A phase I study of bexarotene and rosiglitazone in patients with refractory cancers

Abstract

13094 Background: Preclinical studies have shown that ligands of the PPAR-γ receptor can slow growth and promote differentiation of malignant cells. Rosiglitazone is a high affinity PPAR-γ ligand approved for treatment of insulin resistant diabetes. After binding ligand, the PPAR-γ receptor heterodimerizes with the RXR receptor. Bexarotene is an RXR ligand with cytostatic properties similar to those of PPAR-γ agonists. Ligands of RXR and PPAR-γ are synergistic, causing cell differentiation and apoptosis. Methods: We conducted a phase 1 study of bexarotene and rosiglitazone in patients with resistant malignancies to define the MTD of rosiglitazone in this regimen. Patients received bexarotene 300 mg/m2 daily. The starting dose of rosiglitazone was 4 mg/day. This dose was escalated in five cohorts using the modified Fibonacchi scheme (4–6-8–10–12). Both drugs were continued until disease progression or toxicity. Patients received atorvastatin 10 mg daily, which was increased based on weekly triglyceride levels. Results: We enrolled 23 patients without encountering dose limiting toxicity. After reaching 12 mg/day of rosiglitazone, the study was closed because of laboratory evidence showing no further synergy at higher doses. Patient Characteristics: Histological type: NSCLC 8, sarcoma 3, prostate 2, and others (SCLC, thyroid/breast, mesothelioma, esophageal, breast, colorectal, renal, parotid, transitional cell, cervical). Median cycles of treatment delivered 1 (range 0–4). The primary serious adverse events (grade 3 or 4) were hyperglyceridemia (17%), dyspnea (9%), nausea (9%), dehydration (9%). Two patients expired, both due to disease progression. Responses: CR/PR none, SD 3, disease progression 14, not evaluable 4, patient withdrawal 2. Conclusions: The combination of bexarotene (300 mg/m2/day) and rosiglitazone (12 mg/day) is safe and feasible. No responses were seen in this heavily pretreated population, which is not surprising given that both agents are cytostatic, rather than cytotoxic. This combination might be studied in cutaneous T-cell lymphoma, or as adjuvant/maintenance therapy in other malignant diseases. Other areas for study might include inflammatory diseases where PPAR-γ ligands have activity. This study was supported by Ligand pharmaceuticals. [Table: see text]