THERE ARE MANY ATTRIBUTES of the eye that are attractive with respect to gene therapy. Anatomical features allow delivery of the gene therapy reagent to specific ocular compartments, thereby exposing only particular sets of cells to a small amount of the reagent. The reagent does not escape from the eye, at least in significant quantities, thereby minimizing exposure to other organ systems and the immune system. The effects of vector/gene delivery can be monitored closely, frequently, and noninvasively thanks to the transparent nature of the cornea, lens, and intraocular fluids and the availability of qualitative and quantitative tests of ocular disease progression. The fact that many ocular diseases are bilaterally symmetrical allows comparison of the effects of vector/gene delivery with disease progression in the contralateral eye. Campochiaro and colleagues have taken advantage of these attributes and have described, in a paper published in this issue of Human Gene Therapy (Campochiaro et al., 2006), results of a phase 1 clinical trial in individuals with advanced neovascular age-related macular degeneration (AMD). AMD affects almost 30% of older adults aged 75–85 years (Leibowitz et al., 1980) and an estimated 11 million people in the United States alone (National Advisory Eye Council, 1993). Thus, development of effective therapy represents a great public health need. Choroidal neovascularization (CNV), the pathology found in neovascular AMD, results in retinal injury and blindness. In the Campochiaro et al. dose escalation study, a serotype 5 (Ad5), E1, partial E3-, E4-deleted adenoviral vector carrying human pigment epithelium-derived factor (PEDF; AdPEDF. 11) was injected one time into the vitreous of each of 28 subjects. PEDF is an endogenous protein with potent antiangiogenic activity and is postulated to have an important role in preventing pathological neovascularization. Its ability to suppress ocular neovascularization had been demonstrated in animal models, using both adeno-associated virus (AAV) and adenovirus vectors. The results from the Campochiaro et al. study are extremely encouraging as there were no serious adverse events or dose-limiting toxicities through the highest dose (109. 5 particle units [PU]). Even though this was a phase 1 study and thus not designed to evaluate therapeutic efficacy, there was evidence that injection of doses greater than 108 PU of AdPEDF. 11 resulted in antiangiogenic activity persisting for several months (Campochiaro et al., 2006). Because PEDF is a secreted protein and can diffuse throughout the eye, transduction of cells distant from the site of pathology can presumably result in therapeutic benefit.

The Campochiaro et al. study brings to two the number of phase 1 human clinical trials in which an Ad5-based vector has been safely delivered into the human eye. In a phase 1 study aimed at testing the feasibility and safety of gene therapy for retinoblastoma, Hurwitz and colleagues delivered an adenoviral vector carrying a herpes simplex virus thymidine kinase gene (AdV-TK) directly into vitreous tumor sites in eight different subjects who had failed standard therapies. Each injection was followed by intravenous delivery of ganciclovir. Similar to the Campochiaro et al. study, there was no serious toxicity. Inflammation was reversible even after repeated (up to five different) injections or after using a dose 1.5 log units higher than that used by the Campochiaro group. The seven patients who were treated with doses 1010 VP had resolution of their vitreous tumor and one patient remained free of active vitreous tumor 38 months after therapy (Chevez-Barrios et al., 2005). With the baseline …