Chew EY, Clemons TE, Jaffe GJ, et al. Effect of Ciliary Neurotrophic Factor on Retinal Neurodegeneration in Patients with Macular Telangiectasia Type 2: A Randomized Clinical Trial. Ophthalmology. 2019;126(4):540-549.
Macular telangiectasia type 2 (MacTel 2) is an idiopathic condition without available treatment options which results in slowly progressive bilateral vision loss. In this randomized sham-controlled multicenter clinical trial, patients with MacTel 2 were randomized to surgical placement of an implant which eluted ciliary neurotrophic factor (CNTF) or a sham procedure. CNTF has been shown in animal models to reduce photoreceptor loss and the surgical implant in this study produced CNTF at 20 ng/day from genetically modified human RPE cells. Patients were followed for 24 months post-intervention.
The primary outcome was the difference in area of neurodegeneration represented by area of ellipsoid zone (EZ) loss on OCT. Secondary outcomes included changes in vision, changes in retinal sensitivity on microperimetry, changes in monocular reading speed and visual field changes. Ninety-nine eyes of 67 patients were randomized with 97% of patients completing 24-month follow-up. Mean baseline vision for both groups was quite good at ~76 letters (~20/30 Snellen equivalent). The mean area of photoreceptor loss increased 0.27±0.05 mm2 from baseline to month 24 in the sham group compared to 0.22±0.05 mm2 (p=0.04) in the implant group representing a 31% greater progression of neurodegeneration in the sham group. Additionally, 44% of eyes in the sham group versus 31% of eyes in the implant group had a >35% increase in EZ loss at 24 months compared to baseline (p = 0.06).
At 24 months, vision loss was minimal and similar for both the sham and implanted groups. In the per-protocol analysis, retinal sensitivity loss was greater in the sham group at month 24 (difference in aggregate retinal sensitivity loss between the sham and implant group was 24.39±8.17 dB; p = 0.002). Additionally, mean monocular reading speed at month 24 decreased in sham-treated eyes by a mean of ~14 words per minute compared to baseline while implanted eyes did not experience a change in reading speed (p = 0.02). There was not a statistically significant difference in either mean deviation or pattern standard deviation on Humphrey visual fields between the groups at various timepoints. The implant was generally well tolerated. However, 18.8% of implanted eyes experienced persistent self-reported delayed dark adaptation and miosis – side-effects which may be attributable to CNTF.
Overall, this study has several takeaways. One of the most important is that visual function in MacTel 2 patients is not readily appreciated by visual acuity measurements alone. Baseline reduction in retinal sensitivity measures such as microperimetry and functional tests such as reading speed are telling of the impact of retinal neurodegeneration in this condition. While there was a statistically significant reduction area of EZ loss in implanted-eyes compared to sham-treated eyes both groups did experience significant EZ loss over 24 months. Despite this EZ loss, implanted-eyes maintained their baseline reading speed while sham-treated eyes lost reading speed.
The results of this important phase 2 clinical trial require further validation in phase 3 trials but appear quite promising. Key questions that need to be answered are whether a threshold exists with regards to lesion size in MacTel 2 beyond which the implant may not be beneficial. Additionally, would patients with smaller lesions/milder disease benefit from such an implant? It will also be interesting to follow this cohort for a longer period of time as the study implant has been shown to continue releasing CNTF up to 5 years out. Of additional interest is whether CNTF has a role in other conditions featuring photoreceptor loss such as macular degeneration and retinal dystrophies.
Akshay S. Thomas, MD