Smaller infants, aggressive disease necessitate evolution in ROP treatment

Retinopathy of prematurity, a neovascular retinal disorder of childhood, is a leading cause of visual morbidity worldwide. Currently, ROP is treated with anti-VEGF injections and/or retinal laser photocoagulation, both of which are effective but pose potentially difficult long-term outcomes. Some experts agree that better treatment guidelines and increased patient monitoring can make these early therapies even more effective.

Advances in medicine have allowed infants to be born at younger gestational ages, which has consequently fundamentally changed the disease of ROP, according to Darius M. Moshfeghi, MD, professor and chairman of the retina service at Byers Eye Institute at Stanford University.

ROP manifests when abnormal blood vessels grow and spread throughout the retina. These abnormal blood vessels are typically fragile and can leak, which can scar the retina and pull it out of position.

Traditionally, ophthalmologists would see a classic progression of ROP in infants from stage 1 to stage 2 to stage 3. Now, with infants being born at younger gestational ages, the disease presents differently, he said.

Source: Kathryn Sill, Web and Communications Specialist, Stanford Department of Ophthalmology

“Retinopathy of prematurity is the only disease I’ve seen in my career where the disease itself has changed. ... Most patients I treat now are aggressive posterior ROP. They’re very small, early babies,” he said.

ROP primarily occurs in infants with a low birth weight, less than 1,250 g, or those born before 31 weeks of gestation. According to the National Eye Institute, approximately 3.9 million infants are born in the United States each year; about 28,000 weigh less than 1,250 g, and between 14,000 to 16,000 of these infants are affected by ROP to some degree.

Factors affecting treatment

Treating ROP depends on the zone, the stage, the severity, and the infant’s age and weight.

Three zones define the area of ROP involvement: Zone I encircles the optic disc; zone II is annular shaped and surrounds zone I, extending nasally to the ora serrata; and zone III is the remaining crescent of retina located temporally.

Severity is rated in five stages. According to the American Academy of Ophthalmology, stage 1 is the presence of a line of demarcation between vascularized and nonvascularized retina. Stage 2 is when the ridge of demarcation rises above the retinal surface and small isolated areas of neovascularization may be present on the retinal surface. Stage 3 is when extraretinal fibrovascular proliferation is observed or neovascularization extends from the ridge into the vitreous. If untreated, the disease can progress to stage 4, partial retinal detachment, and stage 5, total retinal detachment.

Retinal laser photocoagulation is performed by a surgeon when the infant is sedated. The procedure reduces proangiogenic factors by destroying ischemic tissue. Early treatment with retinal ablation, as evaluated in the ETROP study, significantly reduced unfavorable outcomes. The study suggested that eyes that are on the threshold of or progress to type 1 ROP should be treated within 48 to 72 hours, whereas a “wait and watch” approach can be recommended for type 2 ROP. The study defined type 1 ROP as zone I, any stage ROP with plus disease; zone I, stage 3 ROP without plus disease; or zone II, stage 2 or 3 ROP with plus disease. Type 2 was defined as zone I, stage 1 or 2 ROP without plus disease, or zone II, stage 3 ROP without plus disease. Plus disease includes the presence of increased venous dilation and arteriolar tortuosity, which can lead to vitreous haze, impaired pupil dilation and iris vascular engorgement.

However, retinal laser photocoagulation has risks. It has been associated with negative sequelae, including visual field loss, and patients may have a greater risk for developing high myopia.

Zone and stage are important factors to consider when developing a treatment protocol, Caroline R. Baumal, MD, of Tufts University School of Medicine, New England Eye Center, said.

“For example, I have seen extremely premature infants presenting with very posterior ROP in zone I and the retinal vasculature is just barely past the fovea. In these infants, I prefer to start with anti-VEGF injection, which will allow continued retinal vasculature growth more anteriorly into zone II,” she said.

The tempo of disease progression, the severity of plus disease and systemic neonatal issues are important to consider when treating ROP, she said.

Whichever treatment an ophthalmologist chooses for ROP, success should depend on the complete regression of all active retinal neovascularization without evidence of retinal detachment or distortion of the posterior pole anatomy. This also includes retinal dragging and ectopic fovea, Baumal said.

Skillful laser therapy

Retinal laser photocoagulation is a skill that is highly variable, Moshfeghi said, pointing to the results of the phase 3 RAINBOW study, which compared outcomes of ROP patients treated with Lucentis (ranibizumab, Genentech) or laser surgery. Patients were treated with either 0.2 mg or 0.1 mg of ranibizumab or underwent laser surgery. The ranibizumab 0.2 mg group had an 80% success rate, the ranibizumab 0.1 mg group had a 75% success rate, and the laser surgery group had a 66.2% success rate.

“That’s catastrophically bad laser outcomes,” Moshfeghi said. “I have to imagine that there was a lot of variability in laser technique.”

Retinal laser photocoagulation is largely dependent on the surgeon’s skill. Those who have been mentored properly or who have the most experience typically have the best outcomes. However, if a surgeon lacks experience, does not sufficiently follow up with a patient or uses an improper technique, the outcomes have a greater chance of failure, Moshfeghi said.

The procedure requires attention to detail and is improved by the use of a widefield camera to ensure all areas of the avascular retina have been treated. After completing the procedure on one eye, Moshfeghi said he photographs the retina with a wide-angle 130° lens and reviews his work to identify any skip areas, treat them and re-photograph to ensure the entire avascular retina has been treated.

“The only way to objectively evaluate the procedure is with a widefield photograph. The photograph doesn’t employ tricks to get around the optics. It uses its optical wide-angle field of use and can immediately show you all the skip areas, whereas scleral depression can inadvertently hide skip areas,” he said.

Before moving on to the fellow eye, surgeons must treat those areas, repeat the photograph, review and go back again to treat any other remaining areas. Only after completing all of these steps can work begin on the fellow eye, Moshfeghi said.

Furthermore, infants should be evaluated 2 to 3 days after laser therapy to ensure that the treatment was adequate and there is no further progression of ROP or an inflammatory response.

Anti-VEGF treatment

Unlike laser therapy, the success of anti-VEGF therapy is not dependent on a surgeon’s skill or technique once the medication is in the patient’s eye, Moshfeghi said.

Anti-VEGFs target and reduce elevated intraocular levels of VEGF. The anti-VEGF medication is delivered by an intravitreal injection at bedside, often without the need for intubation.

Anti-VEGF can be effective in infants weighing less than 750 g or younger than 28 weeks’ gestational age. Laser therapy for infants in this population can be difficult, and sedating them with anesthesia makes the procedure more risky, Moshfeghi said.

However, anti-VEGF therapy is not a “one and done” procedure. Infants treated with anti-VEGF need to be followed weekly to monitor disease recurrence or infection, Clio Armitage Harper III, MD, of Austin Retina Associates, said.

Harper, who is a clinical associate professor at the University of Texas-Austin, said he prefers to use ranibizumab over Avastin (bevacizumab, Genentech) because it is a less robust molecule and dissipates from systemic circulation faster. Anti-VEGF allows the retinal vessels to continue to grow toward the retinal periphery, allowing more viable retina to be useful in the future. Laser photocoagulation, however, destroys a significant amount of retina, especially in zone I or posterior zone II, Harper said.

Anti-VEGF safety

Drug delivery via injection is an area of safety risk in anti-VEGF therapy. At the 2018 ROP Hot Topics meeting in Chicago, Harper presented data from a 2017 Retina study supporting the use of a 32-gauge 4-mm needle to deliver the medication compared with the results of a 31-gauge approximately 8-mm needle in the BEAT-ROP study and a 30-gauge approximately 12-mm needle in the RAINBOW study.

The smaller needle had a lower risk for iatrogenic retinal breaks or injury and lower risk for lenticular injury, whereas a 30-gauge 12-mm needle could possibly penetrate the lens or retina during injection, Harper said.

“It’s all about the safety of delivering the medications. We’re trying to make this as safe as possible for everyone because not everyone has the same experience with performing these injections,” he said.

After anti-VEGF injection, infants are brought back 48 to 72 hours later for an examination.

“Babies can’t tell you if they have an infection. An adult would notice if their eye was red or their vision was affected. An infection will develop in 48 to 72 hours, so it’s important to bring them back in,” he said.

The infants should be brought back for weekly exams until the retinas are fully mature or a fluorescein angiography is performed — by at least 60 weeks of gestational age — to determine if there is leakage, if neovascularization is evident or if flat neovascularization occurs as a white plaque peripherally, Harper said.

Fluorescein angiography can improve the sensitivity of diagnosis of ROP. In a 2015 study published in Ophthalmology, when fluorescein angiography was used in conjunction with color fundus photographs, sensitivity for diagnosis of stage 3 ROP or worse disease improved from 39.8% using just color fundus photographs to 74.1%.

The vast majority of infants who have had anti-VEGF injections remain avascular at 60 weeks of age. This is difficult to assess clinically without the use of fluorescein angiography, Harper said.

“I perform an FA and then treat the remaining avascular retina (about 95% of the time), especially if active leakage or neovascularization is present. It does not have to be dense, confluent laser like you would perform initially for type 1 ROP. It needs to be closer to PRP, like we would do for a diabetic, making sure to laser at the vascular-avascular zone, guided by FA, as this is where the majority of VEGF is produced,” Harper said.

Close observation after anti-VEGF treatment is necessary, Baumal said, because ROP may recur if the retina has not completely vascularized when the anti-VEGF effect “wears off,” depending on the half-life of the agent as well as the severity and zone of ROP, among other factors.

“In our series at New England Eye Center, we found that anti-VEGF injection with a low dose of 0.02 mg to 0.025 mg of ranibizumab led to immediate complete regression of ROP and allowed continued anterior normal-appearing vascularization of the retina. The anterior growth of normal-appearing retina was documented photographically by comparing fluorescein angiograms before and many weeks after ranibizumab injection. If the ROP recurred, it was in a more anterior zone and the options could include reinjection with an anti-VEGF or laser, which would be in a more peripheral location at that time,” she said.

Long-term effects

The long-term effects of anti-VEGF treatment and its systemic safety are still relatively unknown, OSN Pediatrics/Strabismus Board Member R.V. Paul Chan, MD, MSc, MBA, FACS, said.

Anti-VEGF works quickly and causes regression of the disease, but its effect on a patient is hard to predict, Chan said.

“When does recurrence happen? Or when do we see persistent avascular retina? How long do we wait until we think avascular retina should be lasered? There are a lot of unknown issues, and the looming question is one of systemic and long-term safety,” he said.

Anti-VEGFs are being used more frequently to treat ROP, but no one agent is yet approved for that indication. Safety parameters of ranibizumab treatment were reported in the CARE-ROP study, a double-masked prospective clinical trial. Three of the 20 patients in the study died, but the deaths were not thought to be related to the treatment.

In a 2018 study published in Ophthalmology, 20 bevacizumab-treated eyes examined 4 years after injection showed abnormalities at the periphery, including avascular areas, vessel leakages, shunts, abnormal vessel branching and tangles. Abnormalities were also seen in the posterior pole, including hyperfluorescent lesions and absence of foveal avascular zones.

Anti-VEGFs have been used for more than a decade to treat ROP, but the long-term effects of these medications have not yet been observed, Chan said.

For Chan, laser photocoagulation is still the typical treatment for ROP. Zone II stage 3 ROP with plus disease can respond well to laser photocoagulation. Chan considers using anti-VEGF therapy in infants who meet treatment criteria and have significant iris neovascularization, very posterior type 1 ROP or aggressive posterior ROP.

“Another consideration is how systemically well the child is. If the patient cannot tolerate a prolonged laser treatment, then anti-VEGF is a viable option because it is quick and does not cause as much morbidity or stress,” he said.

At this time, most infants with ROP that requires treatment will get laser, although there appears to be some variability in practice patterns around the U.S., J. Peter Campbell, MD, MPH, assistant professor of ophthalmology at Oregon Health and Sciences University, said. However, the benefits of anti-VEGF medication delivery and its refractive outcomes may eventually outweigh the potential long-term concerns of that treatment, he said.

“I think there is a lot we don’t yet know that may move the needle much faster toward the use of anti-VEGF once these questions are addressed,” he said.

Campbell and Moshfeghi are enrolling patients for the PROPER study, which will include 12 institutions evaluating short- and long-term ocular outcomes and long-term neural outcomes in infants with all levels of ROP. The cohort will include infants treated with anti-VEGF, laser, a combination of both or neither, Campbell said.

Both therapies

Notwithstanding the benefits and positive outcomes of anti-VEGF therapy, Moshfeghi said that most of the infants he treats with anti-VEGF eventually undergo laser photocoagulation as well. Between 5% and 10% of these anti-VEGF-treated patients will not need later laser therapy, while 90% to 95% of them will because of ongoing vascular nonperfusion.

Anti-VEGF provides an early benefit as far as vascular maturation goes, but eventually the benefit ends, which could account for failures seen in patients as late as 4 years after treatment, he said.

“Those babies have an induced VEGF dormancy in the eye, and at any point those eyes can reactivate. Not many will, but the problem is, if your baby reactivates, they can go blind. It’s a terrible thing to be blind from birth, but it’s downright cruel to have vision for 4 to 5 years and then have it taken away,” Moshfeghi said.

Better accuracy in screening and grading of the disease may improve treatment paradigms. Different physicians who look at the same infant may diagnose the disease in different zones and different stages, leading to incorrect treatment decisions, Moshfeghi said.

“It tells you that people are very poor at grading the disease in a consistent and reproducible manner,” he said.

A 2016 study in Ophthalmology demonstrated a large discrepancy in ROP classification between experts. In the prospective study, two experts evaluated each of 1,533 images and disagreed on the stage classification in 620 of the comparisons.

Future treatments

Preventive treatments for ROP may be effective moving forward. Drug trials on beta blockers, oxygen saturation curves and omega-3 fatty acids could help move research and new treatments forward, Harper said.

However, because these infants are born at a younger age and smaller size, as early as 22 weeks’ gestational age and as small as 300 g, ROP is more aggressive and more posterior than in infants born at an older age, with younger infants tending to respond less to any treatment, Harper said.

Chan said improved ROP definitions and guidelines may make a difference in treatment. Having a consensus guideline for when to use anti-VEGF and when to use laser photocoagulation would potentially improve outcomes.

“There is also no consensus of what the definition of recurrence is after anti-VEGF. In the literature, people define it in many ways. Is it just a recurrence of plus disease, the recurrence of stage 3? What does it really mean and how do we know when to treat after anti-VEGF if you have to [treat] again?” Chan asked.

In the future, artificial intelligence and imaging analysis systems may give ophthalmologists better predictive models on how infants will respond to different treatments and who is at greatest risk for disease recurrence, he said.

Despite the challenges and potential long-term effects of both anti-VEGF and laser photocoagulation, ophthalmologists need to focus on preventing blindness before worrying about the secondary outcomes of the treatment, Moshfeghi said.

“One of the worst adverse neurodevelopmental outcomes is blindness. We need to take care of the biggest one first, prevention of retinal detachment and blindness,” he said. – by Robert Linnehan

• References:
• Baumal CR, et al. Ophthalmic Surg Lasers Imaging Retina. 2015;doi:10.3928/23258160-20150422-05.
• Good WV, et al. Trans Am Ophthalmol Soc. 2004;102:233-250.
• Facts about retinopathy of prematurity (ROP). nei.nih.gov/health/rop/rop. Accessed May 7, 2019.
• Harper CA 3rd, et al. Retina. 2019;doi:10.1097/IAE.0000000000001996.
• Klufas MA, et al. Ophthalmology. 2015;doi:10.1016/j.ophtha.2015.04.023
• Lepore D, et al. Ophthalmology. 2018;doi:10.1016/j.ophtha.2017.08.005
• Mintz-Hittner HA, et al. N Engl J Med. 2011;doi:10.1056/NEJMoa1007374.
• Novartis to seek ROP indication for Lucentis. www.healio.com/ophthalmology/retina-vitreous/news/online/%7Bf5b0b439-b8b4-40ee-bb9c-d0b1af9a060f%7D/novartis-to-seek-rop-indication-for-lucentis. Published Sept. 24, 2018. Accessed May 6, 2019.
• Roohipoor R, et al. Ophthalmol Retina. 2018;doi:10.1016/j.oret.2018.01.017.
• VIDEO: Anti-VEGF agents allow broad, effective treatment for ROP, but laser still has a role. www.healio.com/ophthalmology/retina-vitreous/news/online/%7bc75aa5cc-237b-4145-8288-8009f7843193%7d/video-anti-vegf-agents-allow-broad-effective-treatment-for-rop-but-laser-still-has-a-role. Published March 27, 2019. Accessed May 3, 2019.
• Wright LM, et al. Retina. 2017;doi:10.1097/IAE.0000000000001561.

• For more information:
• Caroline R. Baumal, MD, can be reached at Tufts University School of Medicine, New England Eye Center, 260 Tremont St., Boston, MA 02111; email: cbaumal@tuftsmedicalcenter.org.
• J. Peter Campbell, MD, MPH, can be reached at Casey Eye Institute – Marquam Hill, 3375 SW Terwilliger Blvd., Portland, OR 97239-4197; email: johnpetercampbell@gmail.com.
• R.V. Paul Chan, MD, MSc, MBA, FACS, can be reached at Department of Ophthalmology and Visual Sciences, Illinois Eye and Ear Infirmary, University of Illinois at Chicago, 1855 W. Taylor St., Room 3.138, Chicago, IL 60612; email: rvpchan@uic.edu.
• Clio Armitage Harper III, MD, can be reached at Austin Retina Associates, 801 W. 38th St., Suite 200, Austin, TX 78705; email: caharper@austinretina.com.
• Darius M. Moshfeghi, MD, can be reached at Byers Eye Institute, Stanford University of School of Medicine, 2452 Watson Court, Palo Alto, CA 94303; email: dariusm@stanford.edu.

Disclosures: Baumal reports she is a speaker for Genentech, Zeiss, Alcon and Novartis. Campbell reports the PROPER study is funded by a research grant from Genentech. Chan reports he is a consultant for Genentech, Allergan, Alcon and BeyeOnics and is on the scientific advisory board for Visionix. Harper reports he is an equity owner in Pr3vent. Moshfeghi reports he is on the board of directors of and has equity in 1-800 Contacts; does consulting for ROP for Akebia; is on the data safety monitoring board at Alcon; is on the scientific advisory board at Allegro; is site PI for Apellis; is on the ROP steering committee at Bayer; is a consultant for Congruence Medical Solutions; is founder and on the board of directors of and has equity in dSentz; is on the steering committee at Iconic Therapeutics; is on the scientific advisory board at Irenix; has equity in Grand Legend Technology; is founder of, has equity in and is on the board of directors of Linc; is on the data safety monitoring board at Novartis; is founder and on the board of directors of and has equity in Pr3vent; is a CME consultant for Prime; is founder and on the board of directors of and has equity in Promisight; is on the scientific advisory board at and has equity in Pykus; is a CME consultant for and on the ROP steering committee at Regeneron; is a ROP expert witness for Shapiro Law; is founder of and has equity in Versl; is a CME consultant for Vindico; and is on the scientific advisory board at and has equity in Visunex Medical Systems.

Click here to read the Point/Counter to this Cover Story.