rescue criteria through the primary endpoint. At week 36, 74.1% of OTX-TKI recipients and 55.8% of aflibercept recipients lost fewer than 15 letters, meeting the trial’s definition of vision maintenance. The comparator did not receive scheduled every-eight-week aflibercept during this period. Dhoot reported floaters in 12.4% of OTX-TKI recipients and nine inflammation events in seven participants; all inflammation events resolved. No treatment- or procedure-related serious adverse events were reported.
Dr. Jennifer Kang-Mieler described aflibercept-loaded microspheres within a thermoresponsive hydrogel. In seven nonhuman primates with laser-induced choroidal neovascularization, treated eyes had less lesion fluorescence and a lower incidence of neovascularization than untreated fellow eyes. Aflibercept remained detectable in aqueous humor through seven months. During discussion, she cautioned against directly comparing the 15 µg dose with a conventional 2 mg injection and noted that the relationship between drug concentration and efficacy needed further study.
The next presentations examined treatment intervals in clinical practice. Dr. William Freeman’s IRIS Registry analysis found that average intervals of at least 13 weeks occurred in 25.8% of nAMD eyes treated with aflibercept 8 mg and 20.8% treated with faricimab. He pointed to differences in fluid tolerance and retreatment criteria as possible explanations for the gap from trial results. The registry’s average interval over follow-up also differs from the interval reached at a particular trial visit.
Dr. Michael Klufas reported an expanded ILLUMIN-8 analysis of approximately 2,600 treatment-naïve eyes and 11,400 eyes switched from aflibercept 2 mg to 8 mg. The median last observed injection interval was 12 weeks in both groups; switched eyes extended by about three weeks. Median visual acuity was maintained overall. Klufas noted that routine chart measurements were not necessarily best-corrected and that medical histories could be incomplete.
Dr. Ferhina Ali presented two-year results from FARETINA-nAMD and FARWIDE-nAMD, retrospective studies in the United States and United Kingdom. Vision was maintained overall as injection frequency declined, with gains among treatment-naïve eyes that started with poorer vision. In FARETINA, the last injection interval increased by approximately four to five weeks in previously treated eyes, including those switched from aflibercept 2 mg.
Dr. Jong Park’s Canadian faricimab series included 482 eyes with nAMD or diabetic macular edema (DME) that had received at least one year of treatment and six injections. By the end of follow-up, mean intervals in switched eyes had increased from 5.9 to 9.2 weeks for nAMD and from 4.2 to 8.5 weeks for DME. At the sixth injection, vision was maintained in switched eyes and retinal thickness had decreased. In discussion, speakers noted that clinicians may cap extension intervals, making observed intervals an imperfect measure of a drug’s maximum durability.
Dr. Jeremy Wolfe reviewed follow-up of up to seven years with the port delivery platform with ranibizumab (PDS) in the Portal extension. CST remained close to baseline, and approximately 95% of patients needed no supplemental treatment within each six-month refill interval. Wolfe contrasted mean BCVA losses of approximately 5 -10 letters over five to seven years with smaller median losses of 2 -5 letters. He also discussed cumulative endophthalmitis risk when comparing long-term PDS treatment with repeated intravitreal injections.
Dr. Yasha Modi presented an interim analysis of injection burden in BELVEDERE, an open-label study of PDS. The analysis included 84 PDS recipients, 43 of whom had required at least nine injections per year before implantation. Visual acuity and CST changed little over the reported 48-week follow-up. Intraretinal fluid decreased in the high-burden subgroup, although subretinal fluid persisted. Modi reported that nearly all patients reached six-month refill intervals without supplementation. When asked about macular atrophy, the presenters noted that seven-year atrophy data were not presented and BELVEDERE follow-up was still limited.
Dr. Diana Do closed with phase 2a results from RIOJA for tiespectus (MK-8748), which combines VEGF antagonism with direct TIE-2 activation. Twenty treatment-naïve participants received three monthly injections. At week 12, mean BCVA had improved by 7.8 letters, CST had decreased by 218 µm, and 85% of eyes had complete fluid resolution on OCT. No drug-related ocular or systemic adverse events were reported. During discussion, Do described a phase 3 noninferiority comparison with aflibercept that would also examine retinal drying, with treatment extension to be assessed later.