Induction chemotherapy with cisplatin and fluorouracil alone or in combination with docetaxel in locally advanced squamous-cell cancer of the head and neck: long-term results of the TAX 324 randomised phase 3 trial

Abstract

Background With a minimum follow up of 2 years, the TAX324 study demonstrated a significant survival benefit of induction chemotherapy (IC) with docetaxel, cisplatin and 5FU (TPF) versus cisplatin and 5fluorouracil (PF) followed by chemoradiotherapy with carboplatin delivered as sequential therapy (ST) in locally advanced head and neck cancer (LAHNC). We report the long term results with 5 years minimum follow-up. Methods TAX324 was a randomized, open-label phase 3 trial comparing three cycles of TPF IC (docetaxel 75 mg/m2 of body-surface area, followed by intravenous cisplatin 100 mg/m2 and 5fluorouracil 1000 mg/m2 per day administered as a continuous 24-hour infusion for 4 days) with three cycles of PF (intravenous cisplatin 100 mg/m2, followed by fluorouracil 1000 mg/m2 per day as a continuous 24-hour infusion for 5 days). Both regimens were followed by 7 weeks of chemoradiotherapy with concomitant weekly carboplatin (AUC1.5). Randomization was performed centrally with the use of a biased-coin minimization technique. At study entry, patients were stratified according to the site of the primary tumor, nodal status (N0 or N1 vs. N2 or N3), and institution. For this long term analysis, data was gathered retrospectively. Overall survival (OS) and progression-free survival (PFS) were the primary endpoints. Data as of December 1, 2008 were analyzed. Tracheostomy and gastric feeding tube dependence were used as surrogates for treatment related long term toxicity. The median follow-up was 72.2 months (mo) (IQR for TPF =33 mo, PF =34 mo and for all pts =34 months). The analysis was based on data from all 501 patients. 61 patients were lost to follow-up and their data as of the initial analysis in 2005 was used. Findings OS was significantly better with TPF versus PF (HR=0.74, 95%CI: 0.58–0.94), with an estimated 5-yr survival rate of 0.52 and 0.42 in the TPF and PF arms, respectively. Median survival time was 70.6 mo (95%CI: 49.0–89.0 mo) with TPF versus 34.8 mo (the 95%CI: 22.6–48.0 mo) in the PF group (p=0.014). PFS was also significantly better with TPF (38.1 mo; 95%CI 19.3–66.1 mo vs. 13.2 mo, 95%CI 10.6–20.7 mo; HR= 0.75, 95%CI: 0.60–0.94). Subjects with hypopharyngeal and laryngeal cancer had significantly superior PFS with TPF (HR=0.68, the 95%CI: 0.47–0.98). No significant difference for dependence on gastric feeding tubes and tracheotomies was detected between the treatment groups. In the TPF arm 3 out of 91 patients (3%) remained feeding tube dependent (no information in 40 cases) while 8 out of 71 (10%) patients required feeding tubes in the PF arm (no information in 30 cases). 6 out of 92 (7%) patients had tracheostomies (no information in 39) versus 8/71 (13%) (no information in 30) in the TPF and PF groups, respectively. Interpretation IC with TPF provides long term survival benefit compared to PF in LAHNC. Patients who are candidates for IC should be treated with TPF.