This study uses the drug PPX (also called Xyotax and CT-2103) in women with advanced colorectal cancer. PPX is an experimental drug that has not been approved by the Food and Drug Administration (FDA). PPX has been shown in the laboratory and in studies in humans to cause some cancer cells to die and some tumors to shrink. Women in some studies with PPX have been shown to live longer than the men that receive the drug. Some studies in humans suggest that estrogen (a hormone found in women) may protect women from getting colorectal cancer and allow women that do get colorectal cancer to live longer than men that do.

The purpose of this study is to see if women with colorectal cancer and a certain level of estrogen experience tumor shrinkage after they receive the drug PPX. This study will also study genes (genes are the cell's blueprint) in participant's tumors and in their blood. Several genes can affect how people's bodies react to the cancer drugs. We want to see if these predict response to the study drugs.

Inclusion Criteria:

• Women with histologically confirmed metastatic adenocarcinoma of the colon or rectum. The primary histologic diagnosis is sufficient if there is clear evidence by imaging and/or markers of metastatic disease sites.
• Must have failed or are intolerant of or ineligible for CPT-11, 5-FU, oxaliplatin, bevacizumab and either cetuximab or panitumumab therapies.
• Tumor must be accessible for biopsy or paraffin embedded tissue must be available for review.
• SWOG performance status 0-2.
• Estradiol levels >30 pg/mL **This may be supplemented by exogenous estrogen (by gel)
• AGC >1,500, platelets >100,000
• Total bilirubin < 3 x upper limit of normal, Transaminase (AST and/or ALT) < 2 x upper limit of normal or < 5 x upper limit of normal in patients with liver metastasis.
• Patients must have a creatinine of < 1.5 x upper limit of normal or a measured or calculated creatinine clearance greater than 35 mL/min obtained within 7 days of first receiving study drug.
• Except for cancer-related abnormalities, patients should not have unstable or pre-existing major medical conditions.
• At least one measurable lesion according to the RECIST criteria which has not been irradiated (i.e. newly arising lesions in previously irradiated areas are accepted). Ascites, pleural effusion, and bone metastases are not considered measurable. Minimum indicator lesion size: > 10 mm measured by spiral CT or > 20mm measured by conventional techniques.
• Have a negative serum pregnancy test within 7 days prior to initiation of chemotherapy (female patients of childbearing potential).
• Life expectancy > 3 months.

Exclusion Criteria:

• History of a malignancy other than colon or rectal cancer, except for adequately treated basal cell or squamous cell skin cancer, in situ cervical cancer, or other cancer for which the patient has been disease-free for five years.
• Estradiol levels < 30 pg/mL, not responsive to supplementation.
• History of previous thromboembolic event, unless patient is on anticoagulation therapy.
• Grade 2 or greater neuropathy.
• Pregnant or lactating woman. Woman or men of childbearing potential not using a reliable and appropriate contraceptive method (either a barrier or hormonal method is acceptable). Patients must agree to continue contraception for 30 days from the date of the last study drug administration. (Postmenopausal woman must have been amenorrheic for at least 12 months to be considered of non-childbearing potential).
• Patients with known brain metastases, unless they are well controlled - i.e. on a stable dose of steroids or if steroid therapy has been completed.
• Patients that have received experimental therapies or other approved bio- or chemotherapies within 30 days of study entry.