QIMR Berghofer Medical Research Institute

The Australian Cancer Study (ACS)

Funded by the National Health and Medical Research Council of Australia (2002-2006) and the Queensland Cancer Fund (2001-2002).

D Whiteman, A Green, P Webb, P Parsons, N Hayward

There are two types of cancer of the oesophagus: squamous cell cancer and adenocarcinoma. To explore the environmental and genetic causes of cancers of the oesophagus, we conducted a population-based case-control study recruiting newly diagnosed patients throughout Australia between 2002 and 2005. We enrolled 367 patients with adenocarcinoma of the oesophagus, 426 patients with adenocarcinoma of the gastroesophageal junction, 309 patients with squamous cell cancer and compared them to 1580 people without cancer for factors including their height and weight, smoking, alcohol, diet, medications and past medical history. We have shown that obesity is a major risk factor for adenocarcinoma of the oesophagus, and that it interacts with acid reflux to greatly increase a person's risk of cancer. Other published findings are that frequent use of aspirin and non-steroidal anti-inflammatory drugs (NSAIDs) greatly reduce the risk of all types of oesophageal cancer, and that the different types of oesophageal cancer are associated with smoking in substantially different ways (a finding of importance for prevention). We have identified a number of DNA repair genes that appear to increase the risk of adenocarcinoma of the oesophagus, and we are continuing to explore associations with a number of other factors. Data from this study is contributing to the international BEACON collaboration.

The Study of Digestive Health (SDH)

Funded by the US National Cancer Institute 2002-2007

D Whiteman, P Webb, A Green, N Hayward, P Parsons, S Pavey in collaboration with P Drew (Flinders University), D Gotley (University of Queensland), G Jamieson, D Purdie, B Smithers (University of Queensland), D Watson (Flinders University)

Nearly all adenocarcinomas of the oesophagus arise from a pre-cancerous condition called Barrett's oesophagus. This condition is becoming increasingly prevalent, yet relatively little is known about its causes or its natural history. We have been exploring the environmental and genetic causes of Barrett's oesophagus through a population-based case-control study which recruited newly diagnosed patients in Queensland and South Australian between 2003 and 2006. We enrolled 479 patients with Barrett's oesophagus and compared them to 646 people without Barrett's oesophagus for factors including their height and weight, smoking, alcohol, diet, medications and past medical history. Using this rich resource, we have shown that leptin (a hormone secreted by fat cells) is significantly higher in men with Barrett's oesophagus than controls, even after accounting for related factors such as obesity. Further analyses will continue.

The ACS Clinical Follow-Up Study

Funded by the National Health and Medical Research Council 2006-2008

D Whiteman, A Green, N Hayward in collaboration with D Gotley & BM Smithers (University of Queensland), D Watson (Flinders University), G Falk (Concord Hospital), G Smith (Royal North Shore Hospital), G Kiroff (Barwon Surgical, Geelong), A Clouston (Histopath, Sydney)

This study collected treatment and health outcomes data for the population-based cohort of patients with oesophageal cancer who took part in ACS. The goal was to identify prognostic and predictive markers to aid patients and clinicians when making treatment decisions, as now exist for breast cancer. Such markers may also serve as novel targets for therapy. The proposed study builds upon the platform of the Australian Cancer Study [ACS], one of the world's largest studies of oesophageal cancer.

The Barrett's oesophagus metabolic study

Funded by the Queensland Cancer Fund 2007-2008

D Whiteman, B Kendall, G Macdonald (Princess Alexandra Hospital)

We have shown that obesity is likely to be a major risk factor for Barrett's oesophagus and oesophageal adenocarcinoma, but how it confers increased risk is unknown. With the high prevalence of obesity in the Australian population, and with predictions of an impending obesity epidemic, we need to identify how increased levels of body fat induce these conditions so that this knowledge can be used for prevention. By comparing patients with proven Barrett's oesophagus to healthy people selected from the general population, we aim to investigate the link between obesity and Barrett's oesophagus. In particular, we are investigating the role of certain hormones (including insulin) that are known to be affected by obesity. Once we have identified those factors, we can then investigate ways of counteracting them with the goal of preventing the development of Barrett's oesophagus and subsequent cancer.

Coeliac disease as a risk factor for squamous cell carcinoma of the oesophagus

Funded by Gallipolli Research Foundation

B Kendall (QIMR/Princess Alexandra Hospital), D Whiteman, G Macdonald (UQ/ Princess Alexandra Hospital)

Coeliac disease (CD) is an immune mediated condition in which inflammation and atrophy of the small intestine occur when genetically predisposed individuals are exposed to dietary gluten. Until recent times, diagnosis of the condition could only be made with an endoscopic small bowel biopsy and the majority of patients diagnosed with the condition had clinical manifestations of the villous atrophy. Recently, serological tests have been developed for the diagnosis of CD. Use of these tests in population studies has shown a much higher prevalence of the condition than previously appreciated. A study from Denmark showed an increase prevalence of CD from 1 in 10,000 pre serology to 1 in 300 with the assay. Multiple serological tests have been developed for the diagnosis of CD with IgA anti-TTG being widely used because of its high sensitivity and specificity. Approximately 2% of those with CD and 0.5% of the general population are IgA deficient. In those cases of IgA deficiency, IgG anti-gliadin antibodies can be used to screen for CD

Cohort studies have shown that people with coeliac disease have increased risk of cancer. The majority of this risk was related to lymphoma and lymphoproliferative disorders, but a number of studies have suggested an increased risk of gastrointestinal cancers including oesophageal SCC in those with CD. To date, no case-control studies have been published examining the relationship between coeliac disease and oesophageal SCC. As custodians of the largest collection of annotated oesophageal SCC serum specimens in the industrialised world, we are well-placed to test this hypothesis definitively. If CD is identified as a significant risk factor for oesophageal SCC, this has potential future implications in terms of population screening for CD, surveillance of those with CD and understanding of the pathogenesis of this terrible disease.

BEACON: The Barrett's oesophagus and adenocarcinoma consortium

Tom Vaughan MD - Fred Hutchinson Cancer Research Center, Seattle

Brian Reid MD PhD - Fred Hutchinson Cancer Research Center, Seattle

Olof Nyren MD PhD - Karolinska Institute, Stockholm

Jesper Lagergren MD PhD - Karolinska Institute, Stockholm

Weimin Ye MD PhD - Karolinska Institute, Stockholm

Marilie Gammon PhD - University of North Carolina

Nick Shaheen MD MPH - University of North Carolina

Anna Wu PhD - University of Southern California

Lesley Bernstein PhD - University of Southern California

Doug Corley MD PhD - Kaiser Permanente, San Francisco

Liam Murray MBBS PhD - Queens University, Belfast

Chris Wild PhD - Leeds University

David Forman PhD - Leeds University

Alan Casson FRCS PhD - University of Saskatchewan, Canada

David Whiteman MBBS PhD - Queensland Institute of Medical Research

BEACON is an international collaborative group established by the US National Cancer Institute to promote research into esophageal cancer. The ACS and SDH studies were invited as founding members of the group.

BEAGESS: Barrett’s and esophageal adenocarcinoma genetic susceptibility study

Funded by the US National Cancer Institute

We are undertaking a large-scale genome-wide association study (GWAS) which takes advantage of the extensive data collected by investigators in the Barrett’s and Esophageal Adenocarcinoma Consortium (BEACON), representing virtually all of the high-quality population-based and other epidemiologic studies of BE and EA in the world. The overall goal of our research is to evaluate the influence of genetic susceptibility on risk of EA and its main precancerous condition, BE. The primary aims are to identify the genetic polymorphisms most strongly associated with risk of these conditions. In secondary aims we will evaluate the extent to which the most significant associations vary according to key environmental and host risk factors for these conditions, including gender, obesity and cigarette smoking. These results will guide additional exploratory analyses examining ways in which individual SNP results might be aggregated to shed light on the importance of specific pathways relevant to the etiology of BE and EA, and to identify possible epistatic effects.

Risk prediction for oesophageal cancer

Aaron Thrift, David Whiteman

While the search for better treatments continues, another strategy to reduce the burden of this cancer is to find people at high-risk, so that they can be monitored more closely by their doctors. This might include preventing the onset of cancer (by removing pre-cancerous tissues in patients with Barrett’s oesophagus), or by making the diagnosis of cancer earlier, in the hope that it will be smaller and more treatable. Doctors have developed tools for predicting risks of heart disease, diabetes and some cancers. The PROBE-NET group based at the Queensland Institute of Medical Research (QIMR) is using data from the ACS to develop tools that predict a person’s future risk of oesophageal adenocarcinoma. Early results have been promising.

Princess Alexandra Hospital – Brisbane

Chief Investigator: Associate Professor Andrew BarbourAndrew Barbour

A comprehensive genomic analysis of oesophageal adenocarcinoma

Investigators: Andrew Barbour, Nicola Waddell, Lutz Krause (QIMR), Derek Nancarrow (QIMR), Damian Hussey (Flinders), David Watson (Flinders), Reginald Lord (UNSW-St Vincents).

Funding: NHMRC project grant

Complete surgical removal of tumour tissue is the mainstay of curative treatment for OAC but cure rates remain poor. Neoadjuvant chemotherapy and chemoradiotherapy regimens have been explored with modest improvements in outcome. Selecting patients most likely to benefit from existing standards of therapy and the search for new therapeutic targets are priorities that could be aided by molecular genetic approaches. Progress in the prognostication and targeted treatment of OAC patients has been hampered by our limited understanding of the genetics of this disease, which is sorely lagging behind other solid tumours.

Briefly, this project aims to perform a comprehensive genomic analysis of OAC including next generation sequencing (NGS), whole genome DNA copy number, whole genome expression and whole genome methylation. This study will improve our understanding of OAC and provide essential information regarding targeted treatment options, where currently very few options exist.

To achieve these aims, we will firstly obtain a detailed catalogue of the genetic aberrations involved in OAC by undertaking NGS in tumours and matched normal tissue samples. We will then integrate NGS data with our existing data of genome-wide copy number, mRNA expression and methylation status so that we can identify genes/pathways involved in OAC that might serve as future drug targets. In addition, this approach might help identify prognostic biomarkers. Finally, we will validate NGS data and biomarkers identified the steps above using 276 formalin-fixed, paraffin embedded OAC tissues from the Australian Cancer Study and other PROBE-NET collaborators.

A randomised Phase II trial of pre-operative cisplatin, 5 fluorouracil and docetaxel or cisplatin, 5 fluorouracil, docetaxel plus radiotherapy based on poor early response to standard chemotherapy for resectable adenocarcinoma of the oesophagus and/or gastro-oesophageal junction

Investigators: Andrew Barbour, Mark Smithers, Bryan Burmeister, Euan Walpole, John Zalcberg, Nigel Spry

Funding: NHMRC Project grant

Surgery forms the mainstay of curative treatment for OAC, however survival remains poor. In an attempt to improve survival for patients with surgically removable OAC pre-operative treatment with either chemotherapy (CTX) or chemoradiotherapy (CRT) has broad support. There are data to suggest that patients whose tumours respond to pre-operative CTX or CRT exhibit better survival than non-responders. It has recently been shown that early responses as detected by PET scan might predict histological response. We are therefore undertaking a randomised clinical trial among those patients whose tumours fail to demonstrate a PET-response after the first cycle of chemotherapy to assess whether changing the therapy induces a histological response.

Our primary objective is thus to determine the histological response rate for the pre-operative treatment groups. Our secondary objectives are to:

I. evaluate the disease-free survival in the treatment arms;

II. evaluate the overall survival in the treatment arms;

III. evaluate the early FDG-PET response to the pre-operative therapy regimen;

IV. determine treatment related toxicity and the effects of treatment on health-related quality of life (HRQoL);

V. evaluate the toxicity of Docetaxel in combination therapy for oesophageal cancer;

VI. bank tumour tissue and blood before and after treatment for future molecular analyses investigating biomarkers of response.

Significance: This study represents a paradigm shift in the way clinicians administer pre-operative therapy for OAC. It represents the first study to focus on metabolic non-responders to pre-operative therapy in a manner that changes their treatment with the aim of improving response rates and survival. This study will be the first to use the combination of DCF with or without radiotherapy for this indication thus generating novel safety and efficacy data. It will determine whether changing therapy can salvage a response to pre-operative therapy and provide valuable data regarding the potential of tailored therapy. The routine pre-treatment tumour and blood banking will give our group a unique opportunity to search for OAC biomarkers or response.