Johns Hopkins clinician-scientists provide regular MRI and endoscopic ultrasound screenings for several hundred people whose family history puts them at risk of developing pancreatic cancer. Illustration shows pancreatic cancer concept.
Researchers in the Johns Hopkins Division of Gastroenterology and Hepatology say the next step is earlier detection in those without familial risk.
When it comes to early detection and treatment of pancreatic cancer, Marcia “Mimi” Canto calls the last two decades “the discovery phase.” She and other Johns Hopkins experts say that, while recent years have seen lifesaving advances in catching pancreatic cancer early, the next phase of research will cast a wider net.
The Johns Hopkins gastroenterologist serves as the director of clinical research for the Division of Gastroenterology and Hepatology. She’s a member of the Pancreas Multidisciplinary Cancer Clinic at Johns Hopkins, whose faculty members — in addition to gastroenterology — come from the fields of pathology, surgery, pain management and diagnostic imaging, as well as medical and radiation oncology.
Johns Hopkins provides regular MRI and endoscopic ultrasound (EUS) screenings for several hundred people whose family history puts them at risk of developing pancreatic cancer. Canto says improvements in MRI and EUS technology have helped physicians spot smaller and smaller tumors.
“No question about it, we’ve been able to save lives,” says Canto. “We’ve found that, if we can find pancreatic cancers early, they’re 80% to 90% resectable, compared with something like 20% when patients start to have symptoms.”
By “early,” Canto says she means stage 1 pancreatic cancer.
“Or even precancer,” she adds. “We can detect precancerous lesions in patients who are part of the surveillance program. But not everyone who gets pancreatic cancer has a familial risk. Those are the cases we want to catch or prevent.”
She points to a 2018 article in the journal Gastroenterology in which she and her colleagues reported on results of a long-term study of patients who undergo surveillance at Johns Hopkins. Canto and co-authors wrote that 85% of patients who underwent surgery to remove stage 1 pancreatic cancer or precancerous lesions survived for at least three years.
“And they’re all still alive,” Canto says with a wide smile.
Despite taking nearly 50,000 American lives each year, pancreatic cancer accounts for only about 3% of cancer deaths annually. Even people deemed to be at a high risk develop it only a small fraction of the time. Screening patients with family history of the disease, while expensive, is worthwhile.
Such screenings in the general population, however, are impractical. Given that only about 10% of pancreatic cancers are hereditary, there is much more room to improve.
Research Suggests MRI and EUS Imaging Lead to Annual Increase in Pancreatic Cancer Diagnoses
Michael Goggins, a professor of pathology and a gastroenterologist, is a co-author on the Gastroenterology article. He points to another journal article for a larger perspective.
He, Canto and biostatistician Amanda Blackford, professor of oncology, epidemiology and gastroenterology Alison Klein, and professor of pathology and oncology Ralph Hruban last year published their study of nearly 100,000 cases in the Journal of the National Cancer Institute. They found the detection of stage 1A pancreatic cancer in the U.S. has increased in recent years, and so has survival, with about 80% of patients surviving at least five years after a stage 1A diagnosis.
Stage 1A pancreatic cancer means the tumor is 2 centimeters or smaller, and that disease has not spread outside the organ.
Goggins says increased detection of early-stage pancreatic cancer has led to a decrease in the percentage of advanced-stage diagnoses.
“Thanks to improved screening tools and expertise, we are detecting pancreatic cancer earlier,” he says. “We found that the proportion of patients diagnosed with stage 1A pancreatic cancer increased, while the average age at diagnosis decreased.”
Goggins credits the ability to find the cancer early to advances in imaging technology.
“In other words, when using MRI or EUS, we can see cancer we couldn’t see before.”
An increase in the number of people enrolled in pancreatic surveillance programs across the country also improved the statistics, says Goggins, who adds that continued advances in the application of artificial intelligence and machine learning will likely improve surveillance methods even more.
The next step, he says, is detecting pancreatic cancer sooner in people who don’t fall into the high-risk category — a prospect he and Canto are actively researching.
“We want to know why people [who don’t have a family history of pancreatic cancer] get it,” he says. “That’s still a tough one. But we had to prove we could do it with the high-risk population first. We still have a lot to do, but what we’ve learned is encouraging.”
The Future of Pancreatic Cancer Research
Last year, a team of Johns Hopkins researchers authored a review article in the journal Modern Pathology that provided their predictions for the next decade of pancreatic cancer research.
- The discovery of genetic variants that make a person more likely to develop pancreatic cancer has had an enormous impact on early detection. The authors believe that, in the next decade, those variants will be further explored and categorized, leading to even more personalized treatments.
- Drug resistance is a barrier to certain therapies aimed specifically at pancreatic cancer. The authors predict that, in the next decade, physicians will prescribe combinations of new targeted therapies, each aimed at a different facet of the disease, making drug resistance more difficult.
- The authors believe that, soon, experts will better understand a critical phenomenon that contributes to the spread of pancreatic cancer: the disease’s invasion of veins in the pancreas. Therapies to slow that process, they say, would lead to a reduction in metastases.
- More specific and more sensitive biomarkers of low-grade precursor lesions, say the authors, will help reduce the number of deaths from the disease.
- Better integration of clinical data, allowing artificial intelligence to discover new patterns that humans can’t spot. The integration of genetics and digital pathology into deep-learning algorithms is, the authors say, “right around the corner.”
Adapted from “The genetics of ductal adenocarcinoma of the pancreas in the year 2020: dramatic progress but far to go,” published in Modern Pathology, July 2020.
Elizabeth Thompson, M.D., Ph.D., Nicholas Roberts, Ph.D., Vet.M.B., Laura Wood, M.D., Ph.D., James Eshelman, M.D., Ph.D., Michael Goggins, M.B.B.Ch., M.D., Scott Kern, M.D., Alison Klein, Ph.D., M.H.S., and Ralph Hruban, M.D.