NEW YORK – A team of researchers from Johns Hopkins University is hoping to take a deep dive into the genetic underpinning of pancreatic cancer using long-read sequencing technology.
Led by Johns Hopkins investigators Alison Klein, Winston Timp, and Michael Schatz, the project, funded by the Lustgarten Foundation, seeks to analyze hundreds of samples from pancreatic cancer patients to help decipher genetic predisposition to the disease, which has been so far inadequately understood.
"Pancreatic cancer is the third leading cause of cancer death in the US," said Klein. "It's a disease late in life, and it is rapidly fatal."
Despite this, the genetic changes underlying pancreatic cancer still remain obscure to researchers. Even though about 10 percent of pancreatic cancer patients have another close relative who also has the disease, Klein said, scientists still fail to grasp which genes are responsible in most of these cases due to a multitude of hurdles.
For instance, because the median age of diagnosis in the US is 71 years, it is challenging to carry out the traditional pedigree-based linkage analysis, where DNA samples from multiple family members with the disease are normally required, Klein said.
Researchers have also tried a slew of other approaches, including DNA microarray analysis, whole-exome sequencing, and short-read whole-genome sequencing. While these technologies have helped uncover various clues about the disease, Klein said, they "were not finding any more low-hanging fruit," leaving the majority of patients' cases unsolved.
"Many of these families have multiple cancers," she added. "We really believe there are genes [with mutations] that we just haven't been able to locate with our current technologies."
One group of genetic variations that the Hopkins researchers are particularly eyeing, for instance, is structural variants (SVs). "SVs are incredibly common in specific regions of pancreatic cancer," said Timp, a biomedical engineering professor who is experienced with long-read sequencing. "I felt like pancreatic cancer might be really ripe for the taking here."
Previously in collaboration with Schatz, a computational biologist who has developed several well-known sequencing analysis tools, Timp's group used long-read sequencing to discover SVs in BRCA1, a gene that is also implicated in pancreatic cancer.
"The most stunning result to me was finding structural variants in BRCA1 in breast cancer patients that were never seen before," Schatz said. The SVs were "hidden away in the repeats, basically impossible to recognize with short-read sequencing."
Using the three-year, multimillion-dollar Lustgarten Foundation award, the JHU researchers plan to scrutinize the genetic code of more than 400 pancreatic cancer samples using long-read technologies, synergizing their expertise in inherited diseases, long-read sequencing, and computational biology.
The project will tap into JHU's National Familial Pancreas Tumor Registry, which is currently maintained by Klein's team. Established more than two decades ago, the registry — one of the largest in the nation of its kind — enrolls almost 8,000 families affected by both familial and single pancreatic cancer cases. Additionally, samples from the Mayo Clinic and the University of Toronto will be included in this study.
The group will first focus on the germline DNA samples from high-risk families — those with an inherited predisposition — given that the tumor tissue samples are harder to obtain for somatic studies. "One thing that's challenging about pancreatic cancer is that because it is late onset and … in an internal organ, the tissue samples are not easy to come by," Klein noted.
As samples are selected, they will be processed and sequenced at Timp's lab. Timp said his team is working to figure out how to carry out the experiment at scale and lock down the protocols for different sample types.
That said, Timp emphasized that experimental design is "key" for the entire project. "We can't just throw long-read [sequencing] at problems and expect them to be solved," Timp said. "There are going to be a lot of signals, but there is also going be a lot of noise."
"The idea was to have the experimental design to help us really focus on what is likely to be the most relevant potential driver mutations," agreed Schatz, whose lab will primarily help tackle the project on the computational end. "I think we have a handle now on common variation, but there are some big challenges for rare variation that just runs in [certain] families, so we're trying to turbocharge some of those methods."
The group has decided to use the Oxford Nanopore Technologies PromethIon platform to generate the long-read data due to its throughput. "We need to get going right away, and at the scale that we're talking about, PromethIon was the only viable option," said Schatz.
However, he noted that "the world did kind of change" last October when Pacific Biosciences launched its new Revio sequencer, which promised to boost the throughput of PacBio sequencing by at least an order of magnitude. With that in mind, Schatz said the team will "keep a close eye on" the new platform and is "eager to explore both [long-read sequencing modalities] moving forward."
After receiving the Lustgarten funding last fall, the group has generated preliminary data for the first dozens of samples. Besides genomics, the collaborators said they are interested in collecting RNA and other data to look at expression changes in the samples.
"We have a laundry list of things that we want to go through as we're generating the data," Timp said. "It's likely we're going to see something. I would be very surprised if we don't." As the new methods and data are generated from the project, he added, the group will also share the results with the community.
Ultimately, the Johns Hopkins investigators hope their endeavors will not only enable them to better understand the etiology of pancreatic cancer — thus paving the way for the development of screening tests and precision treatment — but also establish a framework to use long-read sequencing to study other diseases.
"I think that now we are at this inflection point for long-read [sequencing]," Timp said. "The things that we figure out are going to tell us stuff about pancreatic cancer and also how to approach other human disease questions."