Research Article - Journal of Clinical Ophthalmology (2020) Volume 4, Issue 2

Endolaserless vitrectomy with aflibercept for proliferative diabetic retinopathy-related vitreous hemorrhage (LASER LESS TRIAL):1-Year Results.

¹Southeast Retina Center, Augusta 30909, Georgia, USA

²Morehouse School of Medicine, 720 Westview Drive SW, Atlanta, Georgia, USA

Corresponding Author:

Dr. Dennis M Marcus
Southeast Retina Center 3685 Wheeler Rd
Ste 201 Augusta, GA 30909, USA
Tel: (706) 650-0061
Fax: (706) 650-0064
E-mail: dmarcus@southeastretina.com

Accepted date: June 02, 2020

Citation: Marcus DM, Starnes D, Walia H, et al. Endolaserless vitrectomy with aflibercept for proliferative diabetic retinopathy-related vitreous hemorrhage (LASER LESS TRIAL):1-Year Results. J Clin Ophthalmol 2020;4(2):254-265.

Abstract

Objective: We report the 1-year safety and efficacy results of vitrectomy without endolaser for proliferative diabetic retinopathy (PDR)-related vitreous hemorrhage (VH).

Methods: All eyes received one preoperative and intraoperative IAI (2 mg). The q8week group received postoperative IAI every 4 weeks through week 16 followed by q8week IAI. The q16week group received postoperative IAI every 4 weeks through week 8 followed by q16week IAI. All patients were examined every 4 weeks; PRN IAI for PDR progression or diabetic macular edema (DME) was allowed.

Results: Thirty-one eyes from 40 patients were randomized. Through 52 weeks, endophthalmitis, progression of traction retinal detachment, iris/angle neovascularization and neovascular glaucoma were not observed. The q8week and q16week groups received an average of 8.4 and 5.4 injections, respectively, through 52 weeks. Adverse events at any time through 52 weeks such as worsened visual acuity>30 letters (6 eyes), new rhegmatogenous retinal detachment (1 eye), and recurrent VH (4 eyes) occurred infrequently and were more common in the q16week group. Preoperative average visual acuity (VA) was 37 letters (20/200) for randomized eyes. Endolaserless vitrectomy resulted in statistically significant 52-week visual acuity 33 letter gain to 72 letters (20/40). Visual acuity outcomes favored (not statistically significant) the q8week group where average acuity was 77 letters (20/32) with a 52-week 40 letter gain versus 66 letters (20/50) with a 52-week 24 letter gain in the q16week group.

Conclusion: Endolaserless vitrectomy with aflibercept demonstrates 52-week safety with significant VA improvement.

Keywords

Endolaser, Pars plana vitrectomy, Proliferative diabetic retinopathy, Anti-vascular endothelial growth factor, Vitreous hemorrhage.

Abbreviations

EPDR: Proliferative Diabetic Retinopathy; VH: Vitreous Hemorrhage; DME: Diabetic Macular Edema; BCVA: Best Corrected Visual Acuity; VEGF: Vascular Endothelial Growth Factor; PRP: Pan-Retinal Photocoagulation; PPV: Pars Plana Vitrectomy; IAI: Intravitreal Aflibercept Injection; DRCR: Diabetic Retinopathy Clinical Research; DRVS: Diabetic Retinopathy Vitrectomy Study; HVF: Humphrey Visual Field; OCT: Optical Coherent Tomography; VA: Visual Acuity

Introduction

Intravitreal injection of anti-vascular endothelial growth factor (VEGF) agents is now established as optimal therapy for centerinvolved diabetic macular edema (DME) [1-4]. Anti-VEGF’s ability to reduce retinal neovascularization during DME treatment has led to investigation of anti-VEGF therapy for PDR [5-9] and PDR-related vitreous hemorrhage [10-13]. For over 4 decades, PRP has remained the standard therapy for PDR. However, anti- VEGF therapy has emerged as an alternative therapy for PDR eyes not requiring vitrectomy [5,7-9]. Diabetic Retinopathy Clinical Research (DRCR) Protocol S compared anti-VEGF ranibizumab to PRP for PDR not requiring vitrectomy [5,8].

DRCR Protocol S concluded that their findings supported either anti-VEGF therapy or PRP as viable PDR treatment [8]. The PROTEUS study showed that adding ranibizumab to PRP was more effective than PRP alone in regressing neovascularization in eyes with high-risk PDR [9]. The CLARITY trial further supports anti-VEGF aflibercept therapy in PDR eyes [7]. When PDR was present in eyes being treated for DME in DRCR Protocol T, aflibercept resulted in greater PDR regression rates compared to ranibizumab and Bevacizumab [6].

Given the above information, intravitreal aflibercept injection (IAI) may represent a useful therapy before, during, and after vitrectomy for PDR-related vitreous hemorrhage. We previously reported the short-term 4-month results with endolaserless vitrectomy and aflibercept for eyes with PRP-naïve PDR-related VH [11-13]. Herein, we report 52-week safety and efficacy results for our expanded cohort of 40 eyes.

Materials and Methods

This is a phase I/II, open-label, randomized, prospective, single center interventional study. Eligible subjects were identified and provided with a copy of informed consent. Informed consent documentation and relevant supporting information was submitted and approved by the Institutional Review Board (IRB)/Ethics Committee (EC).

Study population

Adult subjects with Type 1 or Type 2 diabetes mellitus and PDR-related VH requiring vitrectomy. Vitrectomy need was determined by a non-study, standard of care visit prior to the screening study visit. Exclusion criteria are listed in Table 1.

Table 1. Exclusion criteria.

Study design

Only one eye per patient was eligible and was randomly assigned after vitrectomy with equal probability to either a q8week or q16week treatment group. Figure 1 presents the injection and randomization schedule.

Figure 1: Aflibercept (2mg) Treatment Schedule: 21 randomized patients with PRP-naïve eyes undergoing vitrectomy for PDR-related vitreous hemorrhage randomized in 1:1 ratio. PRN: pro re nata; Scr:
screen; SX: surgery; POD1: post-operative day 1; POWK: post-operative week 1.

Intraoperative methods and assessment

Eyes underwent 23-gauge PPV. After removal of VH, eyes were intraoperatively evaluated for randomization eligibility. Eyes ineligible for randomization, as determined intraoperative, did not receive intraoperative IAI, but did receive intraoperative endolaser and were followed postoperatively.

Visit schedule

Figure 2 summarizes the visit schedule.

Figure 2: Visit Schedule. FA: Fluorescein Angiography; FP: Fundus Photography; Scr: screen; SX: Surgery; POD1: post-operative day 1; POWK1: post-operative week 1.

Post-operative PDR treatment and monitoring

PDR was monitored and assessed for progression by postoperative clinical exam and q12 week standard 7-field fundus photography and Optos wide-field fluorescein angiography/photography. If progression of PDR occurred at a visit where IAI was not mandatory, IAI was required at that visit and in 4 weeks. If regression of the progressed PDR was not evident after 2 consecutive monthly IAI, then PRP could be administered. Criteria for PDR progression and regression are summarized in Table 2.

Table 2. Criteria for PDR progression and regression.

DME treatment and monitoring

Eyes with clinical and optical coherent tomography (OCT) DME in the q8week group were to receive appropriate IAI per label as part of the mandatory schedule through 16 weeks. Starting at week 12 and week 20 in the q16 and q8week groups, respectively, eyes were eligible to receive additional 2mg IAI (monthly) for the treatment of DME (Table 3).

Table 3. Criteria for additional monthly IAI for DME.

Outcome Measurements

Primary outcomes

Primary safety outcomes are summarized in Table 4.

Table 4. Primary safety outcomes.

Secondary outcomes

Secondary outcomes are summarized in Table 5. Qualitative angiographic outcomes were evaluated based on size and leakage intensity of hyper fluorescence. Qualitative photographic outcomes were based on photographic DME/exudate size and thickness.

Table 5. Secondary outcomes.

Visual field outcomes were evaluated by the mean cumulative score and change for the combined 30-2 and 60-4 Humphrey visual field (HVF) decibel thresholds from week 4 to week 52.

Results/Observations

Study participants

Forty eyes from 40 subjects were enrolled. Baseline demographics are presented in Table 6. All eyes received preoperative IAI at enrollment. Two intraoperative retinal tears, in two eyes, were treated with intraoperative retinal cryopexy. Five of the 31 randomized eyes required membrane peel for nonmacular traction. Average surgical time was 21 minutes (range: 12-47 minutes) for randomized eyes. Twenty-five of 31 (81%) randomized subjects completed their 52-week follow-up visit. Figure 3 summarizes patient randomization and retention.

Figure 3: Participant randomization and retention. Last observation carried forward (LOCF) from week 48 was used for this participant for week 52 analysis.

Table 6. Baseline Demographics for 31 randomized eyes. One subject declined vitrectomy after enrollment and receipt of pre-operative IAI. This subject continues follow-up for safety as a nonrandomized eye. One subject was unable to undergo vitrectomy after enrollment and receipt of pre-operative IAI due to accelerated hypertension at time of surgery. This subject has been lost to follow-up since the decision to not undergo surgery. Thirty-eight of 40 enrolled eyes underwent vitrectomy surgery. Seven of those 39 eyes were not randomized due to intraoperative evidence of previous peripheral retinal ablation (2 eyes), traction retinal detachment involving the macula (2 eyes), significant nasal fibrovascular proliferation and noncentral nonmacular traction retinal detachment (1 eye), cataracts and retrolental hemorrhage (1 eye), and intraoperative tears (1 eye). These seven non-randomized eyes received intraoperative endolaser and did not receive intraoperative IAI and continue follow-up for safety as nonrandomized eyes.

Primary safety outcomes

Adverse primary safety outcomes are summarized in Table 7. All adverse events are summarized in Table 8. Treatment schedules for two subjects experiencing recurrent vitreous hemorrhages are presented in Figures 4 and 5.

Figure 4: Vitreous hemorrhage case presentation, VH: Vitreous Hemorrhage.

Figure 5: Vitreous hemorrhage case presentation; VH: Vitreous Hemorrhage.

Table 7. Primary safety outcomes results through 52 weeks.

Table 8. Adverse events.

Injection treatment requirement and compliance

An average of 8.4 (range: 2 to 10) and 5.4 (range: 2 to 9) total injections though 52 weeks were administered to the q8week and q16week treatment groups, respectively (Table 9).

Table 9. Follow-up and injection compliance.

Visual acuity outcomes

Endolaserless vitrectomy with IAI resulted in statistically significant VA gain for all eyes and in both groups (Table 10, Figure 6). Baseline preoperative VA letter score for 31 randomized eyes was 37 letters (20/200) (range: 0 to 84 letters). For 26 randomized eyes in both groups with 52-week followup, baseline preoperative VA letter score was 40 letters (20/160) (range: 0 to 84 letters). At 52 weeks, the average best corrected visual acuity (BCVA) score improved to 72 letters (20/40) with an average change in BCVA of +33 Early Treatment Diabetic Retinopathy Study (ETDRS) letters from baseline (range: -10 to 90 letters) (p<0.00004; one tailed t-test). For randomized eyes in the q8week group with 52-week follow-up, average BCVA improved to 77 letters (20/32) with an average letter gain of +40 ETDRS letters (range: -10 to 90 letters) (p<0.0009; one-tailed t-test). For randomized eyes in the q16week group with 52-week follow-up, average BCVA improved to 66 letters (20/50) with an average letter gain of +24 letters (range: -6 to 85 letters) (p<0.0097; one-tailed t-test). Baseline letter score did not show a statistically significant difference between the two groups (35 letters q8week and 42 letters q16week) (p<0.882; two-tailed t-test). Letter score change from baseline favored the q8week group (40 letter gain q8week versus 24 letter gain q16week) but this was not statistically significant (p<0.480; two-tailed t-test).

Figure 6: Average visual acuity.

Table 10. Visual acuity change from baseline.

OCT Outcomes

Baseline OCT was performed at 4 weeks postoperatively in 31 randomized eyes with no statistically significant difference between the two groups but thinner in the q16week group (336 um q8week group and 282 um q16week group) (p<0.408; twotailed t-test). Endolaserless vitrectomy with IAI resulted in statistically significant OCT thinning from week 4 to week 52 for all eyes in the q8week but not in the q16week group (Table 11, Figure 7). For 26 randomized eyes at 52-week follow-up, average OCT central subfoveal thickness (CST) was 264 um (range: 147 to 338 um), which displayed an average thinning of 33 um from the baseline OCT 4-week postoperative visit (p<0.000014; onetailed t-test). For randomized eyes in the q8week group with 52-week follow-up, average 52 week OCT CST was 270 um with an average thinning of 53 um from baseline (p<0.004; onetailed t-test). For randomized eyes in the q16week group with 52 week follow-up, average 52 week OCT CST was 255 um with an average thinning of 12 um from baseline (p<0.165; one-tailed t-test). Fifty-two week OCT thickness change (53 um thinning q8week group and 12 um thinning q16week group) from baseline favored the q8week group (not statistically significant) (p<0.312; two-tailed t-test)

Figure 7: Average OCT values.

Table 11. Average OCT values through 52 weeks post-operatively.

Eight of 31 randomized eyes (5 in q8week group and 3 in q16week group) demonstrated DME intraoperatively. At 4 weeks and 52 weeks postoperatively, OCT CST>300 um was observed in 11 of 31 randomized eyes (7 in q8week group and 4 in q16week group) and 4 of 25 randomized eyes (2 in q8week group and 2 in q16week group), respectively. Clinical DME at 52 weeks postoperatively was observed in 3 of 25 eyes (1 in the q8week group and 2 in the q16week group) (Table 12).

Table 12. Clinical DME evaluation and incidence.

Fundus Photographic (FP) and Wide-Field Fluorescein angiographic (FA) grading

Fundus photographic and wide-field fluorescein angiographic grading is presented in Table 13.

Table 13. Fundus photographic and wide-field fluorescein angiographic grading.

Humphrey visual field outcomes

Sample size for HVF score analysis was a limitation in our analysis. Also, due to equipment malfunction, not all patients were able to undergo visual field testing at their week 52 visit (Table 14). Baseline mean total HVF point score combining 30-2 and 60-4 was 2573 dB [SD, 658 dB] at week 4 for 28 randomized eyes. For the 21 total randomized eyes from both groups with both baseline and 52-week HVF testing, average total score was 2269 dB [SD, 649 dB] at week 52 with an average change of -440 dB (p<0.007; one tailed t-test). Additional secondary treatment outcomes are summarized in Table 15.

Figure 8: Case presentation.

Table 14. Humphrey visual field.

Table 15. Additional secondary outcomes.

Non-randomized eyes Demographics and outcomes for the 9 non-randomized eyes are summarized in Table 16.

Table 16. Baseline demographics, visual acuity, and 52-week outcomes for non-randomized subjects.

Discussion

Our 1-year results indicate safety, moderate-term durability, and significant VA improvement after endolaserless vitrectomy with aflibercept for PDR-related VH. Through 52 weeks, endophthalmitis, progression of traction retinal detachment, iris/ angle neovascularization, and neovascular glaucoma were not observed. Adverse events at any time through 52 weeks occurred infrequently and were more common in the q16week group. VA improved soon after surgery with a 52-week letter gain of 33 letters. VA loss at week 52 occurred in 5 eyes (3 from DME recurrence, 0 from proliferative consequences, and 2 of unclear etiology).

VA 52-week outcomes favored the q8week group with a 40 letter gain versus a 24 letter gain in the q16week group. While the safety and acuity outcomes tend to favor the q8week group, a larger cohort and longer follow-up is required to determine the optimal dosing regimen.

Our 1-year endolaserless PPV VA results are consistent with previous one year results from the DRIVE-UK study which retrospectively evaluated vitrectomy with endolaser for a nonclearing VH subgroup of 60 eyes (excluding traction retinal detachment) [14]. Other retrospective studies of vitrectomy with endolaser for PDR-related VH (without traction) yield no better visual outcomes than our results [14-17]. Aflibercept with endolaserless vitrectomy safety and efficacy results appear comparable to using endolaser while avoiding vitrectomy for the 3 q16week eyes and 1 q8week eye with recurrent vitreous hemorrhage by administering pro re nata (PRN) aflibercept and presumably reducing other endolaser PRP side-effects such as DME exacerbation or visual field compromise. Our visual field results are limited due to small numbers but indicate some overall visual field loss (statistically significant). Given the DRCR Protocol S finding of reduced visual field scores from years 2 to 5 in the ranibizumab group [8], planned longer-term evaluation is necessary in our cohort to document our long-term visual field outcomes. We also demonstrated macular thinning in both groups and a low incidence of OCT>300 um at 52 weeks. While much of our visual gain is likely due to clearance of vitreous hemorrhage by PPV alone, aflibercept-related macular thinning may result in greater acuity gains observed in the q8week group.

Post-operative wide-field fluorescein angiography, employed in our approach, is a key component for neovascularization monitoring. After endolaserless PPV, PDR progression at any time through 52 weeks, was observed in 21% and 29% of q8week and q16week eyes, respectively, despite mandatory aflibercept injections. Interestingly, absence of neovascularization at 52 weeks was found in 71% of q8week eyes but in only 35% of q16week eyes. These findings speak to the issue that while anti- VEGF therapy can cause regression of neovascularization, it is likely that neovascularization will recur without some form of maintenance anti-VEGF. While it has been proposed that PRP may be a more durable treatment for PDR and neovascularization than anti-VEGF therapy, results from our major PDR trials for eyes not requiring vitrectomy have not borne this out. Through 1 year, 65% and 6% of CLARITY study PRP eyes required additional PRP and vitrectomy, respectively [14]. Through 5 years, 54% and 22% of DRCR Protocol S PRP eyes developed vitreous hemorrhage and need for vitrectomy, respectively [8]. The above trials monitored neovascularization via clinical examination and did not monitor with wide-field or standard fluorescein angiography, which may account for potential undertreatment and relatively high rates of vitreous hemorrhage. The DRCR Protocol S required at least 1200 spots of laser in the PRP group which is significantly less than the 1600 spot requirement established by the DRS and ETDRS trials decades ago. Thus, under-treatment with PRP is another possible cause for relatively high rates of proliferative consequences in our more recent PDR trials. This may also limit the support for advocates of a “lighter endolaser PRP” during PPV. In the PROTEUS study, which evaluated PDR eyes with high-risk characteristics not requiring vitrectomy, complete neovascularization regression was observed in 44% in eyes receiving both PRP and ranibizumab versus only 25% in the PRP monotherapy group [9]. Thus far, evidence does not support the notion that adding endolaser to aflibercept or endolaser alone would necessarily provide improved longterm durability or safety for PDR eyes requiring vitrectomy for vitreous hemorrhage. It is also possible that removal of the vitreous scaffold may be somewhat protective for proliferative consequences even without administering endolaser PRP.

Nevertheless, there understandably remains reticence for adopting endolaserless PPV especially for more advanced PDR. Aflibercept demonstrates superior outcomes to other anti-VEGF agents for DME eyes with worse VA and for regression of PDR [1,2,6]. In DRCR Protocol S, eyes with greater baseline severity of PDR demonstrated better visual outcomes with ranibizumab than PRP [18]. These data may reduce concerns for omitting endolaser for more severe PDR. Given the absence of any other anti-VEGF agent use for this endolaserless approach, it remains unclear whether aflibercept is the optimal agent. Concerns are escalated especially for patients at risk for non-compliance with follow-up visits[19] even within a structured clinical trial, such as DRCR Protocol S, only 66% of patients completed the 5-year visit [8]. Our 1-year noncompliance rates are equally concerning. One initially non-compliant patient returned later without visual compromise or complications. Longer-term follow-up of compliant and non-compliant endolaserless patients may help determine critical outcomes after poor adherence to visits. It is also concerning that half of our nonrandomized patients were noncompliant after undergoing PPV with endolaser. In addition to compliance concerns, cost, systemic health, and frequency of visits remain a consideration when choosing a treatment approach [8].

The strengths of our study include the prospective nature, use of ETDRS BCVA, modest but significant number randomized, two clinically relevant dosing regimens, visual field assessment, protocol specific PRN dosing criteria based on clinical, OCT, fundus photographic, and, importantly, wide-field fluorescein angiographic analyses. In addition, our inclusion criteria required that vitrectomy was indicated and planned prior to enrollment. Baseline acuities were wide ranged and our population is likely representative of real-world eyes.

The limitations of our study include a modest number and only 1-year data. Our study received approval to include 40 eyes with 3 years duration. This will address the ultimate safety of both dosing regimens and compliance rates. Other limitations include an absence of a control group such as standard or limited or “lighter” peripheral endolaser PRP, PRN dosing, or comparison to other anti-VEGF agents. Our randomized eyes with “simple” vitreous hemorrhage and exclusion of traction macular detachment may have a favorable prognosis with other therapies. Our study intends to evaluate safety and potential efficacy of this new surgical approach and is not intended in a smaller pilot study to make comparisons to vitrectomy with endolaser, observation, anti-VEGF injections without surgery etc.

Conclusion

Our favorable results do not necessarily support widespread adaptation of endolaserless vitrectomy. Additional study follow- up and widening experience will determine the continued balance and contemplation between anti-VEGF therapy and endolaser PRP [12].

Acknowledgments

Siobhan Ortiz, Thomas Bailey, and Ken Ivey at Southeast Retina Center, provided technical and data support. Dr. Dennis Marcus has full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Meeting Presentation

This work was presented at the American Society of Retina Specialists, Boston, MA-August 2017; Association for Research in Vision and Ophthalmology, Baltimore, MD-May 2017; American Academy of Ophthalmology, New Orleans, LA- November 2017 and American Society of Retina Specialists, Vancouver, BC-July 2018.

Financial Support

Regeneron Pharmaceuticals Inc. was responsible for manufacturing and supplying the drug used in the study. The investigator-initiated study received clinical research grant support from Regeneron to conduct the trial. The sponsor was Southeast Retina Center and Dr. Marcus was responsible for the design of the study, conducting the study, data collection, data management, data analysis, interpretation of the data, preparation, and approval of the manuscript. Regeneron, the funding organization, reviewed the manuscript prior to submission.

Conflict of Interest

Dr. Dennis M. Marcus has served as an advisor or consultant for Regeneron Pharmaceuticals, Genentech/Roche; as a speaker or member of the speakers bureau for Regeneron Pharmaceuticals; received grants for clinical research from Regeneron Pharmaceuticals; and reports pharmaceuticalsponsored clinical research from Allergan, Alcon, Aerpio, Kalvista, Ionis, Mylan, Samsung, Novartis, Opthea, Chenghdu, Clearside, Astellas, Allegro, Alimera, Ophthotech/Iveric, Outlook, Gemini, Genentech, ThromboGenics, Tyrogenex, Graybug, Topcon, Optos, Xplore, Gyroscope, Stealth Spiam, Aerie, Apellis, Roche, Novartis, OHR, Xplore, Regenxbio, Kodiak , Zeiss, and Regeneron Pharmaceuticals. Dr. Harinderjit Singh has received grants for clinical research from Regeneron Pharmaceuticals and reports pharmaceutical sponsored clinical research from Allergan, Alcon, Clearside, Astellas, Allegro, Alimera, Ophthotech Genentech, ThromboGenics, Tyrogenex, Apellis, Roche, Novartis, OHR, and Regeneron Pharmaceuticals. The remaining authors have no relevant financial disclosures. Dr. Lalane has received grants for clinical research and reports pharmaceutical-sponsored clinical research from Allergan, Alcon, Clearside, Astellas, Allegro, Alimera, Ophthotech, Genentech, ThromboGenics, Tyrogenex, Apellis, Roche, Novartis, OHR, Samsung, Ironside, Chengdhu, Mylan, and Regeneron Pharmaceuticals.

References

1. Wells JA, Glassman AR, Ayala AR, et al. Aflibercept, Bevacizumab, or Ranibizumab for Diabetic Macular Edema: Two-Year Results from a Comparative Effectiveness Randomized Clinical Trial. Ophthalmology. 2016;123:1351-9.
2. Diabetic Retinopathy Clinical Research Network, Wells JA, Glassman AR, et al. Aflibercept, bevacizumab, or ranibizumab for diabetic macular edema. N Eng J Med. 2015;372:1193-203.
3. Nguyen QD, Brown DM, Marcus DM, et al. Ranibizumab for diabetic macular edema: results from 2 Phase III randomized trials: RISE and RIDE. Ophthalmology. 2012;119:789-801.
4. 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;123:2376-85.
5. Writing Committee for the Diabetic Retinopathy Clinical Research Network, Gross JG, Glassman AR, et al. Panretinal photocoagulation vsintravitreous ranibizumab for proliferative diabetic retinopathy: A randomized clinical trial. JAMA. 2015;314:2137-46.
6. Bressler SB, Liu D, Glassman AR, et al. Change in Diabetic Retinopathy Through 2 Years: Secondary Analysis of a Randomized Clinical Trial Comparing Aflibercept, Bevacizumab, and Ranibizumab. JAMA Ophthalmol. 2017;135:558-68.
7. Sivaprasad S, Prevost AT, Vasconcelos JC, et al. 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, randomised, controlled, phase 2b, non-inferiority trial. Lancet. 2017;389:2193-203.
8. Gross JG, Glassman AR, Liu D, et al.  Five-Year Outcomes of Panretinal Photocoagulation vsIntravitreous Ranibizumab for Proliferative Diabetic Retinopathy: A Randomized Clinical Trial. JAMA Ophthalmol. 2018;136:1138-48.
9. Figueira J, Fletcher E, Massin P, et al. Ranibizumab Plus Panretinal Photocoagulation versus Panretinal Photocoagulation Alone for High-Risk Proliferative Diabetic Retinopathy (PROTEUS Study). Ophthalmology. 2018;125:691-700.
10. Diabetic Retinopathy Clinical Research Network. Randomized clinical trial evaluating intravitreal ranibizumab or saline for vitreous hemorrhage from proliferative diabetic retinopathy. JAMA Ophthalmol. 2013;131:283-93.
11. Marcus DM, Singh H, Starnes DC, et al.  Endolaserless vitrectomy with intravitreal aflibercept injection for proliferative diabetic retinopathy-related vitreous hemorrhage (LASER LESS TRIAL). J Vitreoretin Dis. 2018;2:127-37.
12. D’Amico, D. From the Editor-in-Chief. J Vitreoretin Dis. 2018;2(3):125-126.
13. Zhao XY, Xia S, Chen YX. Antivascular endothelial growth factor agents pretreatment before vitrectomy for complicated proliferative diabetic retinopathy: a meta-analysis of randomised controlled trials. Br J Ophthalmol. 2018;102;1077-85.
14. Gupta B, Sivaprasad S, Wong R, Laidlaw A, et al. Visual and anatomical outcomes following vitrectomy for complications of diabetic retinopathy: the DRIVE UK study. Eye (Lond). 2012;26:510–6.
15. Hadayer A, Mueller BH II, Adeniran JF, et al. Outcomes of surgically managed proliferative diabetic retinopathy. J Vitreoretin Dis. 2018;2:338-42.
16. Mikhail M, Ali-Ridha A, Chorfi S, et al. Long-term outcomes of sutureless 25-G+ pars-plana vitrectomy for the management of diabetic tractional retinal detachment. Graefes Arch ClinExpOphthalmol. 2017;255:255-61.
17. Ozone D, Hirano Y, Ueda J, et al. Outcomes and complications of 25-gauge transconjunctivalsutureless vitrectomy for proliferative diabetic retinopathy. Ophthalmologica. 2011;226:76-80.
18. Bressler SB, Beaulieu WT, Glassman AR, et al. Panretinal photocoagulation versus ranibizumab for proliferative diabetic retinopathy: Factors associated with vision and edema outcomes. Ophthalmology. 2018;125:1776-83.
19. 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;126:407-13.