Pivotal clinical trials, such as RISE/RIDE¹ and VIVID/VISTA,² have established the critical role of anti-VEGF (vascular endothelial growth factor) intravitreal injections in the treatment of diabetic macular edema (DME), a condition for which the world’s 463 million people with diabetes mellitus are at risk.³ More recently, there has been increasing use of anti-VEGF injections for proliferative diabetic retinopathy (PDR). There is also a growing body of evidence to suggest that anti-VEGF injections may be beneficial in the management of nonproliferative diabetic retinopathy (NPDR) regardless of DME presence or absence. Because patients’ quality of life and visual function can be negatively impacted as DR severity worsens,^4,5 it is important to continue developing evidence-based treatment algorithms that will allow retina specialists to optimize DR disease states in a clinically meaningful way. The latest data concerning the use of intravitreal anti-VEGF for nonsurgical PDR and NPDR will be reviewed here.

ANTI-VEGF FOR PROLIFERATIVE DIABETIC RETINOPATHY

Proliferative diabetic retinopathy is a major cause of visual impairment among American adults aged 20-74 years.⁶ Based on the findings from the DRS and ETDRS studies, panretinal photocoagulation (PRP) became the standard of care for PDR for many decades.^7,8 Early DME studies highlighted that many patients who were treated with ranibizumab were concurrently experiencing considerable improvements in their Diabetic Retinopathy Severity Score (DRSS).^9,10 These observations, along with those from several short-term studies,^11,12 provided the impetus for further exploration into the use of anti-VEGF in PDR.

Protocol S

The DRCR Retina Network Protocol S study randomized 394 study eyes with PDR (with or without center-involving DME [CI-DME]), no prior PRP, and best-corrected visual acuity (BCVA) of approximately ≥20/320 Snellen equivalent to intravitreous ranibizumab 0.5 mg (as frequently as every 4 weeks based on a structured retreatment protocol) or PRP.¹³ Ranibizumab injections for DME were performed in both groups at baseline and at investigator discretion thereafter. The primary outcome measured in Protocol S was mean visual acuity (VA) change at 2 years, and ranibizumab was found to be noninferior to PRP (+2.8 ETDRS letters in the ranibizumab group vs +0.2 in the PRP group). Several secondary outcomes favored ranibizumab over PRP, including the mean treatment group difference in VA area under the curve (4.2, 95% CI +3.0 to +5.4, P<.001), visual field sensitivity (-23 dB vs. -422 dB, P<.001), development of DME (9% vs 28%, P<.001), and rate of vitrectomy (4% vs 15%, P<.001). The 5-year outcomes of Protocol S included 305 participants who received a mean of 19.2 injections in the ranibizumab group and 5.4 injections in the PRP group; the mean change in VA letter score was 3.1 for ranibizumab and 3.0 letters for PRP.¹⁴ The ranibizumab group had lower rates of developing vision-impairing DME, but the degree of visual field loss approached that of the PRP group. Rates of severe vision loss or serious complications of PDR were similarly low in both groups. The authors conclude that both ranibizumab and PRP are viable treatments for PDR.

CLARITY

The CLARITY study was a phase 2b, single-blind noninferiority trial conducted in the United Kingdom.¹⁵ Two hundred thirty-two patients with untreated or post-laser-treated active PDR and baseline BCVA of ≥54 ETDRS letters were randomized to intravitreal aflibercept (3 monthly doses followed by a pro re nata [PRN] regimen) or PRP. One difference between CLARITY and Protocol S was that patients with DME were excluded in CLARITY. The primary outcome was BCVA at 52 weeks with a prespecified noninferiority margin of -5 ETDRS letters. Aflibercept was found to be noninferior and superior to PRP in both the modified intention-to-treat and the per-protocol populations with a mean BCVA difference of 3.9 letters (95% CI 2.3 to 5.6, P<.0001) and 4.0 letters (95% CI 2.4 to 5.7, P<.0001), respectively. There was also lower incidence of macular edema, Esterman visual field loss, and vitreous hemorrhage in the injection group. Patients in the aflibercept group received a median of 1 injection in the 40 weeks after the loading phase. The investigators conclude that patients with PDR who were treated with intravitreal aflibercept had a better outcome at 1 year compared to those treated with PRP.

Considerations in Management Strategies for PDR

The evidence supporting anti-VEGF as an effective treatment for PDR is robust, but treatment frequency and patient compliance are critical factors to consider. A 2-step regression in severity of DR was seen in nearly half of patients in Protocol S, both at 2 years and 5 years, despite a decreasing number of injections over time,¹⁴ and similar improvements in DR levels were also observed in CLARITY.¹⁵ In Protocol S, the mean (SD) number of injections required in year 1 was 7.1 (2.2); in years 2 to 5, patients continued to require 2.9 to 3.3 injections annually on average. In eyes that were able to defer anti-VEGF at any point in the first 2 years, 62% resumed injections within 16 weeks of deferral; almost half of eyes in both groups developed vitreous hemorrhage over the 5 years, emphasizing the importance of regular follow-up.

Unfortunately, many patients with PDR demonstrate poor follow-up compliance; the 5-year completion rate in Protocol S was only 66%¹⁴ and another study reported a lost-to-follow-up (LTFU) rate of 25.4% at 4 years.¹⁶ Patients with PDR who are LTFU after anti-VEGF treatment appear to have worse anatomic and visual outcomes compared to those who are LTFU after PRP.^17,18 Some patients may even suffer irreversible vision loss from complications such as neovascular glaucoma or tractional retinal detachment; thus, patients with PDR treated exclusively with anti-VEGF should be carefully selected.

Panretinal photocoagulation and anti-VEGF are not mutually exclusive and, in the real world, many patients are receiving a combination of both. Whether monotherapy or a combined approach is pursued requires a careful discussion between physician and patient, and factors such as DME status, patient compliance, cost, treatment-associated risks, and presence of macula-threatening tractional retinal detachment should all be considered.

ANTI-VEGF FOR NONPROLIFERATIVE DIABETIC RETINOPATHY

The ETDRS study established PRP as the standard of care for high-risk PDR, but it also evaluated the timing of laser and provided important insights into the rates of DR progression without treatment. In eyes with moderate NPDR, severe NPDR, and early PDR (all without clinically significant macular edema), early photocoagulation and deferred treatment both resulted in low rates of severe vision loss at 5 years.⁷ Considering the potential adverse effects of PRP, observation was followed in most of these cases. However, the progression of DR to vision-threatening levels is not trivial. The 5-year risk of developing PDR was 44.3% in eyes with moderate NPDR (DRSS level 43), 66.4% for moderately severe NPDR (DRSS level 47), and 75% to 81% for severe NPDR (DRSS level 53A-D).¹⁹ Given the high risk of vision loss associated with PDR, the advent of anti-VEGF has led researchers to revisit the prospect of intervening at earlier stages of DR.

RISE/RIDE Open-Label Extension Study

In the RISE/RIDE phase 3 registration trials for ranibizumab in the treatment of DME, more than 36% of ranibizumab-treated eyes demonstrated a ≥2-step improvement in DRSS levels compared to only 5% in the sham group. This response was most pronounced in eyes with DRSS levels 47 to 53 (moderately severe and severe NPDR), with approximately 80% being able to achieve this degree of improvement at 2 years.¹⁰ Similar effects were observed in the VIVID/VISTA phase 3 clinical trials for aflibercept, with more than one-third of eyes treated with monthly aflibercept experiencing a ≥2-step improvement in DRSS levels at the 100-week visit compared to only one-tenth of eyes treated with focal laser.²⁰

The open-label extension (OLE) study of RISE/RIDE sought to evaluate the durability of DR regression with a decrease in the frequency of ranibizumab, although it is critical to note that retreatment was based on DME, not DR.²¹ Of the 759 patients originally randomized in RISE/RIDE, 500 (of which 367 had evaluable DR at months 36 and 48) completed the final 36-month visit and enrolled into the OLE study where they could receive ranibizumab 0.5 mg PRN based on predefined DME retreatment criteria; the average number of injections received in the OLE study was 3.5. The main outcome measure was the change in DR severity from months 36 to 48.

Approximately one-fourth of patients (121/500) did not require any additional ranibizumab injections. Among those not requiring additional ranibizumab (88/367), 57% to 78% experienced DR stability; 0% to 7% experienced ≥2-step improvement; 22% to 36% experienced ≥2-step worsening. Among patients requiring additional ranibizumab injections (279/367), 84% to 94% experienced DR stability; 2% experienced ≥2-step improvement; 3% to 14% experienced ≥2-step worsening. Regardless of change in DR severity, VA improvements were sustained on average.²¹ The investigators concluded that even though more than 70% of patients maintained DR level improvements after switching from monthly to PRN ranibizumab, a substantial proportion also experienced DR worsening. Thus, careful monitoring is necessary, especially of those who receive no further injections.

PANORAMA

PANORAMA is a double-masked, randomized, phase 3 multicenter trial, and it is the only large prospective trial studying the effect of anti-VEGF on NPDR without DME for which data has been reported. This study of 402 patients with moderately severe to severe NPDR (DRSS score 47 or 53), BCVA of approximately Snellen ≥20/40, no CI-DME, and no history of anti-VEGF therapy randomized subjects to sham, aflibercept every 8 weeks (q8) after 5 loading injections, or aflibercept every 16 weeks (q16) after 4 loading injections. The primary endpoint was the proportion of patients improving ≥2 DRSS steps at 24 weeks (for combined aflibercept arms) and 52 weeks (for individual aflibercept arms). At 52 weeks, 79.9% of patients treated q8 and 65.2% of patients treated q16 (P<.0001 for both) improved by ≥2 DRSS steps vs only 15% of patients in the sham arm. Additionally, fewer patients in the aflibercept arms (3% to 3.7% vs 20.3% in sham, P<.0001) developed PDR/anterior segment neovascularization or CI-DME (6.7% to 8.2% vs 25.6% in sham, P<.001).²²

In year 2, patients in the q16 arm continued to receive injections at the scheduled interval while those in the q8 arm were transitioned to a PRN regimen based on investigator-determined DRSS level; if DRSS was ≥35 (mild NPDR or worse), injection was given. On average, patients in the q16 arm received 7.8/9 potential injections while patients in the q8/PRN arm received 10.3/9 to 15 potential injections through week 100. At 100 weeks, 50% of patients treated q8/PRN and 62.2% of patients treated q16 (P<.0001 for both) improved by ≥2 DRSS steps vs only 12.8% of patients in the sham arm. This means that more than 92% of eyes that achieved a ≥2-step DRSS improvement at year 1 maintained these improvements through week 100 in spite of decreased dosing. Mean change in BCVA was similar across all groups. Similar to the 52-week results, significantly fewer patients in the aflibercept arms (6.0% to 8.1% vs 27.1% in sham, P<.001) developed PDR/anterior segment neovascularization or CI-DME (10.4% to 13.4% vs 33.1% in sham, P<.001). In a subgroup analysis of patients in the q8/PRN arm, having 0 injections in year 2 correlated with an increased likelihood of developing PDR/anterior segment neovascularization or CI-DME; this should be interpreted cautiously due to small numbers and potential undertreatment in the PRN arm. The proportion of patients who worsened by ≥2 DRSS steps through week 100 was 20.2%, 4.5%, and 2.4% in the sham, q16, and q8/PRN arms, respectively.²³

Protocol W

This study is being conducted by the DRCR Retina Network, and compares intravitreous aflibercept with sham in the prevention of PDR or CI-DME in eyes with severe NPDR and BCVA approximately Snellen ≥20/25. Those randomized to aflibercept will receive 3 loading doses followed by injections every 4 months for 2 years; patients will be followed for a total of 4 years. Visual outcomes in eyes that receive early aflibercept vs those that are treated only if high-risk PDR or CI-DME with vision loss develops will be evaluated along with secondary objectives such as DRSS scores, macular thickness, and treatment-associated costs. Study completion is anticipated in the spring of 2022.

Considerations in Management Strategies for NPDR

The efficacy of anti-VEGF injections in lowering DR levels is apparent (Figure 1), but 2 major questions are still outstanding: (1) What duration and frequency of treatment is necessary in order to sustain DRSS improvements? and (2) Does the reversal in DR level severity equate to true disease modification? In other words, have patients who have been therapeutically induced to go from a DRSS of 53 (severe NPDR) to a DRSS of 35 (mild NPDR) actually reduced their probability of vision-threatening complications? Because of the limited time horizon and post-hoc nature of many of the studies that have assessed anti-VEGF for NPDR, these uncertainties linger as we await answers from future research.

Both the RISE/RIDE OLE and PANORAMA studies suggest that repeated injections are necessary to maintain benefits. In the second year of PANORAMA, patients who were treated with aflibercept q16 weeks continued at that frequency, but those who were initially treated with aflibercept q8 weeks transitioned to a DRSS-guided PRN schedule. While BCVA was similar between these 2 arms, more patients (62.2%) in the q16 arm improved by ≥2 DRSS steps at week 100 than did patients in the q8/PRN arm (50%) despite the opposite being observed at week 52 (65.2% in q16 vs 79.9% in q8). If anti-VEGF injections are needed indefinitely to perpetuate their positive effects, then careful patient selection will be indicated given the aforementioned challenges with compliance in this population. This will be especially true if there are any untoward consequences of treatment interruptions for NPDR, and the possibility of this remains unclear.

Whether anti-VEGF-driven DRSS regression represents a true reversal of disease is also unclear. In Protocol S, the ranibizumab group initially had less visual field deterioration than did the PRP group, but lost most of that advantage by year 5. Why this happened is not fully understood, but it may point to similar degrees of ischemia between the 2 groups. In a substudy of CLARITY, intravascular oxygen saturation at 52 weeks did not differ between the anti-VEGF and PRP arms.²⁴ In the RISE/RIDE OLE study, patients with injection-induced DRSS ≤43 were compared to those with DRSS ≤43 at RISE/RIDE enrollment and who were randomized to sham. The former exhibited a greater risk of DR progression compared to the latter.²⁵ Further investigation is necessary to elucidate whether anti-VEGF-driven DRSS improvements are more than just phenotypic alterations and actually represent redirection of the DR disease course.

CONCLUSION

The role of anti-VEGF injections in the treatment of PDR and NPDR, even in the absence of DME, is expanding. For the treatment of PDR, anti-VEGF is a valid option, although patient compliance is key; PRP remains an important part of the therapeutic armamentarium. Utilization of anti-VEGF in the treatment of NPDR (without concurrent DME) is a newer concept and would certainly represent a paradigm shift from the watch-and-wait approach that has been the longstanding standard of care. More will be learned with continued research about the durability and disease-modifying effects of anti-VEGF. For now, a personalized decision about incorporating anti-VEGF to treat DR should be made for each individual patient. RP

REFERENCES

1. Brown DM, Nguyen QD, Marcus DM, et al. Long-term outcomes of ranibizumab therapy for diabetic macular edema: the 36-month results from two phase III trials: RISE and RIDE. Ophthalmology. 2013;120(10):2013-2022.
2. Heier JS, Korobelnik JF, Brown DM, et al. Intravitreal aflibercept for diabetic macular edema: 148-week results from the VISTA and VIVID studies. Ophthalmology. 2016;123(11):2376-2385.
3. International Diabetes Federation. Diabetes facts & figures. Accessed June 15, 2020. https://www.idf.org/aboutdiabetes/what-is-diabetes/facts-figures.html
4. Willis JR, Doan QV, Gleeson M, et al. Vision-related functional burden of diabetic retinopathy across severity levels in the United States. JAMA Ophthalmol. 2017;135(9):926-932.
5. Mazhar K, Varma R, Choudhury F, et al. Severity of diabetic retinopathy and health-related quality of life: the Los Angeles Latino Eye Study. Ophthalmology. 2011;118(4):649-655.
6. Centers for Disease Control and Prevention. Vision Health Initiative. Accessed June 15, 2020. https://www.cdc.gov/visionhealth/basics/ced/index.html
7. Early Treatment Diabetic Retinopathy Study Research Group. Early photocoagulation for diabetic retinopathy. ETDRS report number 9. Ophthalmology. 1991;98(5 Suppl):766-785.
8. The Diabetic Retinopathy Study Research Group. Photocoagulation treatment of proliferative diabetic retinopathy. Clinical application of Diabetic Retinopathy Study (DRS) findings, DRS Report Number 8. Ophthalmology. 1981;88(7):583-600.
9. Ip MS, Domalpally A, Sun JK, et al. Long-term effects of therapy with ranibizumab on diabetic retinopathy severity and baseline risk factor for worsening retinopathy. Ophthalmology. 2015;122:367-374.
10. Wykoff CC, Eichenbaum DA, Roth DB, Hill L, Fung AE, Haskova Z. ranibizumab induces regression of diabetic retinopathy in most patients at high risk of progression to proliferative diabetic retinopathy. Ophthalmol Retina. 2018;2:997-1009.
11. Filho JA, Messias A, Almeida FP, et al. Panretinal photocoagulation (PRP) versus PRP plus intravitreal ranibizumab for high-risk proliferative diabetic retinopathy. Acta Ophthalmol. 2011 Nov; 89(7):e567-e572.
12. Tonello M, Costa RA, Almeida FP, Barbosa JC, Scott IU, Jorge R. Panretinal photocoagulation versus PRP plus intravitreal bevacizumab for high-risk proliferative diabetic retinopathy (IBeHi study). Acta Ophthalmol. 2008 Jun; 86(4):385-389.
13. Gross JG, Glassman AR, Jampol LM, et al; Writing Committee for the Diabetic Retinopathy Clinical Research Network. Panretinal photocoagulation vs intravitreous ranibizumab for proliferative diabetic retinopathy: a randomized clinical trial. JAMA. 2015;314(20):2137-2146.
14. Gross JG, Glassman AR, Liu D, et al; Writing Committee for the Diabetic Retinopathy Clinical Research Network. Five-year outcomes of panretinal photocoagulation vs intravitreous ranibizumab for proliferative diabetic retinopathy: a randomized clinical trial. JAMA Ophthalmology. 2018;136(10):1138-1148.
15. Sivaprasad S, Prevost AT, Vasconcelos JC, et al; CLARITY Study Group. Clinical efficacy of intravitreal aflibercept versus panretinal photocoagulation for best corrected visual acuity in patients with proliferative diabetic retinopathy at 52 weeks (CLARITY): a multicentre, single-blinded, randomized, controlled, phase 2b, non-inferiority trial. Lancet. 2017;389:2193-2203.
16. Obeid A, Gao X, Ali FS, et al. Loss to follow-up in patients with proliferative diabetic retinopathy after panretinal photocoagulation or intravitreal anti-VEGF injections. Ophthalmology. 2018;125(9):1386-1392.
17. Obeid A, Su D, Patel SN, et al. Outcomes of eyes lost to follow-up with proliferative diabetic retinopathy that received panretinal photocoagulation versus intravitreal anti-vascular endothelial growth factor. Ophthalmology. 2019;126(3):407-413.
18. Wubben TJ, Johnson MW; Anti-VEGF Treatment Interruption Study Group. Anti-vascular endothelial growth factor therapy for diabetic retinopathy: consequences of inadvertent treatment interruptions. Am J Ophthalmol. 2019;204:13-18.
19. Early Treatment Diabetic Retinopathy Study Research Group. Fundus photographic risk factors for progression of diabetic retinopathy. ETDRS report number 12. Ophthalmology. 1991;98(5 suppl):823-833.
20. Dhoot DS, Baker K, Saroj N, et al. Baseline factors affecting changes in diabetic retinopathy severity scale score after intravitreal aflibercept or laser for diabetic macular edema: post hoc analyses from VISTA and VIVID. Ophthalmology. 2018;125(1):51-56.
21. Sun JK, Wang P, Taylor S, Haskova Z. Durability of diabetic retinopathy improvement with as-needed ranibizumab: open-label extension of RIDE and RISE studies. Ophthalmology. 2019;126:712-720.
22. Boyer DS. Treatment of moderately severe to severe nonproliferative diabetic retinopathy with intravitreal aflibercept injection: 52-week results from the phase 3 PANORAMA Study. Paper presented at: the Association for Research in Vision and Ophthalmology Annual Meeting; April 29, 2019; Vancouver, BC, Canada.
23. Lim, JI. Intravitreal aflibercept injection for nonproliferative diabetic retinopathy: year 2 results from the PANORAMA study. Invest Ophthalmol Vis Sci. 2020;61:1381.
24. Nicholson L, Crosby-Nwaobi R, Vasconcelos JC, et al. Mechanistic evaluation of panretinal photocoagulation versus aflibercept in proliferative diabetic retinopathy: CLARITY substudy. Invest Ophthalmol Vis Sci. 2018;59:4277-4284.
25. Goldberg, RA. What happens to diabetic retinopathy severity scores with less aggressive treatment? A post hoc analysis of the RISE and RIDE open label extension study. Paper presented at: the American Society of Retina Specialists Annual Meeting; July 29, 2019; Chicago, IL.