By Julia Karow
Complete Genomics expects prices for its human whole-genome sequencing services to decrease less rapidly in 2012 than in prior years and to reduce its costs and increase its capacity further.
After pre-announcing in January that it had shipped fewer genomes than expected in the fourth quarter of 2011 (IS 1/10/2012), the company formally reported its earnings last week and provided an update on its plans and expectations for the current year.
Complete shipped around 600 genomes in the fourth quarter and recognized $2.5 million in revenue, or about $4,200 per genome on average, down from $3.8 million in revenue during the prior-year period. Shipments during the fourth quarter were lower than expected because of a delay in the expansion of its sample prep lab.
R&D expenses rose to $8.3 million during the quarter from $5.6 million in Q4 of 2010, and sales and marketing costs more than doubled to $4.3 million from $2 million.
The firm's net loss for the fourth quarter widened to $22.3 million from $10.3 million in the year-ago quarter.
For the entire year, Complete booked $19.3 million in revenue for approximately 3,000 genomes, or $6,400 per genome on average, up from $9.4 million for about 800 genomes in 2010, which translated to almost $12,000 per genome. It delivered genomes to about 90 customers last year, and more than 125 customers in total have used its services so far.
The net loss for 2011 totaled $72.3 million, up from $57.7 million in 2010.
Complete ended the year with $83.1 million in cash, cash equivalents, and short-term investments, and $500,000 in restricted cash. The remaining cash is expected to last the company until the end of the year, based on a forecast cash burn rate of about $20 million per quarter, CFO Ajay Bansal said during a conference call to discuss the earnings. Earlier this week, Complete announced an agreement with investment bank MLV for a potential sale of $30 million of its common stock (GWDN 3/12/2012).
"2011 was both a gratifying and a very challenging year for us," said Complete CEO Cliff Reid during the call.
While the firm took orders for about 8,000 genomes over the course of the year, and several customers published findings from data they received from the company, the price decline for human whole-genome sequencing "was significantly more rapid than we had anticipated," Reid said.
In addition, delays in the expansion of its sample prep lab affected both revenue and shipments at the end of the year. Complete's sample prep is partially automated and partially manual, Reid explained, and turned out to be "much more challenging" to scale up than the firm had thought, though this has now been achieved.
Outlook for 2012
Complete's strategy for 2012 "remains unchanged," Reid said: to ship "the most accurate genomes in the industry" for basic and translational research; to cut its unit cost of sequencing by reducing both its fixed and variable costs; and to enter the clinical market.
Over the last three years, he said, prices for human whole-genome sequencing have come down by a factor of two each year, caused by a "dramatic reduction" of consumables costs.
As those costs have stabilized and now only account for "a relatively small fraction" of the overall cost of sequencing, Complete now expects prices to decline by less than 50 percent per year, though Reid declined to make any precise predictions.
The company started 2012 with orders for 5,800 genomes, mostly from customers in academic and medical research, which represent about $28 million in revenue, translating to an average genome price of $4,800.
While that is more than the $4,200 average price per genome during the fourth quarter of 2011, this does not mean genome prices are going up, Reid said. The order backlog contains a mix of small orders, which are priced higher, and large orders, which are priced lower, and that mix changes from quarter to quarter, he explained. Also, some of the genomes on order have additional services attached to them, for example validation services, which increase the price per genome. Overall, he said, "prices are flat and will come down gently."
Reid acknowledged that in the past, when prices for genome sequencing services were falling "precipitously," the company sometimes renegotiated existing contracts with customers, offering them additional genomes at no further cost. However, as prices begin to stabilize, the firm does not expect price renegotiations to have "a major effect" in 2012.
Customers can also expect a further "stratification of products" that will have different prices, he said. Last year, the company introduced a cancer sequencing service for $12,000 per tumor/normal pair, as well as an 80x version of its standard sequencing service. It sees "additional opportunities" for new products, especially in the clinical market, he said. The company is still evaluating its long fragment read technology in collaborations and plans to launch it commercially for clinical applications in 2013.
Turnaround time is expected to decline further this year, from about 60 days last year. By the end of the year, Complete plans to offer research customers a 30-day to 35-day turnaround time, which Reid said is "very good." Recent announcements by sequencing instrumentation vendors that one-day human genomes will soon be possible, he said, don’t account for the entire process of sample acquisition, sample preparation, sequencing, data analysis, and data validation. "We expect to be able to do all of this, soup to nuts" within 30 to 35 days, he said. For the clinical market, Complete is targeting a two-week turnaround time, which it plans to achieve sometime in 2013, he added.
Last month, Complete Genomics also teamed up with software partners — DNAnexus, Golden Helix, and Ingenuity — whose offerings will complement its own service. Reid explained that Complete provides its customers with variant calls and annotations from public databases along with a set of free basic analysis tools to analyze the data that customers have requested. But he said that it cannot develop many specialty tools and user interfaces and relies on partners to provide these biological and medical analysis tools. The partnership program is "not a revenue generator" for Complete, he said, and more partners will be added in the near future.
Complete is also making "good progress" toward CLIA certification and expects to obtain a CLIA license by the middle of this year, Reid said. Interestingly, not only translational and clinical customers but also research customers have been interested in CLIA sequencing services, he said, because they see this as a "validation of the quality and reproducibility" of the data. "So we think CLIA is not only important to our future as a clinical sequencing company but also somewhat important at least to our research business," he said.
In addition to filling existing orders, the company is working on contracts for new orders, including "multiple projects" that will comprise more than 1,000 genomes each. So far, it has not signed any "giant deals" that would add substantially to the 5,800 genomes on order, but it expects to close and announce several large deals over the course of the year, Reid said.
The company's sales force stood at 20 sales reps as of the end of 2011, with around 15 targeting the research market and the remainder the translational and clinical market. The latter fraction will grow further this year, Reid said, as sequencing services for translational and clinical applications are expected to grow.
Increasing Capacity, Competition
In the first quarter, the company expects to ship between 1,100 and 1,300 genomes, or about 400 genomes per month on average. In addition, it has already received samples for more than 2,000 genomes, which it expects to sequence in the second quarter.
Complete's sequencing capacity will stay ahead of these numbers: by the end of the first quarter, it expects to be able to sequence about 1,000 genomes per month on 24 sequencing instruments.
For the last few months, the firm has been testing a prototype of a new generation of sequencers. It expects to receive the first of six commercial-grade instruments of this type in the second quarter, and to install all six by the end of the third quarter.
These instruments will initially be able to generate a terabase of data per day, equivalent to six human genomes, and are expected to increase the overall sequencing capacity to 2,000 genomes per month by the end of the third quarter.
Reid cautioned, though, that "it's always challenging to bring up a new generation of technology," and promised to keep analysts informed about any "glitches" that may occur during the implementation phase.
He also said that the company does not plan to replace all of its 24 existing instruments with new ones at the moment because that would increase its capacity beyond its current demand for services. "We forecast demand, then put in a capital purchase plan to meet that demand," he said.
Asked about competitors for whole human genome sequencing services, Reid mentioned Illumina and BGI but said that Illumina has been the most important one to date. "In a handful of major accounts, we do compete with Illumina," he said, adding that "if there is a premium on accuracy, we compete very effectively."
He cited a study by researchers from Stanford University last year who compared the two platforms and found that Complete's data is more accurate than Illumina's. However, that publication also found that Complete's platform is less sensitive than Illumina's, potentially missing true variants (IS 12/20/2011).
Have topics you'd like to see covered in In Sequence? Contact the editor at jkarow [at] genomeweb [.] com.