585 Background: For patients undergoing surgery for cancer, it has been suggested that risk-adjustment with cancer-specific variables is needed when evaluating short-term outcomes. Our objectives were to assess the influence of cancer-related variables on postoperative complications and hospital quality comparisons. Methods: Patients from ACS NSQIP and NCDB who underwent colorectal resection for cancer were linked (2006-2008) to create a dataset containing robust information on comorbidities, complications, and oncologic variables. Three hierarchical models were developed predicting the NSQIP outcome 30-day mortality or any serious morbidity using variables from (1) NSQIP only, (2) NCDB only, and (3) a combined model using NSQIP and NCDB. Models were compared with fit statistics and hospital outlier agreement. Results: From 146 NSQIP hospitals, 11401 patients underwent a colorectal resection for cancer, of which, 1954 (17%) experienced a mortality or serious morbidity event. The first five variables selected in the NCDB-only model were Charlson comorbidity score, neoadjuvant therapy use, T stage, primary payer, and M stage (c-statistic, 0.64; AIC, 9886). The first five variables selected in the NSQIP-only model were ASA class, preop sepsis, albumin, surgical procedure, and COPD (c-statistic, 0.66; AIC, 9787). In the combined model, neoadjuvant therapy use was the only cancer-specific variable selected in the top five. The remaining variables were ASA class, preop sepsis, albumin, and wound class (c-statistic, 0.67; AIC, 9455). At the hospital-level, the NCDB-only model identified three high outliers (worse than expected) and one low outlier (better than expected). Both the NSQIP-only and combined models identified the same four high and two low outlying hospitals (kappa: 1.0), which agreed marginally with the NCDB-only model (kappa: 0.59). Conclusions: Addition of cancer-specific variables to NSQIP models slightly improved model fit; however, hospital outcome comparisons were identical. For patients with colorectal cancer undergoing resection, cancer-related factors have limited predictive ability for short-term outcomes and did not influence hospital quality comparisons.

10584 Background: Colon cancer is the fourth most commonly diagnosed cancer and the second leading cause of cancer death in the United States. Mutations of the APC gene are found in the majority of sporadic cases of polyps and cancer and are the cause of familial adenomatous polyposis. The role of inflammation in the progression of polyps to cancer is well established. Cytotoxic T-lymphocyte antigen 4 (CTLA4) is a member of the CD28-B7 immunoglobulin superfamily that mediates non-antigen specific interactions w/ antigen presenting cells and prevents or down-regulates the activation of T-cells in response to a stimulus. Treatment of cancer patients with anti-CTLA-4 antibody (aCTLA4) has shown some promise in melanoma, prostate and ovarian tumors. We have recently observed polyp specific T-cell responses in a mouse model of polyposis and hypothesize that treatment with aCTLA4 will enhance this response, resulting in a reduction of polyps. Methods: Using a well-established murine model of polyposis, APCΔ468, animals were randomized to receive either 100µg monoclonal aCTLA4, 9H10 (BioXCell), or PBS. Intraperitoneal injections were performed every other day for 3 doses. The primary endpoint was number of polyps in the small intestine and large intestine. Animals were sacrificed and polyps were counted by gross inspection at 3-weeks (n=6 for each group) and 6-weeks (n=3 for each group) post injection. Interferon-γ ELISPOT analysis was performed on splenic lymphocytes to evaluate tumor specific immunity. Results: A reduction in small intestinal polyp count was seen in the 3-week aCTLA4 group compared to controls, 48 +/- 13 vs. 100 +/- 12 (p=0.0154). This decrease was sustained at 6-weeks, 46 +/- 6 (p=0.0032). No significant difference was noted in large intestinal polyp counts at 3- or 6-weeks. Preliminary ELISPOT analysis revealed a trend toward increased spots in the aCTLA4 group. Conclusions: Anti-CTLA4 treatment with a monoclonal antibody resulted in a reduction of polyp load in a murine model of intestinal polyposis up to six weeks following treatment. Thus, aCTLA4 may be useful for treatment of intestinal pre-neoplasia. Further studies are needed to assess the immunologic affect of aCTLA4 in the treatment of polyposis and colon cancer.

e16615 Background: We compared 3-year event-free survival of veterans with colon cancer who used Department of Veterans Affairs (VA) and Medicare healthcare to those who used VA only.METHODSWe identified veterans diagnosed with stage I, II, or III colon cancer between 1999 and 2001 in VA central tumor registry and 8 regional NCI SEER registry records. Inclusion criteria included age 66 or older and Medicare fee-for-service enrollment. We searched VA utilization and Medicare claims data for healthcare events signaling cancer relapse or progression, defined separately for each stage. Using extended Cox models, we evaluated relationships between VA-Medicare dual use and relapse/progression events.RESULTSWe identified 1,007, 1,169, and 898 patients diagnosed with stage I, II, and III colon cancer, respectively. Of those, 20%, 15%, and 11% received colon cancer care in both VA and non-VA settings. At 36 months, event-free survival rates were 84% (stage I), 53% (stage II), and 36% (stage III). Among stage I patients, dual users were twice as likely as non-dual users to experience a relapse/progression event (adjusted HR 2.01, CI95%: 1.40-2.88). African American race and distance to a VA outpatient facility greater than 4.6 miles were also associated with higher odds of a relapse/progression event. Among stage II and III patients, dual use was not associated with relapse/progression (adjusted HR 1.22, CI95%: 0.96-1.56 and HR 1.22 CI95%: 0.94-1.58, respectively). For stage II patients, older age, being unmarried, having a greater comorbidity burden, and higher tumor grade were associated with higher odds of a relapse/progression event. For stage III, patients who were older, had a greater comorbidity burden, higher tumor grade, or no adjuvant chemotherapy had higher odds of an event.CONCLUSIONSAmong VA- and Medicare-eligible veterans with stage I colon cancer, dual system healthcare use was associated with increased likelihood of a clinical event suggesting cancer relapse or progression within 3 years. Understanding the specific processes by which dual use leads to cancer relapse deserve further study.

Pancreatic ductal adenocarcinoma (PDAC) is associated with a pronounced collagen-rich fibrosis known as desmoplastic reaction; however, the role of fibrosis in PDAC is poorly understood. In this report we show that collagen can regulate the tumor suppressive let-7 family of microRNAs in pancreatic cancer cells. PDAC cells growing in 3D collagen gels repress mature let-7 without affecting the precursor form of let-7 in part through increased expression of membrane type 1-matrix metalloproteinase (MT1-MMP, MMP-14) and ERK1/2 activation. PDAC cells in collagen also demonstrate increased TGF-β1 signaling, and blocking TGF-β1 signaling attenuated collagen-induced MT1-MMP expression, ERK1/2 activation and repression of let-7 levels. Although MT1-MMP overexpression was not sufficient to inhibit let-7 on 2D tissue culture plastic, overexpression of MT1-MMP in PDAC cells embedded in 3D collagen gels or grown in vivo repressed let-7 levels. Importantly, MT1-MMP expression significantly correlated with decreased levels of let-7 in human PDAC tumor specimens. Overall, our study emphasizes the interplay between the key proteinase MT1-MMP and its substrate type I collagen in modulating microRNA expression, and identifies an additional mechanism by which fibrosis may contribute to PDAC progression.

Introduction: The arachidonic acid pathway enzymes, cyclooxygenase (COX) and 5-lipoxygenase (5-LO), have both been shown to modulate cell growth and apoptosis. Many studies have investigated the effects of COX inhibition on human colorectal cancer, but the role of 5-LO in colorectal cancer development has not been well studied. Interestingly, 5-LO has been shown to be a major mediator of mast cell proliferation and migration. Mast cells have recently been shown to be essential for polyposis. The aim of the present study was to evaluate if 5-LO does indeed have a role in colorectal carcinogenesis, and if so, whether this role may be mediated by mast cells. Methods: To assess the role of 5-LO in colorectal tumorigenesis, we examined the effects of 5-LO gene knockouts on APCΔ468 knockout mice (APC), a model of human familial adenomatous polyposis. APC/5LO(-/-) mice were sacrificed and evaluated at 16 weeks of age. Age-matched APC/5LO(+/+) mice were used as positive controls. Intestinal polyp counts and polyp diameters were measured in both groups. Mice were also injected with BrdU solution 2 hours prior to cull. Polyp mitotic index was then scored by measuring BrdU incorporation using immunohistochemistry(IHC). The mast cell's contributory role in colorectal tumorigenesis was also investigated by isolating total mononuclear cells(MNCs) from the intestines of the APC/5LO(-/-) and APC/5LO(+/+) groups. These MNCs were then grown in a 2-week culture, and the number of mature mast cells was scored. Results: 5-LO null mutation led to a dramatic reduction in the number and size of intestinal polyps. The APC/5-LO(+/+) group developed 69 ± 5 polyps at 16 weeks. In contrast, age-matched APC/5-LO(-/-) mice had 31 ± 6 polyps, an approximately 55% reduction (p<0.01). Median polyp diameter was 1.5 mm in the APC/5-LO(-/-) group, a reduction compared to 1.8 mm in APC/5-LO(+/+) controls(p<0.05). The mitotic index was significantly lower in the APC/5LO(-/-) group compared to the APC/5LO(+/+) group (6.9% ± 0.6 versus 17.2% ± 1.1, respectively; p<0.01). Intestines from the APC/5LO(-/-) group yielded 5,465 ± 594.2 mature mast cells after 2-week culture. This was a statistically significant reduction from the mast cell yield of the APC/5LO(+/+) group [10,700 ± 81.00, p<0.01]. Conclusions: These findings provide direct genetic evidence that 5-LO has an important role in tumorigenesis which in turn, is partially mediated through mast cells. Our data points to 5-LO-selective inhibitors as a potential, novel class of therapeutic agents for colorectal polyposis and cancer.