Massachusetts General Hospital's Diagnostic Molecular Pathology Laboratory and Translational Research Laboratory are planning to implement next-generation sequencing for screening cancer patients.
As a first step, MGH researchers are converting their SNaPshot assay, a multiplex-PCR and CE-based test that screens for hotspot mutations in 14 oncogenes, to NGS. Later, they plan to develop larger panels for clinical use to guide treatment and placement into clinical trials, and also for research use to help identify markers of drug response and resistance.
MGH has been offering SNaPshot through a CLIA-certified laboratory for about three years and uses it to test about 50 to 60 patients per week, Long Le, a pathologist within MGH's Diagnostic Molecular Pathology lab, told Clinical Sequencing News.
Le said that because the assay is CE-based and screens only for hotspot point mutations, it does not pick up indels or copy number variation confidently, so the lab runs other tests alongside SNaPshot to pick up those types of alterations.
For instance, the test cannot detect EGFR deletions or MET and HER2 amplifications. Instead, the lab uses PCR amplicons and CE sequencing in combination with fluorescence in situ hybridization to detect these alterations.
"SNaPshot can kind of hint at that information," Le said, "but then we have to confirm it."
Switching to next-gen sequencing would enable all these alterations to be detected in one test. "We can lump it all into one assay that would encompass what was originally three or four assays," Le said.
Additionally, SNaPshot is "a pretty focused assay, but we would like to open it up," Le said.
The team is now developing its own in-house multiplex-PCR targeted assay that it will run on either the Ion Torrent PGM or the Illumina MiSeq.
The next-gen version of SNaPshot will still cover the same 14 genes, but will expand the number of mutations it detects in genes like TP53 and PTEN. "For certain genes like TP53," said Le, "mutations are scattered throughout," so targeting hotspots will likely miss mutations. For genes like KRAS or BRAF, hotspot sequencing is effective, he said.
"There are only one or two exons in those genes that you should look at," said Le. "Other portions of the gene are not mutated."
Le said that the team considered using either the Ion AmpliSeq Cancer panel or Illumina's TruSeq Cancer panel, but decided to develop its own assay primarily because it wanted more coverage of tumor suppressor genes like TP53 and PTEN. Neither assay covers those genes in their entirety.
The MGH lab currently has a PGM and has been using that instrument to develop and quality-control the assay. Le said that they are considering a MiSeq for the clinical test, however, due to homopolymer issues on the PGM — particularly for genotyping TP53 exons in full. The lab does not yet have a MiSeq but has been running samples on a system housed at the Dana Farber Cancer Institute, Le said.
Le anticipated that an NGS-based SNaPshot test could be up and running out of the CLIA lab in about three to four months. The main challenge will be putting automation and bioinformatics infrastructure in place, he said.
For this, the MGH team has been collaborating with PerkinElmer, testing out equipment such as the NGS Express Workstation for automating library prep for benchtop sequencers, putting in place a LIMS system for tracking samples, and implementing informatics software.
A Broader View
Aside from the SNaPshot assay, Le said the team is also working to develop a larger cancer panel.
"We need a fast turnaround assay," he said. SNaPshot would be "the initial go-to assay. But those that are negative, could get a larger panel," he said.
Additionally, he said, while SNaPshot would be used solely for diagnostic purposes where speed is a main criteria, larger panels could be used for discovery purposes.
"We're trying to find the balance between being as comprehensive as possible and being able to take the information and narrow it down to a concrete statement to guide the care of patients," Darrell Borger, co-director of MGH's Translational Research Laboratory, told CSN.
"What we're trying to do now is … increase the comprehensiveness of our testing," he said.
The next test after SNaPshot, Borger said, would likely home in on 50 to 150 different targets to "really focus on those genes that are well-described as being drivers in cancer," he said.
Additionally, he said, the lab works with clinicians on clinical trials. So, while focused, targeted sequencing could be used to place patients into clinical trials, more comprehensive sequencing could be done retrospectively to try and determine why some patients responded and why others did not, identifying biomarkers of drug response and resistance.
Similarly, sequencing could be used on patients who initially respond to a treatment but later develop resistance to try and identify the acquired alterations that drive resistance, he said.
In these cases, cost and turnaround time are not as important as in a clinical test, he said, and information that is difficult to interpret is not as much of a concern, because the sequencing is not being used to make clinical decisions about patient care in real time.
"We're very interested in developing a platform that interrogates around 800 targets and that would at least initially, be applied much differently," he said. The expanded panel could eventually provide the basis for developing an even broader clinical testing platform, he said.
Borger said that in developing such an assay, the team plans to work closely with clinicians who are running the trials.
It will be important to make sure that the assay includes not only targets for which drugs are already available, but targets for drugs that are entering clinical evaluation for human studies, or projected to be entering phase I clinical trials in the near future, he said.
Borger said the assay would be developed in house, but that the researchers would potentially work with a company that develops capture reagents to help design the assay, and that the MGH lab would bring in next-gen sequencers, rather than outsource sequencing.
Both Le and Borger said that the lab is evaluating several different platforms simultaneously, but has not yet settled on a specific technology. Aside from its in-house PGM, it has been outsourcing targeted sequencing to Dana Farber on the MiSeq and exome sequencing to PerkinElmer on the HiSeq.
However, MGH is not considering doing exome or whole-genome sequencing for clinical purposes, due to cost, turnaround time, and the complexity of the data.
"The goal is to direct what is the best treatment option for patients," Borger said. "It's unclear how whole-exome sequencing would be helpful at this stage." Even in the more focused panels, "there are many mutations that we anticipate finding whose biological and clinical relevance may not be known," he said.
Borger said that the lab would seek reimbursement for its clinical tests. The CE-based SNaPshot assay is already reimbursed at a "high rate," but he did not want to speculate on whether an NGS-based version of the test would be reimbursed at similar rates.