Anti-vascular Endothelial Growth Factor: the Future Treatment of Choroidal Neovascularization in Pathologic Myopia?
Zhu Hong,Wang Feng-hua,Sun Xiao-dong
DOI: https://doi.org/10.3760/cma.j.issn.0366-6999.20130188
IF: 6.133
2013-01-01
Chinese Medical Journal
Abstract:Myopia is the most common refractive disorder. High myopia affects 27%-33% of all myopic eyes in Asia.1,2 The pathologic myopia (PM) is the most severe vision-threatening phenotype of high myopia.3 It is also the second most common cause of choroidal neovascularization (CNV) in Asia. Unlike age-related macular degeneration (AMD) which mostly effecting elders, PM causes severe vision loss in young adults, resulting in a significant impairment of their working ability.4 PM has become the second leading cause of low vision and blindness particularly among those aged at 40-49 years in some Asia countries.3 DEFINITION OF PATHOLOGIC MYOPIA Pathologic myopia5-7 is usually defined with eyes has -6 diopters (D) or more with typical fundus pathologic changes such as peripapillary chorioretinal atrophy, tigroid fundus (choroidal vessels being visible through the retina, often with lacquer cracks in Bruch's membrane), geographic atrophy of the retinal pigment epithelium (RPE) and choroid (diffuse or patchy), posterior staphyloma, lacquer cracks, choroidal neovascularization (CNV), spontaneous subretinal haemorrhages, and Fuchs' spot. Among these changes, CNV is the most vision threatening complication of myopia and develops in around 5.2%-10.2% of high myopic eyes.5-7 In particular, PM imposes an even greater public health burden because of its vision threatening complications such as CNV, retinal degeneration and increased risk of glaucoma, cataract, and retinal detachment.4,5 These vision threatening complications cannot be prevented by optical or surgical correction. EPIDEMIOLOGY OF PATHOLOGIC MYOPIA In recent years, evidences have shown that the prevalence of myopia is increasing rapidly in populations of Northeast Asian, Southeast Asian, Caucasian and Australian Aboriginal origin. In Asia, population-based studies on children and adults have been carried out in both northern and southern India. Bourne et al8 reported that the prevalence of high myopia in Bangladeshi adults was 1.8%. The prevalence of myopia is quite low in comparison to those reported from East Asia. Xu et al9 reported that prevalence of myopia of in urban and rural adults of northern China population of high myopia (<-6.0 D) was 4.1%. In Singapore, Wong10 reported 9.1% prevalence of high myopia among adult Chinese in Singapore. In Taiwan, a national survey revealed that the subjects between the ages of 16 and 18 years have a rate of myopia of 84%. The prevalence of high myopia is 18% among young male and 24% among young female in Taiwan, China. In the west, a large population-based study of people aged 4-74 years in the USA showed that 3.2% had high myopia (-5.01 to -10.00 D), and 0.2% had extreme myopia (<-10.00 D).11 In the Israeli population, the prevalence of high myopia significantly increased from 1.70% in 1999 to 2.05% in 2002. In the Melbourne Visual Impairment Project,12 it was found the prevalence of high myopia 2%, and extreme myopia 0.3%. These problems not only affected in adults, but also in school age children. Myopic refractive error is becoming a major cause of reduced vision in school-aged children. The prevalence of myopia in young adolescent eyes has increased substantially over recent decades. It is now approaching 10%-25% and 60%-80%, respectively, in industrialized societies of the West and East.13,14 Myopia is considered to be a complex disease involving familial inheritance genetic origin and environmental factors. There is compelling evidence to suggest that myopia is a hereditary condition. Evidence is also strongly supported that myopia may be under environmental influences. The rapid increase in the prevalence of myopia over the last several decades suggests that environmental factors are important. It was significantly associated with younger age, urban region, higher education, management occupations and socioeconomic factors. CNV IN PATHOLOGIC MYOPIA PM remains one of the leading causes of CNV in people younger than 50 years and the second commonest cause of CNV after AMD.15 Most myopic CNV is subfoveal or juxtafoveal “classic” lesion with minimal subretinal fluid or exudation.16 Myopic CNV is associated with significant socioeconomic setbacks in many countries, since it is commonly seen in younger working populations.4 It is still unclear the initial stimulation that produce neovascularization in PM. The development of CNV in PM is probably a function of the attenuated blood supply to the thin choroid. Based on previous research with CNV in other diseases, the growth of CNV is most likely the result in the overproduction of VEGF or an imbalance of VEGF levels.17 Natural history studies4,18,19 showed that the visual prognosis of CNV secondary to PM is generally extreme poor though it shows a self-limiting course after ten years follow up. Yoshida and associates4 studied 25 consecutive patients (27 eyes) with myopic CNV who were observed without treatment for more than 10 years. Most of the patients lost visual acuity over the course of the study. At initial examination, 22.2% of eyes had a visual acuity of greater than 20/40, while 29.6% of eyes had a visual acuity of less than 20/200. However, at final examination, only one eye retained a visual acuity of greater than 20/40 and 96.3% of eyes had a visual acuity of less than 20/200 during the follow-up of more than 10 years. At 5 and 10 years after onset, all CNV were observed to have regressed completely, becoming flat and sometimes unrecognizable. Hayashi et al20 observed 57 eyes with myopic CNV who were followed for at least 5 years after the onset of CNV. Among 57 eyes, 14.0% patients had a final visual acuity better than 20/40. 64.9% patients had a final visual acuity worse than 20/200. 91.9% patients developed chorioretinal atrophy. Most studies showed that the finial visual acuity worse than 20/200 was between 60% and 73% with different follow up period. Secrétan et al19 reported the results of 50 laser treated and 50 non-treated patients with myopic CNV after 5-10 years follow-up. The difference between the mean decrease in VA of the treated and non-treated groups was no longer significant after 8 and 10 years. After long-term observation with 325 eyes, Ohno-Matsui5 also found that the incidence was higher (34.8%) among the fellow eyes of patients with pre-existing CNV than among eyes of patients without pre-existing CNV (6.1%). So it is clear that treatments need to be developed to help eyes with myopic CNV avoid progressive visual impairment. TREATMENT OPTIONS FOR MYOPIC CNV There are some options used for treatment of myopic CNV. These include thermal laser photocoagulation, surgical management (surgical excision of CNV and macular translocation), transpupillary thermotherapy (TTT) and photodynamic therapy (PDT) with verteporfin. Laser photocoagulation for CNV has been associated with several potentially vision-threatening complications such as inadvertent foveal ablation, sub-retinal hemorrhage, and retinal pigment epithelial (RPE) tears.21,22 Laser scar expansion is a well-recognized and severe long-term complication after laser treatment in eyes with PM, even in eyes free of CNV recurrence. It was reported that expansion of laser scars occurred in 92% to 100% of myopic eyes treated with various wavelengths. A high rate of CNV recurrence is another major problem associated with thermal laser treatment. Lacquer cracks arising after laser treatment are often observed and may be a risk factor for CNV recurrence. Recurrence rates varying from 31% to 72% have been reported, with recurrent CNV arising most commonly from the edge of laser scars. TTT involves long-exposure, large spot size irradiation of the CNV using an 810 nm diode laser, associated with a lower thermal output than traditional laser photocoagulation. TTT may represent another alternative for the management of myopic CNV.23 Because diode laser energy absorbed depends greatly upon the amount of melanin, the most effective light absorber in the chorioretinal layers and highly attenuated RPE and choroids in PM, it is necessary to get appropriate laser settings for patients of differing racial background case by case. It is easier that TTT may cause damage to the neurosensory retina.24 Ultimately, a randomized prospective trial evaluating TTT for the treatment of CNV in PM of varying racial backgrounds may be necessary. Surgical methods were also developed to removal of the CNV membrane with or without limited or 360 degree macular translocation. Different approaches of submacular surgery for CNV in PM have been evaluated in several studies with differing outcomes.25,26 Machemer and Steinhorst27 first developed macular translocation surgery for repositioning the neurosensory retina overlying the CNV onto an area of healthier RPE and choriocapillaris. Most studies were small case series and had conflicting results. One of the main disadvantages of macular translocation with limited or 360 degree retinotomy is more extensive surgical manipulations. Other complications include retinal detachment, proliferative vitreoretinopathy, postoperative diplopia, recurrent CNV, and severe hypotony. Therefore confirmation of long term efficacy and safety of macular translocation surgery await further assessment, though it is effective treatments for myopic CNV.28 Verteporfin (Visudyne, QLT Inc./Novartis Ophthalmics) PDT has been widely investigated in AMD as well as in PM.29-31 PDT involves the administration of an intravenous photosensitive dye that is thought to accumulate preferentially in the CNV. The CNV is then exposed to non-themal irradiation using a 689 nm diode laser. This wavelength activates the accumulated dye and occludes the CNV. Until now, the veteporfin PDT is the only therapy with evidences from randomized large scale clinical trial. PDT was first successfully used for the treatment of CNV in the treatment of age-related macular degeneration with photodynamic therapy study group (TAP) trials.29 The subsequent verteporfin in photodynamic therapy (VIP) clinical trial evaluated the use of PDT for PM.29,31 In the VIP trial, 120 patients with subfoveal CNV associated with PM were randomly assigned to verteporfin therapy (81 eyes) or placebo (39 eyes). For entry, the greatest linear dimension of the lesion had to be no more than 5400 microns, with a best-corrected visual acuity of approximately 20/100 or better. After 1-year follow up, VIP study Group reported that 58 (72%) of the PDT-treated patients compared with 17 (44%) of the placebo-treated patients lost fewer than eight letters (P <0.01), including 26 (32%) versus 6 (15%) improving at least five letters (>/=1 line). Seventy (86%) of the PDT-treated patients compared with 26 (67%) of the placebo-treated patients lost fewer than 15 letters. The visual acuity of the verteporfin-treated group had a greater chance of remaining stable compared with the placebo-treated group. However, by 2-year of follow up, statistical significance of the treatment benefit was lost. Twenty-nine of 81 PDT-treated patients (36%) compared with 20 of 39 placebo-treated patients (51%) lost at least 8 letters (P=0.11). Schnurrbusch, Pece et al reported similar results.32 In view of the possible ethnic influence on the results of the VIP study, Lam et al30 conducted a prospective, non-comparative study of 31, 11 ethnic Chinese, respectively with sub-foveal and juxtafoveal CNV secondary to PM. In subfoveal group after 24 months of follow-up, the median visual acuity improved by 1.7 lines, although the mean visual acuity remained unchanged at the baseline level; 14 (63.6%) eyes had stable or improved BCVA while 6 (27.3%) eyes gained more than three lines of vision. The mean number of PDT treatments required in the first 2 years, 2.3, was considerably less than the mean of 5.1 treatments required in the VIP study. These studies in Asians showed that visual outcomes of PDT for myopic CNV achieved similar to those in Caucasian populations but with less retreatment. Visual acuity recovery correlates with the number of PDT re-treatments and ages. The results of difference between the TAP study and the VIP study are possibly because the nature history difference between neovascular AMD and PM.31 PDT is most effect for predominantly classic CNV in AMD, the most aggressive form of neovascular AMD. However, only 80% of PM eyes enrolled in the VIP study contained predominantly classic CNV and there's a greater tendency for CNV in PM to be self-limited, so the treatment benefit may be less apparent. In addition, PDT may not occlude the CNV in PM as well as the CNV in AMD. A histopathologic evaluation of CNV after PDT showed only partial vascular occlusion, and Moshfeghi et al33 concluded that PDT therapy did not appear to lead to permanent and complete occlusion of the CNV. The complications of PDT for CNV in PM include delay in choroidal perfusion, occlusion of large choroidal vessels, lacquer crack formation and sub-retinal fibrosis.34 Myopic eyes may be more susceptible to PDT-induced chorioretinal damage. Myopic eyes may be predisposed to post-treatment lacquer cracks, recurrent CNV, and RPE atrophy with sub-equent loss of visual acuity. ANTI-VASCULAR ENDOTHELIAL GROWTH FACTOR (VEGF) TREATMENT AND MYOPIC CNV VEGF-A, also known as vascular permeability factor, is a homodimeric protein that is a potent stimulator of vascular endothelial cell growth, functions as a survival factor for newly formed vessels, facilitates the recruitment of leukocytes by acting as a chemotactic factor, and induces vascular permeability. These biological activities give it a central role in angiogenesis, both in normal and pathologic conditions. Inappropriate over expression of VEGF has been hypothesized to play a key role in the growth of pathologic neovascularization in AMD, diabetic retinopathy, retinal vein occlusion and other vascular eye diseases.17,35,36 Despite differences in pathogenesis among the AMD and PM processes, there is evidence that the concentration of VEGF in aqueous humor increases in PM. It suggests that VEGF may also play a key role in the development of CNV in PM just as that in other neovascular eye diseases.17,35 Available vascular endothelial growth factor inhibitors, include pegaptanib, bevacizumab, ranibizumab, VEGF-trap and conbercept, have been used for eye disease treatment. All the agent can treat CNV via inhibit the activity of VEGF, especially VEGF-A. Although there is no evidence from randomized clinical trials to prove the efficacy and safety of these drugs in treating neovascular PM, many finished retrospective and perspective researches showed promising results. The visual outcome of these therapies has been reported to be better than the natural history of the condition.36-42 Ranibizumab (rhuFabV2, Lucentis; Novartis, Switzerland), a humanized monoclonal antibody fragment, is designed to bind all isoforms of VEGF and block vessel permeability and angiogenesis. It binds and inhibits VEGF165, VEGF121, and VEGF110 and has also been shown to penetrate all layers of the rabbit retina-the first demonstration of retinal penetration of an anti-VEGF therapy intended for AMD, diabetic macular edema and retinal vein occlusion. Figurska et al first reported using ranibizumab treated CNV in two patients with high myopia. After 9 months follow-up in a 25-years old woman's after three injections of Lucentis, visual acuity improved of three lines on ETDRS chart (15 letters), another patient gained 10 letters after two injections.8 After then, there has been number of publications reporting the use of intra-vitreal ranibizumab in the treatment of myopic CNV. These studies are limited to case-series involving relative small number of study subjects, and short follow-up period (3-36 months). In these series with limited follow-up, intra-vitreal ranibizumab was a safe and effective treatment for CNV secondary to PM, resulting in functional and anatomic improvement.37-39,42-45 Recently, Franqueira et al46 reported a 3-year retrospective, nonrandomized, interventional case series. They observed forty eyes of 39 patients with myopic CNV. 15 with previous photodynamic therapy, and 25 naive eyes. Best-corrected visual acuity (BCVA) changes, central foveal thickness (CFT), and number of treatments were assessed, from baseline to month 36. A mean of 4.1 injections were performed in the first year, 2.4 in the second year and 1.1 in the third year. It was concluded that intravitreal ranibizumab seems to be an effective and safe therapeutic procedure to treat CNV in highly myopic eyes, with a high proportion of patients gaining or stabilizing BCVA at a 3-year follow-up. Bevacizumab is a full-length recombinant, humanized antibody approved for the treatment of metastatic adenocarcinoma of the colon, appears to effectively treat choroidal neovascular membranes associated with age-related macular degeneration.47 Since the initial promising results of systemic and intravitreal using bevacizumab on the treatment of CNV in AMD, now bevacizumab is currently being used at multiple centers around the world as an “off-label” treatment of neovascular AMD and other ocular diseases. It was widely “off-label” used to treat different types of CNV and macular edema because its relative cheap cost and well tolerated and efficacy that was proved by the visual acuity and anatomic improvements after injection. Nguyen et al48 reported the first two patients with subfoveal CNV secondary to PM treated with intravenous bevacizumab. These two patients both were presented with recurrent CNV and treated bevacizumab as a salvation therapy. They showed significantly substantial improvement of vision with decrease in foveal thickness and macular volume proved by OCT. There were no ocular or systemic side effects were observed after intravenous bevacizumab. Then Yamamoto et al49, Sakaguchi et al50 reported intra-vitreal injection bevacizumab results. Yamamoto et al49 reported the results of 11 eyes with the history of treated by photodynamic therapy, which three eyes received two bevacizumab injections and eight eyes received one injection. It was showed that visual acuity improved by a mean of +3.5 lines with no injection complications or drug-related side effects observed after intra-vitreal injection bevacizumab after a mean follow-up of 153 (range 35-224) days. The visual improvement was also proved by OCT evaluation that CFT improved from 340 μm (range, 253 to 664 μm) to 234 μm (range, 142-308 μm). The encouraging results are similar with Sakaguchi's results that the BCVA improved two or more lines in six eyes (75%) and remained the same in two eyes (25%), which is recently supported by more cases reports. Neelam et al51 reviewed 32 studies which used pegaptanib,1 bevacizumab22 and ranibizumab.9 In these primarily small series, although mostly with limited follow-up, it was elementarily proved that intra-vitreal bevacizumab is a safe and potentially efficiency treatment for sub-foveal CNV secondary to PM. The mean number of injections per year in PM patients is far more less than that in AMD. Since the adverse events (AE) are still observed in anti-VEGF treatment, less injection might means less incidence of AE. Unlike systemic administration of VEGF inhibitors associated with potential life-threatening complications, such as thromboembolic events, myocardial infarction, malignant hypertension, intra-vitreal administration of VEGF inhibitors minimizes the risk. The ocular complications reported included the risk of developing intraocular inflammation, glaucoma, endophthalmitis, and cataract. Also, in myopic eyes, there is an increase likelihood of developing retinal tear and retinal detachment.52 It was also reported that development and gradual expansion of chorioretinal atrophy around regressed CNV could be observed following successful treatment with VEGF inhibitors, however, the area of chorioretinal atrophy may be smaller in these eyes than eyes treated with PDT.53 Currently, there appears to be a global consensus that the treatment strategy is logical and understandable, despite these shortcomings of present preliminary studies. We are optimistic using VEGF inhibitors as the first line therapy for the CNV secondary to pathologic myopia in the future. But it must be cautious until the numerous questions can be answered by rigorously designed a large series of case-controlled, prospective, randomized, multicenter, long-term follow up clinical trials.