Pacific Biosciences represents one of the big dreams in genomics of the past decade. It rallied big money and prominent people around a one-of-a-kind sequencing technology. The Menlo Park, CA-based company (NASDAQ: PACB) ended up reaching the market, and has helped researchers make some high-profile discoveries, like the origin of the 2010 cholera outbreak in Haiti.
But in its first couple years as a public company, the numbers tell a business story scary enough to send shivers down the spine of the average investor.
Consider PacBio’s profile from two years ago: The company went public amid great fanfare, buoyed by an excited base of researchers that eagerly awaited its $700,000 machine. It raised $200 million, and debuted on the NASDAQ with a market valuation of $800 million. It delivered its first product to the market about nine months later, in late April 2011.
Now look at the picture two years later: PacBio has now burned through more than $500 million of investors’ money through the end of most recent quarter. The company generated $2.8 million in revenue in the three-month period that ended Sept. 30—a 73 percent decline from the previous quarter. The once-loaded company expects to burn through roughly $80 million in cash this year, and end the year with about $100 million of cash left in the bank. It has an installed base of 70 machines around the world, and its once-fat backlog of orders now has just five customers on it, as of its last quarterly report on Oct. 25. Worst of all, this one-time highflier has a market capitalization of less than $70 million. Not only does that mean PacBio’s early investors have lost their shirts, it says current investors essentially think the technology is worthless.
To be sure, PacBio isn’t the only company that has struggled to get a foothold in the fast-moving genome sequencing world. While scientists are excited about the research that’s now enabled by fast/cheap sequencing with instruments from Illumina (NASDAQ: ILMN), Life Technologies (NASDAQ: LIFE) and PacBio, many have seen their purchasing power weaken because of cutbacks and future uncertainty in federal research funding. Tough new competition from Oxford Nanopore and Genia Technologies is threatening to make many of today’s genomic instruments obsolete.
Tycho Peterson, a genomics industry analyst with JP Morgan, rates PacBio stock “underweight” which is one way of essentially saying beware. After the company’s earnings report on Oct. 25, he wrote in a note to clients that “we remain cautious on PACB given limited visibility on demand and more importantly continuing cash burn. We reiterate our Underweight rating, pending dramatic improvements to the PacBio RS platform or a material shift in the strategy.” When I checked with a few good genomics researchers around the world and asked them about the PacBio machine, several said they couldn’t comment because they don’t use it.
Grim as the situation may look from the outside, new PacBio CEO Mike Hunkapiller gives off the sense that he’s seen this movie a couple times before, it’s no big surprise, and all his company needs to do is work hard on executing behind the scenes to get things going in the right direction. Hunkapiller, a gene sequencing pioneer who started working in the field at Caltech and Applied Biosystems in the 1980s, is known for his understated, just-the-facts style. His body language, on a visit to his office Oct. 15, practically screamed, “I’m not a hypemaster.” And while he did nothing to hype PacBio’s technology, he insists he still believes in it, and it just needs a few adjustments to get the company back on track.
When I joked that maybe it’s time to prepare the PacBio obituary, he laughed. “It’s a little early,” he said, before offering up a little historical perspective. “I remember back in the original ABI (Applied Biosystem) sequencing days, about the second year in, we were doing our planning, and the sales force said ‘Well, we’ve now identified all the high-throughput sequencing labs that are ever going to need an automated sequencer.’ That was about 1987. At that point, we had sold maybe 120 units at the time. We ended up selling 200 more than year. And we got up to a couple thousand in the glory days of the original genome work.”
The point of this little anecdote, Hunkapiller says, is that the technology in this business improves fast, the market demand is hard to predict, and essentially all genome technologies are evaluated at first on their promise before the reality of their limitations dawns on people. “The technologies, if they get good enough, and really address scientific problems that people want to address and they can’t get elsewhere, then they wind up getting funded. You just have to get it to that point,” Hunkapiller says. Essentially, it means hunker down in the lean times, conserve cash, and make the machines better. “It’s the only thing you can do,” Hunkapiller says.
Hunkapiller has a lot on the line in this bid to turn around PacBio. He was one of the early venture investors with Alloy Ventures who bet on the company, and lent it a lot of important scientific credibility in its early days. When PacBio’s early marketing stumbles became clear in October 2011, he stepped up from his board seat to take a more active role as executive chairman. He then took over from Hugh Martin as full-time CEO in January.
Hunkapiller’s turnaround plan is based heavily on a lot of behind the scenes execution. The machines that were delivered to customers in the first year had hardware and software bugs. Many of the new machines delivered to customers were spending a lot of time sitting around, not being used. The first version of the microchips used in the machines weren’t being manufactured consistently, causing problems for customers, and prompting PacBio to find a new supplier. Some employees were in need of training. Many customers needed help getting trained on how to use the machines efficiently. And word was getting around the industry about the snags the early adopters experienced, which prompted many other customers to wait for improvements before shelling out $700,000 for a hot new toy.
“We had to go back and really focus on robustness, reliability, training, teaching people how to do sample prep, and how to teach people to deal with an informatics data set that was different than the ones they were used to dealing with,” Hunkapiller says.
So PacBio focused on essentially making its existing customers a bit happier in the first half of 2012. For a couple months, it deprioritized efforts to secure new customers. The hope was that by solving some of the early problems, the company would have a more compelling pitch. “We wanted to lay the groundwork for expanded sales in the second half of the year,” Hunkapiller says. “You never perfect the technology internally. It really requires going through the crucible of fire out there in the real world.”
As it has from the start, PacBio is still attempting to differentiate itself from its competitors on its ability to analyze “long reads” or long stretches of DNA. Today’s standard technologies assemble genomes by essentially stitching together many short stretches of DNA. That approach enables the machines to process high volumes of DNA, improving speed and lowering cost, but it isn’t very good at spotting oddities in certain genomes like strange repeat sections in, certain genomes like that of rice. The PacBio instrument, by looking at longer and longer reads of DNA, is supposed to be able to better capture some of this information, Hunkapiller says.
Of course, PacBio needs people to get excited about a lot more than just using its machine for a few plant or bacterial genomes. Hunkapiller says he sees room for growth in plants and bacteria, but also in human sequencing, especially for complex targeted regions where it might be an advantage to look at longer stretches of DNA.
Hunkapiller isn’t going on the record with any forecasts for how many instruments PacBio might sell in the year ahead. The company announced on Oct. 25 it has a backlog of five orders. If Congress and the President can’t agree on a new round of budget talks, the National Institutes of Health could suffer major cuts as part of the “sequestration” showdown, which would greatly impact PacBio’s academic customer base in 2013.
That’s probably enough to make any sequencing company executive nervous, but the possibility has to be especially chilling for PacBio. The company has about 18 months of cash left in the bank, although it has filed paperwork with regulators to raise another $30 million in periodic stock offerings.
When I asked Hunkapiller if PacBio is headed down the same rocky road that led Complete Genomics to get acquired by BGI-Shenzhen, he insisted he has no plans to throw in the towel.
“We’re not seeking strategic alternatives,” Hunkapiller says. “We’ve still got a fair amount of cash in the bank. The cash takes us, at our current burn rate, through the middle of 2014. We are a public company with the ability to raise additional capital if we need it. We’ve got a substantial and growing installed base of customers that can do science with our technology that they can’t do otherwise. We’re still unique in the long-read space. As the technology gets better from a throughput perspective, we start encroaching on other technologies. We have to execute, but we are pretty confident we can do that.”