Marco A. Zarbin, MD, PhD
In this randomized, multi-center, double-masked 100-week clinical trial, the primary efficacy endpoint — the proportion of eyes with a 2-step or greater improvement in DRSS level from baseline 47 to 53 (moderately severe to severe NPDR) — was met in the aflibercept groups at both weeks 24 and 52.
Intravitreal injections of aflibercept were administered in one of two ways — every 16 weeks after 3 initial monthly doses and one 8-week interval (aflibercept 2q16 group) or every 8 weeks after 5 initial monthly doses, with PRN dosing beginning at week 56 (aflibercept 2q8/PRN group). Sham injections were administered to the control group.
The structure of the dosing schedule is interesting. Except for the loading phase, the 2q16 week posology conforms to the follow-up frequency that normally is adopted for patients with moderately severe to severe NPDR.
At 52 weeks, 65% of eyes in the aflibercept 2q16 group and 80% in the aflibercept 2q8/PRN group compared with 15% in the control group showed a 2-step improvement in DRSS level. At week 100, however, 62% of eyes in the 2q16 group vs 50% in the 2q8/PRN group vs 13% in the control group showed 2-step improvement in DRSS level.
There seems to be a dose response effect with the 2q8/PRN cohort exhibiting better improvement in the DRSS score at week-52 (80%) vs the 2q16 cohort (65%) and the control group (15%) for 2-step improvement in DRSS level. However, there also seems to be a need for ongoing therapy to maintain this benefit since the improvement in the 2q8/PRN cohort decreased once they were on a PRN posology after week-52, which was associated with a lower frequency of injections (average of approximately two during the second year).
Although fewer eyes in the aflibercept-treated groups exhibited vision-threatening complications such as PDR and anterior segment neovascularization (NVI), and/or CI-DME through week 100 (2q16:16% of eyes, 2q8/PRN: 19%, control: 50%), the visual outcomes at week 100 were not different among the groups.
One may question whether the additional cost imposed by this therapy merits the potential benefit it affords. There is a suggestion that there may be a cost/benefit ratio favoring treatment. Specifically, a post hoc analysis identified a greater number of participants experiencing 5, 10, and 15 ETDRS letter losses from baseline in the control group (29% lost 5 letters, 14% lost 10 letters, and 8% lost 15 letters) compared with the aflibercept 2q16 group (9% lost 5 letters, 5% lost 10 letters, and 3% lost 15 letters) and the aflibercept 2q8/PRN group (9% lost 5 letters, 5% lost 10 letters, and 3% lost 15 letters). So, there may be a long-term visual benefit that was not captured during the time intervals reported thus far.
Even among treated patients, there was a tendency for 2-step worsening of the DRSS over time. For example, at week 52, among the 2q16 cohort 1.6% of eyes exhibited 2-step worsening and among the 2q8/PRN, 0% eyes exhibited worsening, but at week 100, among the 2q16 cohort 4.5% exhibited worsening and among the 2q8/PRN group 2.4% exhibited worsening.
Therapy with aflibercept among eyes with moderately severe to severe NPDR is a risk reduction, not a risk elimination strategy, with a need for ongoing monitoring as well as ongoing therapy.
Apart from changes in DRSS level, fewer eyes treated with intravitreal aflibercept vs. sham injections developed PDR, NVI, and/or CI- DME through week 100: 16.3% in the 2q16 group and 19% in the 2q8/PRN group compared with 50% in the control group.
As mentioned above, despite the demonstrated reduced risk of vision threatening complications, some treated patients developed complications, which underscores the need for ongoing follow-up with or without anti-VEGF therapy.
Although there was no difference in the systemic adverse event rate among the cohorts, it is not clear that large scale deployment of anti-VEGF therapy among such patients will not result in a safety signal. Thus, physicians are advised to heed the warnings on the labels of FDA-approved anti-VEGF therapy regarding potential systemic adverse events and to individualize treatment recommendations accordingly.
The take home message of this study is that anti-VEGF therapy can reduce the risk of progression of diabetic retinopathy as judged from DRSS, and also can reduce the risk of vision threatening complications such as PDR, NVI and CI-DME. To be effective, it seems that this therapy must be administered regularly, at least during the first 2 years of treatment, after a suitable loading dose. However, despite prophylactic anti-VEGF therapy benefits on DRSS and the risk of vision-threatening complications/CI-DME, there was no demonstrated benefit on visual outcome in the pre-planned analysis of data. In addition, despite the demonstrated benefits, some treated patients progressed to develop worsening of diabetic retinopathy severity and vision threatening complications/CI-DME, albeit at a lower rate than the control group. Additional follow-up is needed to determine whether the documented anatomical gains from anti-VEGF therapy will result in visual benefit with continued prophylactic therapy.
Marco A. Zarbin, MD, PhD
Professor and chair, Institute of Ophthalmology and Visual ScienceUMDNJ-New Jersey Medical School
Disclosures: Zarbin reports consulting for Boerhinger Ingelheim, Frequency Therapeutics, Genentech/Roche, Iduna, Iveric Bio, Life Biosciences, Novartis Pharma AG, Ophthotech, Perfuse Therapeutics and Selphagy.