D Palmer

Immunohistochemical hENT1 expression as a prognostic biomarker in patients with resected pancreatic ductal adenocarcinoma undergoing adjuvant gemcitabine-based chemotherapy.

Human equilibrative nucleoside transporters (hENTs) are transmembranous proteins that facilitate the uptake of nucleosides and nucleoside analogues, such as gemcitabine, into the cell. The abundance of hENT1 transporters in resected pancreatic ductal adenocarcinoma (PDAC) might make hENT1 a potential biomarker of response to adjuvant chemotherapy. The aim of this study was to see whether hENT1 expression, as determined by immunohistochemistry, was a suitable predictive marker for subsequent treatment with gemcitabine‐based adjuvant chemotherapy.

Evaluation of NQO1 as a predictive biomarker and therapeutic target in metastatic colorectal cancer

Abstract Background A knowledge of the biology of colorectal liver metastases would be invaluable to inform clinical decision making; however, deriving this information from metastatic lesions is not feasible until after resection. We aimed to use proteomic analysis to establish the degree of biological similarity across disease sites and identify and validate biomarkers in the primary tumour that predict response to neoadjuvant chemotherapy in liver metastases. Methods Tissue from primary colorectal tumour and liver metastases (n=16) were subjected to proteomic analysis with isobaric tagging for relative quantification. Data were analysed with ProteinPilot software (Ab Sciex, Framingham, MA USA) with stratification of patients into low or high response to chemotherapy. These biomarkers were investigated by immunohistochemistry on a tissue microarray of 56 patients. Their therapeutic potential was investigated by dosing of SW480 cells with irinotecan or fluorouracil with or without inhibition by small interfering RNA (siRNA) or a competitive inhibitor (dicoumarol). Findings We identified 5766 discrete proteins, of which 2·54% were differentially expressed between primary and metastatic tumours. There were 170 potential response biomarkers in the primary tumour and 27 in the metastases. Lambda-crystallin homolog and NQO1 were common to both tissue types and showed consistent dysregulation. Immunostaining of NQO1 in metastatic tissue was lower in patients responding than in those not responding to chemotherapy (p=0·041), with a significant correlation between primary and metastatic disease sites (r=0·44, p=0·001). Knockdown of NQO1 with siRNA followed by treatment with irinotecan or fluorouracil reduced the IC50 from 100·1 to 49·8 μM and from 200·1 to 25·0 μM, respectively. Treatment of cells with dicoumarol before incubation in irinotecan or fluorouracil reduced the IC50 from 100·0 μM to 50·0 μM and from 183·7 μM to 49·9 μM, respectively. Interpretation We show that proteomic sequencing of matched metastatic colorectal cancer samples is feasible with high coverage. The high degree of similarity between the primary and secondary tumours suggests that the primary tissue is predictive of the metastatic phenotype. NQO1 expression in the primary tumour predicts response to neoadjuvant chemotherapy in liver metastases, and inhibition of this protein at both genetic and functional levels improves chemosensitivity. Funding Cancer Research UK.

PTH-116 Evaluation of the d’amico clinical staging system and the albi grade for prognostication in cirrhosis and hepatocellular carcinoma

Introduction The ability to prognosticate in both cirrhosis and hepatocellular carcinoma (HCC) is vital for correct treatment decisions. For HCC, many staging systems incorporate tumour characteristics but inadequately consider underlying liver dysfunction. Therefore, models that accurately stratify the degree of liver dysfunction, may improve our understanding of its impact on survival in patients with HCC. We aimed to compare current and potential prognostication models of survival for patients with cirrhosis, and to determine their applicability in HCC patients. Method A single-centre retrospective comparative analysis of cohorts of consecutive patients diagnosed with cirrhosis alone was performed against HCC patients with background cirrhosis. The Child-Pugh score, MELD score, UKELD score, the D’Amico clinical stage of cirrhosis (D’Amico G, et al., J Hepatol, 2006) and the Albumin-Bilirubin (ALBI) grade – a composite model of liver function using serum bilirubin and albumin levels (Johnson PJ, et al., J Clin Oncol, 2015), were determined at diagnosis. The patient outcome at 1 year post diagnosis was also determined. The primary outcome was the models’ ability to predict death at 1 year. The impact of liver dysfunction on survival in patients with HCC was also evaluated. Results A total of 117 patients were included in this study (67% male, median age: 57 years). The aetiologies were ALD (52%), NASH (15%), HCV (11%), PBC (5%), HBV (2%), AIH (2%), mixed aetiology (9%) and others (5%). 75% of patients had cirrhosis alone while 25% had HCC. All the evaluated models accurately predicted death at 1 year (all with p < 0.001). The risk of death was greater amongst patients with HCC (p = 0.04; OR 2.6, 95% CI: 1.1–6.3) compared to patients with cirrhosis alone, despite favourable levels of albumin (37 vs. 35 g/L, p = 0.003), bilirubin (25 vs. 87 μmol/L, p < 0.0001) and prothrombin time (12 vs. 16 s, p < 0.0001). D’Amico stage 3–4 also predicted a significantly increased risk of death for cirrhotic patients (p < 0.0001; OR 5.7, CI: 2.4–14). Conclusion All the evaluated prognostication models performed well, though the study was limited by a small cohort of patients with HCC. Despite a better liver function, the survival of patients with HCC was inferior, suggesting that tumour biology has a greater impact overall compared to the degree of underlying liver dysfunction. However, future studies specifically looking at the impact of liver dysfunction on survival in non-surgical treatment of HCC should be considered. Disclosure of interest None Declared.

PTH-272 Modulation of NRF2 alters the responsiveness of colorectal cancer cells to irinotecan

Introduction Chemotherapy is essential to improve outcomes in colorectal cancer (CRC). Irinotecan is a pro-drug commonly used in CRC which requires conversion to the active metabolite Sn38. The cytoprotective protein NRF2 has been associated with chemo-resistance and a worse prognosis in a number of cancers.1It also regulates the activity of enzymes (including CES1 and UGT1A1) that coordinate irinotecan metabolism and efflux. 2NRF2 remains bound to the regulatory protein KEAP1 in its quiescent state. The aim of this study is to establish the value of modulating NRF2 as a novel strategy for enhancing the efficacy of irinotecan-based chemotherapy either through increased conversion of irinotecan to Sn38 or by sensitisation of CRC cells to its effect. Method NRF2 was modulated in the human CRC cell line HCT116 using small inhibitory RNA (siRNA) targeting NRF2and KEAP1 or the chemical inhibitor brusatol and inducer CDDO-Me. Cells were then exposed to irinotecan over a range of concentrations for 48 h. Cell viability was measured by MTS assay and expressed as a percentage of control for the calculation of IC50values. GraphPad Prism 6®was used for all statistical analysis and the generation of dose-response curves. Results Successful NRF2 modulation was confirmed by western blotting for Nrf2 and downstream targets NQO1 and HO-1. The IC50 value of irinotecan in untreated control cells was 90µM. Induction of NRF2 by KEAP1siRNA (IC50 220µM) or CDDO-Me (IC50 325µM) significantly reduced irinotecan cytotoxicity (p < 0.0001). In contrast, inhibition of NRF2 by NRF2siRNA (IC50 11µM) or brusatol (IC50 55µM) had the opposite effect (p < 0.0001). Conclusion Inhibition of NRF2 sensitises CRC cells to irinotecan cytotoxicity and may allow improved response to treatment, or permit lower doses with fewer side effects. We are currently validating these findings in vivo and examining the contribution of processes that mediate the disposition of irinotecan. The availability of NRF2 inhibiting and inducing compounds may allow for successful modulation in the clinical setting. Disclosure of interest J. Evans Grant/ Research Support from: Cancer Research UK, B. Winiarski: None Declared, P. Sutton: None Declared, D. Palmer: None Declared, N. Kitteringham: None Declared. References Kawasaki Y, Ishigami S, Arigami T, Uenosono Y, Yanagita S, Uchikado Y, et al. Clinicopathological significance of nuclear factor (erythroid-2)-related factor 2 (Nrf2) expression in gastric cancer. BMC Cancer 2015;15(1):5 Wang M, Zhu JY, Chen S, Qing Y, Wu D, Lin YM, et al. Effects of co-treatment with sulforaphane and autophagy modulators on uridine 5’-diphospho-glucuronosyltransferase 1A isoforms and cytochrome P450 3A4 expression in Caco-2 human colon cancer cells. Oncol Lett. 2014;8(6):2407–16

PWE-305 Nqo1 is a predictive biomarker and therapaeutic target in metastatic colorectal cancer

Introduction A knowledge of the biological phenotype of colorectal liver metastases would be invaluable to inform clinical decision making; however, deriving this information from the metastatic lesions is not feasible until after resection. We aimed to use proteomic analysis to identify biomarkers in the primary tumour which predict response to neoadjuvant chemotherapy in liver metastases. Method Fresh tissue from both primary colorectal tumour and liver metastases from 17 patients was subjected to proteomic analysis using isobaric tagging for relative quantification, with stratification of patients into those showing low or high response to chemotherapy permitting the identification of potential predictive biomarkers. These markers were subsequently validated by immunohistochemistry on a tissue microarray of 57 patients and further investigated in an in vitro model. Results We identified 5768 discrete proteins, 5 of which predicted histopathological response to fluorouracil-based chemotherapy regimens. Immunostaining of NQO1 in the metastases was lower in those responding to chemotherapy (p = 0.041), with a significant correlation between primary and metastatic disease sites (r = 0.44, p = 0.001). Knockdown of NQO1 with small interfering RNA followed by treatment with irinotecan and 5FU reduced the IC50 from 100.1µM to 49.8µM and from 200.1µM to 25.0µM respectively. Pre-treating cells with dicoumarol (a known competitive inhibitor) prior to incubation in irinotecan and 5FU reduced the IC50 from 100.0µM to 50.0µM and from 183.7µM to 49.9µM respectively. Conclusion These results show that proteomic sequencing of matched metastatic colorectal cancer samples is feasible, with high protein coverage. The high degree of similarity between the primary and secondary proteomes suggests that primary tissue is predictive of the metastatic phenotype. NQO1 expression in the primary tumour predicts response to neoadjuvant chemotherapy in the liver metastases, and inhibition of this protein at both genetic and functional levels improves chemosensitivity. Further validation in an in vivo model is planned. Disclosure of interest P. Sutton Grant/ Research Support from: Cancer Research UK, J. Evans: None Declared, R. Jones: None Declared, C. Goldring: None Declared, D. Palmer: None Declared, N. Kitteringham: None Declared, H. Malik: None Declared, D. Vimalachandran: None Declared.

OC-132 Cardiopulmonary exercise testing increases hepatectomy rate in the elderly

Introduction Hepatectomy offers the only possibility of cure to patients with colorectal liver metastases (CRLM) and although 50% of CRLM patients are >70, only 25% of those undergoing hepatectomy are >70.1 This is likely to be due to the higher perioperative mortality. Cardiopulmonary exercise testing (CPET) can identify patients at higher operative risk.2 We introduced CPET on 1 October 2009. This study assesses its effect on patient selection and outcome. Methods After 1 October 2009 all patients undergoing resection of CRLM meeting one of the following criteria underwent CPET. Criteria Planned extended right/or extended left resection Over 65 Significant comorbidities Data were collected prospectively. Group (A) 1 October 2009–21 February 2011 was compared to a published historical series 1 August 1990 to 1 April 2007 (Group B). We also compared our results to a recently published series from LiverMetSuvey (LMS).1 Results Group A 155 patients, Group B 654 patients. In Group A 64 (41%) patients were 70+, Group B 181 (28%) patients (p<0.0001). Within the over 70s groups here was no difference in male:female ratio or postoperative complications, but there was an increased use of neoadjuvant chemotherapy in Group A 26 patients (40%) vs 34 patients (19%) Group B (p=0.001). In hospital mortality was 1.6% (1 patient) Group A vs 4.9% (9 patients) group B (p=0.46). Comparing Group A to LMS data1 from 1/1986 to 6/2007 (7764 resections). Group A 41% aged 70+ vs 21% LMS (p<0.01). No significant difference in neoadjuvant chemotherapy use (Group A 41% vs LMS 34%) (p=0.22), or 60 day mortality (Group A—1 patient (1.6%) vs 62 (3.8%) LMS) (p=0.51). Conclusion CPET enabled accurate preoperative fitness assessment and led to a significant increase in the number of patients over the age of 70 proceeding to surgery. There is a trend towards lower mortality possibly due to better preoperative assessment. We believe that the number of hepatectomies undertaken for CRLM in patients aged over 70 would rise if CPET were introduced as a standard preoperative assessment tool. Competing interests None declared. References 1. Adam R, Frilling A, Elias D, et al; LiverMetSurvey Centres. Liver resection of colorectal metastasis in elderly patients. Br J Surg 2010;97:366–76. 2. Smith TB, Stonell C, Purkayastha S, et al. Cardiopulmonary exercise testing as a risk assessment method in non cardio-pulmonary surgery: a systematic review. Anaesthesia 2009;64:883–93.

OC-026 Predicting complications in liver surgery

Introduction Cardiopulmonary Exercise Testing (CPET) is a non-invasive test that has been used to identify patients at higher perioperative risk. Studies have found that different CPET variables seem to be more predictive in different patient groups. There is little literature on the use of CPET within the HPB field, and no series concentrating on patients undergoing Liver resection. Our aim was to identify the most sensitive CPET variable for risk prediction in this patient group. Methods From 1 October 2009 CPET was carried out in all patients due to undergo Liver resection meeting one or more of the following criteria (1) planned extended right/or extended left resection (2) over 65 (3) significant comorbidities. Data were prospectively entered into a database. This was correlated with preoperative CPET data and analysed using version 19 of SPSS. Results Between 1 October 2009 and 1 July 2011 188 patients underwent Liver resection, 121 (64%) underwent CPET (Group A), and 67(36%) did not (Group B). Group A were older (mean age 70 vs 54) and had higher complication rates (56% vs 36%) and had longer length of stay (median 7 vs 5) (all p<0.001). The three deaths occurred within group A. Multivariate analysis of Group A including age, BMI, extent of surgery (segments), VO2 at anaerobic threshold (AT), VO2 peak, O2 pulse, and heart rate found that O2 pulse at AT, and HR at AT correlated best with a risk of increased complications. OR O2 pulse 0.86(CI 0.72 to 1.01, p 0.07), HR at AT 1.04 (CI 1.001 to 1.06, p<0.01). Conclusion This is the largest study of CPET in the HPB field, and the only study involving only Liver resection. CPET can be used to identify those at higher perioperative risk, with O2 pulse and HR at the Anaerobic Threshold the most sensitive indicators. The selective use of CPET was justifiable as all patients who died in the postoperative period were identified. Complications still occurred within the non-CPET cohort suggesting expansion of CPET selection criteria may be needed. Competing interests None declared.