Article Text

Download PDFPDF

Early diagnosis of pancreatic cancer; looking for a needle in a haystack?
  1. Marco J Bruno
  1. Correspondence to Dr Marco J Bruno, Department of Gastroenterology and Hepatology, Erasmus University Medical Center, ‘s Gravendijkwal 230, Rotterdam 3015 CE, The Netherlands; m.bruno{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

The paper by Kanda et al highlights a potentially novel method for the early diagnosis of pancreatic neoplasia.1 Although our understanding of the pathological and molecular events of pancreatic cancer has advanced significantly over the past years, we are still far away from a clinically meaningful breakthrough in its treatment as demonstrated by the poor survival statistics which have hardly changed over the past decades.2 ,3 In light of this sobering observation, it is important to consider other ways of trying to lower the mortality of this dreadful disease. Primary prevention through life style changes of established risk factors like smoking has a considerable potential to reduce the number of pancreatic cancer deaths,4 but it is well known that such behavioural changes are difficult to accomplish. Secondary prevention by screening the general population for pancreatic cancer is not feasible at present because of its low incidence and the lack of a non-invasive, reliable and cheap screening tool. However, it is estimated that about 10% of all pancreatic cancer cases are caused by inherited (genetic) factors with some individuals carrying a life time risk up to 55%.5 Affected individuals are known mutation carriers of pancreatic cancer prone hereditary (tumour) syndromes (eg, familial atypical multiple mole melanoma syndrome, Peutz–Jeghers syndrome, breast cancer 1 and 2 (hereditary breast and ovarian cancer)) and, the largest group, individuals with a strong family history of pancreatic cancer, but without a known underlying genetic defect.

The only chance for a cure of pancreatic cancer is offered by surgical resection and the outcome is strongly dependent on the tumour stage. The 5-year survival for patients with a tumour smaller than 2 cm, negative margins and no nodal involvement (T1N0M0 American Joint Committee on Cancer stage IA) after curative resection is 31.4%, with numbers dropping dramatically at higher stages.6 Unfortunately, only a small minority of symptomatic pancreatic cancer patients present at stage IA. However, pancreatic cancers in asymptomatic individuals are likely to be small(er) and early detection and resection should offer a better prognosis. It might even be possible to detect (advanced) non-invasive precursor lesions before they have evolved into a malignancy. The latter is highly desirable as surgical resection of precursor lesions evidently avoids the risk of residual tumour and metastases, thereby dramatically improving survival rates. Recognised and well-defined precursor lesions of invasive pancreatic cancer are pancreatic intraductal neoplasia (PanIN), intraductal papillary mucinous neoplasms (IPMN) and mucinous cystic neoplasms.7 PanIN and IPMN are both intraductal lesions predominately composed of columnar, mucin-producing cells and they occur frequently in individuals at a high risk of pancreatic cancer.8 Importantly, quantitative analysis of the timing of the genetic evolution of pancreatic cancer shows that it takes at least a decade between the occurrence of the initiating mutation and the birth of the parental, non-metastatic founder cell and at least 5 more years for the acquisition of metastatic ability which highlights the broad time window of opportunity for early detection.9

Based on these considerations, the utility to screen high risk individuals seems plausible and promising provided a suitable test is available that is capable of detecting small pancreatic tumours and precursor lesions. During recent years, the application of endoscopic ultrasonography and MRI/magnetic resonance cholangiopancreatography has been tested in various high risk populations around the world. Both imaging modalities detect a high frequency of precursor lesions and in patients selected for surgical resection some of these lesions proved to be advanced neoplastic lesions.5 To predict the presence of advanced neoplasia or malignancy in IPMN we rely on morphological features as demonstrated during imaging investigations using the criteria defined by Tanaka et al which take into consideration cyst size, presence of intracystic nodules and the diameter of the main pancreatic duct.10 ,11 Diagnosis of PanIN lesions is even more challenging as their morphological correlate on imaging is much less apparent and probably only reflected indirectly by features of chronic pancreatitis as detected on endoscopic ultrasonography.

It is for all these reasons that the article by Kanda et al is of particular interest as it highlights a potentially novel method for the early detection of pancreatic neoplasia.1 Their study is an eloquent example of proof of principle testing; it shows that samples of duodenal juice collections suctioned through the endoscope channel after pancreatic stimulation with secretin are an excellent and reliable source of mutant DNA from the pancreas. This is exciting news as the methodology is simple, adds only very little time to the Endoscopic UltraSonograph (EUS) screening investigation and does abolish the need for invasive endoscopic retrograde cholangiopancreatography which is a pivotal requirement in a screening programme.

Their application of GNAS mutation detection, a gene encoding the G protein α-subunit involved in the transmission of hormonal and growth factor signals to effector proteins, per se does not appear particularly helpful from a clinical point of view; although highly specific for IPMN, it lacks adequate sensitivity and its expression is not associated with the grade of dysplasia. Nevertheless, some interesting observations were made. Mutant GNAS concentrations were higher in patients diagnosed with IPMN than in those with diminutive cysts, suggesting that its measurement in some degree reflects the size of the GNAS-mutant clone. Also, individuals enrolled for their family history of pancreatic cancer were more likely to have mutant GNAS detected than those with known germline mutations, indicating that depending on the underlying condition differential mechanisms are involved in cyst development some of which are independent of GNAS mutations. Intriguing and of important scientific interest is the capability of GNAS mutation analysis to ‘detect’ pancreatic cysts even before they are visible on imaging investigations highlighting the sensitivity of this methodology.

Duodenal collections of secretin-stimulated pancreatic juice analysis for pancreatic cancer screening may show great promise in high risk individuals, but it is important to note that its clinical application is much broader. For example, the diagnostic and management dilemmas in individuals presenting with asymptomatic sporadic cysts which are discovered at an increasing frequency due to the abundant use of cross sectional imaging are much more prevalent in daily clinical practice, in particular when dealing with small branch IPMN. The quest is now on for a comprehensive panel of markers of pancreatic neoplasia, in particular for detecting PanIn-3 and IPMN with high grade dysplasia, with discriminative and predictive capabilities that go beyond what already can be achieved by EUS or MRI. Let the search begin….



  • Funding None.

  • Competing interest None.

  • Provenance and peer review Commissioned; externally peer reviewed.

Linked Articles