This story was originally published May 15.
In one of the first published examples of Life Technologies' Ion AmpliSeq-Cancer Panel in the clinic, researchers from the University of British Columbia used the panel to interrogate two tumor samples from an unusual pediatric brain cancer case.
The researchers demonstrated that the technology could deliver results within three days of sample collection using only 10 nanograms of DNA. While the findings were not immediately clinically actionable, the team did uncover a rare mutation to the MET gene.
The team is now testing the panel on around 200 gastric cancer samples and is part of a group that is applying for a large grant to study the implementation of the PGM and Illumina MiSeq into clinical pathology labs.
The pediatric brain cancer case was published this month in the Journal of Neurosurgery Pediatrics.
The patient, a 10-month old infant, developed a rare case of brain cancer. The primary tumor was a grade four glioblastoma, which typically comes with a poor prognosis. It was removed with surgery and the patient was placed on chemotherapy. However, she did not tolerate the drugs well. Three months later, a brain scan detected tumor recurrence so therapy was stopped.
However, 18 months later, not only was the patient still alive, but she also showed signs of developmental improvement. At the same time, the recurrent tumor had grown larger, so doctors surgically removed it.
When the researchers examined the second tumor under the microscope, it appeared to be a type of grade 2 tumor called pleomorphic xanthoastrocytoma, or PXA, which is much less deadly than glioblastoma.
"It's well recognized that low grade tumors can evolve with time and accumulate mutations into a higher grade progression, which is usually, obviously, a bad thing," said Christopher Dunham, senior author of the study and pathologist at the Children's and Women's Health Center of British Columbia. But typically, "you don't see a tumor go from high grade to low grade," he told Clinical Sequencing News.
To try and understand what caused the tumor to become less lethal, the researchers embarked on a genetic analysis of both the primary and relapse tumor.
They first karyotyped both tumors, but did not find any major differences between the two. So the team next used Sanger sequencing to examine individual genes. Recent work has found that between 70 percent to 80 percent of PXA tumors carry the V600E mutation in BRAF, so the researchers examined both tumors for that mutation, but neither had it.
Using Sanger sequencing, they also looked at the IDH1 and IDH2 genes, which have also been linked to pediatric brain cancers. All the tests turned up negative.
Finally, they decided to turn to next-generation sequencing on the Ion Torrent PGM using the 314 chip and the Ion AmpliSeq Cancer panel, which interrogates 190 amplicons in 46 known cancer genes.
Using 10 nanograms of formalin-fixed paraffin-embedded tissue from each tumor sample the researchers were able to achieve an average coverage of 1,000-fold for each sample.
The results confirmed the Sanger sequencing findings of no mutations to BRAF, IDH1, or IDH2, and also identified one rare mutation to the MET oncogene. In the first tumor, the mutation was present only at a frequency of 0.25 percent, and in the second tumor it was present at 5.5 percent. No other mutations were found in either tumor.
"I don't really know what to make of that," Stephen Yip, a coauthor of the study and assistant professor of medicine at the University of British Columbia, told CSN.
The MET mutation in the first tumor sample was at such a low frequency that it was below the threshold to call as a mutation, so it isn't clear if the mutation is real or not, he added. The MET mutation in the second sample was at a high enough frequency to be called as a real mutation, but was nevertheless still very rare.
"I think [both mutations] might be real, but it's really pending validation by a different method," he said.
Additionally, the fact that the same mutation was found in both samples, and at an increased frequency in the second tumor, lends some support to the idea that there was some clonal evolution between the first and second tumor, "but the caveat is that we're dealing with such a low allele frequency that there are a lot of technical issues that need to be resolved before we can say anything definitive," Yip said.
Further studies are being done to try and determine the mutation's biological relevance, he added.
The same mutation has been identified in other cancers, such as thyroid carcinoma and small-cell lung cancer, and other MET mutations have been associated with renal papillary carcinoma, gastrointestinal cancers, and glioma.
Over-expression of the gene has been associated with poor prognosis in adult glioblastoma, but MET mutations have not previously been reported in PXA tumors.
The finding was not immediately clinically actionable, and further work is needed to determine its significance, but the patient is nevertheless "doing quite well" 20 months post-surgery, said Dunham.
PGM in the Clinic
The study was also a good test of the PGM in a clinical setting, Yip said, adding that there are clearly advantages of using next-gen sequencing.
For instance, if the team had used Sanger sequencing on all the individual genes, they would have needed around 1 microgram of DNA for each Sanger sequencing test. But, using the PGM, they needed only 10 nanograms of DNA and were able to cover multiple genes, he said. Additionally, Yip added, "we went from DNA to results in three days time."
Yip said that he is now working on a project to test the AmpliSeq Cancer Panel on 200 gastric tumors, as well as on ovarian and colorectal cancer samples, using FFPE derived DNA, with the goal of bringing the PGM and the cancer panel into the clinical testing laboratory.
Yip added that currently, whole-genome and whole-exome sequencing is still too expensive for the clinic, and said that "having a focused panel looking at specific genes that might be actionable would be a lot more useful clinically."
Additionally, Yip said that the BC Cancer Agency, along with other centers in Canada, is applying for a large multi-center grant from Genome Canada that would explore the use of both the PGM and the MiSeq in a clinical setting. David Huntsman, associate professor of pathology and laboratory medicine at the University of British Columbia, is leading that effort, Yip said.
The grant would look at "the practicality of having a PGM or MiSeq in a pathology lab and having a pathologist running these samples," said Yip. The lab would run either the Ion cancer panel or Illumina's TruSeq Amplicon - Cancer panel, which targets 212 amplicons in 48 genes.
It would look to compare not only the performance of each of the platforms, but also how they fit into the entire clinical workflow and their reproducibility, accuracy, and consistency.
"We want to really get the technologist that runs Sanger sequencing to be proficient at running a [next-gen] panel … and see how it affects workflow," Yip said.
Yip added that his lab, which currently has only the PGM, is also looking to acquire a MiSeq. For high-throughput sequencing, the lab has access to the HiSeq 2000 machines at the Genome Sciences Center.
He said that even though these benchtop sequencers don't allow for whole-genome sequencing, they are ideal for clinical laboratories because they are less expensive and quick, and researchers can do deep, targeted sequencing.