Defibrotide (DF), an Orally Bioavailable Modulator of Myeloma Tumor-Microenvironment Interactions: Molecular Sequelae and Clinical Implications.

Introduction: Defibrotide (DF), a polydisperse oligonucleotide with anti-thrombotic, thrombolytic, and anti-adhesive properties, but minimal hemorrhagic risk, selectively targets microvasculature and is a promising treatment for stem cell transplant-associated hepatic veno-occlusive disease (VOD), a serious complication characterized by endothelial cell (EC) and stromal injury in liver. We previously reported that DF has minimal activity against multiple myeloma (MM) in vitro, but increases MM cell sensitivity to conventional chemotherapy, including melphalan and dexamethasone in vivo. We hypothesized that DF modulates MM cell interactions with local microenvironment, including stroma and ECs, as well as the bioavailability of cytokine/growth factors for tumor and ECs.

Methods/Results: Pharmacokinetics of DF (administered po vs. iv) at clinically-relevant doses (0–48 mg/kg) to 8 week-old male Fisher 344 rats were first evaluated. Peripheral blood plasma was collected (0–8hrs) and HPLC analyses, validated by agarose gel determination (mean and median variability of measurement of 8.1% and 7.9%, respectively), showed that after iv and po administration, peak DF levels (Cmax) were 1253 and 474 μg/mL, and area under the curve (AUC0-240 min) were 14.7 and 17.2 μg*min/Lt. Time to Cmax was 30 min for po administration, with undetectable levels by 240 min after po or iv treatment. In vitro studies at DF concentrations achievable in vivo were next conducted to evaluate effects on growth factors (including VEGF, bFGF, PDGF) critical for tumor neovascularization and pathways regulating bioavailability of extracellular matrix (ECM)-resident growth factors, including expression/enzymatic activity of heparanase, which degrades heparin sulfate (HS), releases HS-bound growth factors and promotes tumor invasion and metastasis. Using a ClRNH32P-OdT18 probe and increasing DF concentrations as a competitor of probe binding to growth factors, we observed high-affinity of DF for certain heparin-binding growth factors, including bFGF and PDGF (calculated Kc values of 30 and 40 nM, respectively), but not others, such as VEGF (Kc1200 nM). Pharmacologically achievable DF levels suppressed in vitro heparanase mRNA (by Real-Time PCR) in U226 MM cells and heparanase enzymatic activity in U266 cell extracts. These data suggest that DF can perturb bioavailability, expression and/or function of diverse positive regulators of neo-angiogenesis. Indeed DF suppressed in vitro formation of vessel-like structures by ECs in corneal matrix assays.

Conclusions: DF, a known anti-thrombotic agent, exhibits, at concentrations achievable after oral administration, diverse biological functions that may suppress tumor-associated angiogenesis. These include direct DF binding to pro-angiogenic growth factors, suppression of enzymes responsible for release of these factors, and thus their depletion from tumor ECM. These studies, combined with our data that DF can counteract stroma-derived protection to MM cells against chemotherapy, provide the framework for an ongoing Phase I/II clinical trial of oral melphalan, prednisone, thalidomide and DF in advanced MM.