Commentary: Antivascular endothelial growth factor and retinopathy of prematurity.

Endophthalmitis following intravitreal antivascular endothelial growth factor (VEGF) drugs is sufficiently rare,[1] but it is unfortunate if occurs bilaterally in premature infants receiving bevacizumab for retinopathy of prematurity (ROP). Although the bevacizumab eliminates the angiogenic threat (BEAT)-ROP study has proven the use of bevacizumab for the use in Zone 1 and posterior Zone 2 ROP, it was underpowered for safety analysis.[2] Many studies have pointed to the immediate and long-term complications of bevacizumab, including recurrence of ROP, vitreous hemorrhage, progression of retinal detachment, rapid myopia progression, and systemic effects such as bowel ischemia and neurodevelopmental delays.[34] A recent Cochrane review concluded that insufficient data preclude strong conclusions favoring routine use of intravitreal anti-VEGF agents in preterm infants with Type 1 ROP.[5] Developing countries like India which are seeing the mushrooming of neonatal Intensive Care Units (ICUs) may witness an epidemic of ROP due to increasing survival rates of extremely preterm infants born even as early as 24 weeks’ gestation.[6] These facts may lead to greater use of intravitreal bevacizumab for treating aggressive posterior ROP (APROP) (Type 1 ROP), with associated complications.

Laser has its own advantages in treating ROP and is still the preferred modality of treatment in ROP involving Zone II anterior and Zone III disease. It can be completed in one sitting, and if done well, it leads to regression of ROP within 6 weeks. Compared to bevacizumab, recurrence rates of ROP are markedly lesser with laser, at follow-up of 5 years, thus avoiding need for periodic evaluation till complete vascularization, as required with bevacizumab.[4] The disadvantages of laser in ROP include a long learning curve, the risk of sedation, and relatively longer duration to complete the laser sitting. In addition, the early treatment ROP trial showed that in Type 1 and Zone 1 ROP, 30% patients had unfavorable outcome.[7] The disease in Zone 1 leads to ablation of almost two-third of the retinal periphery leading to field restriction, inflammation a high incidence of myopia.[8]

The advantages of anti-VEGF injections in ROP are that injections can be administered quickly in both eyes under mild sedation. The procedure can be done on the bedside in the neonatal ICU if required. The response to anti-VEGF injections is rapid with the plus disease disappearing within a day and the vitreous haze clearing with the progression of normal vascularization. However, the ophthalmologist needs to be trained to follow-up these children till the vascularization progresses up to the ora serrata. It is advisable to periodically monitor these children, preferably with photographic documentation, and take them up for laser ablation if vascularization ceases to progress for >2 weeks or if there is development of Stage 3 disease. A caveat is that it may take up to 60–70 weeks’ gestational age for the vascularization to complete; hence, careful observation is mandatory until this occurs. It is challenge to examine bigger babies who may require sedation to allow thorough peripheral retinal evaluation.

The risks of the anti-VEGF should not be forgotten as this paper on endophthalmitis highlights (REF IJO paper).[9] Fortunately, there have been very few such cases reported in literature thus far,[10] yet examining clinician should look for signs of infection at every visit in these premature babies.

Anti-VEGF is a double-edged sword as the neural, vascular, lung dependent of these neonates is driven by the VEGF.[11] The long-term effects of these drugs in infants are unknown, and the use of off-label drugs in these infants can lead to litigation and should always be born in mind. The other big conundrum is which drug and what dose to choose. Avastin has the advantage of low cost but it leaks into the systemic circulation for almost 2 weeks. Ranibizumab has the advantage of shorter half-life and hence reduces the systemic side effects but there is a chance of recurrence of ROP.

The dosage in BEAT ROP for bevacizumab was half of the adult dose 0.625 mg (10,000 higher than the amount of VEGF in the vitreous of these small infants). Hence, a de-escalation dose study needs to be done to see therapeutic effects of 0.25 mg of avastin. There has been a surge in the various ROP trials published; they all have shown meaningful structural regression rates of ROP. Almost 700 infants and 1400 eyes have been treated in these trials from 2012 to 2017. However, as doctors, we should remember our motto of primum non nocere before embarking on treatment. The importance of informed consent cannot be emphasized more in ROP as anti-VEGF use is an off-label use; hence, we need to take the parents in confidence and have a detailed discussion about both the pros and cons of both laser and anti-VEGF. Reactivation of disease, crunch phenomena postinjection, and systemic implications of anti-VEGF with the need for longer follow-up should be clearly described to the parents.

Hence, the anti-VEGF consensus stands as it should not be used as a primary treatment of ROP but be reserved for cases of APROP (i.e., Type 1 ROP) where laser is not possible due to poorly dilating pupils and vitreous haze. Or it can be used after laser in APROP cases where there is a high risk for retinal detachment progression and where there is a risk to foveal development and the disease is recalcitrant.

In the future, laser might not necessarily remain the gold standard indefinitely, especially for posterior progressing and asymmetric ROP. However, the questions remain such as should anti-VEGF be used in the Zone 2 ROP, use it before or after laser, is earlier better, and what is the lowest safe dose.