Tag Archives: Genentech

It Had to be You: Why Roche Was the Lone Suitor for Foundation Medicine

By Steve Dickman, CEO, CBT Advisors

January 16, 2015

Originally published on Xconomy

The buzz from day one of the JP Morgan conference in San Francisco earlier this week was the announcement on Sunday night by Roche that it was acquiring a majority interest in Foundation Medicine (NASDAQ: FMI) for a bit more than $1 billion in cash for a little more than half the company, which translates into $50 a share. Those were just the latest eye-popping numbers from Foundation, which went public in September, 2013, amid warnings of a biotech bubble. From its initial offering price of $18, Foundation proceeded to enjoy a first-day jump all the way up to $35 a share, straining credulity for those investors focused on the fact that the company had not received meaningful reimbursement for its flagship cancer diagnostic product FoundationOne.

It’s sixteen months later and Foundation still has not received the positive coverage decisions from Medicare or major private insurers that it would need in order to even dream of making money on its sequence-based diagnostic test. The stock had ridden down to $23 a share before the acquisition and there was plenty of short interest even at that level (pity those investors who did not cover those shorts on Friday!).

Roche was the only pharmaceutical company in the world that had a rationale for acquiring control of Foundation. It is best positioned to make the acquisition a success.

But Roche still went ahead and bought an unprofitable company for $1 billion in a transaction reminiscent of its purchase of a controlling stake in Genentech in 1990. Aside from the deal structures, which in both cases leave the US management team intact if now reporting to Roche HQ in Basel, there would seem to be virtually no similarity. Roche has moved far beyond its early-1990s status as a small-molecule-heavy European pharma eager to transition into biologics. At the time of the initial Roche transaction Genentech was already a powerful product engine, having developed early protein replacement therapeutics human insulin and human growth hormone with lots more in the pipeline and vibrant science to match. By contrast, Foundation has done little more than make losses on its diagnostics business.

But there is one big parallel between that deal and this one: In both cases, Roche believes that it has seen the future of the pharmaceutical industry. And it can only grasp that future by placing a large and risky bet on a US innovator company. Roche’s thorough transformation into a company invested in targeted therapies driven by disease biology supports my thesis that it was the only pharmaceutical company in the world that had a rationale for acquiring control of Foundation and that it is the one best positioned to make the acquisition a success.

In my view, the key reasons boil down to these:

  • Roche was early and fervent in its embrace of diagnostics as drivers of drug development and sales. I know only one top executive in the pharmaceutical industry who cut his teeth in molecular diagnostics and he did so at Roche– Dan O’Day, who was CEO of Roche Molecular Diagnostics from 2006 to 2010 and is currently COO of Roche Pharma. Once the Foundation transaction is completed, O’Day and two others chosen by Roche will join Foundation’s board of directors. Aside from the personal, Foundation also fell on fertile ground at Roche on the institutional level. Roche had already changed its drug-discovery focus to be more diagnostics-driven than most other pharmaceutical companies on virtually every level. As Roche CEO Severin Schwan declared in 2012: “More than 60% of our pharmaceutical pipeline projects are coupled with the development of companion diagnostics in order to make treatments more effective.” That number has almost certainly gone up.
  • Roche was the pharma that had most thoroughly integrated clinical genome sequencing into its trial protocols, long before it had figured out how best to use the data. In my work with biotech companies, I had been hearing for years how Roche had embraced sequence data as a key success factor for the pharma industry of the future. As soon as the cost of sequencing became halfway affordable (maybe $5,000 to $10,000 per full sequence), Roche began to require genome sequence data as a key data point from every patient in every clinical trial. If there was any doubt about how highly Roche regarded sequencing, its $51-a-share Illumina bid in 2012 dispelled it. (Illumina, whose CEO Jay Flatley said at the time that the bid seriously undervalued his company, now trades at $181). An executive speaking under condition of anonymity who knows Roche Ventures well confirmed that Roche places high importance on sequence data, both data which it has itself collected as well as data being collected by Foundation. As an aside, Roche Ventures had invested in Foundation two rounds before the IPO in 2012 and had no strings attached in the form of a promised acquisition or partnering deal. That investment is another indicator of the value Roche management placed on keeping up with the world of clinical sequencing. That executive told me on Monday that Roche was counting on Foundation’s data scientists to be able to make the most effective use of their own data banks of both sequence data and outcomes data and that the prospect of joining forces was irresistible.
  • Roche realized that it would face competition if another pharma company scooped up Foundation. Roche believes that in genomic profiling, it has identified a common theme for creating value in oncology therapeutics of whatever stripe – targeted therapies like kinase inhibitors; biologics like monoclonal antibodies; and even immuno-oncology approaches like checkpoint inhibitors. Having made this observation, it did not want anyone else to catch up. Combining Roche’s clinical sequence data and its drug pipeline with Foundation’s gene-level and patient-level insights clearly would realize the most powerful synergy. But in the hands of another pharma, the Foundation team and data sets could have posed competition. Ergo, Roche decided to buy it now.
  • Roche is counting on a shift from biochemical targets to genomic profile targets, especially in drugs for solid tumors. Foundation’s “poster child” cancer cases are those in which genetic profiling suggested – against all experience of oncologists and with no evidence from n-of-1,000 clinical trials – that cancer drug X, developed for, say, ovarian cancer, would work best in cancer indication Y (e.g. prostate cancer). Because the patient is desperate, the physician prescribes the drug and voila – there is a response or even a remission. This pattern echoes what has happened in blood cancers such as chronic lymphocytic leukemia (CLL), in which old-fashioned chemotherapeutic agents have been replaced by biologics like rituximab (Rituxan) and are being augmented or, eventually, replaced again by kinase inhibitors like ibrutinib (Imbruvica). This will likely happen in solid tumors as well: chemo as we know it will be scaled back (though, like that other blunt instrument, surgery, it will likely never completely disappear) and physicians will chase cancer cells through various waves of genetic mutations, each of which demands a different targeted therapeutic or biologic to hold it at bay. In that world, the company that is most on top of the mutation patterns and treatment patterns and can incorporate those into both its drug development efforts and its sales pitch, wins, or at least has an edge. Diagnostics will likely be an important part of immunotherapy as well, an area where Roche is currently weak. Right now companies like Juno (NASDAQ: JUNO), Kite (NASDAQ: KITE), Bellicum (NASDAQ: BLCM) and Novartis are taking baby steps with CAR-T. Most companies are focusing on surface antigens like CD19 that are widely expressed and therefore do not require molecular diagnostics.  To realize the full potential of these therapies, companies will need to match patient-specific tumor profiles with panels of off-the-shelf biologic reagents and cell engineering products. That’s where Foundation’s tests might come in.
  • Foundation is setting new standards in cancer genome analysis. Foundation has raised the bar in the accuracy of genome-based tumor profiling (sensitivity, specificity) by something like a factor of three, and built robust and scalable informatics and analytics. Roche was already using the Foundation platform on a limited basis and realized that it was simpler to expand that use rather than to try to copy it.

Since the last few pharma mega-mergers, the industry’s biggest players have gone in wildly different directions. Novartis embraced gene therapy and gene editing. AstraZeneca doubled down on the biologics franchise it obtained with the acquisition of MedImmune. Bristol Myers and Merck have raced ahead in checkpoint inhibitors. Merck, Sanofi and to some extent Pfizer have rapidly expanded investment in “beyond the pill” and “digital interventions” (apps as drugs). And Roche took up diagnostics and genetics. For Roche, drug development, especially in oncology, is all about “genetics-driven medicine,” which in their view requires “genetics-driven drug development” and “genetics-driven marketing.” No one else has placed such a big bet on genetics though all pharma companies are certainly exploring it. For example, AstraZeneca, Johnson & Johnson and Sanofi recently announced a collaboration with Illumina (NASDAQ:ILMN) to develop a “next-generation-sequencing based test system for oncology.” In some sense, if Roche wins this one, others – e.g. those betting on checkpoint inhibitors and CAR-T cells – might lose out.

In CBT Advisors’ world of venture-backed biotech companies, this landscape poses significant challenges. Gone are the days when a biotech’s innovation would be appreciated by as many as five or ten pharma companies at the same time (there are barely fifteen left that regularly carry out M&A) and there could be a big bidding war. The biotechs’ leverage is not what it used to be. Counterbalancing that is the obvious productivity flop in pharma R&D. Biotech is the only place pharma can turn for real innovation. And turn they do, early and often.

This creates a bonanza for firms like mine that assist early, science-driven companies in managing their public positioning and their BD pitch from day one to create the largest possible exits. Now more than ever, the right story sells, just maybe to only one or two bidders. In Foundation’s case, the billion-dollar number was probably what it took to get the company’s pre-IPO investors (who included Google Ventures and Bill Gates not to mention smart funds like Casdin Capital) to give up on at least some of their dreams of long-term returns in exchange for a sweet 10-12x (I’m guessing) on their last pre-IPO investment from early 2013.

From Foundation’s point of view, the deal does three things, all of them good:

  • Cashes out the early investors at a price they can accept.
  • Delays, perhaps indefinitely, the need to break even on selling tests and shifts the focus to drug development and companion diagnostics
  • Relieves the constant pressure to market the company’s analytic services to multiple pharma companies in deals that have been the main source of revenue for Foundation to this point. That pressure was undoubtedly going to become heavier as Foundation’s pharma partners realized that, quarter after quarter, there was no reimbursement coming from Medicare and little from other payers, leaving pharma to provide the vast majority of the company’s source of revenue. (A first small insurer in Grand Rapids, MI, announced coverage of FoundationOne and another Foundation test in October, 2014.)

Back to why the acquirer had to be Roche: remember that over the last 25 years, Roche has had the undoubtedly humbling but ultimately very profitable experience of owning Genentech. Revenues and product pipeline from that acquisition long ago overtook those products from Roche’s own drug development in volume and importance. In some sense, Genentech has come to own Roche. Since Roche is nowhere near as advanced in gene therapy as Novartis nor as advanced in checkpoint inhibitors as Merck and Bristol-Myers, the move to own Foundation is an attempt to be the best it can be as a genetics-driven drug developer and marketer. No other pharma would have seen this deal that way. When and if Foundation’s investment bank called around looking for better offers, I bet no one called them back.

For Roche, the deal will either turn out to be a leap frog or, maybe, a dead end. But if cancer therapies, especially for solid tumors, really do wind up getting developed and marketed in a genome-driven way – and many trends point in that direction – then this move will have turned out to be prescient indeed.

Leave a comment

Filed under Biotech, Diagnostics

Moderna Therapeutics as the next Genentech? Not so fast

By Steve Dickman, CEO, CBT Advisors

December 10, 2012 (slightly shorter version originally published on TechnologyReview.com)

Quick biotech PR tip: When exiting stealth mode, heralding your company as the next Genentech is one way to get above the noise. That was the approach of Moderna Therapeutics, a Cambridge, MA-based startup that announced itself last Thursday, revealing that it had raised more than $40 million and attracted an all-star set of board members and scientific advisers.

Announcing that you just might be on the way to becoming Genentech II raises the bar a wee bit. And, at first blush, Moderna looks like it might even get over that very high bar.

Central Dogma of Molecular BiologyThe concept is intriguing, to say the least. Biology’s central dogma is “DNA to RNA to protein.” Although Nobel Prizes have been won for discoveries that expand upon that central dogma (the discovery of reverse transcriptase, for example), the core approach underlies the first several generations of biotech products. Think EPO, Neupogen or the grandfather of them all, human insulin. You manipulate the DNA in the lab and then express the protein in the production facility. Then you put it in a vial and sell it to the patient, who gets an injection or an infusion. The main role for the dogma’s middleman, messenger RNA (mRNA), is a passive one: get transcribed from the DNA then, in turn, get translated into protein.

Moderna turns the dogma on its head: go straight to the RNA, do some fancy chemical tricks to it and deliver it directly into the body. This makes the patient herself into the production facility. All of us carry around cellular protein factories, known as ribosomes, and if properly activated, those can be harnessed (at a much lower cost) to produce proteins in which we have deficiencies.

One report on Moderna, published on Xconomy, quoted venture investor Noubar Afeyan of Flagship Ventures as saying that the company “builds on lots of things that have been tried before.” One of those things is gene therapy, providing genes (that is, DNA) via viruses or other delivery vehicles and trying to get cells to express those genes. Those approaches, too, tried to use the body as a manufacturing facility. Unfortunately, with some recent intriguing exceptions, most of them have failed.

Aside from this novelty, three things make Moderna so interesting:

Breadth of application

Since the mechanism is potentially so universal, proteins could be produced that address any number of diseases. The company said it will focus first on areas where protein therapeutics are already well-established: oncology supportive care, inherited genetic disorders, hemophilia and diabetes. But the company also claimed that it can also induce production of intracellular proteins that could never be given exogenously due to efficacy or immunogenicity concerns. Should this approach work, and it’s a bit of a long shot, it opens up new areas of application to the pharmaceutical industry.

Repeat dosing

Unlike many gene therapies, which could potentially be curative, in Moderna’s case the patient will need to be dosed with the mRNA over and over again. Think “recurring revenue stream.”

Intellectual property

When Genentech and Amgen were founded, neither one had a monopoly on the production of all human proteins in bacteria. When monoclonal antibodies were invented in Cesar Milstein’s laboratory in Cambridge, UK, Milstein was discouraged from patenting the concept. But in Moderna’s case, filing broad and deep intellectual property was the company’s central focus and a big reason why the company remained in stealth mode for the past two years. This means that even if other companies manage to enable the use of mRNA-based techniques in areas not yet explored by Moderna, the company could still demand royalties.

Yet another reason to pay attention to Moderna: unlike many other biotech companies, Moderna was not based on work published soon after its founding. The original publication that drew interest from Afeyan didn’t involve using patients as protein factories at all. The paper, published by company founder Derrick Rossi in 2010, involved using injected mRNA to produce cells that resemble embryonic stem cells. According to the Xconomy article, Afeyan did not want to invest in a stem cell company, which he perceived as too risky. Instead, he suggested that Rossi use the mRNA as a way to induce protein production in patients. That led to the key experiments, as yet unpublished, that were the basis of the company’s intellectual property and its initial financing. According to Moderna’s triumphant press release, the publications are supposed to come in 2013.

At the same time, there are three big questions:


Isn’t Moderna facing a double hurdle, first in selectively getting into the right kind of cell and then in achieving the right therapeutic dose level? The first of these hurdles represents the same kind of delivery problem that has presented such an enormous challenge to RNA interference (RNAi) companies like Alnylam. For all its promise, RNAi was born amid a hail of questions expressing doubt about delivery. How to use systemic delivery to propel nucleic acid molecules with strong negative charges and potentially vulnerable to ribonucleases into the right cell types in the right organs at high enough concentrations to have a biological effect? That was the question. (The early results, as I viewed them in a cramped biochemical laboratory in Kulmbach, Germany, in 2002, looked blotchy at best.) More than ten long years later, despite some powerful efforts that cleverly take advantage of biological reality, for example, the “leakiness” of tumors, those questions have still not been completely laid to rest.

The other part of the delivery challenge has to do with what happens to the mRNA once it is inside the right kind of cells. How many cells exactly has it penetrated? What are the expression levels over time of the desired proteins on a per-cell or per-tissue basis? Will the levels in one patient be the same as in the next one? Achieving appropriate dosing without setting off alarm bells at the Food and Drug Administration will be tough.

Where are the other investors?

The only institutional investor named in the press release was Flagship Ventures. If other VC firms were involved, one would expect to find them sharing the limelight. So either Flagship decided that what it had in Moderna was so good, it did not want or need to share or other VC funds were approached and said no. It will be interesting to learn over the coming weeks which of these explanations, or which combination of them, pertains.

What’s the value in its first applications?

Let’s assume that the Moderna approach works. Suddenly EPO, Factor VIII and beta-globin can all be produced in patients deficient in these proteins simply by dosing them regularly with mRNA. But so what? There are already therapies on the market that will be doing this. In fact, some of those will be going generic and will be joined on the market by “biosimilars” that will presumably cost less than the existing (expensive) drugs. Furthermore, many of today’s most successful protein therapeutics have been modified (e.g. pegylated) to improve their half-lives. Where would be the advantage of an injection of mRNA over one of protein, especially a second-generation, long-acting protein such as Amgen’s Neulasta®?

Perhaps the advantage would come in proteins that cannot be injected as such because they elicit unwanted immune reactions from patients. But there are not too many examples that come to mind (thrombopoietin is one). That might be one reason why Moderna CEO Stéphane Bancel said that the company would be partnering the largest-market indication areas, like cancer, while retaining only rare diseases (in which intracellular protein production might make sense) for itself.

In summary, Moderna reflects a novel approach. For that, its founders and visionary investors deserve their well-earned day in the spotlight. It is especially commendable that a venture investor in the current no-whip, Splenda-only funding environment would create a good old-fashioned full-fat latte of a biotech company. Funding it exuberantly, vigorously protecting the IP and keeping the shares to yourself are all probably wise moves. But for the rest of us to see Moderna as a new Genentech, Moderna will have to publish in a peer-reviewed journal, partner with a pharmaceutical company or at least explain how it addresses basic questions like delivery and consistent dosing across tissues and patients.

# # #


Filed under Startup