The goal of this clinical research study is to find the highest safe dose of docetaxel in combination with Gleevec (imatinib mesylate) that can be given to men with advanced androgen-independent metastatic prostate cancer that involves bone. Docetaxel is a commercial chemotherapy which interferes with the cancer cell ability to divide and grow.

OBJECTIVES:

1. To define the maximum tolerated dose of weekly docetaxel in combination with fixed-dose oral STI571 in adult men with metastatic androgen-independent prostate cancer (AIPC).
2. To determine the qualitative and quantitative toxicity of the combination of oral STI571 and docetaxel.
3. Evaluate PSA modulation with STI571 alone at thirty days in patients with AIPC.
4. Obtain a preliminary estimate of the response rate in AIPC to the combination of STI571 and docetaxel.
5. Obtain tissue for correlative science studies (these are optional studies).

• Drug: imatinib mesylate
• Drug: docetaxel

Inclusion:

• Patients with histologic proof of adenocarcinoma of the prostate and must have progressed on conventional hormonal therapy.
• Patients must have bone metastases which can be demonstrated by bone scans. Lytic bone lesions should be considered for biopsy if there is a clinical suspicion of histologic conversion to small cell carcinoma.
• Patients must have evidence of progression of disease. PSA- progression is defined as 2 consecutive increments in PSA (an absolute change of at least 1ng/mL) over 4 weeks. An increase by 25% of the product of bidimensional disease qualifies as progression. An increase in the number of metastatic lesions on bone scan qualifies as progression.
• All patients must have a minimum PSA of 1ng/ml.
• Patients on antiandrogens should be discontinued from flutamide or nilutamide for at least 4 weeks and bicalutamide for 8 weeks. If progression is documented as below prior to this time interval, patients are eligible.
• Patients must have a performance status of < 2 (ECOG).
• Patients must have an expected survival from cancer or co-morbidity of at least three months.
• Patients may receive no concurrent chemotherapy, immunotherapy or ketoconazole.
• Patients should not have received prior chemotherapy or radiation within the last 30 days and no Strontium or Samarium within the last 90 days.
• Patients must have castrate serum testosterone levels (< 30ng/dl). For patients who are medically castrated, luteinizing hormone releasing hormone analog must continue to maintain testicular suppression.
• Patients must have adequate bone marrow function defined as an absolute peripheral granulocyte count of > 1,500/mm3 and platelet count of > 100,000/mm3.
• Patients should have adequate hepatic function defined with a bilirubin of < 1.5 mg/dl and AST/ALT < 2X the upper limits of normal.
• Patients should have adequate renal function defined as serum creatinine clearance > 40 cc/min (measured or calculated by Cockcroft and Gault formula) or serum creatinine < 1.5 X upper limit of normal.
• Fully recovered from any previous surgery (at least 4 weeks since major surgery.
• Patients must sign an informed consent indicating that they are aware of the investigational nature of this study, in keeping with the policies of the institution. The only approved consent is attached to this protocol.

Exclusion:

• Patients with severe intercurrent infection.
• Patients whose tumors contain small cell or sarcomatoid elements.
• Patients with NYHA Class III/IV CHF, unstable angina or MI in the last 6 months or evidence of active myocardial ischemia on ECG.
• CNS metastases that are uncontrolled.
• Prior hypersensitivity or dose-limiting toxicity with docetaxel.
• Oxygen-dependent lung disease
• Contraindications to corticosteroids.
• Uncontrolled severe hypertension or uncontrolled diabetes mellitus.
• Second malignancies (except non-melanoma skin cancer) unless disease-free for 3 years.
• Overt psychosis or mental disability or otherwise incompetent to give informed consent.
• Patients with a history of non-compliance with medical regimens or who are considered potentially unreliable.