This study involves two phases. Phase I of this study is designed to find out the maximum dose of paclitaxel poliglumex which can be safely given to subjects when combined with cetuximab and radiotherapy in head and neck cancer. Once the maximum safe dose of paclitaxel poliglumex is found, Phase II of the study will continue to find out whether the addition of paclitaxel poliglumex increases tumor response and survival compared to treatment with cetuximab and radiotherapy alone.

An additional 20 patients have been added, to balance data. These patients must be HPV negative.

Patients with locally advanced (stage III and IV) head and neck cancer are often managed by radiotherapy with or without chemotherapy because most of them have unresectable tumor, require too extensive surgery, or are medically unfit to go through radical surgery. However, the treatment results from conventionally fractionated radiotherapy for locally advanced head and neck cancers are poor in terms of local control and survival. Therefore, combinations of radiation and chemotherapy have been studied to improve treatment results.

Sequential radiation-chemotherapy (most given in neo-adjuvant setting) has been studied extensively in prospective pilot and large randomized trials. So far, a survival advantage over standard radiotherapy has not been demonstrated, but organ preservation has been achieved in many patients. Response rates to chemotherapy are high, and decrease in distant metastases has been demonstrated in some trials. Despite a high response rate in trials comparing neoadjuvant chemotherapy and radiotherapy to radiotherapy alone, improved locoregional control (LRC) has not been shown. Concurrent radiation and cisplatin-based chemotherapy has shown survival advantage over radiotherapy alone in meta-analysis. However, the administration of cisplatin-based chemotherapy is associated with significantly increased local and systemic toxic effects, which may preclude many patients from proceeding with combined therapy. Therefore, there is a great interest in defining an active regimen that does not contain cisplatin.

An alternative approach to concurrent chemotherapy and radiotherapy has emerged with the development of molecular targeted agents. A recently reported randomized phase III study demonstrated improved duration of control of locoregional disease and overall survival with the addition of the antibody against the epidermal growth factor receptor, cetuximab, to definitive radiotherapy in patients with squamous cell carcinoma of the head and neck. Importantly, cetuximab administration did not increase radiation-related toxicity.

The most commonly used chemotherapy other than cisplatin chemotherapy for the treatment of advanced head and neck cancer is paclitaxel. There are many studies showing improvement of tumor control when paclitaxel was added to the radiotherapy. Paclitaxel poliglumex (PPX, CT-2103, Xyotax) is a macromolecule that consists of a biodegradable, water-soluble polymer of glutamic acid, a naturally-occurring amino acid, linked to paclitaxel. Preclinical studies suggest increased tumor uptake of PPX compared with paclitaxel, resulting in enhanced tumor cell kill. PPX may potentiate tumor radiocurability without affecting acute normal tissue injury. Moreover, a synergistic increase in tumor cell death was observed when paclitaxel poliglumex was administered with cetuximab in a preclinical tumor model.

The proposed study will assess the rational combination of PPX with radiotherapy and cetuximab. This regimen is of great interest and has the potential to improve the therapeutic ratio compared with an approach of either cisplatin-based chemoradiotherapy or radiotherapy and cetuximab.

There is also an optional tissue submission component of this study, in which subjects who require surgery following their treatment can give permission for a block of tumor tissue removed at the time of their surgery to be sent to Cell Therapeutics, Inc. (the manufacturer of PPX) for evaluation of PPX accumulation, level of cathepsin B, and estrogen receptor expression. This information will be used to correlate the tumor response and survival of patients in the future.

Since the initiation of this study, the relationship of HPV to head and neck cancer has become very evident. Our initial results have many more HPV positive subjects, and therefore we have added 20 more HPV negative patients to the study, to determine if this status affects the outcome.

Inclusion Criteria:

• Patients with histological proof (from the primary lesion and/or cervical lymph node) of squamous carcinoma of the oral cavity, oropharynx, hypopharynx, larynx, or unknown primary.
• Patients should have stage III or IV disease
• Patients must have ECOG Performance Status of 0-1
• Patients must be >/= 18 years of age
• Patients must have measurable disease
• Patients should have adequate bone marrow function defined as an absolute peripheral granulocyte count (AGC) of >/= 1500 cells/mm3, platelet count of >/= 100,000 cells/ mm3; adequate hepatic function with bilirubin </= 1.5mg/dl, AST and ALT </= 2x the upper limit of normal; serum creatinine </= 1.5mg/dl, creatinine clearance >/= 50 ml/min and INR 0.8 - 1.2
• Patients must sign a study specific informed consent form prior to study entry
• Final 20 subjects must be HPV negative

Exclusion Criteria:

• Histology other than squamous cell carcinoma
• Evidence of metastases (below the clavicle or distant) by clinical or radiographic examinations for phase II study subjects
• History of malignancy other than non-melanoma skin cancer
• Prior chemotherapy or anticancer biologic therapy for any type of cancer, or prior radiotherapy to the head and neck region except for radioactive iodine therapy
• Prior history of allergy or hypersensitivity to cetuximab or paclitaxel
• Weight loss > 10% in the past three months
• Patients with uncontrolled intercurrent disease
• Patients with currently active malignancy
• Pregnant or lactating women
• Female patients of childbearing potential who are unwilling to practice adequate contraception during study treatment and for two months after the last administration of study drug