2018 was undoubtedly one of the most important years ever for biopharma, setting records for valuations and dealmaking. But a paper published in January called some of that exuberance into question, noting that many startups over the years have managed to become unicorns – with valuations of $1 billion or more – despite having little or no published, peer-reviewed data.

Could the emperor have no clothes?

Venture capitalists seem to believe the opposite. They said the high valuations that biopharma startups in particular have garnered in recent years reflect not so much undue exuberance, but factors like more money being available, strong due diligence and cultural cross-pollination between tech and life sciences investing.

“It’s important to remember that it only takes one thing for a company to achieve unicorn status, namely a venture fund willing to place that price tag on it,” said Mike Pellini, former CEO of diagnostics maker Foundation Medicine and managing partner of the venture capital firm Section 32, in a phone interview. “So I think it’s difficult to generalize.”

Biotech investing, particularly in startups, hit multiple records in 2018. According to the Silicon Valley Bank’s annual report, issued in January of this year, biopharma exits – including initial public offerings and mergers and acquisitions – reached a $49 billion deal value, including 18 deals worth $1 billion or more. Healthcare venture fundraising reached $9.6 billion. Biopharma fundraising was double what it was the year before and four times that of devices, diagnostics and tools. Last year also saw the biggest-ever biotech IPO, with Moderna Therapeutics’ pre-IPO valuation reaching nearly $7 billion, according to data provided by PitchBook.

But the paper, which called into question biotech valuations, published in the European Journal of Clinical Investigation, looked at the unicorn phenomenon from a different perspective.

Led by Stanford University School of Medicine researcher Dr. John Ioannidis, the researchers compared 18 current and 29 exited healthcare unicorns – startups reaching valuations of $1 billion or more – with how many peer-reviewed publications they had. They also looked at how many of those publications reached their threshold for being highly cited, with 50 citations or more, a number that Ioannidis wrote in an email was arbitrary, but would place a paper in the elite tier of academic literature. The study used valuation data from CBInsights through November 2017, alongside publication records from PubMed and citation counts from the Web of Science.

The thinking was that peer review is indispensable for validating innovative products and technologies in medicine, and those not subjected to peer review, but based on internal data generation alone, may be problematic and not trustworthy. Ioannidis calls this “stealth research.”

Of the companies listed by PitchBook as having valuations of $1 billion or more before they exited, three were included in Ioannidis’s paper: Aduro BioTech, FibroGen and Juno Therapeutics. However, Aduro and FibroGen both have several highly cited articles. Meanwhile, none of Juno’s 21 publications were reached the authors’ threshold for being highly cited, though the company has presented data on its lead product candidate, the CAR-T therapy JCAR017 (lisocabtagene maraleucel). Those have shown it to have a potentially better safety profile than the two FDA-approved CAR-Ts. Another company in the paper, Enobia Therapeutics, had no publications when Alexion Pharmaceuticals acquired it in 2011 in a deal worth up to $1.08 billion, but its drug, Strensiq (asfotase alfa), has since won FDA approval for hypophosphatasia, a rare metabolic disorder.

But Enobia is not a good example of companies with high valuations despite lack of published data because it was never highly valued, said Jonathan Silverstein. He is the managing partner of venture capital firm OrbiMed and chairman of Enobia at the time of the acquisition, in a phone interview.

“I will say Enobia saved thousands of kids’ lives,” he said, adding that in another market, it would have sold for a lot more. “Analysts estimate [Strensiq] will be a very significant product in the years to come.”

The $1.08 billion deal included a $610 million upfront payment – which CBInsights also gives as Enobia’s valuation – plus up to $470 million in milestones.

It’s important to note that companies with limited or no publication histories in Ioannidis’s study were more prevalent in health technology than life sciences. And the paper acknowledged that publishing is not startups’ primary mission due to time constraints and peer reviewers’ potential resistance to new ideas. But even if the record mostly vindicates the life sciences startups, it still leaves open the question of how such companies attract so much money to achieve unicorn-level valuations.

“It is probably reflecting mostly some insider knowledge – which may or may not be well-informed – marketing and reputation building,” Ioannidis wrote. “But in the absence of science, I am not sure these valuations are objective and secure bets; they probably carry tremendous uncertainty.”

Venture capitalists, of course, begged to differ.

ARCH Venture Partners Managing Director Robert Nelsen said in a phone interview that what matters to his firm are the factors that correlate with a company’s long-term value. Platforms are more valuable than single assets, and important considerations include the company’s growth potential and how transformative the technology could be.

“I like my companies to have just really good science,” Nelsen said. “So when we’re investing in stuff, we ask Nobel Prize winners and a whole bunch of smart people what they think before we make the investment.”

To mitigate potential bias, ARCH also doesn’t pay the scientists it consults, he said. “They have to really like it – they have to be willing to spend time on the science and say, ‘This is really cool.’”

Other than the underlying science, internally-generated data are desirable in the early stages, Pellini said. “Ultimately, what we’d like to see as a company advances is that there is some ‘stealth research’ or ‘stealth data’ that maybe drives the earliest days of the company,” he said. “But as that company matures, I believe the vast majority of them should be able to generate potentially important peer-reviewed manuscripts if they’re going to be able to deliver something important to the marketplace.”

Lack of published data should not necessarily be viewed as a turn-off, he added. “If that fund has the expertise to verify and validate the data in a way that others don’t, and they believe the data warrant a valuation that’s north of a billion, then all the more power to them,” he said.

Nelsen said it is preferable for companies he invests in to keep their work secret from the public for as long as possible, but not from their investors. Indeed, a startup keeping its work secret from ARCH is a “blinking red light,” he said, adding that it was one of the reasons ARCH was among the first to turn down funding for Theranos.

Another growing trend that may explain some of the more aggressive investment is a cultural cross-pollination between tech and life sciences venture capital, Pellini suggested. Tech investors have historically been more aggressive, whereas life sciences and biotech investors have been more methodical.

“But we’re entering an era where many companies intersect,” he said, with an ongoing “sorting-out period” as a crossover takes place between tech and life sciences venture capital. Consequently, the field is seeing life sciences investment that is elevated compared with what’s been seen historically.

“It’s happening as we see more tech investors in the life sciences world – they’re starting to better understand this world as they invest more and more,” he said. “On the life sciences and biotech side, we’re also learning from having tech culture populate our companies, but also our investor base.”

But Nelsen disagreed, saying that tech investing’s cultural influence is more in health technology and in hybrid areas like machine learning. In core biotech, it doesn’t explain the larger rounds.

“The tech people I know who are playing in pure biotech aren’t playing in those large rounds – they’re being driven, I think, by different factors,” he said.

In particular, biopharma startups are pursuing more expensive projects, and there’s more money to support them, he said. With the pace of innovation in biotechnology and the potential impact of their technology, many investors are willing to support them for the longer term, thanks in part to their having bigger checkbooks.

While Nelsen disagreed with Pellini over the impact of tech investors on life sciences, Silicon Valley Bank managing director Jonathan Norris echoed the latter. 

“When I do look overall on the valuation side and quickness, it feels like those tech investors move fast,” he said in a phone interview. He emphasized that it doesn’t mean they aren’t performing due diligence, but that they’re moving in response to competitive market pressure. By contrast, healthcare investors have been accustomed to taking longer to decide on things, and that is creating pressure on them to get into deals.

Norris also pointed to increasing collaboration between healthcare and tech venture capital firms around artificial intelligence and machine learning, particularly in therapeutics and drug discovery, driven in part by the need for each other’s expertise. Consequently, they’re interacting more and benefiting from each other’s expertise.

One example he pointed to was South San Francisco, California-based Senti Biosciences. The company is working to develop cell and gene therapies using synthetic biology, an interdisciplinary field that combines life sciences with principles more commonly found in engineering and computer science. A look at the $53 million Series A round the company raised in last February reveals the kind of collaboration that is developing, Norris said. Tech-focused venture capital firm New Enterprise Associates led the round, while 8VC – whose website touts its “tackling the Bio-IT wave” – participated, along with other tech and life sciences investors.

The term “stealth research” came from a paper Ioannidis published in February 2015 in the Journal of the American Medical Association pointing out Theranos’ lack of published data – eight months before Wall Street Journal reporter John Carreyrou began publishing his articles exposing the company’s fraud. The concern in the more recent paper isn’t necessarily fraud, but trustworthiness when there are no publicly available data. However, venture capitalists stand by their methods for vetting the companies they invest in.

“I think clinical data trumps everything,” Silverstein of OrbiMed said. “If people are fudging the data, you could hire a forensic scientist to look through it, but most people in this industry take everyone at their word and assume they’re telling the truth.”

Photo: StockFinland, Getty Images