Background: The combination of cisplatin (CDDP) and 5-fluorouracil (5-FU) can be regarded as a reference regimen in squamous cell carcinoma of the head and neck (SCCHN). Raltitrexed (Tomudex) is a direct and specific thymidilate synthase (TS) inhibitor, which has shown clinical activity against SCCHN in a previous phase I study, when combined with 5-FU and levo-folinic acid (LFA). Preclinical data support the combination of CDDP and raltitrexed. The aim of the present study was to evaluate the combination of cisplatin, raltitrexed. LFA and 5-FU in a phase I-II study.
Patients and methods: Patients with locally advanced or metastatic SCCHN were treated with a combination of cisplatin at the starting dose of 40 mg/m2. followed by raltitrexed at the starting dose of 2.5 mg/m2 on day 1; levo-folinic acid at fixed dose of 250 mg/m2, followed by 5-fluorouracil at the starting dose of 750 mg/m2 on day 2. Doses of the three cytotoxic agents were alternately escalated up to dose-limiting toxicity (DLT). Treatment was recycled every two weeks and given up to a maximum of eight courses; after chemotherapy, patients with locally advanced disease received a locoregional treatment.
Results: Forty-five patients were entered into the study. Six dose levels were tested. At CDDP 50 mg/m2, raltitrexed 3 mg/m2, 5-FU 900 mg/m2, four out of six patients showed DLT, which was in all cases grade 4 neutropenia. Therefore, this dose level was defined as maximum tolerated dose (MTD). CDDP 60 mg/m2, raltitrexed 2.5 mg/m2, LFA 250 mg/m2, 5-FU 900 mg/m2 was the dose level recommended for phase II. CDDP, Raltitrexed and 5-FU mean actually delivered dose intensities at the selected dose level were 26, 1.05, and 378 mg/m2/week, respectively. Neutropenia was the main side effect and was observed even at the lowest dose levels. Nonhematologic side effects were mild. Nine complete responses (20%) and twenty-one partial responses (47%) were observed, for an overall response rate of 67% (95% confidence interval (95% CI): 51%-80%), according to intention to treat analysis. Fifteen of fifteen patients (100%) treated at the dose level selected for phase II had an objective response (5 complete responses, 10 partial responses).
Conclusions: The results of our dose escalation clearly demonstrate that it is possible to combine CDDP, raltitrexed, and modulated 5-FU at effective doses, without unexpected toxicities. The response data point to an impressive clinical activity, which will be better defined by an ongoing large phase II study.