Diagnosis of pancreatico-biliary malignancy: detection of gene mutations in plasma and stool.

Chromosomal abnormalities, including mutations, deletions and allelic losses of different oncogenes and tumour suppressor genes have been discovered in the DNA of cancer cells and the application of molecular biological techniques now permits identification of these alterations in tumours. Although it has been possible to detect potentially important genetic alterations in tumour material for some years, it is now evident that many neoplasms shed tumour cells into sputum, urine, bile, pancreatic juice, faeces and blood of infected patients. Mutated DNA has also been detected free in the plasma of patients with cancer, and the DNA alternations in plasma are identical to those in the DNA of the primary cancer cells. Thus, the identification of DNA mutations in plasma, pancreatic juice and faeces might be a useful approach for the early detection and monitoring of patients with pancreatic cancer. The K-ras gene is mutated in over 90% of pancreatic cancer. These mutations are well defined, reliably detected by DNA application in assays and occur early in the genesis of pancreatic cancer. K-ras mutations can be detected in cancer tissue and pancreatic duct secretions. K-ras mutations have also been detected in stool of patients with pancreatic cancer. Invasive techniques for obtaining pancreatic juice or pancreatic tissue are undesirable and would certainly be inappropriate for cancer screening. Similarly, there is a lack of enthusiasm for developing diagnostic techniques that involve faecal extractions. Isolation of plasma DNA from pancreatic patients and detection of K-ras alterations with a PCR assay and subsequent product sequencing showed K-ras mutations in the plasma of 17 out of 21 patients (81%), and in cases in which both plasma and pancreatic tissue were available, DNA mutations were similar in plasma and tissue. Plasma DNA alterations were found 5-14 months before the clinical diagnosis of pancreatic cancer in 4 patients. K-ras mutations are also demonstrated in micro-dissected tissues taken from patients with pancreatic hyperplasia, with or without chronic pancreatitis. This has lead to the suggestion that pancreatic cell hyperplasia may be a premalignant condition although the demonstration of K-ras alterations in some cases of chronic pancreatitis has raised doubts about the sensitivity and specificity of K-ras testing for pancreatic cancer. However, the detection of K-ras mutations in plasma may still identify patients with or at risk of developing pancreatic cancer as it may only be in these patients that sufficient quantities of mutated DNA enter and can be detected in plasma. Thus, this non-invasive approach to early cancer detection may be applicable both to diagnosis of the symptomatic patient and for screening. A combined approach with other tumour markers such as p53 gene might increase the sensitivity of the test.