Each year, over 8,000 people die due to oesophageal cancer in the UK. Oesophageal cancer occurs in the food pipe between the mouth and stomach. In recent years, incidents of the disease have risen substantially in the western world and this is thought to be due to lifestyle changes.
Oesophageal cancer does not normally cause any symptoms until a late stage when it has spread, so effective treatment is essential. Current treatment methods are, however, limited due to difficulties identifying the cancer-causing genes. As with other forms of cancer, there are many different genes which can lead to patients developing oesophagus cancer, the majority of which are patient-specific. This makes finding treatments for large cohorts of people especially challenging.
To better understand the disease and develop patient-specific treatment, Professor Francesca Ciccarelli, Faculty of Life Sciences & Medicine and the Francis Crick Institute, and her research team have developed new algorithms based on machine learning which are able to characterise the properties of cancer-causing genes among all faulty genes in a patient and sketch a portrait of what the cancer gene looks like.
The team applied their algorithms to a cohort of oesophageal cancer patients from the OCCAMS consortium, a national network that aims to collect and analyse oesophageal cancers from all over the UK. They were able to identify the cancer-causing genes in each of these patients. Moreover, they could divide the whole patient cohort into six groups that share alterations in similar genes and that acquire genetic fragilities that could be used to treat the disease.
“We are now working towards extending our approach to study the early phases of oesophageal cancer and identify the genetic drivers that initiate the disease. This way we can hopefully improve early detection and suggest patient-specific intervention strategies”– Professor Francesca Ciccarelli
The study was funded by Cancer Research UK and has been published in Nature Communications.