New findings by Johns Hopkins researchers have revealed that pancreatic cancer develops and progresses much more slowly than scientists initially thought. This indicates that there is potentially a very broad window for screening, diagnosis and prevention of the deadly disease, which is the fourth greatest cause of cancer-related deaths in the United States.
According to Christine Iacobuzio-Donahue, associate professor of pathology and oncology at Hopkins, the problem is that the vast majority of people are diagnosed after that window has closed.
The lack of symptoms and of specific imaging techniques makes pancreatic cancer difficult to be detected in its early stages. The new study, however, published October 28th in Nature, suggests that it takes at least ten years for the first mutation in a pancreatic cell to turn it into a full-fledged cancer cell.
After this first cancerous cell appears, it takes an average of another seven years for the single cell to divide into the billions of cells that constitute a tumor the size of a plum, after which at least one of the cells in the tumor is capable of spreading to other organs. Death occurs an average of two and a half years after this metastasis.
“The common and pervasive belief about pancreatic cancer has been that it develops and metastasizes very rapidly and thus screening would be ineffective for this tumor type,” Iacobuzio-Donahue said. “Our findings show that pancreatic cancer is actually quite similar to other tumor types in that it grows slowly over many years, and therefore there is fertile opportunity for early detection while still in the curative stage.”
The team carried out the research using tissue samples collected during the autopsies of seven patients who had died of pancreatic cancer that had spread to other organs. “The metastases from the seven patients in our study were previously included in the pancreatic cancer genome project published by the JHH group in 2008 in Science,” Iacobuzio-Donahue said. “Therefore, we knew all mutations present in the exomic fraction in these seven lesions.”
The samples were taken within six hours of death, so the scientists were able to keep some cells alive long enough to extract their DNA and sequence the series of chemical “letters,” or bases, that make up the genetic code.
They found metastatic deposits in two or more sites in every patient’s body, with the liver, lungs and peritoneum (lining of the abdomen) being the most frequent sites. They then evaluated for the presence or absence of these mutations in all other samples that included the patients’ matched primary carcinoma and other metastases present at autopsy.
“Mutations fell into two categories,” Iacobuzio-Donahue said. “In the first category, we called all mutations that were present in all samples for a given patient – but not normal, of course – as a founder mutation. That means these mutations were present in the founder cell that initiated the infiltrating carcinoma in that patient. In the second category called progressor mutations, the mutations were present in some but not all samples for the patient, so they likely occurred after the carcinoma formed.”
The team used mathematical models to study the timing of pancreatic cancer progression. “We also used the number of mutations in a given patient’s metastasis sequenced in the genome project to derive an estimate of how long it took from the time of the initiating mutation in a normal cell in a given patient until that patient died,” said Iacobuzio-Donahue. “That work was done in collaboration with investigators at the Program for Evolutionary Dynamics at Harvard led by Martin Nowak and indicates it takes at least two decades for the entire process to occur.”
Iacobuzio-Donahue believes screening for pancreatic cancer could be an effective tool towards curing patients with this disease.
“Right now, patients with pancreatic cancer are diagnosed only when they develop a sign symptom that something is wrong and by then the disease is at its most complex and aggressive. Compare this scenario to breast or colon cancer in which patients are routinely screened at a certain age based on what we know of the natural history of those tumor types. Because many breast and colon cancers are diagnosed early, many patients can be cured,” she said.
In another study published in the same issue of Nature, by British researchers at the Wellcome Trust Sanger Institute in collaboration with Iacobuzio-Donahue and her team, tissues and cell lines were used from the same patients as the Hopkins study to look for rearrangements in the genetic code, finding that more than half of specific rearrangements occurred in all metastases and primary tumors.
Other scientists involved in the research were Shinichi Yachida, Siân Jones, Rebecca Leary, Baojin Fu, Mihoko Kamiyama, Ralph H. Hruban, James R. Eshleman, Victor E. Velculescu and Kenneth W. Kinzler of Hopkins, Ivana Bozic and Martin A. Nowak of Harvard University in Cambridge, Massachusetts.
The genome sequencing work was supported, among others, by the National Institutes of Health and the Bill and Melinda Gates Foundation. The study has shed light on the development and progression of pancreatic cancer, and could potentially lead to development of more effective screening methods to identify pancreatic cancer in its early, less lethal, stages.