Determine the safety of a combination of IFN, celecoxib, and rintatolimod for patients with recurrent colorectal cancer. This will also test whether the above combination can help the immune system to fight the tumors. The results will allow the investigators to determine the "preferred" combination for subsequent extended studies.

A previously-demonstrated correlation between the density of CRC-infiltrating effector T cells and long-term outcomes (Galon et al., 2006; Pages et al., 2005) has been established. In preclinical ex vivo studies performed using explants of resected metastatic CRC, the combination of IFNα with nonselective or COX2-selective inhibitors of prostaglandin synthesis resulted in elevated production of the effector T cell-attracting chemokines CXCL10 and CCL5. This was associated with concomitant suppression of the intratumoral expression of CCL22, a Treg-attracting chemokine (Muthuswamy et al 2008 Canc Res, and Muthuswamy et al, submitted to Canc Res 2011). However, in a subset of patients, the optimal results, particularly with regard to CCL5 induction, required additional stimulation by a third agent, poly-I:C (a toll-like receptor -TLR Ligand).

Therefore, the investigators seek to establish the safety profile of a novel chemokine regimen consisting of IFN, celecoxib and poly-I:C. The investigators also hypothesize that the proposed neoadjuvant chemokine modulation treatment in recurrent CRC patients undergoing tumor resection may increase the density of tumor infiltrating lymphocytes (TILS).

In addition, treatment in the neoadjuvant setting will allow a comparative analysis of the effect of chemokine modulation on the local recruitment of effector-type T cells and the de-recruitment of Treg within resected tumor tissues; helping to determine the "preferred" chemokine-modulating regimen for subsequent extended studies. Such prospective studies will focus on using combinations of chemokine modulation and cancer vaccines in patients with CRC. The investigators have, for example, recently observed that αDC1, a new type of DC vaccine (Kalinski and Okada, 2010; Mailliard et al., 2004) is particularly effective in inducing the effector pathway of T cells differentiation. This was manifested by the induction of tumor-killing function and the induction of effector-type chemokine receptors (CXCR3 and CCR5) (Kalinski and Okada, 2010; Watchmaker et al., 2010). Combining the αDC1 vaccine to a safe, tolerable and efficacious CKM regimen may hold promise for patients with poor prognostic CRC.

Inclusion:

• Recurrent and/or metastatic resectable colorectal cancer, including disease within the abdomen and pelvis with no evidence of extra-abdominal metastases. Isolated resectable pulmonary metastasis are allowable in the absence of intra-abdominal metastasis. Intra-abdominal disease includes: isolated hepatic metastasis/metastases (see next inclusion criteria point), isolated peritoneal metastasis, peritoneal carcinomatosis (including patients undergoing cytoreductive surgery alone or in combination with hyperthermic intraperitoneal chemoperfusion - HIPEC), or a combination of hepatic and extrahepatic metastasis.

• Patients with isolated hepatic metastasis must satisfy a Clinical Risk Score of 3 or higher (see Appendix C)

• Eligible patients are expected to have a complete resection based on preoperative imaging. Any patient not found to be able to have complete resection will not be eligible for this study.

• No chemotherapy, radiotherapy, major surgery, or biologic therapy within 3 weeks of protocol treatment

• An ECOG performance status of 0, 1, or 2.

• Age equal to 18 years or older.

• Must have normal organ and marrow function as defined below:

  • Platelet ≥ 75,000/µL
  • Hemoglobin ≥ 9.0 g/dL
  • Hematocrit ≥ 27.0%
  • Absolute Neutrophil Count (ANC) ≥ 1500/µL
  • Creatinine < institutional upper limit of normal (ULN) OR
  • Creatinine clearance ≥ 50 mL/min/1.73 m2 for patients with creatinine levels greater than ULN
  • Total bilirubin ≤ 1.5 X institutional upper limit of normal (ULN)
  • AST(SGOT) and ALT(SGPT) ≤ 2.5 X institutional upper limit of normal (ULN)
  • Serum amylase and lipase within normal limits.

• Patient must be able to understand and be willing to sign a written informed consent document.

Exclusion:

• Patients currently treated with systemic immunosuppressive agents, including steroids, are ineligible until 3 weeks after removal from immunosuppressive treatment.

• Patients with active autoimmune disease or history of transplantation.

• Patients who are pregnant or nursing. Women of childbearing potential (WOCBP) will have to undergo a urine pregnancy test as part of screening.

• Patients with comorbid medical conditions that render them unfit for surgery.

• Metastatic or recurrent disease that is deemed partially resectable or unresectable based on preoperative imaging.

• Metastatic disease outside the confines of the abdomen, pelvis and thorax (e.g bone, brain)

• Cardiac risk factors including:

  • Patients experiencing cardiac event(s) (acute coronary syndrome, myocardial infarction, or ischemia) within 3 months of signing consent
  • Patients with a New York Heart Association classification of III or IV (Appendix A)

• History of upper gastrointestinal ulceration, upper gastrointestinal bleeding, or upper gastrointestinal perforation within the past 3 years. Patients with ulceration, bleeding or perforation in the lower bowel are not excluded.

• Prior allergic reaction or hypersensitivity to sulfonamides, celecoxib, or NSAIDs.

• Patients are ineligible if they plan on regular use of NSAIDs at any dose more than 2 times per week (on average) or aspirin at more than 325 mg at least three times per week, on average. Low-dose aspirin not exceeding 100 mg/day is permitted. Patients who agree to stop regular NSAIDs or higher dose aspirin are eligible and no wash out period is required.