Researchers have identified two novel panels of microRNAs that eventually might be used to screen patients for pancreatic cancer, distinguishing affected patients from healthy individuals and from people who have other diseases of the pancreas, particularly pancreatitis, according to a report published online Jan. 21 in JAMA.
The two microRNA panels were better than standard serum cancer antigen 19-9 (CA19-9) testing at identifying pancreatic cancer within the initial study cohort used to assess hundreds of microRNA candidates. However, CA19-9 had better diagnostic accuracy in the validation cohort used to replicate those results. Combining the results of either microRNA panel with CA19-9 offered the best diagnostic accuracy of all, said Dr. Nicolai A. Schultz, of the department of oncology, Herlev Hospital and Copenhagen University Hospital, and his associates (JAMA 2013;311:392-404).
The discovery may lead to earlier diagnosis of pancreatic cancer at a stage when it is more treatable. At present, only 20% of patients can be operated on with curative intent, because most already have advanced disease by the time they are diagnosed, Dr. Schultz and his colleagues said.
However, the investigators emphasized that their findings are preliminary. "Further research is necessary to understand whether these microRNAs have clinical implications as a screening test for early detection of pancreatic cancer and how much this information adds to serum CA19-9," they noted.
The investigators began their search for relevant biomarkers by screening whole blood samples for the presence of any of 754 microRNAs suspected of being expressed in circulating blood by pancreatic cancers. MicroRNAs are small (17-25 nucleotides long), noncoding, single-stranded RNAs that promote oncogenesis by inhibiting the expression of tumor suppressor genes or by upregulating the expression of oncogenes. They are active in the pathogenesis, progression, and metastasis of pancreatic cancer.
The blood samples had been taken from 409 patients with histologically verified pancreatic cancer and 337 control patients, consisting of 25 patients with chronic pancreatitis and 312 healthy blood donors. Samples were drawn from the pancreatic cancer patients before they underwent any treatment; 44 had resectable tumors, and 365 had unresectable tumors.
The microRNA panels were identified and tested in a three-step process using discovery, training, and validation cohorts of patients.
In the discovery cohort, 38 microRNAs were found to have the potential to distinguish pancreatic cancer patients from other participants. Thirteen of those microRNAs were the strongest predictors and were used to develop the two diagnostic panels, which were then tested in the validation cohort.
The first diagnostic panel, termed index I, comprised 4 microRNAs, and the second panel, index II, comprised 10 microRNAs. Both panels correctly diagnosed pancreatic cancer in 85% of affected patients and correctly diagnosed chronic pancreatitis in 29% of affected patients.
Among control patients, index I correctly ruled out pancreatic disease in 48% and index II in 54%. When CA19-9 results were added to the panel results, index I correctly ruled out pancreatic disease in 98% of the control patients, and index II did so in 89%.
Thus, "the indices do achieve some separation of pancreatic cancer from healthy participants, but not a complete separation," Dr. Schultz and his associates said.
"When this microRNA biomarker study was designed in 2008, the aim was to identify microRNAs in whole blood that could identify pancreatic cancer (local, locally advanced, and metastatic) in individuals thought to be healthy," the investigators explained. "We first wanted to test the extremes (i.e., patients with known pancreatic cancer vs healthy blood donors), because if panels of microRNA could not differentiate patients with . . . high tumor burden from these healthy individuals, it would be difficult to develop a diagnostic microRNA test for patients with small tumors and low-stage cancer.
"In the future, we will assess the accuracy of indices I and II in combination with serum CA19-9 in large cohorts of patients either seen by a family physician or referred to hospitals on suspicion of cancer ... and assess whether it is possible to identify patients with pancreatic cancer at an early stage," Dr. Schultz and his associates said.
This study was supported by Herlev Hospital; the Danish Ministry of Science, Technology, and Innovation; the Danish Cancer Society; Region Hovedstadens forskinings fond til Sundhedsforskning; Aase og Ejnar Danielsens Fond; and Prosektor Axel Søeborg Ohlsen og Else Søeborg Ohlsen Mindelegat. Dr. Schultz and his associates reported holding a European patent application submitted by Copenhagen University Hospital.