Endometriosis is a disease in which the lining of the uterus (endometrium), shed at the time of menstruation, becomes established at sites such as the peritoneum and ovaries. These explants develop a rich blood supply that enables them to survive and grow. We hypothesized that inhibitors of angiogenesis would prevent this growth by disrupting sensitive vessels supplying endometriotic lesions. Vessels sensitive to angiogenic antagonism have few associations with pericyte cells. The vessels supplying human endometriotic lesions were immunohistochemically characterized and found to be predominantly pericyte free. A model in which human endometrium is implanted into nude mice was used to test the effects of two antagonists of the angiogenic growth factor, vascular endothelial cell growth factor A. Soluble truncated receptor (flt-1; P = 0.002) and an affinity-purified antibody to human vascular endothelial cell growth factor A (P = 0.03) significantly inhibited the growth of nude mouse explants. Pericyte-free vessels were shown to supply endometrial lesions in nude mice and were disrupted in lesions taken from soluble flt-1-treated mice. In summary, antiangiogenic agents inhibited the growth of explants in an in vivo model of endometriosis by disrupting the vascular supply, and this effect is likely to apply to the human disease. These findings suggest that antiangiogenic agents may provide a novel therapeutic approach for the treatment of endometriosis.
BACKGROUND: Women with endometriosis have elevated levels of cyclooxygenase-2 (COX-2) in peritoneal macrophages and endometriotic tissue. Inhibition of COX-2 has been shown to reduce inflammation, angiogenesis and cellular proliferation. It may also downregulate aromatase activity in ectopic endometrial lesions. Ectopic endometrial establishment and growth are therefore likely to be suppressed in the presence of COX-2 inhibitors. We hypothesized that COX-2 inhibition would reduce the size and number of ectopic human endometrial lesions in a nude mouse model of endometriosis. METHODS: The selective COX-2 inhibitor, nimesulide, was administered to estrogen-supplemented nude mice implanted with human endometrial tissue. Ten days after implantation, the number and size of ectopic endometrial lesions were evaluated and compared with lesions from a control group. Immunohistochemical assessment of vascular development and macrophage and myofibroblast infiltration in control and treated lesions was performed. RESULTS: There was no difference in the number or size of ectopic endometrial lesions in control and nimesulide-treated nude mice. Nimesulide did not induce a visually identifiable difference in blood vessel development or macrophage or myofibroblast infiltration in nude mouse explants. CONCLUSION: The hypothesized biological properties of COX-2 inhibition did not influence lesion number or size in the nude mouse model of endometriosis.