Over the last few years, next-generation sequencing company Illumina (San Diego) has made a concerted effort to diversify. With a growing pipeline of diagnostics products and two strategic acquisitions (in 2012 BlueGnome, a maker of cytogenetics and in vitro fertilization screening products and this January noninvasive prenatal diagnostics maker Verinata Health), the company is moving away from being strictly an instrument manufacturer, even while its core business remains strong with 2012 revenue expected to be between $1.134 billion and $1.144 billion. Gregory Heath, Ph.D., general manager of diagnostics at Illumina, recently spoke to DTTR about the company’s strategic vision for its diagnostics segment and plans for the division’s growth. Please tell us about Illumina’s strategic decision to diversify and enter the diagnostics market. A little over four years ago Illumina decided they were quite successful in the life science business so they wanted to diversify into other areas, and molecular diagnostics is one of the biggest areas. I joined from Roche, and we started building up the diagnostics business. One of the first things we did was look at strategy and selected a couple of areas where we thought our technology would really be a good fit. Our technologies are high-density microarrays and NGS [next-generation sequencing]. The four areas clinically we have settled on are genetics, with a particular emphasis on reproductive genetics, cancer, infectious diseases, and transplantation. What does your pipeline of diagnostic products look like in each of these selected clinical areas? For the short term the areas we are focusing on are genetic disease and cancer. Within reproductive genetics we recently acquired a company called BlueGnome, which works in the IVF space with preimplantation genetic diagnosis and preimplantation genetic screening. We have a cystic fibrosis test in development, which we are getting ready to submit to the U.S. Food and Drug Administration, and both Illumina and BlueGnome have cytogenetics products that we plan to submit after that. That is an array-based platform, not based on NGS, but it has certain characteristics that are well suited to the cytogenetics market. Within the cancer space we have focused on a couple of things in the short term. We are working with a number of labs on somatic mutations panels. We have also been in discussion with a number of companies for companion diagnostics. We think cancer is a big opportunity in the longer term. In infectious diseases, the technology has really moved fast but it probably needs to be a little cheaper and a little faster. But the coverage is better than what we typically see with hotspot analysis that is done today. . . . For HIV, there I think we have a different approach where the technology will allow you to circumvent hotspot testing. If somebody’s viral load is rising because the virus has mutated and you want to know exactly what that mutation is, you can sequence the whole viral genome. Those applications will come as the technology gets cheaper and faster. The last area we see as an opportunity is transplantation, particularly HLA-typing, looking at bone marrow transplants as well as solid organ. There again, I think we need to be a little cheaper on the blood marrow side and a little faster on the solid organ side, but that is coming quickly. Will the diagnostics division pursue additional acquisitions to drive growth? The business is young and growing. We see growth coming from both [internal development and acquisitions]. If we see a good acquisition that makes sense for the company we are always interested. It has to be a good fit with our technology and in our areas of interest (genetic disease, cancer, infectious disease, and transplantation) where we believe sequencing will play a significant role going forward. BlueGnome made a lot of sense because they had a good software platform, BlueFuse Software, and their IVF side of the business was growing rapidly and it was very compelling. What clinical applications are most likely to routinely adopt NGS-based testing in the near term? Cancer is an area where we have great alignment with the performance characteristics of the platform and the need in the market. For example, by deep sequencing you can tease out the heterogeneity of the tumors. In our CLIA lab when we look at applications, what we tend to see is people who are cancer cases at the end of standard of care—that they’ve exhausted their options and are looking for some alternative, maybe even an experimental protocol they can be placed on. We can add value there. In genetics we see pediatric and adult cases where typically people will go on this diagnostic odyssey where physicians will test them for one thing—kind of piecemeal. When someone comes to our lab they get their whole genome sequenced and the physicians can hypothesis test in silico. They can look at that first gene, immediately discover they were wrong, and move on to the second gene without sending the patient back for additional samples or testing. How will clinical use of NGS-based testing evolve in the next five years? Illumina has a good vantage point because we have such a large footprint on the research side. Everything that is going to wind up in the clinic is going to start with research. Something like 90 percent of all bases sequenced are done on Illumina platforms. We partner with a lot of people and we can see which changes are emerging. Cancer will be an area that in the not-too-distant future there will be many more applications. Pharma is investing heavily in oncology. There are a limited number of drugs available for any particular cancer today, and there are only a handful of genes that currently drive those treatment decisions. I think with whole-genome sequencing we are going to see many, many more markers come to the forefront and, hopefully, many more therapeutics. As the therapeutics proliferate the number of choices physicians have to make goes up, and to inform those choices we think NGS will play a role. Genetic disease testing will continue to proliferate. This is the right technology to do genetic diagnosis, particularly around reproductive genetics. We saw the emergence of noninvasive prenatal testing recently. We are seeing a lot of people on the research front working on immunosequencing, microbiome, and methylation, primarily on arrays. All of those will emerge, and we will see short-term breakthroughs in genetics and cancer and along the way technically infectious disease and transplantation will be possible, but [the technology] probably needs to get a little cheaper and faster to be economically feasible. . . . I see NGS as being applied broadly in the next three to five years. Reimbursement and regulatory issues are frequently cited as challenges diagnostics manufacturers must overcome for successful market penetration. How are these issues affecting NGS-testing? Health care costs are going up. But our costs are going down so we think we are creating a lot of value for the health care system. Within reimbursement I think it depends a little bit on whether there are existing codes and this is a technological substitution for an existing approach, in which case reimbursement may be in place already. Or, if this is a new-to-world test, reimbursement has to be established. We see out-of-pocket pay for these new-to-the-world tests. For a pharmacogenomic example, if somebody has breast cancer the choice may be aromatase inhibitors, which cost $30,000 for a course of therapy, versus tamoxifen, which may be $7,000 for a course of therapy. If you happen to be a poor metabolizer for tamoxifen, you won’t be able to convert it to an active state. You will be on chemotherapy, have all the side effects, but none of the benefits. If I were a physician I would want to know which patients should be on the aromatase inhibitors. If I were a patient, I would definitely want to know that, and if I were a payer there is an economic incentive to look at it. You could easily cover the cost of even a several-thousand-dollar test with that price difference. Regulatory issues will be a challenge, but there are nuances there too. The FDA wants to be considered innovative and bring innovative technology through, but part of the challenge with NGS is, what’s the gold standard you compare to? If you compare it to Sanger sequencing, it is almost like comparing your high-definition TV to a black and white TV and saying it is just as good. It is not really a good comparison. There is also the issue that with high resolution you can see a lot of things that are real changes. Analytically you know the mutations are there, but they may lack clinical significance. So what do you do with those? Clinical science will take some time to catch up and that is what drives physician adoption. But for a lot of physicians today, if their patient is out of options, they may look at NGS for those cases.

Associated Data