Final results of the multicenter randomized phase II PAZOGIST trial evaluating the efficacy of pazopanib (P) plus best supportive care (BSC) vs BSC alone in resistant unresectable metastatic and/or locally advanced gastrointestinal stromal tumors (GIST).

Session Type and Session Title: 
Oral Abstract Session, Sarcoma
Abstract Number: 


J Clin Oncol 33, 2015 (suppl; abstr 10506)
Jean-Yves Blay, Mathieu Molimard, Claire Cropet, Julien Domont, Maud Toulmonde, Emmanuelle Bompas, Philippe Alexandre Cassier, Isabelle Laure Ray-Coquard, Maria Rios, Antoine Adenis, Antoine Italiano, Axel Le Cesne, Olivier Bouche, Olivier Mir, Florence Duffaud, Francois Bertucci, Nicolas Isambert, Aurelie Belleville, Julien Gautier, David Pérol; Centre Léon Bérard, Department of Medicine, Lyon, France; Groupe Hospitalier Pellegrin, Bordeaux, France; Unité de Biostatistique et d'Evaluation des Thérapeutiques - Direction de la Recherche et d l'Innovation, Centre Léon Bérard, Lyon, France; Institut Gustave Roussy, Villejuif, France; Institut Bergonié, Department of Medical Oncology, Bordeaux, France; Department of Medical Oncology, Centre René Gauducheau, Nantes St. Herblain, France; Centre Léon Bérard, Lyon, France; Centre Leon Berard, Lyon, France; Centre A Vautrin, Nancy, France; Medical Oncology Department, Centre Oscar Lambret, Lille, France; CLCC Institut Bergonié, Bordeaux, France; Centre Hospitalier Universitaire Robert Debré, Reims, France; La Timone University Hospital, Marseilles, France; Institut Paoli Calmette, Marseille, France; Centre Georges-François Leclerc, Dijon, France

Abstract Disclosures


Background: GIST is the most common mesenchymal neoplasm of the gastrointestinal tract. Imatinib followed by sunitinib then regorafenib is the usual sequence in advanced setting. P is effective in soft tissue sarcomas; it has never been evaluated in a randomized setting in GIST. Methods: Eligible patients (pts) were randomized (1:1; stratification on number of prior drugs (2 vs ≥3)) to receive P+BSC or BSC. Primary endpoint was Progression-Free Survival (PFS). 80 pts were planned to detect an improvement in the 4-month PFS rate (PFS-4m) from 15% (BSC) to 45% (P+BSC) with 5% two-sided α and 80% power. Secondary objectives included Best Overall Response (BOR), Overall Survival (OS), safety and trough plasma P concentrations (Ct). Results: From Apr 11 to Dec 13, 81 pts (P+BSC: 40, BSC: 41) were randomized. Arms were well balanced; median age: 63y (27-85), 70% males and 54% with ≥ 3 prior drugs. The intent-to-treat analysis based on investigator-assessed progressive disease (PD) showed an improved PFS with PFS-4m of 45.2% (95% CI 29.1-60.0) for P+BSC vs 17.6% (95% CI 7.8-30.8) for BSC (HR: 0.59, 95% CI 0.37-0.96; p = 0.03). 36 pts out of 41 switched to P following investigator-assessed PD (median P duration: 3.5 months (0.1-19), main reasons for discontinuation: PD (52.8%) and toxicity (22.3%)). Centrally-assessed BOR showed stable disease in 84.2% vs 70.7% and PD in 15.8% vs 26.8% of pts in P+BSC and BSC arms. Among all pts treated with P (n = 76), 72.4% experienced grade ≥ 3 related adverse events (AE) (hypertension: 36.8%), including 25% of related serious AE (pulmonary embolism, 9.2%). At the time of analysis, 29 vs 31 pts had died in the P+BSC and BSC arms (OS HR: 0.94, 0.56-1.56). 1 toxic death occurred (hepatic cytolysis). Preliminary results showed a higher median Ct at steady state in pts progression-free at 4 months vs progressive: 25 (2-63) vs 17 (8-35) µg/ml (n = 43; p = 0.08). Conclusions: P combined to BSC improves PFS in advanced GIST resistant to imatinib and sunitinib. In the context of a high switch rate, no difference in OS was found. Further PK analyses are ongoing. Clinical trial information: NCT01323400