Gemcitabine and cisplatin versus vinorelbine and cisplatin versus ifosfamide + gemcitabine followed by vinorelbine and cisplatin versus vinorelbine and cisplatin followed by ifosfamide and gemcitabine in stage IIIB-IV non small cell lung carcinoma: a prospective randomized phase III trial of the gruppo oncologico italia meridionale

Abstract Purpose: we carried out a phase III randomized trial to compare vinorelbine–cisplatin regimen to gemcitabine–cisplatin regimen, and to a sequential administration of gemcitabine–ifosfamide followed by vinorelbine–cisplatin or the opposite sequence of vinorelbine–cisplatin followed by ifosfamide–gemcitabine according to the ‘worst drug rule’ hypothesis in patients with locally advanced unresectable stage IIIB or metastatic stage IV non-small cell lung cancer. The primary endpoint was survival parameters, while secondary endpoints included analysis of response rates and toxicity. Patients and methods: patients were randomized to receive: (a) gemcitabine 1000 mg/m 2 on days 1, 8 and 15 plus ifosfamide 1500 mg/m 2 on days 8–12 with mesna uroprotection (GI regimen) followed by vinorelbine 25 mg/m 2 on days 1 and 8 plus cisplatin 100 mg/m 2 on day 1 (GI→VC regimen); (b) the opposite sequence (VC→GI); (c) vinorelbine plus cisplatin as above described (VC regimen); or (d) gemcitabine 1400 mg/m 2 on days 1 and 8 plus cisplatin 100 mg/m 2 on day 8 (GC regimen). All regimens were given every 4 weeks. All patients were chemotherapy naive and had a ECOG PS 0-2. Results: 400 patients were enrolled into the trial. Interim analysis after inclusion of 243 patients showed that ORR were 19% in the GI→VC arm, 32% in the inverse sequence arm (CV→GI), 42% in the VC arm, and 30% in the GC arm. The VC arm was statistically superior over the GI→VC arm ( p =0.0074), but not over the other regimens. Median TTP was 3.1 months in the GI→VC arm versus 5.0 months in the VC→GI arm ( p =0.014). For these reasons the GI→VC and VC→GI arm were closed since the ‘worst drug rule’ hypothesis was rejected. Accrual in the VC and GC arms continued up to 140 and 138 patients respectively. Final ORR were 44% for the VC regimen (4 CR), and 34% for the GC regimen (1 CR). This difference was statistically significant ( p =0.032). OS was 9.0 and 8.2 months, respectively, with no statistically significant difference. The 1-year survival rate was 24 and 20%, respectively for VC and GC regimens. As expected the incidence of phlebitis was higher in the VC arm, while thrombocytopenia, flu-like syndrome and asthenia were more frequent in the GC arm. Conclusions: the results of this trial indicate that the combination of vinorelbine and cisplatin and that of gemcitabine and cisplatin are equivalent in terms of median TTP and OS, although the vinorelbine–cisplatin regimen is associated with a higher ORR. Both regimens may be considered as reference treatments for future studies. Moreover, our data reject the ‘worst drug rule’ hypothesis of sequential treatments in NSCCL at least with the combination used in this study.