Early-Stage Biotech Value Creation: The Roles of Equity and Partnerships

WARNING TO THE READER: this piece is REALLY LONG.

Its genesis lay in my current efforts to develop a business/financing strategy for Decibel Therapeutics.  In that context, I have been giving thought to how others have been successful—and failed—in such endeavors. 

The piece starts with my thoughts. It is then followed by responses to these thoughts by three great thinkers (and friends) in the biotech industry: Mark Levin, John Maraganore, and Jeff Tong, who reviewed an earlier draft. My replies to their responses are also included. In fact, there are two rounds of responses and replies.

The publication of this piece on a blog, which has a “comment” functionality, is intended to enable additional colleagues in the industry (including you) to join the conversation if they/you so desire. Invariably I find that responses to my thoughts are much more insightful than anything I have to say.

In addition to being very long, the piece features extensive endnotes. I tried to move the bulk of my asides, philosophical musings, anecdotes from the history of biotechnology, etc. to endnotes to make the main text flow better. One can read the text without reading the endnotes. With respect to the historical anecdotes, at least one reviewer who is a biotech history buff loved them (albeit, flipping back and forth between the text and the footnotes is somewhat an act of masochism perhaps only appealing to fans of the writings of David Foster Wallace).  I make no claim to 100% historical accuracy or completeness. They are the best my failing memory has to offer and draw heavily on my first-hand personal experiences.

Thanks to Mark, John, and Jeff for consenting to the inclusion of their comments and for their encouragement to put this out there for consumption and participation by a broader audience.

==============================================

Prologue

Building a great, sustainable biopharmaceutical company requires so many things:

·        Great science

·        Great drug discovery and development

·        Great people

·        Great culture

·        Great vision and strategy

·        Great execution

·        Great investors

This piece addresses NONE of these.

Its sole focus is the STRATEGIC FINANCIAL aspect of building a great, sustainable biopharmaceutical company: how do you put the “gas in the engine” in the period before revenues and profits in a manner that enables financial investors (and your employees who hold stock/stock options) to enjoy a return on their investment (= increase in share price relative to their purchase price).

The appropriate financing strategy will be dictated by internal forces (the nature of your assets and vision) and externalities (the partnering and capital markets). There is no single strategy appropriate to every company in every environment. At the same time, the fact of the matter is that, in the period before you have product revenues, you have only two things to sell. The first is technology and information of current value to the customer (typically, a larger biopharmaceutical company). The second can be called “futures,” and those futures come in one of two flavors: equity (and instruments that look, smell, and taste like equity) and rights (marketing/profit participation rights) in/to the products (and the revenues they will produce) you are seeking to develop. As it were, you are selling participation in the future financial appreciation of the whole and/or of one or more of the parts. (Note: the sale of technology and information of current value to the customer can and most often involves a sale of rights.)

I make no claim of originality for the observations recorded here. In many, if not all, respects, they are obvious. If this rambling has any value, it will be found in having achieved the goal of philosophical inquiry described by Wittgenstein in Philosophical Investigations, Part I, Section 122: “A main source of our failure to understand is that we do not command a clear view…A perspicuous representation produces just that understanding which consists in ‘seeing connexions’.”

============================================

Chapter 1: The Basics

·        Value creation is not about the size of your market cap, or not directly about that. It’s all about share price appreciation.

o   Market cap is about the size of a CEO’s ego.

o   Share price is what pays for your (and your investors’) kids’ college educations.

·        You can create a large market cap without creating shareholder value by taking too much dilution early when your stock price is at its lowest. Therefore, be wary of early dilution.

·        Yet beware that corporate partnerships are not an end in and of themselves and are not a (purely) “non-dilutive” form of cash; they are an implement of strategic value creation.

o   Use them to contain early dilution but only if you can do so in a manner that does not give away the major sources of your future value creation.

·        Ours is a business in which you can spend yourself to death, but you can’t save your way to glory.

·        Therefore, under-spending/under-capitalizing and over-spending/over-capitalizing both represent sources of risk to exceptional value creation.

·        The art lies in balancing the two different sources of capital as they represent two different forms of dilution and in recognizing the relative risk of the two that applies to your specific situation.

·        Identify your specific situation at the given moment in your evolution and act accordingly.

Chapter 2: Product Companies

• In general, if you are a “product-based, single or limited assets” company[1], your primary risks are either over-spending/over-capitalizing early (dilution) or partnering in a manner that sells off your long-term value-creation potential.
• Your value is directly driven by the attainment of meaningful development milestones.

§  Drive your lead asset.

• Do not partner away (the majority of) your rights early; use equity as funding, otherwise you will have given away the source of your longer-term value and be “back on the hamster wheel” trying to develop the next shot on goal.
• Raise sufficient capital early; devote this early/expensive cash to driving to the value-inflection development milestones; but do not “over” capitalize too early.
• Be sure to be realistic and not find yourself—when the inevitable hiccups occur—with the gas tank approaching “E”…venture capitalists in such situations often morph into vulture capitalists.
• On the back of good data, raise more, large amounts of equity at higher prices: now is the time that the risk has shifted from over-capitalization to under-capitalization; keep doing so as you knock off the data-based milestones.

o   Achieve out-scale return via:

§  Get sold.

§  Get to market with your product.

§  Expand and diversify and grow:

·        Develop additional indications of the lead molecule.

·        Diversify via adjacencies, e.g., new formulations/analogs with better pharmaceutical properties of lead; attack same/related pathways as lead’s mechanism of action.

·        Forward integrate to the market.

Chapter 3: Platform Companies

Part I: Introduction: A Typology of Platform Companies

·        There are two genera of platform companies and, within the second genus, there are two species:

o   Genus (1): Product (Therapeutic Modality) Platform Company

§  Genus (1) encompasses companies that pioneer novel therapeutic modalities and whose proprietary competitive edge lies in the ability to generate new and better products of that modality.

o   Genus (2): Novel and High-Value Insight Platform Company

§  Species (A): Insights into pathways, targets, etc. and their relationship to disease.

§  Species (B): Insights based on a deep, disease-specific biology and related translational research/medicine expertise.

Part II: Creating Value in a Genus 1 Product (Therapeutic Modality) Platform Company[2]

• In general, if you are a Product (Therapeutic Modality) Platform Company, your primary risk is under-spending/under-capitalizing early:
• Your value is directly driven by your first-mover advantage.

o   You need to drive as many shots on goal as possible.

o   You need to stay ahead of the pack in the perfection of your platform/product engine: be, continue to be, and be perceived as the leader.

• Because you have a potential “embarrassment of riches,” you can partner early and often to raise non-equity capital because whatever product rights (= source of long-term value) you monetize, there will be more where those came from.
• Early deals can give away all product rights, e.g., to a disease area, maintaining only a downstream economic interest (typically, milestones and royalties).
• In the next step, the ability to “belly to the bar” for greater downstream financial participation (cost and profit share) will likely feature without commercialization rights.
• In the following step, some level of commercialization rights (co-promotion/ geographic splits) will be added to the cost and profit share.
• In a possible next chapter, the deals may add on full commercialization rights to one or more of the products coming from the collaboration (e.g., via a picking mechanism).
• It is the author’s belief that you are better off sharing in all of the fruits of the collaboration with profit share and retention of a geography and/or co-promotion than in going with a picking mechanism to provide forward integration to commercialization ability. Picking is a binary bet…spread the risk.[3]
• The combination of a high stream of non-equity capital and clarity that you are using that to build out the leading platform while, at the same time, you have retained enough opportunities for forward integration into downstream value, will drive your stock price up.
• Now is the time to use the equity markets as your primary source of capital to fuel your retained product opportunities.[4]
• Counterpoint: Why can’t a Genus 1 Product (Therapeutic Modality) Platform Company choose never to forward integrate and just remain a product engine for others who make and sell the end products, collecting high-margin royalties/profit shares?
• The model here is “Intel Inside.”
• In theory, the model makes sense.
• In practice, there are only two companies I can think of in the history of the industry with a multi-billion dollar market cap that have done it, ISIS/Ionis and GenMab.
• Even ISIS/Ionis has held onto some product rights and is, via a majority-owned subsidiary (Akcea), forward integrating into commercialization.
• Moderna is the current version of the ISIS/Ionis model; here, too, it has held onto some product rights.
• While in theory the “Intel Inside” model makes sense, the important differences between a computer chip and a drug suggest why the theory has a fundamental flaw.
• It is not, or not simply, that the probability of success (POS) of making a drug, and then a second drug, is so much lower than the POS of making the Generation 1 chip and then a (significantly better) Generation 2 chip; and it is not, or not simply, that the time from idea to product is so much longer and more costly for a drug.
• Assume, for the sake of argument, that the POSs, development times, and costs are the same for developing a chip and for going from a Generation 1 chip to a Generation 2 chip as it is to develop the first drug and then go on to develop the second drug.
• The key difference is that the Generation 1 chip can be “inside” multiple different products in multiple different verticals; this is not the case for a drug.
• So, Isis/Ionis and GenMab stand out as the exceptions to the rule.

Part III: Creating Value in a Genus 2 Novel and High-Value Insight Platform Company[5]

Species (A): Insights into pathways, targets, etc. and their relationship to disease

·        As with Genus 1 Product Platform Companies, in general if you are a Genus 2, Species A Platform Company: Novel and High-Value Insights (claiming insights into pathways, targets, etc. and their relationship to multiple diseases), your primary risk is under-spending/under-capitalizing early.

• Your value is directly driven by your first-mover advantage.

o   You need to drive as many shots on goal as possible.

o   You need to stay ahead of the pack in the perfection of your platform/product engine: be, continue to be, and be perceived as the leader.

·        The partnering strategy employed for the Genus 1 Platform Company is exactly the same: partner broadly, increasingly forward integrate into rights/profit participation, and then forward integrate into proprietary product development.

·        HOWEVER, Genus 2, Species A Platform Companies face a set of challenges not faced by Genus 1 Platform Companies. These arise intrinsically from the fact that the output of the Genus 2, Species A Platform Companies is data/information/insights, NOT, as with the Genus 1 Platform Companies, New Chemical Entities (NCEs) and biologic therapeutics.

o   The history of data in the biopharmaceutical industry is the history of its commoditization.[6]

o   Companies whose “life’s blood” is drugs/products have a vested interest in rendering data “pre-competitive” (or, at least doing so after they have had proprietary access for a time). They win based on their products; they don’t want to be held captive by the owner of the information.

o   A more restrictive intellectual property (IP) environment: gone are the days when a transcriptional profile showing over-expression of a gene in a diseased tissue (or a genetic mutation in the diseased state) could get you an issued patent of the logical form, “a method of treating disease X comprising modulating target A by any means” (with dependent claims stating that the “means” could be an antibody, an antisense, an RNAi, a gene therapy, a small molecule, etc.).

o   Moreover, the demands of the customer base became more expansive over time. Whereas, in the 1990s, most big pharma customers were willing to agree to terms that restricted their use of the data to the discovery and development of small molecule drugs (because that is all they did), in the present all pharmas/big biotechs would demand the right to exploit the data for all therapeutic modalities.

o   Finally, the Genus 2, Species A Platform Company has and builds its expertise in the generation, curation, and interrogation of the data. It does not come into existence or can afford to cultivate/invest in significant capabilities in drug discovery/development in one or more therapeutic modalities, or in-depth biology/translational capabilities, in one or more diseases.

§  The Net Result: the Genus 2, Species A Platform Company ends up abandoning its data/information business to build and become a drug discovery and development business.[7]

·        Counterpoint: Why can’t a Genus 2, Species A Platform Company (novel and high-value insights) choose never to forward integrate and just remain an information company whose customers make and sell the end products, collecting high-margin royalties/profit shares and/or just selling the information at a high margin?

o   The model here would be, e.g., Bloomberg.[8]

o   But, now, consider the Bloomberg business model and the respects in which it differs. 

§  Time value of information: the bond or currency trader making multi-hundred million-dollar trades lives or dies based on moment-to-moment information. Last time I checked, the drug discoverer can wait a week (a month? six months?) for the latest genetic sequence or proteomic profile.

§  Value of that information: for the trader, that information, and its currency, is the key to making the investment which is the money-making activity. For the drug discoverer, the information is just the start of a multi-year, multi-disciplinary journey lasting a decade or more to the money-making activity (= discovery, development, approval, and marketing of the drug). Hence, a subscription model, not a royalty/profit-share model of payment, is the only one that will sell.

§  Number of high-paying customers: a Bloomberg terminal subscription costs $30,000/year per user (not per firm). In 2016, there were 325,000 terminals in use. How many pharma and biotech subscribers, in a best case, would there be at what user fee?

o   In at least one case, namely deCode, the information was of such a special, and potentially powerful, nature (the genetically homogenous population of Iceland plus the Icelandic national database of EHRs), that a single company, Amgen, saw enough value to purchase it for exclusive use.

o   23andMe is a deeply instructive, unfolding model well worth reflection.[9]

Species (B): Insights based on a deep, disease-specific biology and related translational research/medicine expertise

·        If you are a Genus 2, Species B Platform Company: Insights based on a deep, disease-specific biology and related translational research/medicine expertise, you face a potentially very challenging path with elements shared by the Genus 2, Species A Platform Companies but a tougher row to hoe.

·        On the one hand, your primary risk is under-spending/under-capitalizing early.

• Your value is directly driven by your first-mover advantage with your (novel) insights into the disease: you need to invest.

§  You need to stay ahead of the pack in the perfection of your platform/ product engine: be, continue to be, and be perceived as the leader.

• From very early on, you also need to invest in the drug discovery capabilities necessary to capitalize on your insights: you need to drive downstream of the insights.
• However, equity markets will have trouble according you value without tangible drug candidates.
• Hence, you need to partner early so as not to experience over-dilution early.
• However, unlike the Genus 2, Species A Platform Company, you are not intrinsically able to partner broadly.
• You can do a single-player, multiple-target deal, but it is very hard early on to take on multiple partners.
• Therefore, pray that your biology is viewed as a “gotta have” by players in your disease space and that there is at least one player who is relatively price insensitive because of a significant strategic commitment to the area in which you are working.
• If you secure a large, foundational corporate partnership, you will have to retain significant rights because you will not have anywhere near the number of shots on goal as the Genus 2, Species A Platform Company.[10]
• This will often require “bellying to the bar” to help shoulder development costs.
• So, you need to take the corporate partner dollars early while retaining significant rights, you need to execute, and, as candidates emerge which capital markets will value, you move into the equity raising mode, at higher prices.

o   With success, you can look to broaden your biological focus into adjacent areas that leverage the capabilities and expertise you have built.[11]

Epilogue: The Role of In-Licensing Product Candidates

The foregoing has systematically left to the side the role of in-licensing product candidates in early-stage biotech company value creation. We tend to think of in-licensing as the domain of big pharma, with biotechs doing, not buying novel discovery. In fact, in-licensing product candidates has played a major role in many, if not most, successful biotech companies. Here are some cursory thoughts on the matter.

·        For Product Companies, the product candidate that forms the basis of the company is, in fact, most often in-licensed, either in the fully blown form that is taken into the clinic or as the active pharmaceutical ingredient (API) (or near-final API) that with some tweaking or formulation for better pharmaceutical properties is taken into the clinic.

·        For a Genus 1 Product (Therapeutic Modality) Platform Company, product in-licensing is much less likely to feature, at least early in its history.

o   This is almost definitional as the company is pioneering the new therapeutic modality.[12]

o   Over time, as others begin to practice the new modality, product in-licensing will assume a more prominent role.[13]

·        For a Genus 2, Species A Platform Company: novel and high-value insights into pathways, targets, etc. and their relationship to disease in general, product in-licensing is less likely to feature early in the company’s history.

o   The fundamental thesis of the company is its ability to identify and elucidate the role in disease of new, previously undrugged targets/pathways; hence, any in-license will be in its nature serendipitous: an existing molecule, designed to modulate a different target/pathway or disease so happens to modulate the novel target/pathway.[14]

o   The company will have focused its resources on its platform and is unlikely to have invested in building the capabilities (e.g., non-clinical and clinical development) necessary to prosecute an in-licensed molecule.

o   Moreover, in the early stages, the fundamental value thesis is the platform. The presence of an in-licensed molecule can come to dominate the market’s perception of the company; the company becomes a binary bet with pressure to decrease investment in the platform in favor of investment in the in-licensed molecule.

§  Hence, if the full value of the platform is to be appreciated by the market, the company needs to work overtime to keep the market focused on the value of the platform.[15]

§  If/when the promise of the platform fails to be (fully) realized, product in-licensing can prove to be the difference between failure and success.[16]

·        For a Genus 2, Species B Platform Company: Insights based on a deep, disease-specific biology and related translational research/medicine expertise, product in-licensing is also less like to feature than in a Genus 1 Product Platform Company but more likely to occur than in a Genus 2, Species A Novel and High-Value Insight Platform Company.

o   The company will have some of the relevant capabilities/knowledge, in the form of in-depth, disease-specific biology and translational research, to enable it to prosecute the earlier stage of development of an in-licensed product candidate.

o   Nevertheless, as with the Genus 2, Species A Platform Companies, the company will have to work overtime to ensure that the market does not equate the company’s value with (the fate of) the in-licensed molecule.

Responses and Replies

JEFF TONG

Hi Steve,

Well thought out, as always.

I know this is a prelude to Decibel, so this example will take you somewhat on a tangent, but for completeness, Illumina arguably is the most successful pure-play “Platform” company in a long time (that also interestingly isn’t forward-integrating into therapeutics).

I’m sure it’s implicit, but I also think it’s worth explicitly pointing out that the financing/ capitalization strategy has to be intimately tied to the maturity and value creation curve of the science / platform / etc... and therefore one can only judge the risk of under- or over-capitalization as a function of how and on what the money would (or wouldn’t otherwise) be spent.

A graph that, for me, sums up the inter-connectedness of the two has always been the following:

HOW one gets to the ‘major value event’ makes a difference...while the company is the same market cap at the end, the share price can be dramatically different. And that’s a function of both the capital-raising strategy and the value-creation curve. On the former, Bruce Booth did a nice dissection of Juno vs. Kite showing the difference in capitalization strategies. Both incredible financial and scientific/clinical successes of course, but nonetheless still a material difference between the two outcomes (https://lifescivc.com/2018/01/two-carts-two-charts-dissecting-returns-t-cell-therapy-ma/).

Jeff

RESPONSE TO AND INTERCHANGE WITH TONG

HOLTZMAN

Thanks Jeff. I would suggest that Illumina is an equipment (and related reagents) company… they were never a “platform” company in the sense I defined these (re Genus A = therapeutic modality platform; Genus B = information/insights applicable to biology/disease per se [Species A] or specific disease biology [Species B]).

If drugs are the gold, then the companies of which I am speaking all end up as gold prospectors because their platforms become commoditized (particularly the Platform B companies). Illumina is Levi Strauss, selling the tools and jeans to the prospectors…and staying ahead of the competition by making better and better tools and supporting services (something Affymetrix failed to do).

Interesting question: both information and hardware become commoditized over time. So, how does one stay ahead in those games? And why is it that a hardware player (in the form of Illumina) has done so successfully while no information company in our industry has been able to do so. I think that comes back to number of users (= potential customers).

TONG

True - but there is indeed an example of an information company uniquely being worth something too: Foundation Medicine.

HOLTZMAN

With respect to Foundation: I believe that, in its initial incarnation, they saw themselves as a Genus 2 company, with potential for Species A (info for disease per se) but starting with a Species B (cancer) approach, that would sell high-value information to big pharmas (initial Roche deal). In this sense, they were like Myriad. However, again like Myriad, they morphed into a Diagnostic Service Business (by analogy, Illumina is a Research and Diagnostic Product Business). (In contrast, Genomic Health, right from the get-go, set out to be a Diagnostic Service Business…but Foundation ends up in the same place, and bought by a larger Diagnostic Service Company).

================================================

MARK LEVIN

Hey Steve and Jeff

I love this conversation!!

A couple added points. Mine will be much shorter than Steve’s and are inspired by my favorite philosopher Nitschke (as in Ray) who had a wonderful perspicacious insight into the biotech field.

POINT ONE

A company Steve, John, Kevin, and I know well, Millennium Pharma, raised $8.35M in VC dollars total at $1 per share/post $13M and sold for $100 per share/$9B (two 2x splits in company’s history).

A good case for raising a small amount of VC dollars and doing partnerships early on if you have a Broad Platform company with multiple opportunities.

If the VCs held on, they received a 100x on their investment at the sale of Millennium to Takeda. IPO investors at $12 a share (I think-$225M post) received an 8x if they held on to acquisition.

Millennium was hatched in the genomics craziness of the 1990s and was a platform company first, a few years later a target, then biology/disease company, and eventually a product company based on product company acquisitions. Although the VCs and IPO investors did well, many investors in the late 1990s did not do as well/actually poorly based on the value of the company rising to $19B during the genomics craziness.

Investors at $19B received a negative 50%+ return if they held on to sale. Post the crash in the early 2000s, the company’s value plummeted to approximately $2B or so. The investors who invested at this low point and held on to exit, received a 4x +.

Lot of great messages here

0)     If you can raise $8.25M in VC dollars total and complement with “non-dilutive” partnerships to get to public markets, you are off to a great start.

1)     Large Series A investments from the VC community are expensive and if done, need to be complemented with real value from the VCs. If you can start with a small investment and immediately complement with a major pharma (non-dilutive partnership), great strategy.

2)     Broad biotech platform companies with multiple disease applications in “revolutionary new areas” can form very substantial “non-dilutive” partnerships and raise money (debt and equity) at non-dilutive prices (GENOMICS, RNAi, GT, CRISPR, CART….).

3)     Teams that include Steve (Holtzman), John (Maraganore), Kevin (Starr), Bob (Tepper) and others are necessary to evolve/iterate/build early platform companies to targets, then diseases, then acquisitions ESPECIALLY when your platform is 15 years early, to ensure products in a reasonable timeframe.

4)     Use your stock price to acquire product-related/complementary assets when your value is disproportionally overvalued compared to others.

5)     Don’t invest in a $19B biotech company with one molecule in early-stage clinical trials; you are overpaying.

6)     Breakthrough products are key to REAL long-term value creation for the patient and the investor.

POINT TWO

This is where our VALUE CREATION MODEL came from at TRV and one we insist on following for all our companies in order to maximize value creation for patients and shareholders.

POINT THREE

Sticking to your investment model can be painful. We talked to Rick Klausner and the President of the Hutch almost a year before Juno was founded. We were asked to invest but passed.

They were going to raise $200M in Series A and we were concerned about the dilution/Value Creation Model.

Thanks Steve for your insights - I love it!!

mark

RESPONSE TO LEVIN

HOLTZMAN

First off Mark, it would appear that somewhere in the last decade you have learned how to unlock CAPS LOCK. (I guess that proves that, indeed, Nothing’s Impossible.)

As you imply, what became the “TRV Model”—a platform company which enters into one or more broad, “foundational” partnerships early in its existence—was grounded in your start-up experiences before and at Millennium. What I think is worth noting, and acknowledging as truly original, was that model when you initiated it at Millennium.[17]

The first generation biotechs (the recombinant protein companies) had broad product platforms that could have provided the basis for such broad foundational partnerships. However, what history shows is that they chose, instead, to partner product candidate by product candidate.[18] For example, a Genentech could have entered into a single, broad partnership with a diabetes company for all secreted proteins relevant to diabetes or a single agricultural health company for all secreted proteins potentially relevant to the area (e.g., bovine and porcine somatotropin, the bovine and porcine interferons, etc.).

Similarly, one of your earliest start-ups, Cell Genesys, developed the first genetically engineered mouse capable of generating fully human antibodies, and established a broad patent position on the technology covering any such mammal produced through germline homologous recombination as well as any (human) antibody generated by such a mammal.[19] Cell Genesys (Abgenix) could have entered into a broad foundational partnership, either covering all antigens/hMAbs relevant to a disease area[20] but did not, instead doing antigen/product by antigen/product deals while forward integrating on a specific product of its own, namely a hMAb directed to EGFR.[21] In the event, it did not.

What I would argue was your second radical innovation at Millennium was broad foundational partnerships based not on (potential) products but on novel biological insights (targets). In other words, Millennium was not a Genus 1 Product Platform Company; it was an extremely broad—and arguably the first—Genus 2, Species A Platform Company. Its “disease-by-disease” large-scale partnering model was (at least to me) interesting for helping to elucidate a distinction that gets lost in speaking about the “genomics” revolution of the 1990s and the companies from that era.

There were in fact two distinct classes of “genomics” companies.

The first comprised the large-scale sequencing companies (e.g., HGS and Incyte). For them, “genomics” meant high-throughput DNA sequencing. The outputs of sequencing (= the genes), through new bioinformatic techniques, could be grouped according to biochemical class, e.g., GPCRs, ion channels, enzymes; however, sequencing in and of itself provided no disease annotation.

The second class comprised the large-scale genetics companies (e.g., Millennium, Sequana, and Myriad [with deCODE following a few years later]). Genetics, of course, starts with a phenotype (such as a disease phenotype). The genetics companies then used the new large-scale technologies to isolate in the genome (what was called “positional cloning”) the gene or genes that were responsible for, or contributed to, the disease phenotype.

Thus, for Millennium, “genomics” was not, or not simply, high-throughput DNA sequencing. We came to define genomics very broadly as the application of high-throughput process technologies (robotics), micro-miniaturization (microfluidics), and the advanced information sciences to the elucidation of biology. Hence, beyond large-scale sequencing and SNP analysis,[22] Millennium also implemented large-scale transcriptional profiling and large-scale proteomic analysis in the context of specific disease-area programs.[23] It also meant that Millennium built up multiple disease biology groups in areas such as diabetes, obesity, immunology, cardiovascular, oncology, and CNS. It was this combination of technologies and expertise in multiple diseases that enabled Millennium to enter into its string of multiple, large disease-area alliances, a partnering approach not available to the large-scale sequencing genomics companies.

That said, since large-scale sequencing featured among the “genomics” technologies put in place by Millennium, by 1997 this provided the basis for Millennium to establish large alliances that were not based on a disease area. Thus, Millennium established a subsidiary, Millennium BioTherapeutics,[24] which had the rights to exploit genes encoding proteins elaborated on the cell surface to develop therapeutics comprising secreted protein and MAbs. “MBio” entered into a foundational alliance with Eli Lilly to share the output of these efforts.[25] The parent company, Millennium, retained the right to exploit all other (biochemically classed) genes that came from MBio’s sequencing efforts.

The establishment of that massive-scale sequencing capability for MBio (with half the funding provided by Lilly) and the sequence database it generated, then enabled Millennium to enter into two large, seminal, “second-generation” alliances. The first, also in 1997, was with Monsanto. Monsanto was in the initial phase of buying up seed companies for all of the major crops. Germline “gene therapy” is the heart and soul of plant genetic engineering. Monsanto needed to rapidly sequence the genomes of all of the major crops in order to not be pre-empted by its competitors. The fastest way to do this was, effectively, to enlist Millennium to do it for them. However, Millennium decided that it did not want to get into the agricultural arena, even via a subsidiary. So, instead, Monsanto established, literally next door to Millennium, a wholly owned research subsidiary (Cereon) to which Millennium transferred its sequencing and bioinformatics technologies for exclusive use in the field of agriculture.

The sequence database generated by MBio, and Millennium’s retained right to use the database for drugs other than secreted proteins and MAbs, then enabled (in 1998) the large alliance with Bayer. In this alliance, Millennium provided to Bayer (hundreds of) potential targets for small-molecule drug intervention defined by biochemical class (e.g., GPCRs, intracellular enzymes, and ion channels).

Why do I recite this history of Millennium first- and second-generation partnerships, particularly to you who lived and led the company?[26] I do so because I fear that in speaking of “The TRV Model” (based in “The Millennium Model”), too many current generation start-up biotech leaders (and their venture backers) fail to appreciate the subtlety and art necessary to practice “the Model”; as it were, they think it is rote. In reality, using large partnerships to create out-size value requires a keen sense of the external partnering environment, identifying the potential partner(s) which at that specific moment in its/their history has/have come to perceive a critical need for what you have, and crafting deals in which you, while you sell rights, retain the potential for value creation by, for example, retaining ownership of the knowledge. That, not a cookie-cutter model, is your legacy in this arena.[27]

==============================================

From John Maraganore

From: John Maraganore <jmaraganore@alnylam.com>
Sent: Tuesday, July 3, 2018 6:10 AM

All,

Great comments from Mark and Jeff and, as always, great insights and framing from Steve. (Steve- I still regard your funding scheme slide that you shared with me 15 years ago as the “bible” of biotech capitalization!)

I would just add the following points:

For the product platform approach, I’m very biased.... but the greatest value creation requires significant product rights retention. I agree that an effective way to fund is with partnerships that involve either geographic splits or profit sharing. However, it’s critical that the platform value is adequately recognized by the partner (e.g., SNY-REGN) to obtain the right economics and governance. Indeed, the greatest downside is loss of full control, which is an extremely expensive cost (sometimes priceless)! Alternatively, financing the company with a robust equity market and retaining full and global control of assets can be a vastly preferable approach, but does require an efficient product development/commercialization strategy such as rare diseases. Regarding the “Intel Inside” strategy, I’ve soured greatly on that approach over the years. It leaves a ton of value “on the table,” as best represented by the relative valuations (and share prices) of IONS and ALNY. –Just my $0.02

RESPONSE TO MARAGANORE

First, at least in my book, with what you have accomplished at Alnylam, you have proven to be the single greatest student and practitioner of the model.[28] How you have used partnerships to build Alnylam and, in particular, the evolution of your partnership structures, I find to be deeply instructive.

An important evolution in the history of the pharmaceutical industry was occurring shortly before the launch of Alnylam, a change which you recognized and capitalized on to the max. Leaders of big pharma, which were prototypically small-molecule chemistry companies, were recognizing that they had largely “missed the boat” on protein therapeutics and MAbs, which were now becoming billion-dollar drugs. As it were, swearing that they would never let this happen again, they sought to ensure that they would never be squeezed out of, and would make sure to get an early foot in the door on, any newly emerging therapeutic modality. At that moment in history, RNAi was just coming on the scene. The latent demand existed for Alnylam as the first mover to corner the market.

Unlike in the early days of recombinant proteins and MAbs, where the practice of the technologies required significant proprietary know-how, once published, RNAi was fairly straight forward to master. Therefore, the key for Alnylam was to broadly capture ALL of the relevant IP, effectively gaining patent coverage on any/all RNAi drugs. You rapidly achieved this goal. The net result: big pharma had to “come through” Alnylam if it was not to risk being squeezed out of the modality.[29]

This, I would argue, is what allowed Alnylam to establish its foundational deals with Novartis and then Merck.[30] These partnerships, which featured very large, up-front and near-term non-contingent funding, enabled Alnylam to cement its first-mover advantage.[31] They were the ladders you climbed from the dark cave (= negative cashflow) of start-up and then kicked away as you emerged into the sunlight of forward integration into proprietary product development.[32]

Since those early foundational deals, Alnylam has masterfully intertwined the art of combining raising equity dollars when capital markets are favorable with entering partnerships with increasing retained product rights.

With the recent approval of its first product (big congratulations), the challenge in front of Alnylam for the next several years is not to have its value equated with the DCF of the first product or two and/or EPS growth. What will power Alnylam into a future of truly out-sized value creation and sustainable independence is the renewable pipeline of more and more products. In your words John: “don’t turn the corner too soon.”

PS: Thanks for remembering the slide (which was jointly created with Jeff Tong) from lo those many years ago. For a stroll down memory lane, it is reproduced below.

======================================

Responses and Replies: Round 2

Jeff Tong

A few additional thoughts

“Intel Inside” companies-- I think Medarex and Abgenix are worth mentions here. That technology is now embedded today in multiple commercial products, and the productivity of the platform (+ the value of their own internal pipelines at the time) ultimately led to their acquisitions. An interesting case study in value creation - had Medarex not been acquired, what would it be worth today? (they contributed BMS’s anti-CTLA4 and PD1 programs). In contrast, had Abgenix fully played out the Vectibix story, it might have gone the other way (not a commercially successful, big product). At the end of the day, it’s still breakthrough medicines that drive the realization of long-term breakthrough value. But there are ups/downs along the way where many companies will exit along the way. And of course, public shareholders can enter/exit anywhere during the ride too by simply buying or selling shares. And conversely, what would BMS look like today w/out Medarex? Would it even still be independent? And Takeda without the Velcade/proteasome franchise & Entyvio?

Holtzman

In my response to Mark, I do discuss the Abgenix case (re: initially “Intel Inside” in the Japan Tobacco deal, followed by forward integration of Vectibix leading to the Amgen acquisition). I didn’t mention Medarex (albeit, I did note in passing its European spin-off, Genmab), the latter of which has arguably been an even more successful example of the “Intel Inside” strategy than Isis/Ionis.[33]

Tong

Your Footnote 11: Yes, many things had to go Agios’s way. One additional one that is important is that Celgene returned the rights to AG-120 (IDH1) after paying for nearly all of its development. This now provides Agios a fully owned product for which it has submitted an NDA many years before it would have otherwise been able to get to with its non-Celgene pipeline.

Holtzman

A picking mechanism that worked in the small company’s favor by being abrogated…

Tong

Footnotes 15/16: of interest, the acquisition of Leukosite actually had two unexpected fruits. The first of course was Velcade which Millennium enjoyed. The second turns out to be Entyvio which Takeda (post-Millennium acquisition) is now enjoying.

Holtzman

Another interesting tale.[34]

===================================

Levin

Hey Steve

I will never forget a couple interesting moments that fits into an overall comment I will make.  When we were negotiating with Wyeth to hold onto significant rights in the CNS area, we were in your office late one night doing an NPV calculation on the project. We, of course, came up with a very significant NPV - very negative. And the next day, went back to Wyeth and changed our strategy so we could use their dollars (not the rights) to maximize more valuable programs.

A few years later, we did a detailed NPV model around our entire Millennium pipeline along with a Monte Carlo analysis on likely pipeline scenarios over a 10 yr period. All scenarios indicated unlikely products in the next decade. Yet, our public value was extremely high. This discussion was key in launching our product acquisitions efforts ASAP.

NPVs on early stage companies, of course, are flawed yet at their highest level, integrated with Informed Monte Carlo analysis and practical R and D input, can give critical strategic input into BD and an overall company strategy.

So, my overall point, when building a Product Platform Company, a company should put in place an Overall Value Creation Model Pipeline tied to NPV of the Pipeline, tied to financial requirements, tied to BD strategy (know what hold, know what to deal, and for what value) and financing strategy in order to maximize value to patient and share price.

Always remember it is only a model and is only as good as the input, but using the input as a key tool in the overall gestalt of things will drive better decisions

HOLTZMAN

On NPV Analysis: You’ve all previously read my view on the role of NPV, etc. analysis for decision making; see, “ON NPV, DCF and IIR(ELEVANCE)”, BioCentury, September 26, 2016, and also available on this blog: https://leadershipandbiotechnology.blogspot.com/.

--It is fascinating to me that the majority of comments I got on this piece from industry colleagues focused on the first half of the article (re: the proposed alternative financial analysis of RU, RD, and WIYD) and also on the question of the cost of capital to the rich and poor man when playing the lottery. To me, the much more interesting set of reflections, the ones that provide the most food for thought, begin on page 2 with, “The Moral of the Story” where I address the dichotomy between a career vs. a vocation and a cast of mine that seeks a refuge from responsibility in the illusion that human action can be dictated by mathematical/logical analysis. --By the way, with respect to the quote from the piece that BioCentury chose as the headline pull-quote [“We are tricked into confusing the extraordinary precision of financial analysis with accuracy.”]: Dr. Tong, while you were in constant discussion about such things from 2001-2009 (and pretty much ever since, on reflection, it was you not I who coined this pithy summation…belated apologies for my lack of attribution to you in that article.

On That Night in My Office: for John and Jeff: we had, in rapid succession (early-1994 through mid-1996), knocked off partnerships with Roche (Diabetes/Obesity), Lilly (Oncology), Astra (Asthma), and Lilly (Cardiovascular), and had gone public. We were negotiating with Wyeth (CNS) and the question was whether to replicate the earlier deals (they fully funded discovery, they get the targets for small molecule discovery [with us getting royalties], and we retained the protein, MAb, antisense, and diagnostic rights) or, instead, it was time for us to “belly to the bar,” assuming more of the costs/risks of early discovery in exchange for more rights/a bigger piece of the action. We had postured this to Wyeth and, after much consternation, they agreed to it. In the “Oh shit!” moment Mark describes, Levin and Holtzman (aka the blind leading the blind) stood at a whiteboard doing the NPVs that Mark describes that led us to conclude that Millennium had not yet reached a point in its evolution where we could/should take on that risk. –So, I got on a plane the next day to Philly with my tail between my legs begging to return to the original deal structure. (Note: a year later, via the MBio subsidiary and its foundational deal with Lilly, we then did start “bellying to the bar” for half the costs of discovery [albeit, the cash to do so came from Lilly’s equity dollars for 20% of the sub].)

Other ground-breaking conclusions Mark and I reached that night:

 * never take equity dollars: they are always dilutive
 * never take debt: you always have to pay it back
 * never sell away rights: they are the source of your long-term value
 * never use your own money for early R&D: it is too risky and NPV negative

–From this we derived the “Gift Model” of Early Biotech Financing: find someone willing to give you large gifts of money with no strings attached: no stock, no repayment, no marketing/ profit participation rights. Arguably, that model has come to pass in three forms: (1) what I earlier called the “Rent a Program Model” (develop the program in a partnership using the partner’s money and then get it back for free when a new head of R&D takes over at the partner and orphans your program in his need to establish an identity in opposition to his predecessor; (2) the Gates Foundation; and, (3) “dumb money”, invested at ridiculous valuations in overheated equity markets.

More seriously, this goes to the whole issue of how you sequence the deals, or more accurately the nature of the deals that you do as your company evolves and the interplay of the same with equity…which is really what this whole dialogue has been about.

---

[1] Paradigm (“successful”) product companies would include Cougar, Medivation, Pharmasett, Pharmacyclics, and (in process) Sage and Intarcia. Note: the overwhelming majority end up being sold as, essentially, single-asset plays.

[2] Paradigm novel therapeutic modality platform companies would include the pioneers in rDNA technology for therapeutic proteins (Genentech, Biogen, Amgen, G.I.), the antibody companies (Centocor, Abgenix, Regeneron, Adimab), biologics from genomics (HGS, MBio, G.I. [discoverese], 5 Prime), antisense (ISIS/Ionis, Hybridon, Gilead [in its initial incarnation]), mRNA (Moderna), gene therapy (Generation 1: Cell Genesys, Somatics, GenVEC, GTI; Generation 2: Bluebird, RegenX, Uniqure), CAR-T (Juno), and CRISPR (Editas). (The companies that produced, at scale, the first-generation kinase inhibitors (Vertex [Novartis relationship] and Exelixis [GSK relationship]) are a variation on this.)

[3] Offhand, the author cannot think of any company of this type that “won” via a picking mechanism. On the other hand, Genentech (Roche), Amgen (Kirin), and Regeneron (Sanofi) have done just fine sharing all and splitting by geography.

[4] It is worth noting that, in the 1980s/early 1990s, the first-generation biotech companies had available to them alternative financing vehicles. Initially, these took the form of R&D Limited Partnerships, imported to the biotech industry from the world of gas and oil exploration by Bob Swanson (founding CEO) and Tom Kiley (founding GC and VP Corporate Development) of Genentech, that were heavily used by Genentech, Centocor, and others. These disappeared when part of the Tax Reform Act of 1986 eliminated the R&D tax credit. The next incarnation of the alternative, off-balance sheet vehicle was the special-purpose entities (with names such as SWORDs and SPARCs), invented by Stelios Papadopoulos and his crew at Paine Webber Development Corp, and heavily exploited by G.I., Centocor, Genzyme, and others. These financing vehicles allowed the companies to maintain and finance their product candidates with less dilutive equity-like dollars. In every case I can think of, when the product developed using the alternative financing was successful, the rights to the product were reacquired with shares in the “parent”; however, in virtue of the success of the product, the equity had appreciated significantly in value. As it were, equity was raised “today” at “tomorrow’s” higher, and thus less dilutive, share price.

[5] Paradigm Novel and High-Value Insight Platform Companies, Species A (insights into pathways, targets, etc. and their relationship to disease) would include the first-generation, large-scale genetics/genomics companies (Millennium, HGS, Incyte, and Celera). Variations on this theme are (or will prove to be) the new generation of “big data” plays (e.g., a database of the single-cell, RNA Seq profiles of every cell, under every possible condition; a database of proteomically defined pathways; the “cancer atlas”; etc.). 23andMe is a variation (and an instructive variation); see discussion below.

Paradigm Novel and High-Value Insight Platform Companies, Species B (insights based on a deep, disease-specific biology and related translational research/medicine expertise) would include the early CNS companies (Athena, Cambridge Neuroscience, Cephalon, Regeneron [in its first generation as a neurotrophic growth factor company]); a more recent example would be Agios (aberrant metabolism of the cancer cell); more recently still, Tango (synthetic lethality in cancer).

[6] For example, no sooner were the initial genetics/genomics companies (HGS, Incyte, Millennium) off and running than big pharma, led by Merck, formed a sequencing consortium. Similarly, NIH massively increased funding of the Human Genome Project to ward off the emerging threat of Venter/Celera. In the current period, Regeneron and GSK have formed a “syndicate” with other companies to fund sequencing the UK BioBank.

[7] Hence, HGS becomes Benlysta; Millennium becomes Velcade; Incyte hires Paul Friedman and become a small molecule, cancer, and immunology drug discovery/development company; Celera buys Axys (Mike Venuti) to become a small molecule drug discovery company.

[8] Indeed, Celera described itself as “The Bloomberg of Biotechnology.”

[9] 23andMe charges $100 (low price) to the consumer (large potential market) for its data. At, say, 5 million users (a number they have yet to attain), it is a $500 million/year revenue business, not a high-margin, multi-billion dollar biopharmaceutical company. Interestingly, 23andMe fueled its growth by adding their ancestry business: people (other than Trump supporters) like to know where they are from/their geographical/ethnic heritage, how Neanderthal they are, who their relatives are, etc. In this realm, 23andMe faces emerging competition (e.g., Ancestry.com) in what will prove to be a race to the bottom featuring falling prices and increasing (expensive) direct-to-consumer TV advertising. Nevertheless, from what I can tell, 23andMe sees this “consumer” business as a means to an end. That end is the “resale” of the genetic data, combined with phenotypic information (and the ability to recontact its customers to collect more phenotypic information), to the biopharma industry for drug discovery/development purposes. In other words, they are pursuing the business model of the early genetics/genomics company. Until the recent large deal with GSK (July 2018) I would have said that that business is not flourishing: very few biopharmas had been willing to pay the high prices (including downstream participation) being sought by 23andMe. And, interestingly, following the pattern set by the early genomics companies, we have seen 23andMe forward integrate into proprietary drug discovery. In March 2015, 23andMe announced the appointment of Richard Scheller (former head of discovery at Genentech) as CSO and head of their new proprietary drug discovery unit. -- Plus ça change, plus c’est la même chose.

Whither the new, emerging generation of “big data/AI/machine learning biotech” plays? —This story is still being written. Having a different ending than its predecessors will, I believe, be a function of whether a step change in computational power is occurring that transforms the key value driver of drug discovery and development from science plus art to technology plus engineering/computation. While the liberal-artsy in me screams that this day will arrive on the same day that computers write Shakespeare, as a self-aware technological Luddite, I know what I don’t know; hence, I make no prediction.

[10] As advised earlier, use a picking mechanism at your peril.

[11] The “poster child” Genus 2 Species B Platform Company  in the last decade is Agios which has, for the most part, successfully implemented the strategy. It is useful to see how many things had to break Agios’ way to achieve what it has. First, Celgene was in a very special period, and played a very important role, in the ecosystem of young cancer biology companies, of which Agios was the prime example. Second, when the first projects/biological approach at Agios failed, it had the great fortune to have on the team the company’s founder, Michael Su, who is an extraordinary leader and drug hunter. He identified the new biology for the company (IDH) and then drove the initial product discovery/ development programs. Next, the first two drugs, both now approved, emerged from the first high-throughput screens in the first three runs, featuring good potency, selectivity, and pharmaceutical properties. They had a world-class drug discover chemistry lead (Janeta Popovici-Muller) who made the molecules into drugs (and eventually medicines) and drove the program team. Finally, though saddled with a picking mechanism, they managed to get Celgene to give them an early pick.

[12] Paradigm examples: Genentech, Biogen, and Genetics Institute (recombinant proteins); Centocor (MAbs); ISIS/Ionis (antisense); Genetic Therapy Inc., Cell Genesis, Merlin, and Somatix (first-generation gene therapy companies; Alnylam (RNAi); and Editas and Intellia (CRISPR).

[13] This is most obvious in the case of the first-generation, recombinant protein/antibody companies. For example, Genentech and Biogen’s marketed products and pipelines today feature many that were in-licensed. However, the phenomenon also can be seen in the new wave of gene therapy companies. Precisely because the Therapeutic Modality is not fully “novel,” there are product candidates—in other companies or the academy—available for in-licensing. Examples include Voyager (Genzyme’s Parkinson’s program) and Spark (UPenn’s RP program).

[14] A caveat: some newer platform companies of this type (e.g., CAMP4, which is elucidating genetically implicated pathways with novel disease associations) explicitly are seeking to in-license existing product candidates for “re-purposing.”

[15] The obvious example of this was the dilemma faced by Millennium when it acquired Leukosite in late 1999. Leukosite brought with it CAMPATH for the treatment of CLL which was approaching submission for approval. Millennium could not afford to let its value be viewed through the lens of CAMPATH as: (1) its approval, or failure to be approved, was a binary event; and (2) the product was projected to be a “small” product, which alone could never justify Millennium’s valuation. Hence, Millennium downplayed CAMPATH as a product and, instead, encouraged the market to see the acquisition of Leukosite as an acquisition of clinical development capabilities which Millennium would need as it began to forward integrate with the products that would soon be emerging from its platform. This strategy proved successful. In the immediately ensuing genomics bubble of 2000, Millennium was able to maintain its reputation as the leading genomics (not CAMPATH) company and, in the course of that year, completed two stock offerings which, together, raised ~$1 billion.

[16] Millennium again provides the paradigm case, albeit somewhat serendipitously. The acquisition of Leukosite with CAMPATH (described in the preceding note), also brought with it “PS-341,” a molecule which Leukosite itself had come into 3 months before via its acquisition of ProScript. PS-341 played essentially NO role in the Leukosite acquisition. In fact, the molecule was almost designated for the chopping block because there was little faith that an inhibitor of the (biologically fundamental and ubiquitous) proteasome could have a meaningful therapeutic index without deleterious side effects; and, furthermore, that a molecule featuring a boronic acid (the active ingredient in rat poison) would survive tox. In the event, Julian Adams refused to let PS-341 die until leadership became supportive. PS-341, of course, has come to be known as “Velcade.”

[17] In fairness, in the same timeframe (1993/1994), HGS implemented a similar strategy with its large, foundational SmithKline partnership.

[18] Or a couple of products at most; see Amgen-Kirin which encompassed both EPO and GCSF.

[19] Of course, the technology was spun off into a subsidiary, Abgenix…which, in the passage of time, became much more valuable than its parent.

[20] In one sense it did, albeit by geography, via its broad deal with Japan Tobacco.

[21] Which came to define Abgenix (see the main text for the challenge for a platform company of not becoming defined by its first product) and was essentially the sole basis of its acquisition by Amgen.

[22] The community did not yet have SNPs (much less whole genome sequencing); so, the positional markers used to map the loci of “disease genes” were (known as) “micro-satellites.”

[23] Since there were not yet commercial suppliers of transcriptional profiling and proteomic technologies (e.g., Affymetrix was just getting off the ground), Millennium had to develop these technologies itself. (This led to the interesting business challenge, by the company’s third anniversary, of financing an annual CAPEX budget of $50 million/year.)

[24] To which we recruited two extraordinary young leaders: John Maraganore and Kevin Starr.

[25] Which, to my knowledge, was the first deal featuring a “picking” mechanism. Not surprisingly, given that Maraganore and Starr are wheeler-dealers of the first order, MBio began trading with Lilly its “draft picks” for future proteins to be named later in the draft…

[26] Note: I have left out the formation of Millennium’s other spin-out, Millennium Predictive Medicine—which exploited Millennium’s retained rights to use all of the technologies and information generated in all of its alliances in the field of pharmacogenetics, pharmacogenomics, and diagnostics (and entered into major alliances with Becton Dickinson, BMS, and Roche). I have also left out the third-generation alliances in immunology (with Aventis) and diabetes and obesity (with Abbott), which were based on the earlier alliances (with Astra and Roche, respectively) in these areas which, by 1999/2000, had expired. –The biotech business model of “rent a program” is, to my mind, underappreciated.

[27] I can’t tell you how many brilliant 28-year-old MD/PhD and first-time company CEOs have offered to educate me on the critical need for start-ups to do a “Celgene-Agios deal.”

Since you started your response by citing your favorite philosopher, Ray Nitschke, PhD Linebacking, University of Green Bay, in this context let me leave you with a relevant quote from his German cousin, Friedrich Nietzsche, PhD Philology, Universities of Leipzig and Basel: “One repays a teacher badly if one always remains nothing but a pupil.” –Thus Spake Zarathustra

[28] Among the other first- or second-generation lineal descendants of Mark Levin who proved to be masters, my list includes: Laurence Reid (major deals at Millennium Predictive Medicine with Becton Dickinson, BMS, and Roche, and Maraganore’s partner in crime for a number of the Alnylam partnerships); Kerry Reinertsen (Laurence’s partner on those Millennium Predictive Medicine deals and leadership of subsequent major deals for Regeneron); Alan Crane (key leadership role on Millennium’s deals with Monsanto, Bayer, Aventis, and Abbott); Jeff Tong (key leadership role on Infinity’s major deals with MedImmune and Purdue/Mundipharma and, most recently, launch of Ambys with blow-away, foundational Takeda deal); and John Evans (key role in Infinity-MedImmune alliance and leadership of Agios-Celgene partnership).

[29] Lost in the history of our industry is the fact that in the late 1980s/early 1990s, there was one head of R&D in big pharma, namely Max Link at Sandoz, who saw the future importance of the emerging new modalities. He took Sandoz into major early deals with acquisitions of pioneer companies in the fields of gene therapy, antisense, cellular therapy, and xenotransplantation. A key reason for his early move into these areas was Sandoz’s leading role and, effectively sole pharma presence in the field of organ transplantation via Sandimmune (cyclosporine). In a similar, but much less expansive and visionary way, Baxter’s move into xenotransplantation, via its acquisition of DNX’s xenotransplantation program (to form Nextran), was motivated by its presence in renal/peritoneal dialysis and the potential for the disruptive impact of genetically engineered pig kidneys for xenotransplantation.) In the event, Link has proven correct, albeit some 30-40 years early.

Moving to the current day, big pharma/big biotech’s imperative not to find itself squeezed out of an emerging, new therapeutic modality is recapitulated and manifest in the large deals done with the (newer generation of) gene therapy, CAR-T, and gene editing (e.g., CRISPR) companies.

[30] That foundational deal with Merck should not be confused with the subsequent deal with Merck in which Alnylam purchased Sirna.

[31] At the time of the announcement of the Novartis deal, Alnylam made much of the fact that there was a “picking mechanism” and that, therefore, it had retained significant product rights for future value creation: Alnylam wasn’t merely doing contract research for potential downstream milestone and single-digit royalties. In fact, however, this was, I would argue, largely optics for the capital markets. While, to my knowledge, Alnylam has never disclosed the mechanics of the Novartis dea- picking mechanism, it was nowhere near what MBio had with Lilly (alternating picks from the get go). Depending on with whom you speak, Novartis had somewhere between the first 15 and 40 draft picks…so much for Alnylam’s retained product rights. (To which, Maraganore could reply, “See, NFL, 2000 Draft, Pick 199”.)

[32] If curious, to unpack this sentence, see Wittgenstein, Tractatus Logico-Philosophicus, Proposition 6.54 (The Wittgensteinian Ladder) and Plato, The Republic, Book VII (The Allegory of the Cave).

[33] Following the histories and fates of the first generation, pioneer transgenic animal companies (which, among other things, developed genetically engineered mice capable of producing fully human antibodies) is quite instructive. The first transgenic animal company, DNX (initially known as Embryogen) got off the ground in 1985/86. The next two, both formed in late-1987, were Cell Genesys (which eventually spun off Abgenix) and GenPharm. DNX was “low profile,” initially funded with < $500,000 by an angel investor with a 32-year-old first-time President. Cell Genesys and GenPharm were high-profile start-ups featuring backing from all-star biotech VCs (e.g., Mayfield, Kleiner Perkins, and Avalon) and featured a number of Genentech alums in senior management positions. All three companies had programs to develop transgenic mice with “human immune systems” capable of generating fully human MAbs. DNX called its mouse the HuMab Mouse (never trademarked, see below). DNX’s approach involved dissecting out from human chromosome spreads the three chromosomal regions encoding the heavy and light chains and introducing these into mouse zygotes via pronuclear microinjection. Had the approach worked it would have produced a mouse with a human immune system in one fell swoop. –Thousands of chromosomes and zygotes later, with no success, DNX dropped the program.

In contrast, Cell Genesys’ Xenomouse and GenPharm’s HuMab Mouse (it was never clear whether they simply copied the name from DNX or they happened upon the same name independently), were created by the more laborious approach of homologous recombination employing Bacterial Artificial Chromosomes (BACs) to deliver fragments of the coding regions. Since a BAC could only deliver a portion of the relevant loci, this meant that multiple mouse lines had to be established together carrying all the relevant genes and then backcrossed over multiple generations to get to a single mouse (line) carrying the entire repertoire. –In this case, the (two) tortoise(s) beat the hare.

DNX went on to establish two business lines: the first, through the acquisition of pre-clinical pharmacology/toxicology, provided services to pharma/biotech companies including drug testing in novel transgenic animal models of human disease (e.g., transgenic mice with human cholesterol profiles) and a transgenic animal production service. The second business continued in the tradition of the HuMAb mouse effort to use transgenic animals to generate homegrown pharmaceuticals, re: transgenic pigs with recombinant human hemoglobin for use as an acellular oxygen carrier (aka blood substitute) and transgenic pigs with organs that could be used for xenotransplantation to humans without immunological rejection. DNX went public in 1991 but by mid-1993 it became evident that its stable but low-revenue (~ $30 million/year)/low-margin testing business and cash-burning, high-risk transgenic pig program were incompatible. The pig programs were sold to Baxter in early-1994 where the xenotransplantation program made great progress until, a few years later, it was stopped in its tracks with the discovery of porcine retroviruses (a total show-stopper coming not that many years in the wake of the HIV epidemic). The testing business was eventually sold to a CRO with the transgenic animal production service subsequently being further sold to Xenogen…and then finding its final resting home at Taconic Farms.

Cell Genesys went public in early 1993. However, by 1996, it spun out Abgenix with the Xenomouse technology in order to focus on its gene therapy and cancer vaccine programs. As noted, Abgenix deployed the technology internally to develop Vectibix which then became the focus of the company leading to its acquisition by Amgen in 2006 for ~ $2.2 billion. Cell Genesys, with serial failures in its retained programs, went through many mergers/ acquisitions eventually being sold to BioSante in 2009 for $38 million.

With respect to GenPharm, as it so happened GenPharm and Cell Genesys shared a scientific consultant. Shortly after GenPharm filed with the SEC to go public in 1994, Cell Genesys slapped it with a lawsuit for trade secret theft and the filing of patents based on the stolen technology. The litigation overhang killed GenPharm’s IPO. GenPharm countersued for damages. While eventually (1997) it won a settlement of $15 million from Cell Genesys/Abgenix, it was a pyrrhic victory. On the verge of bankruptcy, the company was sold to Medarex later that year for $65 million. Medarex then went on to pursue an “Intel Inside” model entering into multiple corporate partnerships to generate fully human MAbs for their partners, including its work for BMS that resulted in PD-1 and CTLA4 products and BMS’ purchase of Medarex in 2009 for >$2 billion. In 1999, Medarex established a European affiliate, GenMab, that also pursued the Intel Inside model with multiple partnerships,.with royalties. Today, with royalties from Arzerra (marketed by GSK) and Darzalex (marketed by J&J), GenMab stands as probably the most successful “Intel Inside” model company with a market cap of ~$10 billion.

And then another young biotech company came along in the early/mid 1990s with a “better” mouse. It deployed it early on in the Intel Inside model as a source of current revenues, but it also used it internally to develop its own proprietary products. –That company goes by the name Regeneron…

---Bonus points to the reader who can name at least 4 rock bands, other than Crosby, Still, Nash and Young, that resulted from the backcross of The Buffalo Springfield X The Birds. Extra bonus points if you are under 40 years old (excluding my son Bryan for whom this music was the lullabies to which I played him to sleep).

[34] As Levin and Maraganore will attest, while Millennium accorded zero value to LKST-341 (Velcade), it did have interest in LKST-02. However, the molecule was in a very early stage, was hung up and going sideways in an unproductive partnership with Genentech, and was being produced in a cell line that was going to have to be rederived, sending the (newly produced) antibody back into pre-clinical studies. Nevertheless, the program refused to die, the partnership with Genentech was terminated, the new cell line and antibody were generated, and in 2014 (14+ years after the Leukosite acquisition by Millennium and 6 years after Takeda’s acquisition of Millennium), the molecule formerly known as LKST-02 was approved by the FDA and brought to market by Takeda as Entyvio…