Skip to main content
Premium Trial:

Request an Annual Quote

Startup Aims to Translate Hopkins Team's Cancer Genomics Expertise into Patient Care


Researchers at Johns Hopkins University who helped pioneer cancer genome sequencing have launched a commercial effort intended to translate their experience into clinical care.

Personal Genome Diagnostics, founded in 2010 by Victor Velculescu and Luis Diaz, aims to commercialize a number of cancer genome analysis methods that have been developed at Hopkins over the past several decades. Velculescu, chief scientific officer of PGDx, is director of cancer genetics at the Ludwig Center for Cancer Genetics and Therapeutics at Hopkins; while Diaz, chief medical officer of the company, is director of translational medicine at the Ludwig Center.

Other founders include Ludwig Center Director Bert Vogelstein as well as Hopkins researchers Ken Kinzler, Nick Papadopoulos, and Shibin Zhou. The team has led a number of seminal cancer sequencing projects, including the first effort to apply large-scale sequencing to cancer genomes, one of the first cancer exome sequencing studies, and the discovery of a number of cancer-related genes, including TP53, PIK3CA, APC, IDH1 and IDH2.

Velculescu told Clinical Sequencing News that the 10-person company, headquartered in the Science and Technology Park at Johns Hopkins in Baltimore, is a natural extension of the Hopkins group's research activities.

Several years ago, "we began receiving requests from other researchers, other physicians, collaborators, and then actually patients, family members, and friends, wanting us to do these whole-exome analyses on cancer samples," he said. "We realized that doing this in the laboratory wasn't really the best place to do it, so for that reason we founded Personal Genome Diagnostics."

The goal of the company, he said, "is to translate this history of our group's experience of cancer genetics and our understanding of cancer biology, together with the technology that has now become available, and to ultimately perform these analyses for individual patients."

The fledgling company has reached two commercial milestones in the last several weeks. First, it gained CLIA certification for cancer exome sequencing using the HiSeq 2000. In addition, it secured exclusive licensing rights from Hopkins for a technology called digital karyotyping, developed by Velculescu and colleagues to analyze copy number changes in cancer genomes.

PGDx offers a comprehensive cancer genome analysis service that combines exome sequencing with digital karyotyping, which isolates short sequence tags from specific genomic loci in order to identify chromosomal changes as well as amplifications and deletions.

The company sequences tumor-normal pairs and promises a turnaround time of six to 10 weeks, though Velculescu said that ongoing improvements in sequencing technology and the team's analysis methods promise to reduce that time "significantly." It is currently seeing turnaround times of under a month.

To date, the company has focused solely on the research market. Customers have included pharmaceutical and biotech companies, individual clinicians and researchers, and contract research organizations, while the scale of these projects has ranged from individual patients to thousands of exomes for clinical trials.

While the company performs its own sequencing for smaller projects, it relies on third-party service providers for larger studies.

PGDx specializes in all aspects of cancer genome analyses, but has a particular focus on the front and back end of the workflow, Velculescu said, including "library construction, pathologic review of the samples, dissection of tumor samples to enrich tumor purity, next generation sequencing, identification of tumor-specific alterations, and linking of these data to clinical and biologic information about human cancer."

The sequencing step in the middle, however, "is really almost becoming a commodity," he noted. "Although we've done it in house, we typically do outsource it and that allows us to scale with the size of these projects."

He said that PGDx typically works with "a number of very high-quality sequence partners to do that part of it," but he declined to disclose these partners.

On the front end, PGDx has developed "a variety of techniques that we've licensed and optimized from Hopkins that have allowed us to improve extraction of DNA from both frozen tissue and [formalin-fixed, paraffin-embedded] tissue, even at very small quantities," Diaz said. The team has also developed methods "to maximize our ability to construct libraries, capture, and then perform exomic sequencing with digital karyotyping."

Once the sequence data is in hand, "we have a pipeline that takes that information and deciphers the changes that are most likely to be related to the cancer and its genetic make-up," he said. "That's not trivial. It requires inspection by an experienced cancer geneticist."

While the firm is working on automating the analysis, "it's not something that is entirely automatable at this time and therefore cannot be commoditized," Diaz said.

The firm issues a report for its customers that "provides information not only on the actual sequence changes which are of high quality, but what these changes are likely to do," Velculescu said, including "information about diagnosis, prognosis, therapeutic targeting [information] or predictive information about the therapy, and clinical trials."

So far, the company has relied primarily on word of mouth to raise awareness of its offerings. "We've literally been swamped with requests from people who just know us," Velculescu said. "I think one of the major reasons people have been coming to us for either these small or very large contracts is that people are getting this type of NGS data and they don't know what to do with it — whether it's a researcher who doesn't have a lot of experience in cancer or a clinician who hasn't seen this type of data before."

While there's currently "a wealth in the ability to get data, there's an inadequacy in being able to understand and interpret the data," he said.

Pricing for the company's services is on a case-by-case basis, but Diaz estimated that retail costs are currently between $5,000 and $10,000 per tumor-normal pair for research purposes. Clinical cases are more costly because the depth of coverage is deeper and additional analyses are required, as well as a physician interpretation.

A Cautious Approach

While the company's ultimate goal is to help oncologists use genomic information to inform treatment for their patients, PGDx is "proceeding cautiously" in that direction, Diaz said.

The firm has so far sequenced around 50 tumor-normal pairs for individual patients, but these have been for "informational purposes," he said, stressing that the company believes the field of cancer genomics is still in the "discovery" phase.

"I think we're really at the beginning of the genomic revolution in cancer," Diaz said. "We are partnering with pharma, with researchers, and with certain clinicians to start bringing this forward — not only as a discovery tool but eventually as a clinical application."

"We do think that rushing into this right now is too soon, but we are building the infrastructure — for example our recent CLIA approval for cancer genome analyses — to do that," he added.

This cautious approach sets the firm apart from some competitors, including Foundation Medicine, which is about to launch a targeted sequencing test that it is marketing as a diagnostic aid to help physicians tailor therapy for their patients. Diagnostic firm Asuragen is also offering cancer sequencing services based on a targeted approach (CSN 1/12/12), as are a number of academic labs.

Diaz said that PGDx's comprehensive approach also sets it apart from these groups. "We think there's a lot of clinically actionable information in the genome … and we don’t want to limit ourselves by just looking at a set of genes and saying that these may or may not have importance."

While the genes in targeted panels "may have some data surrounding them with regard to prognosis, or in relation to a therapy, that's really only a small part of the story when it comes to the patient's cancer," Diaz said.

"That's why we would like to remain the company that looks at the entire cancer genome in a comprehensive fashion, because we don't know enough yet to break it down to a few genes," he said.

The company's proprietary use of digital karyotyping to find copy number alterations is another differentiator, Velculescu said, because many cancer-associated genes — such as p16, EGFR, MYC, and HER2/neu — are only affected by copy number changes, not point mutations.

Ultimately, "we want to develop something that has value for the clinician," Diaz said. "A clinician currently sees 20 to 30 patients a day and may have only a few minutes to look at a report. If [information from sequencing] doesn't have immediate high-impact value, it's going to be very hard to justify its use down the road."

He added that the company is "thinking very hard about what we can squeeze out of the cancer genome to provide that high-impact clinical value — something that isn't just going to improve the outcome of patients by a few months or weeks, but actually change the outlook of that patient substantially."

The Scan

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.

Coronary Artery Disease Risk Loci, Candidate Genes Identified in GWAS Meta-Analysis

A GWAS in Nature Genetics of nearly 1.4 million coronary artery disease cases and controls focused in on more than 200 candidate causal genes, including the cell motility-related myosin gene MYO9B.

Multiple Sclerosis Contributors Found in Proteome-Wide Association Study

With a combination of genome-wide association and brain proteome data, researchers in the Annals of Clinical and Translational Neurology tracked down dozens of potential multiple sclerosis risk proteins.

Quality Improvement Study Compares Molecular Tumor Boards, Central Consensus Recommendations

With 50 simulated cancer cases, researchers in JAMA Network Open compared molecular tumor board recommendations with central consensus plans at a dozen centers in Japan.