Retinal Fluid and Visual Acuity After PDS with Ranibizumab, With Veeral Sheth, MD, MBA

At the 2025 Association for Research in Vision and Ophthalmology (ARVO) Annual Meeting, Veeral Sheth, MD, MBA, director of clinical research at the University of Retina and Macula Associates and clinical assistant professor at the University of Illinois, presented a volumetric analysis of retinal fluids from the Archway phase 3 trial evaluating the Port Delivery System (PDS) with ranibizumab in patients with neovascular age-related macular degeneration.

The post hoc analysis used a pre-trained deep learning model to segment subretinal and intraretinal fluid (SRF and IRF) on optical coherence tomography scans.1

“What our model did was get rid of the noise, because with fluctuations, you get a lot of noise,” Sheth told HCPLive. “And one of the key things is filtering out some of that noise. If you leave it to us just individually plot it out, it can be very labor-intensive, very cumbersome. This is where advancements in technology have really helped us. We can use these algorithms to get rid of the noise, and then we can really start looking at deviations in a slightly different way.”

In the phase 3 Archway trial, patients were randomized 3:2 between the PDS with ranibizumab 100 mg/mL, with established 24-week refills (PDS Q24W) and ranibizumab 0.5 mg intravitreal injection every 4 weeks (RBZ Q4W). Sheth’s study highlighted eyes with imaging data of quality adequate for volumetric analysis of retinal fluid. A variability metric robust to absent values and outliers was calculated for the central subfield thickness (CST), volume of SRF and IRF in the central subfield for each eye.1

Sheth and colleagues found no significant difference in fluid fluctuation from baseline to week 96 across both treatment arms (PDS Q24W, RBZ Q4W, mean +/- [standard deviation] SD, CST (µm): 5.2 +/- 5.1, 4.9 +/- 6.5l; IRF (nL): 0.4 +/- 2.0, 0.5 +/- 2.6; SRF (nL): 2.1 +/- 3.8, 1.4 +/- 3.2, all P >.05).1

When the team categorized by IRF fluctuation, best-corrected visual acuity (BCVA) was substantially higher (P <.001) in eyes in the lowest quartile than in the highest (lowest vs highest IRF fluctuation quartile, mean +/-SD, BCVA (letters): 76 +/- 10.8, 68.0 +/- 17.0). However, when segmented by SRF fluctuation, median BCVA was not significantly different (P >.05) in eyes in the lowest and highest quartile (lowest vs highest SRF fluctuation quartile, mean +/- SD, BCVA (letters): 72.9 +/- 14.3, 74.8 +/- 11.0).1

Sheth and colleagues’ results showed that BCVA is significantly higher in eyes with less IRF fluctuation than with more.

“I think that really helps support some of the data, the building of a body of evidence that intraretinal fluid fluctuations or even residual intraretinal fluid really does matter, and has an impact on patients’ vision,” Sheth said. “I think we should look closely at intraretinal fluid in patients visit to visit and really look for treatments that address that variability or fluctuation in the intraretinal fluid.”

References

1. Sheth, V, et al. Volumetric analysis of retinal fluids in Archway phase 3 trial of the Port Delivery System with ranibizumab (PDS) in patients with nAMD. Abstract presented at Association for Research in Vision and Ophthalmology in Salt Lake City, UT, from May 3 - May 8, 2025.