Skip to main content
Premium Trial:

Request an Annual Quote

Targeted Sequencing Panels Gaining Ground for Cancer Dx


By Monica Heger

Even as the costs of whole-genome sequencing continue to drop, targeted approaches continue to make inroads into the clinical realm, with panels that sequence commonly mutated genes in cancer becoming a key tool in cancer diagnostics and prognostics, and helping to select patients for clinical trials or specific drug regimens.

A number of academic and commercial efforts are underway to develop sequencing panels with tens to hundreds of genes implicated in cancer. For example, Foundation Medicine aims to launch a CLIA-certified commercial assay that would sequence 176 known cancer genes (CSN 6/8/2011), and the UK government is funding an initiative that would make targeted sequencing-based panels for cancer patients part of routine health care (CSN 6/21/2011).

Targeted sequencing may be especially useful for cancer because studies have found that many mutations thought to be unique to a cancer from a particular tissue are in fact found across a range of cancers, so one targeted sequencing panel could be used for all cancer patients. Additionally, researchers are continuing to gain an understanding of specific mutations, and what they mean for prognosis and treatment response.

Nikhil Wagle, a fellow at the Dana-Farber Cancer Institute, told Clinical Sequencing News that targeted sequencing is ideal for cancer because it can be done "for a reasonable price in the clinical setting" and can be limited to known, clinically actionable genes.

Wagle is working with a group at Dana-Farber and the Broad Institute that is developing a targeted panel to sequence the coding regions of known cancer genes. "The information we get, we know what to do with," he said.

"It would be great to sequence each individual tumor genome. But that would be extraordinarily expensive, and you'd get all this information that you didn't know what to do with," he added.

Wagle and his team are currently testing a panel of 138 genes using Agilent's SureSelect in-solution enrichment technology and sequencing on the Illumina HiSeq, which allows them to pool and barcode samples, also helping to drive down costs.

While the researchers are not looking to launch a commercial test, Wagle said the hope is that they would eventually be able to use it on patients who are seen at Dana-Farber.

The team has tested it in a pilot study of 10 formalin-fixed, paraffin-embedded breast and colon cancer tissue samples. They achieved an average of 102-fold coverage and detected 125 single nucleotide variants, 17 small insertions and deletions, and 47 copy number alterations. The alterations included known mutations in KRAS and PIK3CA; nonsense mutations in the tumor suppressor genes APC, MSH2, SMAD2, SMAD4, TSC1, and TP53; and a deletion in BRCA1.

Additionally, some of the mutations suggested novel therapeutics, such as PI3K inhibitors, TOR inhibitors, PARP inhibitors, and FGFR inhibitors.

"We were able to identify at least one potentially actionable alteration in each of the samples," Wagle said. Results of the study were presented at the American Society of Clinical Oncology meeting in June.

The National Cancer Institute is also considering targeted sequencing of cancer patients, according to Paul Meltzer, the NCI's head of molecular medicine, who recently spoke at the Next Generation Sequencing and Genomic Medicine Summit in San Francisco.
To sequence cancer patients in a clinical setting, turn-around time and data interpretation are two critical considerations, he said.
"Patient care is a real-time activity. Robust, actionable data is needed in a timely fashion at a reasonable cost," he said.

One thing the NCI is considering is creating a "hierarchy of mutations" while doing sequence analysis. For example, in a non-small cell lung cancer patient, it might make sense to first look for mutations to EGFR, and then to RAS and RAF, and finally to PTEN and PIK.

[ pagebreak ]

"While going through the analysis, the patient can be assigned to the best available drug," he said.

Research vs. Clinical Setting

The NCI has already begun sequencing clinical samples of cancer patients, usually from FFPE samples, Meltzer said. So far they have been using the Illumina platform, but are also exploring other technologies, such as the Pacific Biosciences RS, Illumina's MiSeq, and the Ion Torrent PGM.

With the Illumina HiSeq, and doing exome sequencing, turnaround time is about 25 days, he said. Currently, the team is "focused on trying to break the time barrier."

Aside from testing other platforms, Meltzer said the team is considering more targeted sequencing in order to make sequencing of cancer patients clinically applicable.

"There is a dichotomy between sequencing and the genome we know how to interpret," Meltzer said. For a clinical sample, Meltzer said the NCI team is exploring whether exome sequencing, transcriptome sequencing, or targeted sequencing of known targets with relevance to therapy is the best way to go.

"When you look at these exomes, you find lots of mutations, but you just don't know what to do with them," he said. On the other hand, "you also find mutations that aren't supposed to be in that tumor, that you do know something about."

Wagle agreed that deciding on the number of genes to sequence for a test is difficult, and even though his team has settled on 138 genes for now, it doesn't preclude adding other genes as more data is gathered.

Sequencing other genes would be "great for research, but we wouldn't know what to do with them," he said. There is a difference between sequencing for discovery and sequencing in a clinical setting, Wagle added. For clinical purposes, it is important to "start with the genes we know what to do with," he said, while for discovery, "you have to cast a wider net."

Wagle said that his team is currently evaluating its test on additional tumors, looking at the "landscape of mutations in various tumor types," and conducting both retrospective and prospective trials to look for biomarkers that correlate with drug response, prognosis, and drug resistance.

"My hope is that we get this technology to the point where it is robust, trustworthy, CLIA-approved … and that we'd be able to use it in patients."

Standard of Care

Developing targeted sequencing tests for cancer patients is catching on in the UK and Canada, where government-funded initiatives aim to make such tests the standard of care for cancer patients.

In the UK, three pilot projects using next-gen sequencing are being funded. One is being spearheaded by Life Technologies, which is developing panels on the Ion Torrent PGM; another is headed up by Source BioScience, which is aiming to develop panels on Illumina's MiSeq; and the third by Oxford Gene Technologies, which has partnered with a healthcare software developer to develop a "clinical decision tree" based on mutation status.

The UK's Technology Strategy Board is requiring that all tests costs less than £300 ($450) each, have a clinically relevant turnaround time, and must screen for at least 22 specified, clinically actionable mutations in nine known cancer genes.

Meanwhile, in Canada, the Ontario Institute of Cancer Research is testing a 19-gene sequencing panel on the Pacific Biosciences RS machine on patient samples with metastatic or recurrent disease (CSN 8/3/2011).

The aim of the pilot study is to make targeted sequencing of cancer patients the "standard of care," said John McPherson, director of OICR's Cancer Genomics Program.

"Personalized cancer medicine is where oncology is heading and it's certainly a pressing need," Wagle added.

Have topics you'd like to see covered by Clinical Sequencing News? Contact the editor at mheger [at] genomeweb [.] com.

The Scan

Study Links Genetic Risk for ADHD With Alzheimer's Disease

A higher polygenic risk score for attention-deficit/hyperactivity disorder is also linked to cognitive decline and Alzheimer's disease, a new study in Molecular Psychiatry finds.

Study Offers Insights Into Role of Structural Variants in Cancer

A new study in Nature using cell lines shows that structural variants can enable oncogene activation.

Computer Model Uses Genetics, Health Data to Predict Mental Disorders

A new model in JAMA Psychiatry finds combining genetic and health record data can predict a mental disorder diagnosis before one is made clinically.

Study Tracks Off-Target Gene Edits Linked to Epigenetic Features

Using machine learning, researchers characterize in BMC Genomics the potential off-target effects of 19 computed or experimentally determined epigenetic features during CRISPR-Cas9 editing.