A strategy to discover circulating angiogenesis inhibitors generated by human tumors.

The phenomenon of inhibition of tumor growth by tumor mass has been studied in many experimental animal systems and has been observed in several clinical scenarios. Not until the recent discovery of angiostatin, a circulating angiogenesis inhibitor generated in the presence of a murine Lewis lung tumor, has a satisfactory mechanism been proposed to explain this phenomenon. Thus far, no other animal or human tumors are known to generate angiostatin. In this study, we utilized a mouse corneal neovascularization model to detect circulating inhibitors of angiogenesis generated by three human tumors grown in immunodeficient mice: (a) the PC-3 human prostate carcinoma; (b) the CCL188 human colon carcinoma; and (c) the UBC urinary bladder carcinoma. Mice bearing these three primary tumors demonstrated significant inhibition of angiogenesis in the cornea induced by a pellet containing basic fibroblast growth factor. Corneas of mice bearing s.c. prostate and colon carcinomas showed significant inhibition of vessel length, clock-hours of neovascularization, and vessel density. However, corneas of mice bearing s.c. bladder carcinomas demonstrated significant inhibition of vessel density only. Three colon carcinomas (clone A, CX-1, and MIP101), the MDA-MB-435S breast carcinoma, the MM-AN melanoma, and the JE-3 choriocarcinoma did not significantly inhibit corneal neovascularization.