Genomic Health, Taking a Hit on Bottom Line, Bets Next-Gen Sequencing Will Yield Next-Gen Diagnostics
Genomic Health built itself into a profitable business in molecular diagnostics by using genetics analysis instruments that were state-of-the-art in the middle part of the last decade. But as the Redwood City, CA-based diagnostics company (NASDAQ: GHDX) looks further into the future, it has come to a clear conclusion.
Out with the old, and in with the new.
Genomic Health is looking for a way to rekindle its rapid growth of the past few years, and betting on new technology is a big part of the strategy. The company said last week it generated $178 million in sales in 2010, and achieved its first full-year profit since it was founded in 2000. But investors didn’t see much to cheer about here, because revenue grew only 19 percent over the prior year, and the company forecasts are that growth will level off at around 12 percent to 18 percent growth next year. Wall Street sent the stock down 5 percent the next day.
Certainly, Genomic Health has short-term things it can do to maximize its bread-and-butter Oncotype DX test by expanding into international markets, and by branching beyond breast cancer into colon and prostate tumor tests. But as those products run their course over time, Genomic Health wants to be in position to roll out new tests based on the breakneck pace of innovation of the past couple years in fast/cheap gene sequencing. So it’s investing in new diagnostic tests will look for patterns in a lot more genes than today’s molecular diagnostics can, and will make discovery of new tests cheaper, says founder and executive chairman Randy Scott. The spending is a drag on Genomic Health’s financial performance today, but the bet is it will pay big dividends further in the future.
“We are heavily invested in R&D programs, like next-gen sequencing, that are going to impact our bottom line for the next five years,” Scott says. “But we think it will help us build this eventually into a multi-billion dollar business over the course of the next 10 years.”
You need to dive a little bit into the history of technology at Genomic Health to understand what’s going on here. The company developed its first Oncotype DX test based on an instrument known as a real-time PCR machine, which uses the TaqMan chemistry sold by Life Technologies’ Applied Biosystems unit. The instrument, which amplifies DNA samples for analysis, was used to help discover the original 21-gene signature that Genomic Health’s Oncotype DX uses to predict the chance a woman with breast cancer will relapse, and whether she will benefit from a round of preventive chemotherapy. The RT-PCR platform was the best choice at the time, Scott says, because it had “really exquisite sensitivity, reproducibility, and precision.” The machine has been so good on those parameters, Genomic Health continues to use it for running the commercial tests it performs today.
But the more genes you want to analyze, the more expensive real-time PCR becomes. To recoup its sizable investment, Genomic Health to set a price of $4,000 for its one-time test, which has forced the company to run a lot of health economic analyses to justify the price. And as the company began to reach the limits of PCR technology, it started looking for a replacement platform, Scott says.
Given that several companies are now talking about sequencing entire genomes for $10,000 today, and potentially $1,000 or less in the not so distant future, it shouldn’t be a surprise that a company like Genomic Health is particularly interested. Over the past year, Genomic Health has been switching over its discovery research work to the new fast/cheap sequencing machines from San Diego-based Illumina, Carlsbad, CA-based Life Technologies and its interesting new Ion Torrent machine, and a new instrument in the works from Menlo Park, CA-based Pacific Biosciences. Different machines are good at different tasks, and Genomic Health is testing them all out, Scott says. “We’re agnostic on the technology,” he says.
There are a number of reasons to make the switch, Scott says. The company often gets tiny tissue samples to work with, from, say, prostate biopsies, and wants to analyze as much of the genetics of that sample as possible in a single run of a machine, he says. Looking at a much broader swath of the genome, including parts that scientists previously thought didn’t do anything important, actually yields new insights based on correlations between genetics and the clinical state of the tumor, Scott says. And when sequencing is dirt cheap—on pace to become essentially free over time, he says—that enables scientists to perform repeated sequencing runs over hundreds of genes to basically ask multiple variations of questions that couldn’t be asked in a cost-effective way before.
“It’s an incredibly exciting time,” Scott says. “Every time you think you’ve seen it all in DNA sequencing, somebody shows you a new technology that takes it a leap forward. It’s amazing how fast the technology is moving.” He adds: “The Moore’s Law effect is going to drive the cost of sequencing into the ground. What it means is we can use it for discovery, and have a much higher probability of actually understanding disease at a molecular level, at a level of detail we never could before.”
All this sequencing might sound exciting to a scientist, but I get asked all the time what we’re actually going to do with this data, so I asked Scott too. Getting a genome sequenced will eventually be the easy part, leaving lots of hard questions for software engineers about how to store, analyze, and visualize this data effectively. Genomic Health is focusing its R&D on these kind of questions, and is working on iPhone, iPad, and Android-based mobile apps to help various consumers of genomic data get meaningful answers from the samples of DNA they sent in, Scott says.
Even when a nifty app shows something happening in a visual way, someone will need to explain in a clear way to physicians, and insurers, Scott says.
That’s the part where Genomic Health is betting its experience with Oncotype DX will really pay off in the future, Scott says. As the sequencing technology moves ahead, Genomic Health hopes to be one of the few companies with a broad and deep network of contacts with all the key constituencies that need to buy into the idea of genomic-based personalized medicine—physicians, patients, and payers. While startup companies might discover novel diagnostic tests using the same instruments as everybody else, they won’t be in as strong a position to commercialize those innovations as a company like Genomic Health, which has been selling its molecular diagnostic since 2004, Scott says.
“We want to be at the customer interface. It’s about how do you translate this information, how do you use it in the context of your healthcare?” Scott says. “While we are best known for developing our own tests, we also see a day when we commercialize tests developed by a wide variety of other groups that may never touch our R&D labs. But they will need an interface to bring that data to the customer.”
Wall Street doesn’t appear to like it, and Scott certainly knows that, but he’s unabashed about saying the company needs to invest now if it wants to prosper later. It will be fascinating to see how long investors are willing to sit tight for this, before getting really antsy about seeing this investment pay off.
“Our plan is actually not to grow profits aggressively over the next two years, but to invest heavily back into the business, and the development pipeline, because frankly we don’t see much competition there,” Scott says. “We think this is the time to be investing before the rest of the world sees what we believe will be a very large opportunity.”