How To Avert Biopharma's R&D Crisis

Using target-based drug discovery to produce blockbusters on demand has led the biopharma industry down the wrong path. It’s time to get back to basics and focus on breakthrough science. The author suggests a formula for success.

Using target-based drug discovery to produce blockbusters on demand has led the biopharma industry down the wrong path. It's time to get back to basics and focus on breakthrough science. Here's a formula for success.

by Bernard Munos

Pharma's attempts to rationalize R&D have plunged the industry into a severe crisis, but it's not too late to beat back the commoditization of the industry.

Drug companies need an innovation model that works, and is geared towards breakthroughs, not blockbusters.

Scientists must be free to pursue unfettered science, and funding should be restricted to breakthroughs, which will help make innovation affordable again.

Disruption should be part of the mix, and should be used constructively to open opportunities, and help produce a plan that will convince investors that better days are indeed ahead.

2010 was not a good year for Big Pharma innovation. Only six new drugs were licensed by FDA to the 12 largest pharmaceutical companies, which marks another low in the long declining trend of new drugs brought to market. ( See Exhibit 1.) Ironically, companies' pipelines are bulging. According to the clinical trials web site of the National Institutes of Health, over 3,000 molecules are in development in more than 16,000 clinical trials. But only a trickle from this large pool is ever submitted for approval, with regulatory applications for new molecular entities dropping from around 45 in the mid-90s to just 23 in 2010. Adding to the pain, success rates have plummeted across companies' pipelines, and now stand around 5% for a compound entering clinical development, down from 20% a decade ago.

To address these challenges, pharma leaders have implemented a series of policies, which, though well-intended, have had the unfortunate consequence of taking R&D off course and precipitating the most severe innovation crisis in industry history. Efforts to produce blockbusters on demand have not delivered; the Taylorization of research has poisoned creativity; portfolio management has failed to mitigate risk; and mergers and acquisitions have wasted half-a-trillion dollars of investors' money with no tangible benefits. The combined impact of these missteps has brought the industry to the verge of irrelevance. ( See Exhibit 2.) If sales of patented drugs only meet 10% of Americans' pharmaceutical needs, as is forecast to be the case by the middle of the decade, does the industry still matter?

And yet, it shouldn't be this way. Even as the industry faces commoditization, science marches forward, producing novel tools and discoveries, such as synthetic biology, tissue engineering, stem cells, and nanotechnology, that are waiting to be translated into useful therapies. In bygone eras, pharmaceutical companies would have seized these opportunities, and turned them into breakthrough treatments. Today, however, much of drug research has been redirected toward incremental, low-risk approaches, or marketing support relabeled as R&D. Despite much talk about a return to innovation, the data from the NIH clinical trials website show that many Big Pharmas routinely sponsor dozens of Phase III trials (as many as 60 in some cases) for their most popular drugs. This is well beyond what is required to ascertain the efficacy and safety of the drugs, and suggests that the purpose of many expensive trials may be more to prop up the sales of aging products than create new cures.

To avoid marginalization, and reclaim its spot as one of the great contributors to human welfare, the biopharma industry must change course and embrace an innovation model that works. Historically, biomedical innovation has flourished along four different pathways: high-risk, unconventional research; patient-oriented research; user-driven innovation; and disruptive thinking. Any one of these could be harnessed to create the next decades' new medicines.

For most of the 20th century, biopharmaceutical R&D embraced risky, unconventional science. This approach paid off handsomely, producing the many breakthroughs that have defined drugmakers' cultures and decorate their corporate lobbies. These include, among others, Eli Lilly & Co.'s production of insulin in 1924; the industry's efforts to make antibiotics at the beginning of World War II; the embrace of biotechnology; and the mastering of monoclonal antibodies.

Another model, patient-oriented research, was pioneered by NIH after WWII. It relies on astute clinical observations by physician-scientists, which can in turn lead to novel hypotheses and treatments for diseases. Over the years, it has led to its share of breakthroughs such as the infectious nature of peptic ulcers, the use of nitrogen mustard in treating cancer, the biology of folic acid and development of methotrexate, and the use of thalidomide as an anti-angiogenic agent. Surprisingly, little patient-oriented research is done nowadays, even at NIH, which has moved upstream into basic biology. Perhaps its laborious nature and lengthy timelines have caused this disaffection, although the internet could do much to revive it.

According to a 2006 study published in the journal Pharmacotherapy, however, the bulk of therapeutic innovations seem to come from physicians trying to help patients for whom the standard of care has failed. Combining their knowledge of disease, patient, and pharmacology, they customize treatments using drugs approved for other purposes. Indeed, hardly a week goes by without an announcement that a molecule approved 30 or 40 years ago has been found to work for an unrelated indication. Still, despite these successes, little has been done to harness this user-driven innovation, which Eric von Hippel, PhD, a professor of technological innovation at Massachusetts Institute of Technology's Sloan School of Management and a coauthor of the study, calls the "dark matter of innovation."

Finally, as scientists at DARPA, the innovation engine of the US military, have shown repeatedly, sometimes the most stunning advances comes from doggedly pursuing goals that cannot be reached with incremental approaches. DARPA's small band of 140 "mad scientists" work on a shoestring budget, with no facilities of their own, and share an obsessive focus on achieving the impossible. Over the decades, this approach has delivered spectacular results, such as the creation of global-positioning technology, night vision, and the Internet. Among their life science programs: the development of prosthetic limbs that are controlled by the brain; the creation of therapies to instruct the regrowth of limbs lost to injury; and even projects to shrink the drug R&D timeline from 10 years to 14 weeks.

Absent from this list is the target-based drug discovery model -- the one currently favored by the drug industry despite its meager output. It rests on an elegant idea, but, unfortunately, one whose time has not yet come. Picking a drug target from a cluster of pathways is possible, but predicting what happens when one modulates it is far more difficult, because it is very rare that inhibiting or activating one cellular pathway won't spark a cascade of biological effects elsewhere. Since many of these effects are caused by biological processes that have yet to be discovered, target-based drug discovery is unlikely to live up to its promise until these knowledge gaps have been eliminated. The target-based model is also ill-suited to translate discoveries that are not linked to discrete targets, such as harnessing the power of stem cells to treat cardiovascular disorders or Parkinson's disease.

Regardless of the innovation model they choose, pharma leaders must refrain from regimenting the R&D process. Successful science cannot be shoe-horned into a one-size fits all approach, as it is replete with unique challenges that do not lend themselves to process optimization. Indeed, the technical hurdles associated with turning the discovery of monoclonal antibodies into commercial products are vastly different from the challenges associated with making stem-cell-based therapies or RNA interference medicines. To succeed, scientists need the freedom to stray into uncharted territory and try unorthodox approaches. The management processes in force must be loose enough to accommodate ambiguity, the crossing of boundaries, and orthogonal thinking.

Pharma leaders should also accept that the elaborate models to predict blockbusters and net-present-values have an abysmal record, and are useless to guide R&D spending. They are even pernicious as they are used to justify steering funds toward "safe" incremental projects of dubious clinical utility. Rather than look for blockbusters, drug companies should concentrate on finding breakthroughs. Breakthroughs do not need to be predicted because they represent such therapeutic leaps that they stand out, with compelling data quickly convincing physicians, patients, and regulators of their clinical utility. That consensus in turn paves the way for their commercial adoption. In addition, breakthroughs, by definition, often have a higher probability of success, and their clinical superiority can be demonstrated with much smaller trials and lower R&D costs.

It is no coincidence that Novartis AG, the company that has adopted a breakthrough strategy and banned the use of sales forecasts in early research, has become the most innovative company in the industry, introducing 20 new drugs in the last 15 years, about 50% more than its nearest competitors. There are sometimes concerns that not enough breakthroughs can be found to support such a single-minded focus. This overlooks the fact that if drug companies only fund breakthrough therapies, their supply will expand to meet demand -- or much money will be saved. To put it differently, the industry only produces marginal innovation because it funds it.

But producing innovation is only part of the challenge. It should also be affordable. As the number of new drugs dwindles, their fully allocated R&D cost soars, and is now well into multiple billions of dollars. Despite the industry's claim, process optimization tools such as six sigma have not been very useful in keeping these costs in check, but a breakthrough approach would help. According to Hans-Georg Eichler, MD, Senior Medical Officer at the European Medicines Agency, half of Phase III trials fail for lack of safety or efficacy. Clearly these compounds did not progress on the basis of their superior clinical profile. Had drug companies required them to first satisfy a breakthrough benchmark they would have saved vast amounts of money.

Alternative research models can also help. Precompetitive collaboration slashes the cost of upstream science. Lilly's Chorus model, for instance, cuts the time and cost of reaching proof of concept by 80 %. ( See "Lilly's Chorus Experiment," IN VIVO , May 2007 (Also see "Lilly's Chorus Experiment" - In Vivo, 1 May, 2007.).) Multiple open-innovation platforms show cost savings of up to 90%. Yet, according to a recent survey published by Chemical & Engineering News, most drug companies continue to cling to their fully integrated models.

In the mid-1990s, the pharmaceutical industry took a wrong turn. Many of its research divisions lost their independence and the freedom to pursue unfettered science. They were asked instead to be responsive to the market, and produce blockbusters that would sustain their companies' marketing franchises. To facilitate the attainment of these goals, R&D groups were reorganized to resemble assembly lines. The routinization that followed has debased innovation, and put the industry's future in jeopardy. Bringing breakthrough innovation back to drug R&D requires a return to evidence-based innovation models. The industry must abandon unsupported assumptions about how to produce innovation, and embrace what research has shown to work. Exhibit 3 summarizes some of the key priorities for that journey. It is unlikely that most companies can retool themselves before patent cliffs take their toll. But by starting on the journey, they might persuade badly battered investors to stick with them until the end of the trip.

Bernard Munos

Bernard Munos is a former advisor for corporate strategy at Eli Lilly and is now the founder of the InnoThink Center for Research in Biomedical Innovation. Email the author at: [email protected].