Gastrointestinal Cancer - Wiley Online Library

bs_bs_banner

doi:10.1111/jgh.13518

Gastrointestinal Cancer An unusual polyp: rectal melanoma K CHAKRADEO,*,† G LAMPE,‡ S SAFA* *Department of Gastroenterology, Logan Hospital, Meadowbrook, †School of Medicine, University of Queensland, St Lucia, ‡Department of Anatomical Pathology, Princess Alexandra Hospital, Woolloongabba, Queensland, Australia A 67-year-old Caucasian man was referred for gastroenterology review following a 6-week history of minimal bright red rectal bleeding thought to be associated with a known history of haemorrhoids and intermittent constipation. There was no history of weight loss or abdominal discomfort. He was an ex-smoker with a 40-pack-year smoking history and was on long-term methotrexate therapy for psoriatic arthritis. His brother had been diagnosed with colorectal cancer at age 75 years. Perianal examination discovered a soft palpable rectal mass. Blood tests were grossly unremarkable, including normal iron studies. Urgent colonoscopy was performed revealing a 35-mm protruding, pedunculated and darkly pigmented polyp in the rectum. This was resected and retrieved for histopathological review. Histology showed a spindle cell tumour with variable amounts of brown pigmentation. Immunostaining found tumour cells to be positive for S-100 and CD117. Immunostaining was negative for AE1/AE3, desmin and BRAF. Histological appearances were not suggestive of perineural or lymphovascular invasion. Findings were consistent with a diagnosis of malignant melanoma of the rectum. A CT scan of the abdomen and pelvis confirmed no radiographic evidence of loco-regional or distal metastatic disease. Our patient was referred for immediate colorectal surgical review for further ongoing management. Rectal melanoma is a rare diagnosis seen in patients presenting with minimal bright red rectal bleeding. Anorectal mucosal melanoma accounts for approximately 0.05 percent of all colorectal malignancies and 1 percent 1 of all anal canal cancers. It carries a very poor prognosis with 5-year sur2 vival of 12–18%. Early detection of this malignancy plays a pivotal role in optimising patient outcomes. References 1. Cagir B, Whiteford MH, Topham A, et al. Changing epidemiology of anorectal melanoma. Dis Colon Rectum 1999; 42:1203. 2. Meguerditchian A, Meterissian S, Dunn K. Anorectal melanoma: diagnosis and treatment. Dis Colon Rectum 2011; 54(5):638-644. doi:10.1007/dcr.0b013e31820c9b1b.

Tumour markers using SurePath liquid-based cytology preparations from endoscopic ultrasound guided fine needle aspiration of solid pancreatic lesions J COLLINS,* H DIXSON,* P BAIRD,† C MEREDITH* *Gastroenterology and Hepatology, Bankstown-Lidcombe Hospital, Bankstown, †Cytology Department, Laverty Pathology, Ryde, NSW, Australia Introduction: SurePath® (SP) is a liquid-based collection system that prepares a cellular pellet for cytology. The pellet is large enough to prepare multiple slides for immunohistochemistry (IHC) giving a huge advantage over conventional smear preparations (CSP) and other liquid-based preparations. Tumour markers have an important and evolving role in the diagnosis, treatment and prognosis of many solid organ tumours. Ki-67 is a specific marker of cell proliferation, CDX2 is a transcription factor which

is expressed by GI adenocarcinoma and p53 is a transcription factor in a tumour suppression gene. Mutations in p53 can lead to malignancy. Aim: The aim is to determine whether tumour markers can be identified using IHC on SP preparations from fine-needle aspiration (FNA) of solid pancreatic tumours and to examine the pattern and potential clinical application. Methods: CSP and SP collections were obtained by FNA on solid pancreatic lesions referred for endoscopic ultrasound. A cytopathologist was not present during the collections. CSP slides were prepared from 3 FNA, and the final FNA was placed in an SP vial. SP-prepared cytology specimens were reviewed independently by two experienced cytopathologists. IHC was performed for Ki-67, CDX2 and p53 and the results tabled. Results: Sixty-two patients were included in the study. The cohort had a mean age of 68 years; 50% female. Adenocarcinoma (AC) was identified in 32 CSP and 40 SP specimens; 8 CSP reported as atypical were positive by SP. IHC staining for the 3 tumour markers was obtained on 18 AC positive specimens. Ki-67 was present in 18, 15 showed positive staining for CDX2 and 13 were positive for p53.

Pos Neg

Carcinoma Atypical NET GIST Negative Lymphoma HCC Total

Ki-67

CDX2

p53

18 0

15 3

13 5

CSP

SP

32 11 6 1 10 1 1 62

40 3 8 1 8 1 1 62

Conclusion: This pilot study confirmed that tumour markers can be determined using IHC on SP preparations, but the potential clinical benefit cannot be determined from this small sample. SP was more sensitive in detecting AC and NET compared with CSP without rapid on-site evaluation by a cytopathologist. Ki-67 was present in all AC positive samples tested and can be helpful in separating reactive lesions from malignancy. Not all specimens had CDX2 and/or p53. The clinical role of these markers is yet to be determined with a larger study, but they can be helpful in confirming an early diagnosis and may be used to guide future treatment.

Getting the cancer screening message to the low socio-economic community “Gastrointestinal Cancer Conference – Colorectal Cancer Screening Conference” DM FLORENCE Kyabram District Health Service Background: Research conducted in 2012 identified Central West Victoria as an underscreened demographic (Health and Cancer Screening in Kyabram: a community perspective). A two-year project in this region (the ‘Community Action Research Project’) is trialling strategies to increase participation in mammograms, pap tests and the National Bowel Cancer Screening Program, predominately targeting members of the low socio-economic community. The National Bowel Cancer Screening Program (NBCSP) is in its infancy in comparison to the papscreen and breastscreen programs; therefore, a strong focus is required to target screening rates for colorectal cancer.

Journal of Gastroenterology and Hepatology 2016; 31 (Suppl. 2): 53–64 © 2016 The Authors. Journal of Gastroenterology and Hepatology © 2016 Journal of Gastroenterology and Hepatology Foundation and Wiley Publishing Asia Pty Ltd

53

Gastrointestinal Cancer

Aim: The three objectives of engagement included the following: • raise awareness of cancer screening, • support training of health and community workers and • improve accessibility of cancer screening services Method: Qualitative and quantitative data collection was obtained using survey data collection with a cross section of the community and health professionals to obtain feedback on current themes, myths, screening habits, and health promotion messages and resources for bowel screening. Results: Quantitative data were obtained from approximately 100 surveys distributed. A total of 79 responses were received from the staff of Kyabram District Health Service in October 2015. Data from this survey showed that 45.6% did not know how often someone should have a bowel screen and 22.8% said that they find difficulty accessing bowel screening services. In regard to completing the NBCSP kit when received in the mail, only 37.5% stated that they have, leaving 42.5% either not completing it at all or only completing it sometimes. A further 59.5% of respondents also said that they did not receive enough media/printed information regarding screening. Qualitative data received in anecdotal accounts from direct consumer engagement identified gaps in knowledge and misconceptions about how to screen. This leads to the development of several resources: An example of one resource produced is the creation and trial of a bowel kit demonstration and PowerPoint presentation. This visual and sensory demonstration has been trialled at Men's Health Nights and other community group meetings. Ongoing evaluation strongly demonstrates that this method of community interaction is not only successful but also sustainable via a ‘train the trainer’ model. As a rural setting, some myths to be dispelled centred around the disposal of the ‘catching’ paper in septic tank systems and the ease of completing the test (no need to ‘poo in a jar’). As many locals have levels of low literacy, the need for a pictograph presentation and hands on demonstration of the kits proved to allay any misconceptions or concerns. Conclusion: This project highlights the need to close the health gap for our at-risk consumers, low socio-economic demographic, by determining ways to convey the importance of the screening message that was both meaningful and purposeful to this cohort. The project was able to, from consultation, trial and feedback, develop resources which target the at-risk consumer groups through the development of health promotion resource templates, health education templates and community engagement strategies which can be scaled to other health services and community activities.

Funding Funding partner was Cancer Council Victoria, and project funding was received from the Department of Health & Human Services.

Service and quality improvement – are we ‘hit and miss’ with gastric and oesophageal cancer detection? Y DING,*,† L LIN,* A WEST,*,† H MOATTAR,* A ALGHAMRY,*,† J CROESE,*,†,‡ M APPLEYARD,*,† R HODGSON,*,† A VANDELEUR,* J THOMAS,*,† R GUPTA,*,† A KAUR,* Z SAMIDURAI,* ENDOSCOPY NURSES COLLABORATIVE (ENC),* T RAHMAN*,†,‡ *Department of Gastroenterology & Hepatology, The Prince Charles Hospital, Brisbane, †University of Queensland, St. Lucia, Brisbane ‡James Cook University, Cairns, QLD Introduction: Upper gastrointestinal (GI) malignancies are notorious in their lethality due to their asymptomatic onset, late presentation and poor

54

survival. Literature defines a cancer detected within three years after an endoscopy as a ‘potential miss’, and a cancer detected within one year after an endoscopy as a ‘definite miss’. This study aims to audit the missed upper GI cancers in our local service. Methods: Forty-three patients were diagnosed with upper GI cancers on endoscopy between January 2013 and November 2015. Demographic data (age, height, weight and sex), indication and histopathology were recorded. The endoscopy database search was carried out to find any previous endoscopies performed within three years prior to the cancer diagnosis. Results: Of the 43 patients (M:F = 28:15, mean age = 70:78), five (11.6%) were ‘definite misses’, with one having had an endoscopy within 24 months also. Of the five, two gastric adenocarcinomas and three oesophageal cancers were missed. The indications included swallowing difficulty (14/43, 32.6%), iron deficiency anaemia (7/43, 16.3%), haematemesis/melaena (6/43, 14.0%), abnormal imaging studies (5/43, 11.6%) and others. The types of observed upper GI cancers constituted oesophageal cancers (25/43, 58.1%, of which 20 were adenocarcinomas), gastric adenocarcinomas (17/43, 39.5%) and one lymphoma. Conclusion: The incidence of missed upper GI cancers at our local centre is similar to the reported value of 5–13%. Nonetheless, it is paramount to remain alert when investigating recurrent upper GI complaints as red flag symptoms only account for a proportion of presentations. Early detection with upper GI endoscopy is the key.

Disclosure of interest None declared.

A systems approach to improving cancer screening outcomes through quality improvement strategies N FIRMAN, J RADREKUSA Murray PHN, Bendigo, Victoria, Australia Background and aims: Victorian bowel cancer data from 2007–2013 shows the Loddon Mallee Region (LMR) had the highest age standardised bowel cancer incidence rate of 43.9 per 100,000 with 1955 new diagnoses. Mortality data from the same period show that the LMR had the second highest age standardised bowel cancer mortality rate of 14 per 100,000 with 728 deaths. Higher rates of positive screening, coupled with lower rates of follow-up colonoscopies, have been recorded for Aboriginal and Torres Strait Islander people (ATSI) and people who lived in regional and remote regions and areas of low socio-economic status. Although bowel cancer may be present for many years before symptoms manifest, up to 90% of cases can be successfully treated if detected in the early stages. Currently however, less than 40% of bowel cancer is detected early, and only 36% of Australians participated in the free National Bowel Cancer Screening Program (NBCSP). NBCSP participants currently make up only approximately 6% of total demand for colonoscopy services (public and private) in Victoria. With the NBCSP progressing towards full implementation by 2020, it is estimated this will increase to approximately 20%. Victoria conducts more than double the number of colonoscopies per incident case of colorectal cancer compared with other regions, which is increasing waiting times and placing great demand on local services. Increasing use of FOBT as a primary test should reduce the need for colonoscopy and contribute to improved access for patients with the greatest need. Two one-year pilot projects (2014–2016) set in the Loddon and Murray regions of Victoria aimed to implement multi-component, and replicable, whole of systems approaches to bowel cancer prevention and screening in community settings and general practice. Methodological approach: The target populations and settings included under-screened population groups (men, ATSI, low SES, CAPD and



people with a disability), all primary health settings (including general practice and community health), local colonoscopy services (particularly public services) and health workforce. To ensure reach, sustainability and replicability, the primary strategies focused on embedding approaches at the systems level through the development of policies, guidelines, referral processes and training; and the coordination of clusters, networks and forums. Primary care interventions, which focused on the pivotal role of general practice in cancer screening, included promoting the use of data systems and clinical software, implementing quality improvement strategies in general practice (including business analysis methods and processing mapping), developing and implementing clinical referral and management pathways, training and educating general practice professionals, addressing quality issues in referrals to colonoscopy services and developing and implementing local referral protocols/pathways and resources in partnership with colonoscopy services. The project was supported by mixed methods in a program logic evaluation framework. The framework, encompassing a broad range of both quantitative and qualitative data, will facilitate both a formative and summative evaluation – looking to define not only the impacts and the results of the initiative but also to identify the key lessons learned about how to design, manage and implement a complex inter-related program of measures delivered across multiple health settings and sectors. Outcomes measured include increased awareness and knowledge of health workforce, increased FOBT orders, improvements to FOBT coding and increased utilisation of reporting through clinical software, data cleansing, improved local referral pathways and clinical workflows, uptake and embedment of quality improvement strategies, and timely access to colonoscopy services. Analysis of the data by October 2016 will provide a more comprehensive picture of short-to-medium-term local screening participation trends and identifying pathways for improving quality of care. Conclusion: This project provides a mixed-methods framework in which evidence-based bowel cancer screening and quality improvement activities can be embedded and normalised within the general practice environment to assist in strengthening service system integration across local health and community services and improve population health outcomes.

Can scoring systems predict outcomes in hepatocellular carcinoma treated with TACE? C HAIFER,* R PRENTICE,* A MOSS,*,† I KRONBORG* *Department of Gastroenterology, Western Health, Victoria, † Department of Medicine, Melbourne Medical School, Western Precinct, The University of Melbourne Background: Transarterial chemoembolisation (TACE) is the standard of care for intermediate stage hepatocellular carcinoma (HCC) and is the most commonly used treatment for HCC worldwide. A proportion of patients have a dismal prognosis despite intervention, and these patients may not benefit from further intervention. Recently, two different scoring systems have been created to distinguish those patients who would benefit 1 2 from further TACE after an initial treatment. The ABCR and the ART scores have been validated in European cohorts and use different prognostic variables (Barcelona Clinic Liver Cancer score, Alpha-Fetoprotein, Child-Pugh score, AST and response to treatment) to identify those who will not benefit from further treatment. Aim: We hypothesise that by applying these scoring systems to real-world patients treated with TACE, we will identify a group of patients with poor prognosis in which further treatment may not be beneficial. Methods: We retrospectively analysed patients with HCC treated with TACE between 1/2/2012 and 1/3/2016 at Western Health. The ABCR and ART scores were calculated between four and eight weeks following the initial TACE treatment. Overall survival (OS) and Kaplan Meier curves were calculated.

Results: Thirty-six patients were analysed, 27 (75%) were male with a median age of 70 years (range 50–84 years). The causes of underlying liver disease were viral hepatitis in 17 (47%), alcoholic liver disease in 13 (36%), non-alcoholic fatty liver disease in 4 (11%), haemachromatosis and autoimmune hepatitis in the remaining two patients. The median Child-Pugh score at baseline was 6, and the Barcelona Clinic Liver Cancer score was A in 16 patients (44%), B in 13 (36%) and C in 7 (19%). Ten patients had prior treatment with either resection, microwave ablation or TACE. Median OS from initial TACE was 14.6 months. When applying the ABCR score, using a threshold of 2, there was a statistically significantly difference in OS with a median OS of 12.1 vs 20.4 months (p = 0.003) (see figure). When applying the ART score with a threshold of 1.5, there was a statistically significant difference in OS with a median OS of 16.2 vs 28.8 months (p = 0.015). Conclusion: Calculating ABCR and ART score after initial TACE appears to identify a group of patients with a poor prognosis that may not benefit from further invasive treatments. References 1. X. Adhoute, G Penaranda et al. Retreatment with TACE: the ABCR score, an aid to the decision-making process. J.Hepatol 2015 Apr; 62 (4):855–862. 2. W. Sieghart, F Hucke et al. The ART of decision-making: retreatment with transarterial chemoembolization in patients with hepatocellular carcinoma. Hepatology. 2013 Jun;57(6):2261–2273.

The changing face of pancreatic neuroendocrine tumours: data from a large single centre cohort S HEW* AND D CROAGH† Departments of *Gastroenterology and, †Upper GI/HPB Surgery, Monash Health, Melbourne, Australia Introduction: Pancreatic neuroendocrine tumours (PNETs) are a rare form of pancreatic neoplasm. Whilst sharing histological similarities with neuroendocrine tumours at other sites, they have a distinct biology and clinical manifestation. The management of PNETs remains controversial, especially in whether to surveil small lesions and of the role of resection in metastatic non-functioning PNETs. We present data from the largest single centre cohort of PNETs in Australia. Methods: Data from consecutive patients referred to Monash Health, a quaternary pancreaticobiliary referral centre, from 2011 to 2016 with PNETs, were prospectively collected and analysed. Patient demographics, tumour characteristics and staging were collected. Where applicable, procedural


55


details and complications, length of stay, and overall morbidity and mortality data were collected. Results: Forty-three patients with a median age of 64.5 years (range 40.8– 84.1 years) were analysed. All the PNETs analysed were non-functional in nature, and in 27 (62.7%) patients, the suspicion of PNET was raised as an incidental finding on cross-sectional imaging. Thirty-two (74.4%) patients were evaluated with endoscopic ultrasound (EUS). The majority of tumours were localised to the pancreas (83.7%). Eleven (25.6%) patients with tumours of a median size of 13 mm and Ki67 proliferation index of 2% were surveilled. None of these patients developed progressive disease at a median follow-up of 16 months. Twenty-one (48.8%) patients had resection of their PNETs, with procedure-related mortality of 0%. Six patients (14%) with metastatic disease proceeded to chemotherapy. Conclusion: The modern face of PNETs in Australia is changing: PNETs are being identified early on cross-sectional imaging and EUS, and small, low-grade lesions can be safely surveilled without tumour progression. Surgical resection is a successful outcome in a large number of patients.

One in 11 colorectal cancers found in adults younger than the age of screening, and patients are more likely to present with left-sided and more advanced disease B JIDEH,* T YANG,† I TURNER,* S AL-SOHAILY* *Department of Gastroenterology and Western Sydney University, Campbelltown Hospital, and †Department of Anatomical Pathology, Western Sydney University and University of New South Wales, Liverpool Hospital, Sydney, NSW, Australia Introduction: Colorectal cancer (CRC) is the second most commonly diagnosed cancer in Australia in both men and women, with an estimated 17 000 new cases in 2015. Large epidemiological studies in the United States have demonstrated a rising incidence of CRC in adults younger than the screening age of 50 years, and these patients are more likely to present with leftsided and more advanced disease. It has been estimated that despite an aging population, by the year 2030, more than 1 in 10 colon cancers and nearly 1 in 4 rectal cancers will be diagnosed among individuals aged