Pancreatic cancer is a highly fatal disease with less than a 10% of 5-year #survival rate. The high mortality rate is mainly caused by the lack of highly sensitive and specific tools to detect the disease in an #early stage, and therefore most of the patients are diagnosed in advanced tumor stages. The key to improving survival of pancreatic cancer rests upon early detection of this neoplasm when surgery is possible and hence, it is still a potentially curable stage. According to GLOBOCAN 2018, it is predicted that pancreatic cancer will soon surpass breast cancer as the third leading cause of cancer deaths in the European Union. Its incidence has increased approximately 10% over the last 10 years.
Over the past decade, advances in diagnostic approaches, perioperative management, radiotherapy techniques, and systemic therapies for advanced disease have made relevant but only modest incremental progress in patient outcomes. New strategies for screening high-risk patients to detect pancreatic tumors at early stages are desperately needed to make a clinically significant impact. Here is where deep learning and big data come into play. The potential uses of artificial intelligence (#AI) are boundless and AI-based tools can revolutionize the medicine as we know it, providing early diagnostic methods able to precisely identify in initial stages lesions that can evolve to #cancer.
Modifiable risk factors have been associated with the development of pancreatic cancer, including smoking, alcohol, obesity, and type 2 diabetes. Indeed, not only diabetes has long been correlated with the development of pancreatic cancer, but also pancreatic cancer is also thought to be a risk factor for developing diabetes.
5-10% of all pancreatic cancers are estimated to be attributable to inherited risk factors, most of them resulting from mutations in tumor suppressor genes (e.g. STK11, BRCA2, CDKN2A).
Chronic pancreatitis is also a well known risk factor for developing pancreatic cancer with a 40% lifetime risk of this cancer in patients with hereditary pancreatitis syndromes. Patients with three or more first-degree relatives with pancreatic cancer have a 14 to 32-fold increased risk of developing pancreatic cancer. The disease frequently causes few, if any, symptoms before it develops to an advantaged stage. Unfortunately, those who develop symptoms often have non-specific complaints: back pain, nausea, abdominal fullness or change in stool consistency. Symptoms often attributed to alternative, benign causes, which can delay diagnosis. The clinical features that occur with the highest frequency at the time of diagnosis include abdominal pain (40-60%), abnormal liver function tests (13-20%), jaundice (~30%), new-onset diabetes (13-20%), weight loss (~10%), etc.
Most pancreatic cancers are characterized as ductal adenocarcinoma and thus represent malignancy of the exocrine pancreas whereas a minority represent neuroendocrine tumors. Pancreatic ductal adenocarcinoma (PDAC) may arise from #precursor lesions, termed pancreatic intraepithelial neoplasias (PanINs) or from cystic neoplasms such as intraductal papillary mucinous (#IPMN) and mucinous cystic neoplasms (#MCN). These three morphologic forms of non-invasive pancreatic neoplasias are, by far, the most common lesions.
Due to improvements in #radiological techniques, asymptomatic pancreatic precursor lesions are being increasingly discovered in the general population. These lesions present an enormous challenge in terms of accurate diagnosis and management stratification. However, these precursor lesions represent a unique and growing opportunity to improve pancreatic cancer survival by identification and treatment of this lethal disease at a potentially #curable stage.
We need to stratify those patients with precursor lesions at high-risk of developing pancreatic cancer. Only in that way, high-risk individuals could benefit from the identification and resection of these pancreatic cancer precursors, preventing the progression to invasive cancer.
Adapted from Mizrahi, J. D., Surana, R., Valle, J. W. & Shroff, R. T. Pancreatic cancer. Lancet 395, 2008–2020 (2020).