Citations

522 Background: Metformin inhibits mTOR and downstream pathways involved in cell proliferation and survival by inducing AMPK activation. Although its use has been associated with decreased risk of colorectal cancer (CRC), the effect of metformin on cancer related mortality is unclear. We performed a meta-analysis of published observational studies evaluating metformin use in patients with CRC and its association with overall and CRC-specific survival. Methods: Through a systematic search of PubMed and conference presentations (cut-off: August 2013), we identified 8 observational studies (18,537 total patients with CRC; 6242 CRC patients with DM) reporting the association between metformin use and survival after diagnosis of CRC. Summary hazard ratio (HR) with 95% confidence intervals (CI) was estimated using the random effects model, and heterogeneity was measured using the inconsistency index (I2). Results: On meta-analysis, metformin use was associated with a 34% reduction in mortality in CRC patients as compared to non-use (adjusted HR, 0.66; 95% CI, 0.54-0.81), albeit with high heterogeneity (I2=66.9%). After exclusion of one study, which contributed to considerable heterogeneity, a 28% reduction in mortality was considered a more conservative, consistent effect estimate (adjusted HR, 0.72; 95% CI, 0.64-0.81), with minimal heterogeneity (I2=16.2%). On restricting analysis only to patients with DM, use of metformin (vs non-use) was associated with decreased all-cause mortality (6 studies; adjusted HR, 0.63; 95% CI, 0.49-0.81) and CRC-specific mortality (3 studies; adjusted HR, 0.66; 95% CI, 0.50-0.87). Conclusions: Based on meta-analysis of observational studies, the use of metformin in diabetic patients with CRC appears to be associated with reduced all-cause mortality and CRC-specific mortality. Further prospective investigation of metformin’s effect on outcome in CRC patients is warranted.

492 Background: Beclin 1 interacts with its cofactor UVRAG to promote autophagy, a cellular degradation process induced by cellular stress. While cells with defective autophagy are prone to genomic instability, it is unknown whether Beclin 1 and/or UVRAG can regulate DNA damage/repair. Methods: Colorectal cancer cell lines (HT-29, DLD1) with Beclin 1 or UVRAG siRNA were irradiated ± 5-FU. DNA damage (pATM, pH2Ax, 53BP1, RAD51), cell death (annexin V, caspase-3, clonogenic survival), and centrosome number (γ-tubulin staining), were determined. Beclin 1 protein expression was analyzed in pretreatment endoscopic biopsies from 96 rectal carcinoma patients who received radiation + 5-FU followed by surgery. Results: Beclin 1 or UVRAG knockdown was shown to enhance DNA double-strand breaks (DSBs) and cell death induced by radiation ± 5-FU. Suppression of either gene increased the percentage of irradiated cells with nuclear foci expressing 53BP1 (nonhomologous end joining), but not RAD51 (homologous recombination). Beclin 1 was shown to regulate UVRAG, and a UVRAG deletion mutant defective in Beclin 1 binding decreased protection against radiation-induced DSBs. Knockdown of Beclin 1 or UVRAG increased centrosome number in irradiated cells. While autophagy inhibition by ATG5 or LC3 siRNA enhanced radiation-induced DSBs, they did not alter centrosome number suggesting independence from autophagy. In rectal carcinomas, high vs low level Beclin 1 expression was detected in 56 (58.3%) and 40 (41.7%) tumors, respectively. Pre-treatment Beclin 1 was not associated with clinicopathological variables. However, high vs low Beclin 1 predicted a significantly reduced pathological response (macroscopic vs. microscopic or no residual tumor) to chemoradiation (high: 14.2% vs. low: 40%; p = 0.017). High vs low Beclin 1 was independently associated with poorer overall survival (HR, 4.30; 95% CI, 1.42-16.47; p = 0.009). Conclusions: Beclin 1 and UVRAG confer protection against radiation-induced DNA damage and maintain centrosome stability. The ability of Beclin 1 overexpression to confer resistance to chemoradiation, suggests its potential utility as a predictive biomarker in rectal cancer.

451 Background: Classifications for molecular subtypes of colorectal cancer based on microsatellite instability (MSI), the CpG island methylator phenotype (CIMP), and somatic mutations in BRAFV600E and KRAS have been proposed to reflect distinct pathways of colorectal tumorigenesis. We assessed the relationship between these subtype classifications and colorectal cancer survival in a population-based study. Methods: Tumor-markers were evaluated in incident colorectal cancer cases diagnosed between 1997-2007 in western Washington State. Cases were classified into pathway-based subtypes according to the following tumor-marker combinations: (1) traditional pathway [microsatellite stable / low MSI (MSS/MSI-L), non-CIMP, BRAF-wildtype, KRAS-wildtype, n = 631]; (2) MSI-high serrated (MSI-high, CIMP-positive, BRAF-mutated, KRAS-wildtype, N=100); (3) MSS/MSI-L serrated (MSS/MSI-L, CIMP-positive, BRAF-mutated, KRAS-wildtype, n = 55); (4) alternate pathway (MSS/MSI-L, non-CIMP, BRAF-wildtype, KRAS-mutated, n = 353); and (5) MSI-high familial (MSI-high, non-CIMP, BRAF-wildtype, KRAS-wildtype, n = 50). Multiple-imputation was used to classify tumor-marker status for cases with missing data on one to three markers. Differences in survival across subtypes were assessed through multivariable-adjusted Cox regression. Results: Relative to cases of the predominant traditional pathway subtype, cases with MSS/MSI-L serrated and alternate pathway tumor subtypes experienced statistically significantly worse disease-specific survival [hazard ratio (HR) = 2.25, 95% confidence interval (CI): 1.50-3.36 and HR = 1.39, 95% CI: 1.12-1.71, respectively]; cases with MSI-high familial tumors had the most favorable disease-specific prognosis (HR = 0.28, 95% CI: 0.12-0.64). With respect to overall survival, associations were similar but slightly attenuated. Conclusions: In this large, population-based study, colorectal cancer subtype classifications based on integrated pathways were associated with marked differences in survival, highlighting the significance of molecular heterogeneity in colorectal cancer.

42 Background: Methods of stratifying esophageal adenocarcinoma patients into prognostic groups are needed, as are new insights into genetic determinants of disease behaviour. Prognosis is likely to have non-negligible genetic influences, as mediated by host responses to tumor, resistance to therapeutic side-effects, and/or an influence on tumor development. Prior studies have used candidate-gene approaches. We took an alternative approach, using an unbiased, genome-wide approach, and novel analytic methods that may be better able to detect multi-gene interactions, which may contribute the majority of genetic effects for many clinical phenotypes. Methods: Germline DNA from a Toronto-based cohort of EAC patients (n=270) was analyzed by Omni1 Quad microarray as part of the BEAGESS initiative. Quality control and analysis was performed using PLINK, R, and GenABEL software packages. A Cox proportional hazards (CPH) model for progression-free survival tested each polymorphism for independent effects at a genome-wide significance level of P < 1E-07, adjusting for population stratification. While classical analysis has limited ability to detect gene-gene interactions, a Random Survival Forest algorithm was used to detect effects based on the complex interactions among top 1,000 polymorphisms by p-value ranking. Results: After data cleaning and standard GWAS quality control procedures, there were 735,309 SNPs and 245 patients remaining for analysis. The CPH model, adjusted for population stratification, produced a satisfactory Q-Q plot, and showed one SNP (rs7844673, Chr 8) that was significant at p=7.8E-8. In addition, Random Forest based variable selection produced a set of 20 polymorphisms that (1) reproduced 86% of the predictive ability of the full 1000 variables, and (2) also included the #3 ranked polymorphism by CPH modeling (rs9290822, Chr 3) upstream of the IGF2BP2 gene. Conclusions: A genome-wide approach has discovered two previously undescribed SNPs with a potential influence on EAC prognosis via a combination of independent and interactive effects. Validation in an independent cohort is currently being pursued.

396 Background: The 12-gene Oncotype DXColon Cancer Assay is clinically validated as a predictor of recurrence risk in stage II colon cancer patients following surgery. We previously reported that the 12-gene assay led to 45% change in physician treatment (Tx) recommendations in a prospective study in MMR-P stage IIA colon cancer patients. Here, from the same prospective study, we report the influence of the 12-gene assay on patient Tx decisions, physician confidence, concordance in Tx choice between physicians and patients, and patient decisional conflict. Methods: Consecutive patients with resected stage IIA colon cancer who were candidates for adjuvant chemotherapy were enrolled by 105 physicians from 17 sites. Patient's tumor specimens were assessed by the 12-gene assay (RT-PCR) and MMR (IHC). Prior to and after receiving these results, patients completed surveys including (1) their Tx decisions (observation (Obs) vs. 5FU-monotherapy (5FU) vs. 5FU + oxaliplatin (Oxal)) and (2) indicators of decisional conflict. Results: 190 of 221 patients enrolled were evaluable including 139 who were MMR-P. Pre-assay: 46% of patients chose Obs, 3% 5FU, 7% Oxal, 2% other and 41% were undecided. Post-assay: 75% chose Obs, 12% 5FU, 11% Oxal, 3% other (undecided was not a possible response). Post-assay 129 (96%) of 135 definitive Tx decisions (Obs, 5FU, or Oxal) were concordant between patients and physicians compared to 49 (66%) of 74 definitive decisions pre-assay. In the majority of cases, patients (85%) and physicians (69%) reported that the assay influenced their Tx decisions. Assay results increased physician confidence in Tx recommendations in 126 (84%) and provided additional clinically relevant information to physicians in 129 (86%) of cases. Patient decisional conflict was significantly lower after assay results (p<0.001). Conclusions: In this prospective study, quantitative recurrence risk information provided by the 12-gene assay influenced Tx decisions for a majority of patients and physicians, increased physician confidence, improved concordance in Tx choice between patients and physicians and decreased patient decisional conflict.

326 Background: Review initial outcomes for patients enrolled on the University of Florida Proton Therapy Institute PC01 protocol for patients with unresectable pancreatic cancer. Methods: The protocol received Institutional Review Board (IRB) approval in 7/2007. The first patient was enrolled in 4/2010. The accrual target was 66 patients to test if the Serious Adverse Event (SAE) ratecould be reduced from 15% (expected) to less than 5%. Protocol therapy consisted of proton therapy to a planning target volume (PTV) dose of 59.4 Cobalt Gray Equivalent (CGE) at 1.8CGE per fraction with concomitant oral Capecitabine (1000mg PO BID 5 days/week on radiation treatment days). Only gross disease was targeted. Results: 12 patients were enrolled. 1 patient died of a gunshot wound 5 days after starting treatment so only 11 patients are reported. Median follow up from start of treatment: 13 months (range 2 to 21 months). Median age: 68 (range 51 to 86) years. Gender: Males 5, Females 6. Race/Ethnicity: 9 White, 2 Hispanic. Overall survival: 90% at 6 months and 60% at 12 months. Freedom from progression: 80% at 6 months and 53% at one year. Local control: 100% at 6 months and 86% at 12 months. Freedom from metastasis: 80% at 6 months and 67% at 12 months.No patient experienced any grade 3 or greater toxicity during treatment or the follow up period. Grade 2 toxicity was limited to a single patient experiencing grade 2 fatigue. Median weight loss over the course of treatment was 1.7 (range: loss of 5.7 to gain of 4.9) Kg. 4 patients were deemed to have had an adequate radiographic response to radiotherapy so as to justify surgical exploration. Conclusions: Proton therapy to 59.4 CGE with concomitant Capecitabine was well tolerated with no grade 2 or greater gastrointestinal toxicities. 4 of 11 patients achieved a radiographic response allowing for attempted surgical resection. The study was closed to accrual on 9/1/2013 due to recognition that slow accrual made achievement of the primary study endpoint unlikely. At the same time, the lack of any meaningful gastrointestinal toxicity suggests significant opportunities for treatment intensification when proton therapy is used in this setting. Clinical trial information: NCT00685763.

11 Background: Acid-suppressive medications, particularly proton pump inhibitors (PPI), may modify risk of esophageal adenocarcinoma (EAC) in patients with Barrett’s esophagus (BE). We performed a systematic review and meta-analysis of studies evaluating the association between PPIs and histamine receptor antagonists (H2RA) and risk of EAC or high-grade dysplasia (BE-HGD) in patients with BE. Methods: Through a systematic search up to June 2013, we identified 7 observational studies (5 cohort studies and 2 case-control studies; 2,813 patients with BE, 317 cases of EAC and/or BE-HGD, 84.4% PPI users) reporting the association between PPIs or H2RA and EAC in patients with BE. Summary odds ratio (OR) with 95% confidence intervals (CI) was estimated using random effects model, and heterogeneity was measured using the inconsistency index (I2). Results: On meta-analysis, PPI use was associated with 71% reduction in EAC risk in patients with BE (adjusted OR, 0.29; 95% CI, 0.12-0.71). There was a trend toward a dose-response relationship with PPI use for >2 years protective against EAC [3 studies; PPI use >2 years vs. <2 years (as compared to no use): OR, 0.45 (0.19-1.06) vs. 1.09 (0.47-2.56)]. Considerable heterogeneity was observed in the overall analysis (I2=81%). On restricting analysis to 5 cohort studies, use of PPIs was consistently associated with a lower risk of EAC and/or BE-HGD (adjusted OR, 0.33; 95% CI, 0.19-0.58; I2=9%). H2RA use was not associated with decreased risk of EAC in patients with BE based on 2 studies (adjusted OR, 1.15; 95% CI, 0.77-1.72). Using a 67% summary risk reduction (derived from cohort studies) of EAC and/or BE-HGD with PPI use in patients with BE, and observed cumulative incidence rates of EAC and/or BE-HGD in patients with BE overall as 10.2 per 1,000 patient years, we estimate the number needed to treat with PPIs to prevent 1 case of EAC or BE-HGD in BE patients at 147. Conclusions: Based on meta-analysis of observational studies, the use of PPI, but not H2RA appears to be associated with a decreased risk of EAC and/or BE-HGD in patients with BE. PPI use should be considered in BE, and chemopreventive trials of PPIs in patients with BE are warranted.

1 Background: Endoscopic resection (ER) is used to treat early esophageal cancers (EAC). With the advent of techniques such as ESD that can remove large lesions en bloc, it is important to determine if performing piecemeal resections of intramucosal cancers is detrimental. Aim: To evaluate the response rates in patients with esophageal intramucosal adenocarcinoma (IMC) who had a single ER (SR) compared to those treated with piecemeal resection (PR). Methods: This is a cohort study of patients being followed in our BE unit. Patients in a prospectively maintained database with EAC from Jan 1998 to Apr 2012 were assessed. IMC was defined as lesions that were intramucosal without extension into the submucosa. All T1a cancers were verified by two expert gastrointestinal pathologists. We compared the tumor characteristics and recurrence rate between the two groups of SR and PR of IMC. Results: A total of 197 IMC (T1a) lesions in 138 patients were identified. 88(44.7%) lesions were performed SR and 109(55.3%) were PR. Mean follow up time were 45 months (IQR 24-70.5) in SR and 39 months (IQR 26-52, P=0.067) in PR. Complete remission were found of 92.1% (81) in SR and of 93.6% (102) in PR (P=0.788) patients. The recurrence rate was 7.9% in the SR patients, and 6.4% in the PR patients (P=0.699). Median time to progression was 12 months in the SR group (IQR 8-23.5) and 11 months in the PR (IQR 8-22) (P=0.765) group. The mean survival duration was 116.3 months in the SR and 89 months in the PR (P=0.079). No statistical difference in 5-year survival rates was found between SR (83.2%) and PR (80.9%) (P=0.082). Cox proportional hazard model analysis showed local recurrence after EMR was not related to number of specimen resected or technique. Conclusions: PR appears to be as effective a therapy as en bloc for esophageal IMC. Using PR does not appear to affect clinical outcomes.

94 Background: Octogenarian breast cancer (BrCa) patients accept less aggressive BrCa treatment due to decreased life expectancy, increased comorbidities, and high likelihood of death from other causes. Unfortunately little data exist stratifying octogenarian outcomes by disease risk. We sought to characterize treatment and recurrence patterns in patients >80yo. Methods: Retrospective review identified 432 women >80yo treated surgically for stage 0-3 BrCa between 11/99-8/11. We gathered clinicopathologic data and classified patients by disease risk as DCIS only, low risk (<2cm and ER positive and node negative), or high risk (>2cm or ER negative or node positive). We compared recurrence rates and estimated survival by Kaplan-Meier curves. Results: Among the 432 women, disease was found by mammogram in 86%, treated with BCT in 64%, and predominantly ER-positive (87%). We classified patients as having DCIS only (N=61), low risk (N=205), or high risk (N=166) disease. We identified the following deviations from standard treatment guidelines: 68% DCIS BCT and 38% high risk BCT patients did not have radiation therapy, 25% low risk BCT patients had surgery only, 51% low risk patients did not take adjuvant hormonal therapy, and 40% high risk patients had no adjuvant chemo/hormonal therapy. At 5 and 10 years the overall estimated survival was 63% and 31%, respectively. Overall, 19/432 (5%) patients developed recurrence (table 1). Patients needing mastectomy for high risk disease had significantly higher risk of recurrence than high or low risk BCT patients (p=0.02). Of the 19 recurrence patients, 7/19 (37%; 1 DCIS, 1 low risk, 5 high risk) occurred despite standard multimodality treatment, while12 (63%; 3 low risk, 9 high risk) had the initial tumor treated less aggressively due to patient choice (n=9) or medical co-morbidities (n=3). Conclusions: Significant deviations from treatment guidelines occur in women >80yo. Those with high risk disease should be counseled accordingly and encouraged to receive adjuvant treatment as two thirds of women >80yo will live at least 5 years. [Table: see text]