Up until now, the money Illumina (ILMN 0.89%), Roche, Life Technologies (NASDAQ: LIFE), and Pacific Biosciences of California (PACB 7.88%) have made selling DNA sequencers has come from research labs. Academics need DNA sequencers to do basic research to understand how genetic variation affects biologic processes.
That basic research has translated into the clinic at an alarming rate, producing the next generation of DNA sequencing demand.
You can see it in the acquisitions
Roche's bid for Illumina appeared to be mostly driven by bringing sequencing to the clinic. Roche has a strong hold in diagnostic testing, and Illumina's sequencing technology is superior to Roche's. But Roche played hardball and didn't want to overpay for the technology.
Last month, Illumina decided it could fill some of the gap on its own, purchasing BlueGenome, a leader in cytogenetics. The company sells tests that look at the DNA to identify genetic abnormalities that lead to cancer and other issues. Currently, those abnormalities are identified by binding probes to the DNA to identify duplications and fusion of chromosomes, but we're not too far off from where cancer patients just routinely get their DNA sequenced to identify the abnormalities.
Blue Genome also has a test to look for abnormalities before in vitro fertilization. Sequencing might be harder there, because of less DNA, but these issues are often overcome eventually.
It sure looks to me like Illumina bought BlueGenome more for its ability to sell and run the clinical test than for the test themselves, which may be obsolete in a few years given the rapid decrease in the cost of sequencing. We will get to the point where running individual tests like Sequenom's (NASDAQ: SQNM) MaterniT21 PLUS will be silly because the entire genome can be analyzed for the same cost.
Diagnosing
Earlier this month, Life Technologies hooked up with CollabRx (NASDAQ: CLRX) to use the company's interpretive analytics to help develop tests for cancer diagnostics. CollabRx combines a patient's data from multiple sources to help the doctor develop a treatment plan. The genotype of a tumor tells you a lot about what drugs might be able to kill the tumor, but the genetic variation still needs to be taken in context with other pieces of information.
Getting the sequence and knowing what to do with it are two different things. One startup sequencing company, Knome, has begun selling a $125,000 supercomputer, so hospitals can analyze patients' DNA sequences directly. Having the process in-house might speed up diagnosis, but the appeal seems to also be about avoiding confidentiality issues for the patients when the data is shared externally. Either way, the fact that hospitals are buying the station is a sign that they see a future in using patients DNA sequences when diagnosing patients.
On the cusp (still)
We've been on the verge of pushing genomics into diagnostics for years. Illumina added a new division for diagnostics back in 2008. Four years later, we're further along, but we're still a ways away from where DNA sequencing is a mainstream test done by most doctors.
The limiting factor is a combination of usefulness and cost, both of which are headed in the right direction. We're learning more about how genetic variations affect patients' physiologies. And the discoveries are accelerated as more people get their genomes sequenced.
Costs are coming down, which will drive adoption, but I worry it'll also put pressure on margins. Complete Genomics (NASDAQ: GNOM), a service company that sequenced genomes for researchers, couldn't find a way to make a profit and eventually sold out to China-based BGI for a song. The pricing pressure by the service organizations will eventually work its way up to the companies making the sequencers, especially if there's no perceived advantage of one over the other.
Initially, companies should be able to make up for the low profits with high volume. There are an awful lot of people who haven't had their genomes sequenced. And there's the aforementioned cancer and prenatal diagnostics.
But the growth won't last forever, because genome sequencing isn't like a cholesterol test that has to be done at regular intervals. Get sequenced as a baby (maybe in utero), and you might not have to be sequenced again unless you develop cancer. At some point, the rate of growth will match the birth rate.
