Study Results Suggest CNV Lesions May Contribute to Development of Type 1 MA in Patients With nAMD

A post-hoc analysis of data from the TREX-AMD (Treat-and-Extend Age-Related Macular Degeneration) trial has determined that eyes with neovascular age-related macular degeneration (nAMD) and treated with ranibizumab, develop more macular atrophy (MA) lesions within baseline (BSL) choroidal neovascularization (CNV) regions than outside those regions.

The analysis, completed on behalf of the TREX-AMD study group and led by Nizar Saleh Abdelfattah, MD, a clinical research fellow at the University of California Los Angeles' David Geffen School of Medicine, also determined that baseline-MA is likely to be located within regions of CNV of nAMD patients prior to treatment with anti-vascular endothelial growth factor-A (anti-VEGF) agents. This discovery suggests that CNV lesions may be contributing to MA development in eyes with nAMD.

The TREX-AMD trial is a 3-year, ongoing, multicenter, randomized, controlled clinical trial designed to assess the effectiveness of a treat-and-extend (TREX) treatment regimen with the intravitreal anti-VEGF agent ranibizumab (0.5 mg) versus equal dosages of a monthly treatment regimen in treatment-naïve AMD patients. Data from this extensive study, which includes ophthalmological imaging (monthly fluorescein angiography [FA] scans, color fundus photographs, spectral-domain optical coherence tomography [SDOCT] scans, fundus autofluorescence (FAF) scans) for 60 patients, have been used in several post-hoc analysis studies.

Abdelfattah and colleagues stated that this TREX-AMD associated analysis used "multimodal imaging to better characterize MA with a focus on the correlation between baseline CNV location and the areas of MA at baseline through 18 months of follow-up study." The team included 26 study eyes (10 from the monthly control group, and 16 from the TREX group of the TREX-AMD study) that "exhibited evidence of CNV at baseline and MA at month 18."

The investigators determined that 65% (n = 13) of the study eyes showed some topographic correspondence between CNV and MA regions at baseline, with 46.2% of those eyes (n = 6) showing partial overlap and 53.8% (n = 7) of those eyes showing "complete" overlap of MA within CNV regions. After quantifying area of MA (AMA), the team reported that eyes with only classic CNV (n = 4) saw an AMA overlapping of 43.05%, eyes with only occult CNV (n=6) saw an AMA overlapping of 96%, and eyes with both occult and classic CNV (n = 3) saw an AMA overlapping of 50.3% of occult CNV versus 49.7% of classic CNV. Abdelfattah and colleagues pointed out that "though numerically occult CNV showed more overlap with the area of atrophy than classic CNV (66.86 + 38 mm² vs 47.3 + 32.8 mm²), [there was] no statistically significant correlation between CNV subtype and area of overlapping atrophy."

The team also reported that all 26 eyes exhibited MA at 18-month follow-up and 84.6% (n = 22) "showed evidence of overlap between regions of CNV at baseline and regions of MA at month 18."

Of the 22 eyes exhibiting overlap, that overlap was complete in 40.9% of eyes (n = 9) and partial in 59.1% of eyes (n = 13). By 18-month follow-up, those patients with occult CNV component saw greater overlap (40.9%, n =9) than those with classic CNV component (36.4%, n = 8) or mixed occult and classic CNV component (22.7%, n = 5).

The data indicated that "occult CNV saw more overlapping with the area of atrophy than classic CNV (47.3 + 32.5 mm² vs 40.4 + 31.6 mm²)," according to the study; but, "there was no statistically significant correlation between CNV subtype and area of overlapping atrophy."

A total of 50% (n = 3) of eyes that showed no overlap between baseline MA and baseline CNV (n =6) developed new MA and 100% (n = 6) developed MA within regions of CNV between baseline and 18-month follow-up. The area of that overlap was higher in eyes with occult CNV (mean AMA 72.5%) than eyes with only classic CNV (mean AMA 78.5%).

The investigators hypothesize that the mechanisms of MA development might be different in areas within and outside of CNV regions. In particular, MA corresponding to CNV "may be more likely owing to the destructive effects of the CNV lesion itself," and that areas of CNV may be more "sensitive" to scarring and angiogenic imbalance leading to MA than areas unaffected by CNV, the authors write. The CNV lesion itself could be, according to the authors, "an important contributor to MA development" but further research is needed to confirm the relationship between CNV and MA, particularly in association with the mechanisms of MA development.

"Topographic Correspondence of Macular Atrophy With Choroidal Neovascularization in Ranibizumab-treated Eyes of the TREX-AMD Trial" appears in the August 2018 issue of the American Journal of Ophthalmology.