A phase II trial with rosiglitazone in liposarcoma patients.

Agents of the thiazolidinedione drug family can terminally differentiate human liposarcoma cells in vitro by activating genes responsible for lipocyte differentiation. One study has shown clinical activity of troglitazone treatment in liposarcoma patients. We sought to find further evidence for this result. In all, 12 patients with a liposarcoma received rosiglitazone 4 mg b.d. They were followed clinically and with repeated biopsies for histological and biological studies. At the molecular level the mRNA translation of three genes that are induced by this treatment (peroxisome proliferator-activated receptor gamma (PPARgamma), adipsin and fatty acid binding protein) was determined. Nine patients were eligible for evaluation. One patient had to stop treatment due to hepatotoxicity. The mean time to progression was 6 months (2 - 16 months), with one patient still on treatment. We did not see any significant change in histologic appearance of the liposarcomas by the treatment. The level of gene expression changed significantly in two patients, but this did not result in a clinical response. Based on this study, rosiglitazone is not effective as an antitumoral drug in the treatment of liposarcomas. Increased PPARgamma activity does not correlate with the clinical evolution.