Drugbaron Blog

June 4, 2015 no comments

A DrugBaron Glossary

To mark the fifth anniversary of this blog, as well as the launch of the new website, DrugBaron has collated some of his favorite phrases, each encapsulating an important concept in pharma R&D. Most of these terms were initially coined by DrugBaron in these pages, but have entered the wider biotech lexicon, and frequently feature in discussions of R&D strategy on Twitter, in articles and at meetings.
Here then is the definitive guide to the meaning behind the label:
Acceptable Failure     There is more than one way to fail, and they are not all equivalent. In drug development, sometimes the candidate fails for reasons that could never have been predicted.   The team did all the right things. With luck they managed to demonstrate the hidden problem with the asset quickly and cheaply. Failure of this kind, which DrugBaron labeled “acceptable failure”, should be embraced, even celebrated (though not actually encouraged, since even acceptable failure creates losses rather than value).
In contrast, failure due to poor experimental design, bad implementation or other avoidable mistakes by the team trusted to progress the asset, is the cardinal sin. There are enough unavoidable reasons for programs to fail without adding wholly avoidable ones.
The challenge for the R&D management is to distinguish the two. Teams always argue the failure was unavoidable, and the complexity of biology always provides a hiding place. And the best solution to that problem is to create the right incentive structure, overcoming the inherent cultural bias towads continuing projects.
Asset-centric     Asset-centric is a term coined by my partners at Index Ventures to describe companies that are focused on the development of a single asset (companies that have comprised the majority of Index investing since 2005).

And that’s all it means. It doesn’t mean the asset is more important than the people. Quite the reverse – the limiting factor for pursuing an asset-centric strategy is the availability of great teams who can take those single assets forward.
There are two principle advantages for asset-centric companies. First, with only one asset there is no insurance policy. No second asset to fall back on the event the lead fails. That sharpens the focus of the team (whose returns depend entirely on the lead asset) and prevents wasted dollars developing an asset that would never have made the grade as a lead asset.
Second, with only one asset to develop it makes no sense to build specific infrastructure.   Asset-centric companies are therefore typically also virtual (or at least semi-virtual), with all the advantages that implies.
A decade into investing according to the assent-centric principles, enough returns have been realized to start to judge the value of the approach in the real world. With thirteen high-value exits from Index Ventures alone (including, most recently, the sale of XO1 to Janssen) asset-centric has already proved its worth and clearly here to stay.
Asset Favouritism     The perceived benefit (but actually hidden penalty) associated with the following the crowd. So many times we see everyone seeking a solution to the same problem all in parallel. From one perspective, that makes total sense when a breakthrough has made a particular problem more soluble. But from a market and commercialization perspective, it is exactly the wrong the thing to do. The chance of wining may be a little higher, the spoils of victory will be shared much more widely.
The original example DrugBaron cited was more than 200 product candidates in phase 2 for rheumatoid arthritis which he contrasted with the 6 active clinical programs in sepsis. Does the perceived increased chance of success in RA really outweigh the increased competition?
Arguably, today the same applies to immune-oncology assets. There is no doubt that co-opting the immune system to attack tumours has the potential to revolutionise outcomes in cancer. But with hundreds of competing technologies jostling for position, its unclear if there will be an overall winner – and if there is not, the spoils will be shared between many players. The optimism AstraZeneca expresses about its PD1 pathway antibodies exemplifies this perfectly: they project a $6-7billion peak annual sales for their antibodies, which seems perfectly reasonable for the dominant exemplar of this mechanism of action, but with two anti-PD1 antibodies already approved (Opdivo™ from BMS and Keytruda™ from Merck) that seems optimistic.
Busters     We are all familiar with the term “blockbuster” to refer to drugs that sell at least $1billion a year. R&D strategies across the industry remain clearly focused on delivering such blockbusters – whatever the cost.
Busters, by contrast, are the polar opposite. These are drugs that get approved but despite expensive launches fail to achieve material sales. For companies with one or only a handful of products, they can live up to their name quite literally, as Dendreon discovered with Provenge™.
The root cause of the problem lies in an outdated view of approval as the route to commercial success. Before the turn of the millennium, any approved drug was virtually guaranteed to provide a healthy return on its development and launch costs. But things have changed. For many indications, the bar has been raised considerably with decent treatment options available at generic prices. Unless the competitive advantage of the newcomer is compelling, approval, based on safety and efficacy versus placebo, may be wholly insufficient to secure any meaningful sales at all.
Death Spiral Pharma      DrugBaron coined this term to describe publicly-traded pharma companies who get caught in the trap of playing safe. Because drug development is inherently risky, and involves taking risks you didn’t even know were there, failure comes with the territory.   The only way to avoid failure is by doing things that have been done before.
But while the “me too” approach largely eliminates the risk of technical failure, it only swaps it for the risk of commercial failure. And commercial failure is worse than technical failure because it occurs later, when more capital has been expended (see Busters above).
For some public companies, though, that apparently unfavorable swap is a swap worth taking. Commercial failure comes later than technical failure, so it kicks the problem down the road, and perhaps more tellingly it occurs slowly over months and years, rather than in a single press release announcing a failed trial.
The death spiral kicks in when, after a period of repeated failure in the clinic, management dare not be exposed to another expensive flop. They avoid that by eliminating technical risk, exchanging it not so much for commercial risk as for inevitable commercial failure – but at least a few years from now. The term was initially coined to describe the situation at AstraZeneca (LSE:$AZN) when they spent a billion each on a prescription fish oil and a fifth-to-market long-acting beta agonist – assets that simply couldn’t fail in the clinic, but (at least in DrugBaron’s view) seemed equally unlikely ever to succeed in the marketplace. Today, it might also apply to Eli Lilly, GSK and arguably even Pfizer – although Merck (who were so labeled in the original 2013 article) may have escaped the trap since appointing Roger Perlmutter to head up R&D.
Idea Bubble     Assets classes are not the only things that can suffer bubble-like behavior – when prices become completely detached from value. Ideas can exhibit the same tendancy, when the assumption of validity of a hypothesis becomes completely detached from the data.
DrugBaron coined the term to describe the persistence of the amyloid hypothesis in Alzheimer’s Disease, which has survived repeated failures in huge, late-stage clinical trials with antibodies that clear amyloid in the CNS. Despite these failures, adherents (DrugBaron used the term acolytes to reflect the near-religious belief) question everything except the hypothesis, wondering if a different antibody, a different dose or a different patient population might yield a different response. The rational observer would of course consider such possibilities, but consider also the possibility that the central hypothesis is simply wrong.
Biology (and hence drug development) is particularly susceptible to such idea bubbles because of the sheer complexity.   It is always possible to construct an argument to explain away any particular observation. And it is these shifting sands that allow long-held hypotheses to survive the assault of even compelling counter-evidence. Much like post hoc data analysis, though, changing the goalposts after the new data emerges is only fooling yourself, and paves the way for ever greater waste of R&D dollars pursuing a doomed hypothesis.
Incremental Innovation     Drug development is no different to any other technology: innovation proceeds mostly by evolution with only an occasional dose of revolution.   Unlike most other technologies, though, the ultra-long development cycle of pharma R&D mandates at least the hope of revolution when starting out on the long and winding path.
Its easy to think that small changes (like reformulated drugs, for example) can yield high returns when you see global players such as AstraZeneca buying companies like Omthera (with its fish oil product) for a billion dollars. But such deals depend on being very close to market, when the acquirer can finely judge the commercial proposition following launch.
For an early stage project, in discovery or preclinical, such meager differentiation makes no sense at all. Being five to seven years from launch makes it impossible to determine if such a wafer-thin USP will still make commercial sense in that distant marketplace. And the chances are it will not, because the bar for commercial success is rising all the time as new products come to market and older products become generic (and hence much cheaper).
Incremental innovation may seem safer (and if you are in a death spiral, it may be the only option) because the technical risk is lower. But the correspondingly higher commercial risk makes it a poor trade. If your preclinical product candidate doesn’t have the potential to change the world, its time to go look for something better to work on.
Kill the Losers     DrugBaron coined the term “kill the losers” to contrast with the “pick the winners” strategies adopted by some pharma companies, following advice from expensive (and mis-guided) consultants. The essence of “pick the winners” is to select early the few projects with the best chance of success and invest heavily in them from day one. The supposed advantage of such an approach is that it dedicates the resources to the best projects and allows control of R&D investment simply by regulating the number of projects selected.
However, its value is predicated on the assumption is possible to pick winners. The trouble is, early-stage drug development operates in a “low validity environment” – in other words, there simply isn’t enough information available to reliably judge which projects will succeed (given enough time and resource) and which will fail. Its not a matter of being smart enough – its just unknowable.  DrugBaron likened the problem to that of escaping from a maze in a game of Dungeons and Dragons.
The alternative is to adopt the opposite strategy and deliberately not select projects early. Instead keep as many going as possible by restricting the available cash devoted to each. The only experiments done in each project are those necessary to decide whether to progress to the next stage. Any that lose their shine are immediately stopped.   As a result, the decision on which projects survive is delayed until more information is available – a process of natural selection rather than central command-control.
Implementing “kill the losers” in practice is harder than it sounds. Human factors and cultural bias make killing projects very difficult. Unless appropriate incentive structures are in place, starting with lots of projects simply ends up with too many progressing, insufficient resources in any of them and zombie-projects shambling along consuming unproductive capital.
Unearned premium     The economics of drug discovery depends on patents to protect expensive new investions. Patents yield monopolies, which allow inventors some protection from competition, and hence to charge higher prices than would otherwise be possible. This provides the return on investment in the original R&D.
The problem is that pricing in monopoly situations is always challenging, but it is doubly so in healthcare. Normally, the price a patent-holder can charge is limited by the degree of improvement he offers over other (potentially unpatented and cheaper) solutions for the same problem. People will pay a little more (a premium) for a vacuum cleaner thats, say, lighter or quieter – but not much more because an old-style product still does a pretty good job.
Not so in healthcare, it seems.
The public, at least in rich countries, demands the best medicine no matter what the cost (usually because the patient pays little or none of the cost directly – with the tab picked up by private or public insurers). This allows companies with patents to charge a much larger premium than the often incremental improvement may merit.
This is particularly visible with cancer drugs, where the best agent might give 2 or 3 months extra life (or maybe even extra time before further progression of the cancer, without extending survival at all) – and yet command million-dollar premiums.
These premiums are not objectively justified by the degree of benefit, and hence earned the tag from DrugBaron of “unearned premium”.
Unearned premiums are restricted to cancer. Crestor™ resuvastatin from AstraZeneca, a cholesterol-lowering agent, may have the largest unearned premium of all, sustaining $5billion in annual sales, with little or no additional benefit of the now-generic atorvastatin.

Yearly Archive