By Monica Heger
This story was originally published Jan. 9.
Knome has hired a new CEO as part of its strategy to expand its whole-genome sequence interpretation services into the clinic.
Additionally, the company also announced this week that Johns Hopkins University has selected it to provide interpretation for 1,000 whole genomes as part of a study on asthma.
New CEO Martin Tolar has a background in the biopharmaceutical industry and has held senior positions at Pfizer, as well as biotech firms Normoxys, and CoMentis. He told Clinical Sequencing News that providing genome interpretation services for pharma and biotech companies would be the next market for Knome.
The Johns Hopkins contract will help demonstrate to pharmaceutical companies that Knome has the capacity to handle the volume of patients that would be part of a clinical drug trial, said Tolar.
For the asthma genome project, Knome will use its KnomeDiscovery tool, which includes not only variant calling, but full interpretation of the proteins and pathways involved in disease. Knome had to scale its technology in order to handle the large amounts of sequence data that the study will generate, Tolar said. He anticipated that the interpretation of all 1,000 genomes would take around one year.
Additionally, Tolar said the company has already started collaborating with some undisclosed pharma companies and would soon be announcing "new hires to help execute a strategy for the pharma, biotech, and medical" markets.
Currently, Knome's main business is academic collaborations, which will generate the bulk of the company's anticipated $20 million in revenue this year. However, the academic market is becoming a "mature" market for the firm, Tolar said. The next phase is to expand to pharmaceutical and biotech companies, which represents a $1 billion market, and eventually to medical institutions looking to use whole-genome sequencing for diagnosis, disease risk prediction, and ultimately to guide treatment, which would be an even bigger market, he said.
Pharma and Biotech Market
There are two places in the drug-development process where Knome's services could be valuable to pharma and biotech companies, Tolar said.
The first is in the transition between phase II and phase III clinical trials. Phase II trials usually involve between 50 and 100 patients and cost between $40 million and $50 million. Moving into phase III, however, requires thousands of patients and hundreds of millions of dollars, said Tolar, adding that companies at this stage are "making a huge bet."
As a result, "if you can improve the risk profile at that point, and say, 'These are the appropriate patients [to include in the phase III trial] because the drug is going to work better, or is going to have less side effects,' that's a huge value added," he said.
The second place where sequencing and interpretation is important is when the drug is in late stages of development or already on the market and has been shown to have problems like adverse side effects in patients. Such side effects often have a genetic basis, Tolar said.
Some drugs currently on the market "could be much, much larger products if you could understand the side effects," he added.
Despite Tolar's optimism, some pharmaceutical companies have doubts about the value of next-gen sequencing for clinical trials. For example, Dominic Spinella, Pfizer's head of translational and molecular medicine, told CSN last year that if sequencing was incorporated after a trial started, it was too late (CSN 6/21/2011).
According to Spinella, sequencing patients after a phase II trial is useful for generating a hypothesis about which genetic markers are predictive for response, but those markers must be tested before they are used to select patients for a phase III trial. Verifying that hypothesis takes time, which can eat away at the patent life of the drug. Alternatively, if the phase III trial moves ahead without the markers and fails, a company is not likely to re-do the trial with the markers. Even if the trial succeeds, a company would be reluctant to incorporate markers that would decrease its potential patient market, he said.
Tolar argued, however, that both the cost of sequencing and time it takes to sequence and interpret genomes is declining, and that introducing sequencing after a phase II trial could save drug companies significant amounts of money.
For instance, he said when he was working at Pfizer, the company spent more than $1 billion developing a drug that would have replaced Lipitor, its blockbuster cholesterol-lowering drug. However, it "blew up at the last minute because there was a side effect that increased mortality." At the time, there was no way to recover from the trial, but if Pfizer could have predicted which patients would experience the side effect, the drug could have moved forward, he said.
Additionally, said Tolar, interpreting the genomic data has proven to be a major hurdle for pharmaceutical companies that would otherwise adopt the technology. And, since interpretation is where Knome's expertise lies, Tolar thinks the service will be attractive to pharma and biotech.
Knome currently offers two interpretation products: KnomeBase, priced at $750 per genome or exome, and KnomeDiscovery, which starts at $1,500 per exome and $2,000 per genome. Sequencing can be provided at an additional cost and Knome contracts sequencing services out to BGI, Illumina, Complete Genomics, and PerkinElmer. In the case where the client does its own sequencing, Knome's analysis is compatible with all major sequencing platforms.
KnomeBase is an automated tool for variant calling geared toward institutions that do not have bioinformatics capabilities but have experience in interpreting genomic data. KnomeDiscovery, meantime, is a "full solution provider," Tolar said.
This offering is targeted at pharmaceutical companies who do not have genomics expertise and want to understand the impact of a particular drug, for instance. This solution not only calls variants, but also provides protein and pathway analysis, Tolar said.
Medical Institution Market
Tolar said that Knome has also identified an opportunity among medical institutions that want to start offering clinical whole-genome sequencing.
He said that while he does not expect this market to take off for another several years, he's been "surprised" by the early interest that some of the "big-name medical institutions" have shown in this area.
These institutions are initially looking to do whole-genome sequencing for clinical diagnosis, particularly for unknown diseases, pre-pregnancy carrier screening, and prenatal health, he said.
Close behind those applications would be whole-genome sequencing of healthy patients for disease risk prediction and whole-genome sequencing to identify effective therapies.
Some companies and universities are already developing sequencing tests for cancer patients to guide therapy, but one main problem, said Tolar, is that there are not that many available drugs that target specific genetic mutations.
One issue the company will likely face in the near future is how the US Food and Drug Administration regulates next-gen sequencing. Currently, sequencing-based tests are mainly offered as lab-developed tests in CLIA-certified and CAP-accredited environments, which the FDA does not regulate. These tests are mainly targeted sequencing tests, with only a handful of institutions currently offering clinical whole-genome sequencing.
Tolar said that he has already had discussions with the FDA about potential guidelines the agency might implement for clinical genome interpretation. Tolar expects that not only will the sequencing instruments themselves need to be certified, but so too will the analysis pipeline.
This would involve showing all the steps involved in the interpretation process — from generating the sequence data to the specific prediction. Additionally, it would require some metric of reliability and measure of confidence in the result.
The information will have to be "traceable back [to the sequence] for each patient," he said.
Competition
Knome is not the only genome interpretation company looking to expand into the clinical market. In fact, as sister publication BioInform recently reported, a number of informatics vendors are vying for the edge in sequence interpretation for clinical purposes.
For example, Cypher Genomics, a spinout from Scripps Translational Science Institute and the Scripps Research Institute, will provide clinical annotation for 1,000 genomes being sequenced by Complete Genomics as part of the Scripps Health System's Wellderly Project (CSN 10/5/2011)
Additionally, informatics firm GenomeQuest markets a tool known as GQ-Dx, a clinical decision support system that allows users to interpret data from whole-genome, exome, and targeted sequencing studies for clinical applications.
Omicia this year developed an algorithm dubbed VAAST, for Variant Annotation, Analysis, and Search Tool, and is planning a platform for clinical annotation of sequence data called the Genome Analysis System (CSN 6/29/2011).
And Personalis, a startup formed from Stanford Researchers, plans to focus on medical interpretation of genomes initially for research, and eventually for clinical applications (CSN 9/28/2011).
Have topics you'd like to see covered by Clinical Sequencing News? Contact the editor at mheger [at] genomeweb [.] com.