Inhibition of Selectin-Mediated Pathologies by Clinically Tolerable Levels of Heparin - Marked Differences in Efficacy of Various Clinically-Approved Preparations.

Heparin has been in clinical use as an anticoagulant for many decades, by virtue of its ability to enhance antithrombin activity. However, heparin is a natural product, comprising a complex polydisperse mixture of highly sulfated glycosaminoglycan chains, only some of which bind antithrombin. We have reported that unfractionated heparin has an unrelated biological effect in inhibiting P- and L-selectin binding to their natural ligands (Koenig et al., 1998, J Clin Invest, 101, 877–889). Meanwhile, retrospective analyses of human clinical data indicate that heparin therapy increases survival in various cancers (e.g., Kakkar et al., 1995, Int J Oncol, 6, 885–888). Clinical trials using Vitamin K antagonists showed no major effect on survival in most carcinomas (Zacharski and Ornstein, 1998, Thromb Haemost, 80, 10–23). Thus, the heparin effect on cancer may not be due primarily to its anticoagulant function, but rather to inhibition of P- and L-selectin binding to tumor cell ligands. Indeed, our studies in mice showed that unfractionated heparin is effective in attenuating selectin-dependent metastasis of carcinomas by blocking P- and L-selectin binding to its natural ligands on tumor cells (Borsig et al., 2001, Proc Natl Acad Sci USA, 98, 3352–3357). Heparin can also attenuate inflammatory pathologies in various animal models by blocking L- and P-selectin action (Wang et al., 2002, J Clin Invest, 110, 127–36). Most of the above studies used unfractionated heparin (UFH). Unlike UFH, low molecular weight heparins (LMWH) have fewer side effects and increased plasma half-life, making them favored in clinical practice (Hirsh et al., 2001, Chest, 119 Suppl., 64S–94S). Further benefits are claimed for a synthetic heparinoid pentasaccharide that specifically binds to antithrombin (Gordois et al., 2003, J Thromb Haemost, 1, 2167–2174). Initial studies suggest that LMWHs may indeed have beneficial effects in malignancies (Kakkar et al., 2004, J Clin Oncol, 22, 1944–8). Translating these findings into clinical practice first requires experimental evaluation of the potential for clinically acceptable levels of heparin to block P- and L-selectin and attenuate selectin dependent pathologies (Varki and Varki, Semin Thromb Hemost. 28: 53–66). We have compared the ability of various heparins and heparinoids to inhibit binding of LS180 cells (a human colorectal adenocarcinoma cell line expressing selectin ligands) to P- and L-selectin in vitro. While UFH was the best inhibitor, only one other preparation had substantial selectin inhibitory activity. Disaccharide compositional analysis of the different LMWH preparations did not show major differences, indicating that variation in ability to inhibit selectins is due to high-order structural differences and/or length. In preparation for pre-clinical studies in mice we have also analyzed different routes of administration (IV, IP, and SubQ) and followed the pharmacokinetics of various injected preparations by monitoring anti-Xa levels. All routes of delivery showed that heparins are cleared more rapidly in mice than in humans. Having determined the optimal dosing and route of administration to sustain clinically accepted levels of various heparins in mice, we are now carrying out functional in vivo assays (metastasis, inflammation, interaction of tumor cells with platelets) to determine if in vitro selectin inhibitory activity correlates with the ability to attenuate selectin-mediated pathologies.