The growth of new vessels in the cornea typically results from pathologic etiologies that cause acute or chronic inflammation, stem cell deficiency, or hypoxia. In the United States, herpes simplex and bacterial keratitis are the most significant causes of neovascularization from inflammation resulting from infection. Infectious etiologies such as trachoma and onchocerciasis, however, are internationally important. Trachoma is estimated to affect 400 million people and to blind approximately 6 million people worldwide (9). Onchocerciasis affects 50 million and is estimated to blind 1 million people (10).

Stem cell deficiency leading to neovascularization of the cornea can result from trauma, alkali burns, or inflammation from diseases such Stevens-Johnson syndrome and cicatricial pemphigoid, or in association with congenital anomalies such as aniridia. Although contact lens wear is probably the most frequent cause of neovascularization from hypoxic injury to the cornea, the incidence of visual impairment from this is low. Contact lens-induced neovascularization depends on the oxygen permeability (Dk) of the contact lens material as well as duration of wear, and it often regresses with discontinuation of contact lens use (11, 12, 13). Recent advances with high-Dk lenses and daily-wear lenses are expected to decrease the incidence of corneal neovascularization secondary to contact lens wear.

Angiogenesis is thought to be initiated by a disequilibrium in the balance between angiogenic and antiangiogenic factors. Collectively termed the angiogenic switch by Hanahan and Folkman (14), this theory arose from a study of tumor models and observations that certain tissues with constitutive expression of proangiogenic molecules [e.g., vascular endothelial growth factor (VEGF)] can remain without new blood vessel growth, whereas other tissues and assay systems could undergo steps in angiogenesis simply with the addition of these same molecules. In this latter group, however, angiogenesis could be blocked with increasing amounts of angiogenesis inhibitors (e.g., thrombospondin-1) (15,16). It is thought that this critical balance, which may be tissue specific (i.e., high inhibitory levels in some, low
levels of both in others), may explain why different levels of angiogenic and antiangiogenic molecules are necessary to initiate the process in any given tissue (14). The metalloproteinases (MMPs), tissue-inhibitors of metalloproteinases (TIMPs), angiostatin, endostatin, prolactin, thrombospondin-1, arresten, tumstatin, canstatin, and pigment epithelium-derived factor are all thought to be involved, either in whole or in part, with the “angiogenic switch.”

The paradigm of cryptic inhibitors of angiogenesis emerged with the discovery that certain ubiquitous extracellular matrix proteins (e.g., fibronectin, plasminogen, collagen) could be degraded by proteolytic enzymes (including MMPs) into fragments with potent antiangiogenic properties (e.g., angiostatin and endostatin-like molecules) (14). Coupled with the presence of angiogenic factors [e.g., basic fibroblast growth factor (bFGF), VEGF] nested within the extracellular matrix, a paracrine balance of factors in the angiogenic switch mechanism is also thought to play an important role in homeostasis and normal physiologic regulation (14). Matrilysin (MMP-7) may play such a role in the cornea, allowing for what some have termed the angiogenic “privilege” of the cornea (17,18).

When a disequilibrium in the angiogenic switch favors angiogenesis, the cascade of steps that follows includes (a) the disruption of the basement membrane, (b) endothelial cell migration, (c) endothelial cell proliferation, (d) endothelial cell differentiation and tube formation (19), (e) vascular pruning by leukocytes (20), and (f) maturation of the blood vessels with the recruitment of pericytes (21,22) (Fig. 5-1). There is evidence that circulating bone marrow-derived endothelial precursor cells are also involved and incorporated into new vessels (22,23) and that placental growth factor (PlGF) may play a role in recruiting them (24). VEGF-A plays a vital role in the switch mechanism and steps (a) through (e). The disruption of the basement membrane is titrated by a balance between urokinase-type plasminogen activator, MMPs, cysteine proteinase systems and plasminogen activator inhibitors, TIMPs, and cystatins (25). Endothelial cell migration is facilitated by the expression of integrins (α5β3, α5β5). The maturation of the blood vessels by pericytes/smooth muscle cells appears to be mediated by angiopoietins, platelet-derived growth factor (PDGF),
and transforming growth factor (TGF)-β, with PDGF playing a critical role in pericyte recruitment. Both in vitro and in vivo assays for angiogenesis have been developed to observe the effects of various factors on each of these crucial steps, and are briefly elaborated in the following section.