We have previously shown that angiogenesis inhibition and tumor regression can be accomplished by combinations of heparin or heparin fagments with cortisone [Folkman, J., Langer, R., Linhardt, R. J., Haudenschild, C. & Taylor, S. (1983) Science 221,[719][720][721][722][723][724][725]. Oral heparin was also effective in combination with cortisone. We now show that a single oral dose of [35S]heparin or [3H]heparin (15,000 units/kg) results in continuous release of radioactive material into the bloodstream for at least 12 hr. This is associated with the presence of anti-factor Xa activity at a level of "0.1 unit/ml. The radioactive material is Identified as oligo-, di-, and monosaccharides by its behavior in chromatographic systems, its possession of anti-factor Xa activity, and the effect of treatment with bacterial heparinase. The heparin fragments are extensively metabolized to fragments without anti-factor Xa activity that are readily subject to urinary excretion.We have previously demonstrated that the combination of heparin, or a heparin fragment, and cortisone inhibited angiogenesis, including the chicken chorioallantoic membrane and the rabbit corneal pocket assay (1). When the combination was used to inhibit tumor growth by parenteral administration to mice, a high rate of hemorrhagic complications was encountered because of the anticoagulant effect of heparin. Other methods of heparin administration were tried, and it was found that the anti-angiogenic effect was maintained when heparin was given orally even though the anticoagulant complications did not occur. This oral regimen inhibited tumor growth and metastasis in some animal tumor models. In an initial experiment to study the absorption of oral heparin, mice were given [35S]heparin in their drinking water. Serum obtained 24 hr later yielded 0.4% of the radioactivity that was present in the drinking water. However, gel filtration chromatography failed to demonstrate radiolabeled heparin fragments with biological activity (1).The anti-angiogenic activity of oral heparin was somewhat surprising. The literature contains numerous reports of attempts to administer heparin orally (2-4), with little success in achieving clinically useful anticoagulant activity. Attempts have been made to improve enteral absorption in a number of ways: (i) use of adjuvants such as EDTA (5), citric acid (6), or dimethyl sulfoxide (7); (ii) oral administration in micellar solutions (8) or in liposomes (9); (iii) rectal administration with bile salts (10); or (iv) administration of low molecular weight heparins (4). No such technique has yet achieved clinical usefulness.As a result of finding anti-angiogenic activity with oral heparin plus cortisone without anticoagulant activity, we decided to examine the fate of radiolabeled heparin given enterally. Furthermore, by examining two differently labeled heparins (35S and 3H), we sought to gain information about metabolism of the drug by desulfation and by depolymerization. We now show that a single oral dose of hepari...