Trial Information

A Phase II Trial Evaluating Weekly Docetaxel and Capecitabine in Patients With Metastatic or Advanced, Locally, Recurrent Head and Neck Cancer

Approximately 28,900 patients will be diagnosed with squamous cell cancers of the oral
cavity and pharynx in the year 2002. Of these, an estimated 7,400 patients will present
with metastases or develop recurrent disease, which is not amenable to surgery or radiation
therapy. Palliative chemotherapy is thus the only treatment option. Currently,
combinations of cisplatin and 5-fluorouracil are used as first line treatment strategies,
with median times to progression of 2.5 to 3 months and median survival rates of 5 to 7
months. The time to achieve maximum response with combination therapy is on average 4
months.

Taxanes, with their unique mechanism of microtubule stabilization, have demonstrated
response rates similar to standard, first line combination regimens. Several phase II
studies have evaluated the efficacy of single agent docetaxel in head and neck cancer
patients. Cumulative response rates were approximately 30%, with higher response rates
observed in patients receiving no prior chemotherapy. Docetaxel has also been used in
combination with cisplatin and cisplatin and 5-fluorouracil. Although response rates with
such combination regimens were superior to the use of single agent docetaxel, grade 3 and 4
toxicities were also more prevalent.

Capecitabine (Xeloda®), a fluoropyrimidine carbamate, is an oral prodrug, which is converted
in tumor tissues to 5-fluorouracil through multienzymatic activation. Capecitabine
(Xeloda®) has documented activity in breast and colorectal cancers and is widely
administered. Because 5-fluorouracil has efficacy in the treatment of head and neck cancer,
it is reasonable to consider that such tumors will respond to capecitabine. To date, there
are no published trials using capecitabine (Xeloda®) in the treatment of metastatic head and
neck cancer patients. However, clinical trials are ongoing in the U.S. and Europe with
promising results.

In preclinical models, a synergistic interaction between capecitabine and docetaxel has been
documented. One possible explanation for the synergy is that docetaxel up-regulates tumor
levels of thymidine phosphorylase, the enzyme essential for the activation of capecitabine
and 5'-dFUrd to 5-fluorouracil. Clinically, O'Shaughnessy, et al. recently reported
improved survival with docetaxel/capecitabine combination therapy in patients with
metastatic breast cancer, who previously failed anthracycline-containing chemotherapy. In
this phase III study, patients were stratified according to previous exposure to paclitaxel
and then randomized to capecitabine (Xeloda®) (1250 mg/m2 twice daily, days 1-14) plus
docetaxel (75 mg/m2 day 1, repeated every 21 days) versus docetaxel alone. Grade 3 and 4
toxicities were more common in the docetaxel/capecitabine combination arm. Capecitabine
(Xeloda®) and docetaxel were interrupted and the dosages reduced by 25% in patients who
experienced a second occurrence of a given grade 2 toxicity, or any grade 3 toxicity,
suggesting that the starting dosages were perhaps too high.

The role of chemotherapy in metastatic head and neck cancer is limited to palliation of the
symptoms of disease. Platinum and 5-fluorouracil combinations remain standard first line
treatment strategies. The taxanes have been shown to have similar efficacy to such first
line regimens and are often used as salvage treatment for patients with metastatic disease.
Given that docetaxel has documented clinical efficacy in head and neck cancer and that
there are preclinical data to suggest synergy with docetaxel and capecitabine, it is
reasonable to consider using these agents in combination to treat head and neck cancer
patients. Moreover, capecitabine and docetaxel have distinct mechanisms of action and no
overlap of key toxicities. A recent phase I/II study by Tonkin, et al. in metastatic breast
cancer patients demonstrated activity and less toxicity when docetaxel 30 mg/m2/week (day 1
and 8 q21 days) was combined with capecitabine 1800 mg/m2/day (14 of 21 days). In another
phase I study by Nadella, et al. weekly docetaxel (36 mg/m2 ) was combined with 14 days of
capecitabine (up to 1500 mg/m2/day) over a course of 28 days. Antitumor responses were
observed in patients with breast, colon, and bladder cancers. Hence, we propose this study
whereby patients with previously treated, metastatic/recurrent head and neck cancer will
receive treatment with docetaxel and capecitabine.

To reduce the potential for toxicity, we will use a modification of the Nadella regimen.
Docetaxel will be administered weekly at a dosage of 30 mg/m2 for 3 out of every 4 weeks and
capecitabine will be administered at a flat dosage of 2000 mg per day (1000 mg p.o. b.i.d.)
for two weeks out of every 4 weeks. The justification for using a flat dosage of
capecitabine versus a calculated dosage is based on pharmacokinetic data that show no change
in clearance of capecitabine with changes in BSA. We plan to use a fixed dose of 2000 mg qd
(1000 mg q am and 1000 mg q pm). Fixed dosing of capecitabine is convenient and feasible, as
shown in a prior University of Michigan study in breast cancer patients. In another study
Schott, et al. informally piloted the combination of weekly docetaxel 36 mg/m2 and 1500 mg
twice daily (3000 mg/day) x 14 days capecitabine in metastatic breast cancer patients, and
found it to be without unexpected or untoward side effects. Additionally, to take advantage
of the time course of upregulation of TP in the preclinical models, the capecitabine dose
will be given on days 5-18. In a flat dosing scheme, the Nadella regimen would have
administered an average dose of 2125 mg qd for 14 days, assuming an average BSA of 1.7 m2.
We plan to round this dosage downward to 2000 mg per day x 14 days; therefore, our regimen
will use a slightly lower dosage of capecitabine. We feel that our proposed slightly lowered
dose (closer to Nadella phase I dosing vs. Tonkin) of capecitabine is justified for the
following reasons:

1. The Nadella study was performed in a group of patients with solid tumor malignancies
that were refractory to conventional therapy or for whom no effective therapy existed.
ECOG performance status (PS) was 1 or 2 in 5/17 (30%) patients, 10/17 patients had
received 2 or more lines of previous chemotherapy, and 7/17 patients had received
previous radiotherapy. Based on data from previous treatment of head and neck cancer
patients at the University of Michigan, the patient population to be enrolled in this
trial is expected to be 60% PS 0 and 40% PS 1, and some will have received prior
chemotherapy and/or radiotherapy. Like the Nadella patient population, a majority of
our patients have been pretreated and are of poorer health.

2. Dose interruptions and modifications are built into the protocol so that appropriate
changes in treatment can be made in patients with Grade I or II toxicity, before the
toxicity becomes Grade III or greater. Since the docetaxel is given weekly, and the
capecitabine is administered daily, if patients are experiencing toxicity within a
cycle, the dose of either can be held or modified.