Texture analysis of pancreatic tumour lesions and mimickers

  • Research type

    Research Study

  • Full title

    Texture analysis of pancreatic tumour lesions and mimickers

  • IRAS ID

    247074

  • Contact name

    SIONG-SENG LIAU

  • Contact email

    ssl30@medschl.cam.ac.uk

  • Sponsor organisation

    Cambridge University Hospital NHS Foundation Trsut

  • Duration of Study in the UK

    2 years, 10 months, 27 days

  • Research summary

    Pancreatic cancer (PCa) is the fourth leading cause of cancer-related deaths in the world. Despite advances in medicine in the last decades with new treatments and better diagnostic tools available, the mortality of pancreatic cancer remains overall unchanged. Currently diagnosis is made based on cross sectional imaging (CT and MRI) with histological confirmation made by invasive procedures such as an endoscopic ultrasound fine needle aspiration (EUSFNA): a telescope test, using ultrasound guidance, to obtain a sample of cells using a needle. However, correct diagnosis, staging and aggressiveness of the disease is still insufficiently predicted by these techniques. For example, differentiation of pancreatic cancer from benign diseases such as chronic pancreatitis can be challenging, leading in some instances to incorrect invasive treatments such as surgery.

    Moreover, other cancers can also metastasise (migrate) to the pancreas, ie renal cancer. This only happens in some patients, but it can have significant implications in treatment. Currently, no technique is able to predict the risk of the development of metastasis in the pancreas. This would provide critical prognostic information.

    There is therefore a critical unmet need for better ways to non-invasively assess PCa at the different stages of diagnosis. The development of new ways to perform imaging analysis of modalities such as CT and MRI allow for a large quantity of information, which is invisible to the human eye, to be extracted from these images. Importantly, this computational image analysis requires little investment, and may provide a cost-effective method to improve cancer treatment and detection.

    We propose a retrospective study for imaging texture analysis using imaging modalities of patients at different stages of the disease; from initial diagnosis and differentiation from other benign diseases to metastatic disease to and from the pancreas. Retrospective and prospective validation with histological, immune, genomic and available blood biomarkers will be done. The main aim is to improve the diagnosis of pancreatic cancer, staging, survival and prognosis.

  • REC name

    East of England - Cambridge South Research Ethics Committee

  • REC reference

    18/EE/0227

  • Date of REC Opinion

    19 Sep 2018

  • REC opinion

    Further Information Favourable Opinion