a revolution in r&d - boston consulting group · pdf filefor a complete list of bcg...
TRANSCRIPT
AmsterdamAthensAtlantaAucklandBangkokBerlinBostonBrusselsBudapestBuenos AiresChicagoCologneCopenhagenDallasDüsseldorfFrankfurt HamburgHelsinki
Hong KongIstanbulJakartaKuala LumpurLisbonLondonLos AngelesMadridMelbourneMexico CityMilan MonterreyMoscowMumbaiMunichNew DelhiNew YorkOslo
ParisSan FranciscoSão PauloSeoulShanghaiSingaporeStockholmStuttgartSydneyTokyoTorontoViennaWarsawWashingtonZürich
www.bcg.com
BCG
A Revolution in R&D
H O W G E N O M I C S A N D G E N E T I C S A R E T R A N S F O R M I N G T H E B I O P H A R M A C E U T I C A L I N D U S T R Y
BCG REPORT
The Boston Consulting Group is a general management consulting firmthat is a global leader in business strategy. BCG has helped companiesin every major industry and market achieve a competitive advantage bydeveloping and implementing unique strategies. Founded in 1963, thefirm now operates 51 offices in 34 countries. For further information,please visit our Web site at www.bcg.com.
The Boston Consulting Group publishes other reports
that may be of interest to senior health care executives.
Recent examples include:
The Pharmaceutical Industry into Its Second Century:
From Serendipity to Strategy
A report by The Boston Consulting Group, January 1999
Ensuring Cost-Effective Access to Innovative Pharmaceuticals:
Do Market Interventions Work?
A report by The Boston Consulting Group and Warner-Lambert,
April 1999
Patients, Physicians, and the Internet: Myth, Reality, and Implications
A report by The Boston Consulting Group, January 2001
Vital Signs: The Impact of E-Health on Patients and Physicians
A report by The Boston Consulting Group, February 2001
Vital Signs Update: The E-Health Patient Paradox
A BCG Focus by The Boston Consulting Group, May 2001
Vital Signs Update: Doctors Say E-Health Delivers
A BCG Focus by The Boston Consulting Group, September 2001
In addition, BCG’s Health Care practice publishes Opportunities for Action
in Health Care, essays on topical issues for senior executives.
For a complete list of BCG publications and information about
how to obtain copies, please visit our Web site at www.bcg.com.
www.bcg.com
PETER TOLLMAN
PHILIPPE GUY
JILL ALTSHULER
ALASTAIR FLANAGAN
MICHAEL STEINER
A Revolution in R&D H O W G E N O M I C S A N D G E N E T I C S A R E T R A N S F O R M I N G T H E B I O P H A R M A C E U T I C A L I N D U S T R Y
N O V E M B E R 2 0 0 1
2
© The Boston Consulting Group, Inc. 2001. All rights reserved.
For information or permission to reprint, please contact BCG at:E-mail: [email protected]: 617-973-1339, attention IMC/PermissionsMail: IMC/Permissions
The Boston Consulting Group, Inc.Exchange PlaceBoston, MA 02109USA
Credits: Left cover photo by Bob Waterston, Washington University, St. Louis, Missouri. Used by permission. The photo shows a bird’s-eye view of one room in the DNA sequencing facility at the Whitehead Institute Center forGenome Research.
Table of Contents
ABOUT THE AUTHORS 4
FOREWORD 5
EXECUTIVE SUMMARY 6
INTRODUCTION 9
CHAPTER 1: THE IMPACT OF GENOMICS 11
Preface 11
The Opportunities 12
The Challenges 18
A Final Word 21
CHAPTER 2: THE IMPACT OF GENETICS 24
Preface 24
Disease Genetics 27
Pharmacogenetics 33
A Final Word 39
CHAPTER 3: MANAGERIAL CHALLENGES 41
Preface: Looking Back and Looking Forward 41
Strategy—Searching for Genomic Competitive Advantage 41
Putting the Strategy into Operation 49
A Final Word 56
CONCLUSION 57
METHODOLOGY 59
3
About the AuthorsPeter Tollman is a vice president in the Boston office and leads BCG's biopharmaceutical R&D business.Philippe Guy is a senior vice president in the Paris office and leads the worldwide Health Care practice. JillAltshuler is a manager in the Boston office and a key contributor to BCG’s genomics initiative. AlastairFlanagan is a vice president in the London office and leads the U.K. Health Care practice. Michael Steineris a senior vice president in the Munich office and leads the German Health Care practice.
AcknowledgmentsSarah Cairns-Smith (Boston) pioneered BCG’s investigation of genomics. Samantha Gray (Boston) has madesignificant contributions throughout the research and writing phases of the report.
The authors would like to thank the advisory team: Oliver Fetzer (Boston), Hamilton Moses (Washington,D.C.), Niko Vrettos (Düsseldorf), and Craig Wheeler (Boston). The authors would also like to acknowledgethe contributions of the project team: Dierk Beyer (Frankfurt), Markus Hildinger (Boston), Raphael Lehrer(Washington D.C.), Nancy Macmillan (Boston), Jonathan Montagu (London), and Joanne Smith-Farrell(Washington, D.C.).
For Further ContactThe authors welcome your questions and comments. For inquiries about this report or BCG’s Health Carepractice, please contact:
Alastair Flanagan, London e-mail: [email protected] Guy, Paris e-mail: [email protected] Lubkeman, Los Angeles e-mail: [email protected] Steiner, Munich e-mail: [email protected] Reeves, Tokyo e-mail: [email protected] Tollman, Boston e-mail: [email protected]
About the Authors
4
Foreword
To meet growth targets, pharmaceutical companies are going to have to increase R&D productivity. By a for-tunate coincidence, that crisis in expectation is being counterbalanced by a surge of opportunity. Recentyears have seen astonishing advances in technology and explosions of data, which are driving two waves ofchange through the industry—a genomics wave and a genetics wave—and radically reshaping R&D methodsand economics in the process. Biopharmaceutical R&D is moving into a new era: almost every link in thevalue chain has the potential for tremendous boosts in efficiency or success.
But these advances are not assured. Technological hurdles have yet to be overcome, particularly in the genet-ics wave. Moreover, because the productivity boosts are likely to be unequal and uncoordinated, the valuechain itself will demand reconfiguring. And so too, in consequence, will many traditional operational pro-cedures and organizational structures. The repercussions of genomics, in other words, are going to reach thefurthest recesses of corporate constitution and culture. A true revolution, in short—and one that is alreadywell under way.
BCG has evaluated deeply the economic and business implications of these disruptions. To bolster our inter-nal understanding, we gathered information and perspectives in an extensive program of interviews withleading R&D scientists and executives. Our findings—based on the combination of these interviews, eco-nomic modeling, and client casework—form the substance of this report. Its three sections are devotedrespectively to the impact of genomics, the impact of genetics, and some of the strategic and operationalimplications for biopharmaceutical firms.
The first two sections have already been published separately. They generated considerable publicity, and—more important—considerable comment. We now look forward to your further responses to the report as awhole.
Philippe GuySenior Vice President
5
In the pharmaceutical industry’s struggle to reachthe levels of growth expected of it, one of its keyaims will be to increase R&D productivity. And a keymeans of meeting this challenge is to adopt some ofthe new technologies and approaches broadlydefined as genomics.1 That is bound to be a com-plicated, perilous, and often painful process, but ifcompanies get their strategy right and overcomethe obstacles, they could, in the best case, as muchas halve the cost of drug development.
The report is divided into three parts.
The Impact of Genomics
The first great advance of the genomics era is intechnology—above all, the integration of new high-throughput techniques with powerful new comput-ing capabilities. The new technologies are activethroughout R&D, most immediately at the drug dis-covery stage, and promise to enhance productivityby boosting efficiency.
The staggering investment needed to develop adrug—$880 million and 15 years is the pre-genomics average—could be reduced by as much as$300 million and two years by applying genomicstechnologies. Productivity gains would be realizedat every step in the value chain. Potential obstacles
abound, however. In particular, two broad chal-lenges must be met to realize the savings:
• Target quality must be maintained. Pursuing newtarget classes could involve unfamiliar costs ini-tially, and these could delay the rewards—thoughonly temporarily. But to jeopardize target qualityby withholding that early investment would be torisk higher failure rates downstream, and thatwould involve far greater costs in the end.
• Bottlenecks must be eased. Owing to the uneven-ness of the efficiency gains at different steps inthe value chain, the pipeline’s flow will beimpeded at various chokepoints. If the requisiteaction is taken, an even flow should be restoredand the promised rewards should be safeguarded.
The Impact of Genetics
The second great advance of the genomics era is inthe quantity and quality of data. From the data,invaluable information about individuals’ geneticvariation can be extracted and exploited. In phar-maceutical R&D, genetics will be applied particu-larly to two tasks: identifying genes whose carriersare susceptible to specific diseases (disease genet-ics); and subdividing patients in clinical trialsaccording to variations in drug response (pharma-
6
Executive Summary
1. Genomics in its narrow sense contrasts with genetics. Roughly, the former concerns itself with the common “standard” genetic makeup, the latter withthe distinctive genetic makeup of individuals. But in its broader sense, genomics includes genetics. In this report, the context makes clear which sense isintended.
cogenetics). The productivity gains will be realizedmostly in later phases of the value chain, throughthe boosting of success rates.
This genetics wave is still gathering strength, but indue course could make an even greater impact onR&D than the genomics wave. In an ideal scenario,the savings would exceed half a billion dollars perdrug. Several troubling hurdles would have to benegotiated first, however. These include:
• Scientific and technical hurdles. For geneticsapproaches to work, the disease susceptibility ordrug response has to be genetic in nature. Thegene in question has to be identifiable and mustlead to a drugable target and/or be found in timeto streamline trials.
• Economic and market hurdles. The cost of con-ducting genetics studies will need to drop, andthe opportunity cost of a restricted label couldoffset the potential market upside of pharmaco-genetics.
Beyond these hurdles, other challenges will need tobe addressed:
• Difficult investment decisions will have to bemade, weighing high risk against potentially highrewards. Companies will need to decide exactly
how to participate in genetics—whether to investin genetics approaches, and how deeply, consis-tent with their level of risk tolerance.
• Unprecedented coordination between marketingand R&D will be necessary. Marketing will need tohave a say in deciding which markets and whichgenetic diseases R&D should concentrate on, andwill need to become involved earlier than ever.
• Careful attention will need to be given to ethicalconsiderations. Companies will have to ensureprivacy of genetic material, and be prepared toaddress any concerns the public may have.
Managerial Challenges
With the new wealth of options and the increasedinterdependencies across the value chain, strategicissues will prove more complex than in the past.Likewise operational issues: many traditional waysof doing business will be disrupted by genomicstechnologies, and companies may need to restruc-ture fairly drastically.
The range of strategic options available to a company will be dictated by the company’s startingposition—its size, beliefs, aspirations, and capabili-ties. Given the magnitude of the opportunities and
7
the risks involved, momentous investment decisionswill need to be made, and at the very highest levelsof the organization. And R&D executives will face adaunting new set of management responsibilitiesand challenges. These include:
• Selecting an appropriate research focus—nolonger just the therapeutic area or disease state ofinterest, but also such dimensions as target classand treatment modality
• Choosing which technologies to implement andwhen and how to implement them—in-house, orthrough partnering or licensing
• Rebalancing the value chain—partly by reallocat-ing resources but mainly by redesigning processesand more actively planning and managing capacity
• Establishing a unified informatics infrastruc-ture—including a centralized knowledge manage-ment system
• Establishing the new organization—creating newinterfaces within the R&D department, betweendepartments, and even between corporations
• Revising decision-making procedures—fullyexploiting the latest data in order to select themost promising targets and compounds to movethrough the pipeline and to optimize their rela-tive resourcing
• Reinforcing these various reforms by engagingthe emotional and behavioral issues as keenly asthe operational ones
All things considered, companies cannot standaside. Certainly there are risks in signing up for therevolution, but there is also a great risk in ignoringit—the risk of becoming uncompetitive. The revo-lution is real, and will leave no one untouched.
8
9
Introduction
Throughout the pharmaceutical industry, execu-
tives are worried. They fear they will not be able to
meet the double-digit annual growth expectations
implied by high market capitalizations. The requi-
site new drugs will not be forthcoming: R&D just
cannot deliver them all.
One standard response to this problem is to scale
up—that has been the basis of many a recent
merger—but while scale can pay off in commercial-
ization, global development, marketing, and distri-
bution, it is unlikely that scale alone can solve the
R&D problem. Another standard response is to buy
in drug candidates. Such a Band-Aid approach can-
not work indefinitely, and is a risky one anyway,
given that the price of these deals will continue to
rise as demand for them grows.
The only sure way to address the problem is toincrease R&D productivity. And the way to ensurethat is either to increase efficiency (lower cost orhigher speed) or reduce failure rates along thevalue chain. Many companies have increased pro-ductivity over the past decade, specifically byreengineering the development phase. That opti-mization may be reaching its limits, however. As forthe discovery phase, it has long been less amenableto such improvements. So the problem of produc-
tivity persists. Traditional approaches cannot pro-vide an answer, but genomics can. (See Exhibit 1.)
It will not be easy, of course. There are some diffi-cult obstacles en route—difficult, but not insur-mountable. By making informed strategic choices,companies can overcome the obstacles and reap theproductivity rewards. Those that embrace the revo-lution most boldly could potentially halve the costand time it takes to develop a new drug—if theymeet certain challenges successfully.
EXHIBIT 1GENOMICS IMPROVES R&D PRODUCTIVITY
Total cost (time) per step
Cost (time) spent
Failed targets/candidates
Successfuldrug
Reducefailure
Improve efficiency
Total cost to develop a drug
SOURCE: BCG analysis.
11
Chapter 1: The Impact of Genomics
Preface
As the science of genomics has advanced, so has thedefinition. When the term was coined in 1986, itreferred mainly to the study of the mammaliangenome—specifically, the mapping, sequencing,and analyzing of all its genes. The scope soonexpanded, focusing not just on the genes’ structurebut on their function as well. More recently, thescope of the term has broadened further, focusingno longer just on knowledge of the genome but alsoon the exploitation of that knowledge, especiallyfor health care.
Going beyond dictionary definitions, our interest isin what genomics means for the economics of phar-maceutical R&D. On the basis of our extensiveresearch and many discussions with prominent peo-ple throughout the industry, we suggest character-izing genomics, for the purposes of this study, as theconfluence of two interdependent trends that arefundamentally changing the way R&D is conducted:industrialization (creating vastly higher through-puts, and hence a huge increase in data), and infor-matics (computerized techniques for managing andanalyzing those data). The surge of data—gener-ated by the former, and processed by the latter—isof a different order from the data yields of the pre-genomics era.
To elaborate. The new high-tech industrializationhas increased the efficiency of certain activitiesbeyond recognition. Instead of assigning individualscientists to work manually on modest individualexperiments, companies now invoke automation
and parallel processing to conduct experimentsmuch larger in scale and complexity, and at a muchfaster pace.
Look around this lab—you have to search highand low to find a human heartbeat. Now robotscan do the menial things we did in grad school.
—Research leader, leading biotech company
The data that emerge are immensely greater bothin quantity and in richness. Enormous databases—detailing gene expression, for example, or homolo-gous genes across species, or protein structures—afford unprecedented comprehensive views ofbiological processes. Increasingly, researchers canunderstand properties of the system rather thanjust individual parts, and that holds out the promiseof a more rational approach to drug discovery.
The new technology of informatics serves to handleand process all these data. Without it, the datawould remain raw material. Informatics was nur-tured by several coinciding factors: the ever-acceler-ating power of computers, refined algorithms, theintegration of data and technology platforms, andthe versatility of the Internet. The effect is thatoverwhelming masses of information can now bemarshaled, managed, and analyzed as never before.Data are transformed into knowledge.
We could never have achieved drug developmentthat fast with traditional techniques. No way—without the computers we didn’t have a chance.
—VP of chemistry, biotech company
The Opportunities
What is the impact of genomics on the economicsof R&D? To what extent will genomics improve pro-ductivity overall, and what will its effects be whenapplied at various points of the value chain? Whatother incidental advantages might genomics bringin its wake?
These crucial questions have received a great dealof attention of late, and a wide variety of responses.To address the questions in a rigorous, fact-basedway, we built an economic model of the entire R&Dvalue chain, grounded in a program of discussionswithin the industry (more than 100 meetings withmore than 60 scientists and executives from nearly50 companies and academic institutions.) (See themethodology section at the end of this report.)
Realizing SavingsBefore genomics technology, developing a newdrug has cost companies on average $880 million,and has taken about 15 years from start to finish,that is, from target identification2 through regula-tory approval. (See Exhibit 2.) Of this cost, about 75percent can be attributed to failures along the way.
By applying genomics technology, companies couldon average realize savings of nearly $300 millionand two years per drug, largely as a result of effi-ciency gains. That represents a 35 percent cost and15 percent time savings. (And those are the savingspossible with technologies that are available today;when new or improved genomics technologiesemerge, the savings will be even greater.) If compa-nies wish to stay competitive, they have no choice:they must implement genomics technologies. (SeeExhibit 3.)
Doing so, however, will hardly produce such hugesavings immediately, or automatically. It will take afew years, and many deft decisions, for the savingsto be realized. The early years of implementationmay in fact involve an increase in costs as the learn-ing curve is negotiated for novel targets—specifi-
cally, as the necessary quality controls are estab-lished—and as major strategic decisions (aboutpersonnel and processes, for instance) are con-firmed or revised.
More on these challenges later. But first, we willtake a closer look at the long-term upside, detailingthe savings at various steps along the value chain.
12
EXHIBIT 2DRUG R&D IS EXPENSIVE AND TIME-CONSUMING
Cost: $880 million total
Approximate cost ($M)
165
20540
120
90
260
Time: 14.7 years total
Approximate time (yrs)
1
2 0.4
2.7
1.6
7
Development
Preclinical Clinical
Chemistry
Screening Optimization
Biology
Target ID Target Validation
SOURCES: BCG analysis; industry interviews; scientific literature; public
financial data; Lehman Brothers; PAREXEL’S Pharmaceutical R&D
Statistical Sourcebook 2000.
NOTE: Cost to drug includes failures. Target identification includes initial
experiments that companies may have outsourced to academic research
institutions.
2. Includes initial experiments to identify potential targets. Traditionally, companies have sourced much of this research from academia.
Target Discovery/BiologyThe identification of targets is being industrial-ized—through the use of technology such as genechips to perform gene expression analysis, forexample—and then further enhanced by bioinfor-matics. Scientists can now use a single gene chip tocompare the expression of thousands of genes, indiseased and healthy tissue alike, all at once, andcan then use informatics technology to find follow-
up information, on these or related genes, in data-bases around the world. (Target validation, how-ever, seems difficult to industrialize, owing to the“slow” biology of whole-animal systems stillinvolved, and is not yet showing significant produc-tivity gains.)
In all, the potential savings per drug are on averageabout $140 million and just under one year of timeto market, achieved entirely through improved effi-ciency. That would add about $100 million in valueper drug (assuming an “average” drug with peakannual sales of $500 million). So for this step in thevalue chain, productivity would increase vastly: itwould be six times as high as before, assuming thesame level of investment. A sixfold increase in thenumber of potential targets!
Several companies have already benefited hand-somely from this windfall. Take the case ofMillennium, which was an early adopter of industri-alized biology. The company, anticipating an over-abundance of targets, established a business modelin which it sells off much its output and uses thatincome to fund internal research. Starting from itsearly genomics platform, Millennium has strategi-cally acquired or partnered with other platformcompanies to establish an integrated drug discov-ery value chain. From the other perspective, phar-maceutical companies such as Bayer and Aventishave made deals with Millennium, in the expecta-tion of profiting from the new abundance of targetsthey can choose to pursue.
Lead Discovery/ChemistryChemistry is being revolutionized by in silico (thatis, computer-aided) technology—specifically, vir-tual screening supported by chemoinformatics. Invirtual screening, potential lead chemicals areassessed with computer algorithms to test how likelythey are to interact with a target. Chemoinformaticsprovides the necessary platform for virtual screen-ing, using data and analysis from high-throughputscreening (HTS) and other chemistry activities.This approach increases efficiency by focusing com-pound synthesis, reducing the number of assays,increasing the parallelization of screening steps,
13
EXHIBIT 3GENOMICS CAN YIELD SIGNIFICANT SAVINGS
Cost to drug
Cost ($M)
Time to drug
880Pre-genomics
740Post-genomics target ID
610Plus in silicochemistry
590Plus preclinical and clinical advances1
Time (years)
Pre-genomics
13.8Post-genomics target ID
13.0Plus in silicochemistry
12.7Plus preclinical and clinical advances1
ID
14.7
1,0008006004002000
151050
Development
Preclinical Clinical
Chemistry
Screening Optimization
Biology
Target ID Target Validation
SOURCES: BCG analysis; industry interviews; scientific literature; public
financial data; Lehman Brothers; PAREXEL’S Pharmaceutical R&D
Statistical Sourcebook 2000.
1Includes surrogate marker savings from early elimination of unpromising
candidates, not from early FDA approval; does not include potential savings
from pharmacogenetics.
and generally helping to optimize screening. Thepower of this approach is expected to increase dra-matically with the availability of larger data sets forrefining the predictive algorithms. (At the moment,however, in silico chemistry has one notable short-coming: it looks as if it will be suitable for onlyabout 30 percent of targets—the rest fail to yieldthe requisite structural information—and eventhen might prove difficult to apply until lead opti-mization. Our savings are calculated for those tar-gets where in silico technology can be applied.)
The potential savings are on average about $130million and nearly one year per drug. That wouldadd about $90 million in value per drug. For thisstep of the value chain, then, productivity woulddouble, assuming the same level of investment.
As a beneficiary of these advances, a good case inpoint is Vertex. Starting from an IDD (in silico drugdesign) platform in chemistry, the company hasgone on to develop an integrated value chain in itsown right. In silico models have allowed more effi-cient design of small-molecule drugs than a purelytraditional approach, and the company’s discoveryfocus has been on certain target classes that benefitmost from proprietary in silico technologies. Thisapproach has met with considerable success, culmi-nating in one of the biggest biotech alliances so far(with Novartis, and worth $813 million). Vertex canfairly claim to have the strongest small-moleculedrug pipeline within the biotech industry. With onedrug on the market and twelve candidates in devel-opment, it compares favorably with some of the bigpharmaceutical pipelines.
Serious money can be saved for the target classeswhere in silico chemistry works.
—Director of chemistry, major pharmaceutical company
DevelopmentThree key genomics advances look set to increasecapacity here. In silico ADME/tox (absorption, dis-tribution, metabolism, and excretion/toxicity) andhigh-throughput in vitro toxicology are revolution-izing the preclinical phase through their power topredict drug properties. And surrogate markers
(physiological markers that correlate with elementsof drug response), applied in both preclinical andclinical trials, evaluate drug effects more efficientlythan before: they are quick to identify failing com-pounds, and once regulatory approval is granted,will be used to identify passing compounds too.
In combination, the potential savings available inthe short term are on the order of $20 million and0.3 years per drug. That would add about $15 mil-lion in value per drug. But these approaches willbecome even more valuable as clinical data on therelationship between genes, gene expression, anddisease accumulate and regulatory agencies beginto accept clinical-marker data: the potential savingscould rise to $70 million.
These technologies are being adopted by forward-looking chemistry companies, and are enablingthem to pull certain preclinical activities into thechemistry part of the value chain. For example,ArQule has recently acquired Camitro to incorpo-rate an integrated in vitro and in silico ADME/toxplatform into its own set of capabilities.
These are not the only advances likely to transformproductivity during the development phase. Phar-macogenomics—through its power to identify sub-groups of patients who respond differently to adrug under study—offers the promise of streamlin-ing clinical trials; we explore this topic in moredetail later. Beyond genomics (and beyond thescope of the current report), “e-technologies,” suchas electronic patient recruitment and monitoringvia the Internet, are expected to speed up thelaunch and completion of clinical trials.
Beyond the Traditional Value Chain: Chemical GenomicsThe various productivity gains just outlined occurwithin specific steps of the value chain. But supposeyou could transcend the traditional value chain, orrefashion it to streamline R&D. That is one of therevolutionary prospects now opening up. The key ischemical genomics, and the way it will dissolve theold boundaries is by introducing into the valuechain a kind of parallel processing. (See sidebar,“Chemical Genomics—Forward or Reverse.”)
14
One immediate result would be to process the glutof identified targets more quickly: instead of join-ing the logjam at the validation stage, a great manyof them can now be diverted directly to screening.If they fail there, they can be discarded right away,and thus simply bypass most of the validation stagealtogether. In other words, screening moves up thevalue chain to rest alongside validation, in a paral-lel rather than consecutive position. By bracketingthe industrialized steps of target identification andchemical screening, chemical genomics has giventhe value chain a remarkable makeover.
The key is to move lengthy, messy biology far down-stream where you know it’s worth pursuing. Manytargets aren’t drugable, so just validate the smallerdrugable subset.
—SVP of discovery, leading biotech company
The effect of this new value chain is dramatic: timeto drug is cut by a further two years (that’s on topof the year already saved by using genomic targetidentification). On the other hand, there is a large
increase in cost, offsetting all cost savings from tar-get identification. But the tradeoff is still positive.In a highly competitive market, where new entrantsare continuously eroding share, chemical genomicscan add more than $200 million in value per drug.(In less competitive conditions, the value addedmay be as little as $20 million.)
No doubt chemical genomics costs more—but youtake the loss to gain the speed. Time is money.
—SVP of discovery and technology, major pharmaceutical company
One important drawback of chemical genomics isthis: it is limited mainly to known target classes.With targets of unknown function, results becomevery difficult to interpret. The proxy assays used forscreening—heat-stability assays, for instance—tendto yield both false positives and false negatives.
Nevertheless, chemical genomics is already beingpursued throughout the industry. Several big phar-maceutical companies have adopted it, and geno-mics companies such as Aurora Biosciences3 and
15
C H E M I C A L G E N O M I C S — F O R W A R D O R R E V E R S E
When companies say they are pursuing chemicalgenomics, they are usually referring to large-scalereverse chemical genetics. (That is how the term isused in our report.) This approach involves findingchemical compounds that bind to a known target.Companies often perform this task for entire targetclasses; it is especially popular for protein classesthat are known to be highly drugable, such as G-protein coupled receptors (GPCRs). The assay forbinding does not need to provide functional informa-tion relevant to a specific disease state—biologicalfunction can be assessed in validation experiments.
The alternative is forward chemical genetics. Thisapproach begins with functional knowledge. Alibrary of compounds is screened in an assay thattests for changes in a specific biological function.
The intention is to screen a library against all ex-pressed genes in the system under investigation.This approach has the tremendous advantage of al-lowing the identification of targets without any pre-sumptions as to their function. Additionally, thesetargets can help to elucidate the mechanism of dis-ease, thereby revealing other potential targets in rel-evant pathways. The drawback is that forward chem-ical genetics has not yet been industrialized, andthroughput levels are therefore very low. Accordingto our model, implementing it today would increasecosts to more than $1 billion per drug, owing to theuse of “slow” biology, which is needed to set up thescreening assays in chemistry. The expert estimate isthat forward chemical genetics is still as much asfive years away from being economically feasible.
3. In July 2001, Aurora Biosciences was acquired by Vertex Pharmaceuticals.
Cellomics are well positioned to exploit the ex-pected resulting demand for screening resources.Aurora is a likely winner in the race to resolve chem-ical genomics-related bottlenecks, since it boastssome of the most advanced screening and assaytechnologies in the industry. It has an unusual busi-ness model, in that it provides tools and discoveryservices but does not engage in any drug discoveryof its own.
* * *
So much for the imminent efficiency savings acrossthe R&D value chain. They are hardly the end of
the story, of course. Other technological advancesare bound to improve R&D productivity further indue course. Important emerging technologiesinclude proteomics, partial target inhibition, andstructural biology. (See sidebar, “Technologies inWaiting.”)
Improving Decision MakingThe economics of R&D hinge on success rates, andsuccess rates depend largely on a cascade of deci-sions that have to be made again and again:whether or not to pursue a target or lead, and if so,how—to what extent and with what approach.
16
In this report we have focused on the technologiesand approaches that are having the greatest impacton R&D economics today. Several other exciting ad-vances appear likely to make a comparable impactbeyond the next three to five years (too far ahead forinclusion in our analysis for this report), in particu-lar, the use of proteomics in target identification,conditional gene inhibition in target validation, andindustrialized structural biology in screening anddrug design.
Proteomics is the study of protein expression andprotein-protein interactions. Its aim is an understand-ing, and ultimately exploitation, of protein function.
Identifying proteins through sequence or structurehomology has recently become much more efficient,thanks to bioinformatics’ role in analyzing large-scale experiments. One example of a genomics com-pany applying proteomics is Oxford Glycosciences,which is engaged in identifying targets and surrogatemarkers, both in collaboration with pharmaceuticalcompanies and in an independent pipeline. But pro-teomics is not really industrialized yet, and has highhurdles to overcome before it is.
We examined the economics of proteomic expressionstudies using two-dimensional gel analysis, followed
by identification of interesting proteins through massspectrometry.
Under optimal conditions today, this approach hasthe potential to save about as much in cost asgenomics-based approaches do, though not as muchin time (about six months less). As the technologybecomes industrialized, proteomics could well sur-pass genomics-based approaches, but that is stillseveral years away.
The aim of the second promising technology weinvestigated, conditional gene inhibition, is to over-come a common problem in target validation. Hereis the background. A standard technique for targetvalidation uses “target knockouts.” The potential tar-get is removed, or “knocked out,” from an animal atconception; this results in the total inhibition of thetarget’s function from embryo to adult. The trouble isthat drugs work differently. Very seldom do theyinhibit target function fully, and they are taken onlyafter genes have already fulfilled their developmentalrole in utero. So the use of target knockouts as a tar-get validation technique does run the risk of creatingfalse negatives (in some cases indicated by death,because of the unnatural disruption of embryonicdevelopment). What is needed instead of total gene
T E C H N O L O G I E S I N W A I T I N G — O T H E R T E C H N O L O G I E S E X A M I N E D , B U T O M I T T E D F R O M O U R R E P O R T
Genomics may offer an opportunity for companies
to make the correct decision more often than
before. For one thing, genomics can ultimately pro-
vide more, better, and earlier information, and
good information translates ultimately into high
success rates. For another, the implementation of
genomics approaches will force companies to
rethink their internal decision-making processes.
Genomics-based information, together with the
ability to mine it productively, gives a company an
enormous advantage. Such a company will now be
able to make and execute decisions on targets and
leads with greater speed and consistency than
before. Guided by more rigorous selection criteria,
the company should go on to improve its success
rates and hence its productivity.
A mere 10 percent improvement in accuracy of
decisions at any stage would confer disproportion-
ately large benefits. Consider, for example, all the
target/lead pairs that fail just before clinical trials:
if a company were able to decide in just one out of
ten such cases against pursuing the target in the
first place, it would save as much as $100 million
per drug on average. As for INDs that fail clinical
17
inhibition, therefore, is conditional gene inhibition,which mimics the partial inhibitory effect of a drug.Several promising approaches have emerged, includ-ing forward genetics, chemical genomics, andmolecular switches that modulate gene expression,but their practicality has still to be proved.
Examples abound of genomics companies engaged indeveloping these target-validation techniques. Lexi-con Genetics, Exelixis, and Ingenium, for instance,are using mass mutagenesis on animals such as miceand zebrafish. In a more focused project, Hypnion isusing forward genetics and other approaches tounderstand sleep-wake disorders in mammals.
What benefits lie in wait? By eliminating the falsenegatives associated with the current knockout tech-nique, these new technologies could double or eventriple the number of validated targets, and in thatway save up to $200 million per drug. At themoment, however, these new kinds of validation(with the exception of certain chemical genomicsapproaches, discussed in the main text) are stillmainly limited to “slow” biology.
Finally, structural biology is used for generating andanalyzing the three-dimensional structure of targets
for virtual screening, and is essential to in silico drugdesign. Unfortunately, it currently entails proteincrystallization (to prepare the proteins for visualiza-tion by X-ray diffraction), which is a difficult, labor-intensive manual process. Speedier alternatives,such as NMR spectroscopy, cannot predict overallprotein shape adequately, being restricted to proteinsubsegments. As a result, in silico modeling remainslimited in its applicability: the algorithms cannotboast really high precision for target classes whereno example structures are available.
Several projects, both public and private, are underway to upgrade structural biology platforms to thepoint where they will achieve industrialized scale.Among the private endeavors is the Novartis Insti-tute for Functional Genomics, founded by Novartis toidentify and characterize targets using high-through-put technologies. In the biotech field, StructuralGenomix aims to become a platform provider andgenerate revenues by selling protein structures; thecompany may also decide to exploit its data in-house, and extend into in silico drug design. But itmight be several years before technologies havedeveloped far enough for the necessary scale effectsto be realized.
trials, if the company were able to decide in just one
out of ten such cases to abandon development ear-
lier, it could save an additional $100 million per
drug.
Improving decision making to that extent will takemore than simply acquiring and implementing thenew genomics technologies and approaches. It willtake some serious strategic rethinking too, and pos-sibly major organizational changes. Whether to keepall activities in-house, or seek partners, or buy in tar-gets or leads. How to redistribute resources, reassignpersonnel, and revise lines of communication andchains of command. Such operational and organiza-tional quandaries will be addressed in detail in thefinal chapter of this report.
We implemented a fast-in/fast-out decision policyabout projects—if we didn’t have optimal condi-tions met in 18 months, we killed it. That made allthe difference.
—Former executive, leading pharmaceutical company
Even the basic business skill of decision making,then, is not immune to the influence of genomicstechnology. Whatever other benefits it brings,genomics serves as a wake-up call across the indus-try, even for companies trying to shelter from thegenomics revolution.
The Challenges
Although implementing genomics offers compa-nies great opportunities, it also presents them withformidable challenges. One of these is to ensurethat the quality of the pipeline remains uncompro-mised. Another is to put the new technologies intoefficient operation.
Maintaining QualityIf the potential productivity gains are to be fullyrealized, the post-genomics R&D pipeline will needto retain or improve its pre-genomics quality. Anydecline in quality—the quality of targets andleads—would obviously have an adverse effect onproductivity. The main threat to quality derivesfrom the unorthodoxy, the unfamiliar nature, of so
many new targets. Entire target classes, previouslyunknown, will need investigating. The temptationto pursue leads prematurely is bound to arise, andquality control will need to be rigorously enforcedto uphold the pipeline’s usual success rates.
In any given experiment, 70 percent of what I see iscompletely new. It could be a gold rush, or it couldbe junk—-there’s no way to tell until I sit at thebench and do more work.
—Director of research, leading biotech company
To appreciate the threat accurately, we need aproper definition of the term quality.
The “intrinsic quality” of a target or lead amountsto its likelihood of success, which is based on factorssuch as clinical relevance and drugability.Companies can do little to alter this type of quality.The “provisional quality” (or “informational qual-ity”) of a target or lead is based on the amount ofdata available on it at any given time—how much isknown about its clinical relevance, drugability, andso on. (This informational quality helps to predictsuccess rates, but does not influence them.)Companies can alter this type of quality, by spend-ing appropriately, and in that way can improve theirability to predict downstream success rates.
This distinction is crucial. But it has at times beenoverlooked, resulting in some confusion in theindustry. A widely publicized concern has been thatnovel targets identified through genomics wouldtend to be of inherently lower quality than pre-genomics targets, and thus more likely to fail atsome costly phase downstream. That inference is anoversimplification, and is misleading.
Certainly genomics proposes many more novel tar-gets (as much as 60 to 70 percent of potential tar-gets, in our interviewees’ experience, may belongto previously unknown target classes), and theirinformational quality at that early stage is duly mod-est. But that says nothing about their intrinsic qual-ity. Any prudent company, no matter how bold, willstrive to learn more about novel targets beforedeciding to pursue them downstream. In our analy-sis, investments made to raise a novel target’s infor-
18
mational quality to the level of a known target’swould be more than recouped in due course.
The overall cost of these novel targets—raisingtheir informational quality and then pursuing themdown the value chain—is bound to rise initially.However, within three to five years from the initialdiscovery of a target in a novel class, according toour model, the overall cost increase per novel-classdrug could return to average.
Where do the added costs come from? And whatmust happen to offset them?
The Cost of Quality ControlOur model predicts that the typical increase will beabout $200 million and more than one year perdrug (that is, a total cost of $790 million versus$590 million, and a total time to drug of 13.8 yearsversus 12.7 years). The increase is mainly attributa-ble to the extra time needed to understand targetfunction and develop appropriate assays in targetvalidation and screening; also, to the need to screena higher proportion of compounds, since an appro-priate subset of a larger library cannot be selectedin advance.
Chemical optimization costs would increase only ifthe novel target required a novel compound (by nomeans a necessary requirement, though certainly apossible one occasionally). Our model examinesthis worst-case scenario explicitly. If a novel targetdoes happen to require a novel compound, or acompound unfamiliar to the medicinal chemists,the potential efficiency loss causes a furtherincrease of $290 million and more than two yearsper drug (that is, a total cost of about $1.1 billionversus $590 million, and a total time to drug of 15years versus 12.7 years). The additional increaseshere would be due to the extra time needed now formedicinal chemists to learn how to modify the com-pound and attain specific properties through trialand error. But this worst-case scenario should notbe very common.
Moving further still down the value chain, to thepreclinical and clinical phases, costs are notexpected to increase. The downstream success rate
for novel compounds or targets should turn out tobe much the same as that for known compounds ortargets, as long as the same standards are applied.There should be no significant increase in toxicityor decrease in efficacy, other than in very unlikelycircumstances—for instance, if existing animalmodels somehow proved less suitable, or if drugsfor novel target classes were to interact with meta-bolic pathways in utterly unfamiliar ways.
Offsetting the CostsRaising the informational quality of novel targetsinvolves a heavy investment, but it is a wise in-vestment. And a fairly quick one: knowledge aboutone novel target quickly elucidates other poten-tial targets in the same class. Thanks to feedbackloops, knowledge increases geometrically. As moreis learned, the level of investment can tail off accordingly.
In any case, the alternatives to making that earlyinvestment in informational quality are far fromattractive. On the one hand, dropping the targetswould be terribly short-sighted: companies wouldbe forgoing the opportunity to discover and exploituntapped sources of revenue. On the other hand,pushing novel targets onward without adequateinformation on them would almost certainly resultin a higher failure rate downstream, with all theassociated implications for cost. An increased fail-ure rate of just 10 percent across chemical opti-mization and all of development would on averageincrease costs by about $200 million per drug.
To sum up, then: costs incurred early in the valuechain (by information gathering) look preferableto those that would otherwise be incurred later (asthe result of a higher downstream failure rate). Allthe more so, given that the early costs should soonbegin falling (investment in information is almostalways associated with an experience curve): asnovel target classes become increasingly familiar, itwill become increasingly efficient and economicalto pursue new targets within those classes. So withproper handling, the burden of that early costincrease is just a short-term one, and the productiv-ity of genomics-driven R&D should soon return
19
almost to that of more familiar target classes. Weestimate the time required for this is about three tofive years from the discovery of a novel target,which is the amount of time it should take to com-plete validation and early screening (assay develop-ment). (See Exhibit 4.)
Putting the New Technology into Operation It is one thing to acquire and install new capabili-ties and another to get them to function as they aremeant to. The challenge of making genomics tech-nologies operational has two major components:easing the bottlenecks that will develop, and resolv-ing the personnel conundrums that are sure toarise.
The Problem of Bottlenecks The bottlenecks result, in effect, from the uneven-ness of the efficiency gains at different points in thevalue chain. (See Exhibit 5.)
Consider the sixfold increase in target identifica-tion described above. This escalating quantity oftargets could turn out to be not so much a gloriousprofusion as an exasperating glut. Unless there issome corresponding increase in the capacity toprocess them downstream, these targets will simply
loiter at their source in a wasteful logjam. Or con-sider chemical genomics. Implementing thisapproach will build up huge pressure on screeningresources: sending unvalidated targets for screen-ing could involve a 120-fold increase. So too withefficiency gains at other points in the value chain:without the necessary downstream adjustments,bottlenecks will inevitably develop.
Our capacity to do functional experiments wascompletely choked by potential targets.
—VP of discovery, major pharmaceutical company
But the problem is a dynamic one, and accordinglyvery awkward to deal with. Ease one bottleneck andyou often create another downstream. Or ease ittoo much and you convert it into a bulge—over-
20
EXHIBIT 4IMPACT OF QUALITY ON COST TO DRUG
Mix of novel and known targets
Approximate cost ($M)
Novel targets only
Novel targets and novel compounds only
Benefits of experience over 3-5 years
Pre-genomics Post-genomics
1,080
880
590
590
790
SOURCES: BCG analysis; industry interviews.
EXHIBIT 5UNEVEN PRODUCTIVITY GAINS CREATE IMBALANCE
2,4001
108138
72
307
Increased productivity
Number today to get one drug
Required productivity2
Not to scale
Potentialtargets
Validatedtargets
Leadcandidates
Drugcandidates
INDs Drug
Targets Compounds
Poten-tial
targets
400
ID Development
Preclinical Clinical
Chemistry
Screening Optimization
Biology
Target ID Target Validation
SOURCES: BCG analysis; industry interviews.
NOTE: Does not include impact of pharmacogenetics, to be addressed in
next installment.
1Number of targets identified by investing same resources post-genomics as
pre-genomics.
2Productivity required to exploit all potential targets from target identification.
resourced in relation to the flow from upstream,and hence wasteful once again. It will take someadroit adjustment of resources and processes alongthe value chain to restore a smooth flow.
This imbalance will affect incumbents—integratedcompanies with established value chains—worst ofall. They have resources and processes in place;changes are likely to be difficult and disruptive. Toimplement the new genomics technologies is trou-blesome enough, but then to have to redistributeresources along the entire value chain will take realdetermination. (To other companies, by contrast,bottlenecks might represent very favorable oppor-tunities. In particular, genomics companies couldbenefit. (See sidebar, “Upstart Start-ups—theCompetitors Classified.”)
The Human FactorTo flourish in the new genomics era, and possiblyeven to survive, companies are going to have toengage the new realities. It will not be easy. Some ofthe new technologies will tend to overstretch oreven defy existing capabilities and organizationalstructures. All along the value chain, processes andresources are going to have to be adjusted.
The resources in question include human resources,and retrenching, reassigning, or supplementing tal-ented personnel is a far from straightforward proce-dure. But it will have to be done. Organizationalrestructuring is likely to entail distressing upheavalsfor corporate culture and personnel alike. Thestrategies adopted for managing it will require con-stant monitoring and fine-tuning. New modes ofcross-functional collaboration may need to be insti-tuted, new incentives offered, and so on.
I spend half my time looking for talent that isn’tout there, and the other half worrying where theywould fit if I found them.
—Research director, leading biotech company
* * *
In sum, implementing genomics technology will bevery tricky. It will almost certainly require a holistic,
cross-value-chain perspective. We will discuss poten-tial solutions to these operational challenges in thethird chapter.
A Final Word
By engaging affirmatively with the brave newgenomics world, companies are making it possibleto increase R&D productivity substantially. They willbring to bear an array of industrialized processes,informatics, and rich data sets—a formidable com-bination that promises to boost efficiency, and evenimprove success rates, all along the value chain.
Here we have discussed both the opportunities andthe challenges that arise when a company adoptsand implements genomics technologies that areavailable today.
The opportunities add up to potential savings ofnearly $300 million per drug—about one-third ofthe cost—and the prospect of bringing each drugto market two years sooner. The challenges includemanaging quality control and dealing with unfamil-iar operational predicaments: bottlenecks along thepipeline and a host of organizational difficulties.
But for companies that choose not to meet thegenomics revolution head on, the challenge is evengreater: they will be unable to compete. These com-panies do more than leave money on the table.They face the inevitability of being left behind.
To reap maximum benefit from the new technolo-gies, companies will need to scrutinize their re-sources, processes, and policies throughout thevalue chain. Pharmaceutical and biotech managerswill need to ask themselves some taxing questions asthey begin to formulate their genomics strategy:
• Which specific genomics technologies andapproaches make the most sense for our com-pany? What investments and capabilities would beneeded to integrate these new technologies andapproaches successfully?
• What capabilities do we already have? Whatinvestment are we prepared to make to acquirethose we lack?
21
22
U P S T A R T S T A R T - U P S — T H E C O M P E T I T O R S C L A S S I F I E D
As the e-commerce revolution has demonstrated,disruptive technologies tend to spawn start-ups thataim to exploit the disruption, either as suppliers to,or as replacements for, incumbents disoriented by achanging world. In the case of genomics, the in-cumbents are easy to identify: the traditional phar-maceutical companies and the larger fully integratedbiotech companies. But who are the start-ups?
Genomics companies can be classified in threebroad groups, on the basis of how closely or dis-tantly they are related to the actual developing andmarketing of drugs.
The group furthest away contains the companiesthat supply enabling technologies, in the form ofhardware or software. These companies resemblethe merchants of the California gold rush who soldpickaxes to the miners, or more recently, companiessuch as Cisco and Sun Microsystems that have beenproviding the necessary infrastructure for the multi-tude of e-commerce practitioners. Examples of suchcompanies are PE Biosystems, the supplier of high-throughput sequencing machines, and Affymetrix,the preeminent gene-chip manufacturer and supplier.
The second broad group contains the companiessupplying information and knowledge, includingthose companies that generate proprietary databasesand offer access to them through subscriptions orfee-per-use business models. One of the best-knownexamples is Celera, which sells subscription-basedaccess to human and animal model-sequence data.The group also includes companies that are attempt-ing to integrate and exploit those databases to con-duct in silico R&D. An example is LION Bioscience,which integrates information from public and private
sources into a single platform to make targets andleads easier to identify and analyze.
Finally, there is the group of companies that developand sell more traditional “physical” drug intermedi-ates—targets and lead compounds. We call theseplatform and orchestrator companies.
Platform companies deploy proprietary technology inthe quest for promising targets and leads. One suchcompany is Aurora Biosciences, which has devel-oped proprietary high-throughput screening technol-ogy to exploit an opportunity in screening and chem-ical genomics. Another example is MorphoSys,which has developed a platform for rapid develop-ment of high-affinity antibodies, for use in target val-idation and therapeutic antibody discovery.
Going one step further are orchestrator companies,which string together adjacent platforms to createoptimized segments of the R&D value chain. As theorchestrators extend their value chain, they can selldrug candidates that have progressed further andfurther downstream—and have thus become moreand more valuable. Although these companies arestill selling only intermediates, they show every signof graduating into fully integrated drug companies.Millennium has already made that transition: con-centrating initially on genomics target discovery, ithas subsequently developed a full R&D pipeline inits own right.
What is the outlook of each of these groups? Thefirst two (the pure suppliers, either of enabling tech-nologies or of information and knowledge) wouldappear to be well positioned if they target areas ofscarcity (that is, bottlenecks) with proprietary prod-
• Where is industrywide scale to be found ratherthan just company-level scale? Which capabilitiesshould we therefore develop in-house, and whichthrough partners?
• How will any new technologies affect the rest ofthe value chain? How can we optimize decisionmaking and information flow up and down thevalue chain?
• What are the implications for the organization ofthe changes we wish to make? How feasible is thenecessary restructuring? And what would be themost efficient way to carry it out?
These questions can be addressed by thorough,thoughtful analysis. Key investment decisions willbe required, as well as a carefully planned imple-mentation program to ensure that the value ofthose decisions is captured.
In the next chapter, we turn to genetics and analyzeits likely impact on R&D productivity. In the finalchapter, we will examine more closely the strategicchoices and operational implications of the variouschanges in prospect.
23
ucts, or if they have enough clients to achieve scaleefficiencies reachable only by supplying multiplecompanies. But so far, most of these companieshave struggled to find a sustainable, profitable busi-ness model.
Meanwhile, the third group seems in the most prom-ising position. The pharmaceutical business remainsattractive, with margins averaging more than 80 per-cent, so it is easy to see why so many genomicscompanies aspire to become drug companies. But
there are many hurdles en route, and overambitiouscompanies risk tripping over them. Although many ofthese companies may fail, those that succeed willhave a transformational impact on the industry.
Moreover, the traditional drug companies seem to bemoving in the opposite direction, increasingly out-sourcing portions of their R&D value chains. What isgoing on? We will try to answer that question in thethird chapter, when we examine these trends inmore detail.
Preface
Having discussed the genomics wave in the previouschapter, and the way that it promises to enhanceR&D productivity, we now turn to the genetics wave.Several broad differences suggest themselves imme-diately. Where the genomics wave is technology-driven, the genetics wave is better viewed as data-driven, exploiting the known details of the humangenome and individual variations within it. Wherethe genomics wave brings benefits mainly at thedrug-discovery and preclinical phases, the geneticswave will prove its worth in both the earliest phaseand the later phases of the value chain—target dis-covery and the clinic. Where the genomics waveenhances R&D productivity mainly by securinggreat improvements in efficiency (with only modestimprovements, if any, in success rates), the geneticswave could boost success rates dramatically as well.
One further difference should be mentioned:where our model for the genomics wave was put for-ward with considerable confidence, our model forthe genetics wave is more tentative. At this earlystage, any assessment of genetics’ impact on the eco-nomics of R&D is bound to be provisional. Certainlygenetics has huge potential: if all goes according toplan, it will change R&D productivity beyond recog-nition. But between that potential and its full real-ization lie several years and many obstacles.
The potential consists in tremendous savings. First,genetics can bring about great efficiency gains bymaking it possible to shorten or even bypass varioussteps in the value chain. Second, genetics holds the
prospect of transforming success rates: failures inthe R&D pipeline currently account for 75 percentof the total cost to drug. But offsetting such oppor-tunities, dangers loom large. Riding the geneticswave involves a greater risk than riding thegenomics wave alone—though it is more exhilarat-ing and, if the risks are successfully negotiated, ulti-mately more rewarding. How to choose between dis-cretion and valor is a crucial strategic decision thatcompanies will have to make.
In analyzing the economic implications of genetics,this chapter of our report considers the effect onlyon pharmaceutical R&D. But genetics is likely toaffect health care far beyond R&D, in both theshort and the long term. In the short term, newmarket opportunities should arise in the formerlysleepy diagnostics sector. (Drug companies may ormay not be able to exploit these opportunities: seesidebar, “Diagnostics—an Opportunity Too Good toMiss…and Perhaps Too Good to Grasp.”) In thelonger term, genetics is likely to transform thedelivery of health care. Increasingly, diseases will beredefined into various subtypes—a refinement thatshould facilitate more appropriate care and more“rational” drug design. The combination of newdiagnostics, new disease definitions, and new tai-lored drugs should prove a winning one, and maywell usher in an era of individualized medicine.
R&D remains the focus of our analysis here, how-ever: specifically, the wide range of economic reac-tions that R&D might show under the impact of thenew genetics information. We discuss the tremen-dous opportunities as well as the accompanying
Chapter 2: The Impact of Genetics
24
risks inherent in genetics-based R&D, and explorevarious ways of managing them.
Two Kinds of Genetics ApproachesThere are two relevant approaches to considerwhen assessing the economic impact of genetics on
R&D: disease genetics and pharmacogenetics. Theyoperate at different stages of the value chain.Disease genetics is invoked earlier, during the discov-ery phase: it involves the search for genes that makepeople susceptible to particular diseases, with theaim of then finding targets. Pharmacogenetics is the
25
D I A G N O S T I C S — A N O P P O R T U N I T Y T O O G O O D T O M I S S …A N D P E R H A P S T O O G O O D T O G R A S P
It will be several years before genetics fulfills itspromise. In the meantime, however, companiesmight begin to enjoy a preliminary reward, in theform of diagnostics—essentially a byproduct of theirbroader genomics research programs. Certainly diag-nostics is the subject of great expectations, thoughwhether and how soon it will meet them remains tobe seen.
Many research projects in genomics and geneticswill devise diagnostic tests as a matter of course—inparallel with research or simply as a preliminarystep, perhaps—without portraying them that way.Diagnostic tests can be understood in a fairly broadsense here. Disease genetics, for example, in identi-fying a target, is in effect finding a marker of diseasesusceptibility. Expression profiling, in identifying themolecular differences characterizing a disease’s dif-ferent subtypes, is pointing the way to differentiatedand fine-tuned therapies. And pharmacogenetics, inidentifying variations in drug response among vari-ous patients, could be helping to suggest the mostsuitable drug for them.
The opportunities inherent in diagnostics will appealto drug companies at several levels. First, costs arelow. The intellectual capital needed to develop a di-agnostic test comes free, courtesy of existing re-search in drug discovery and development; validationstudies can be run in parallel with drug efficacy stud-ies, or perhaps can even simply borrow their resultsand extrapolate from them; and as for safety studies,diagnostic tests don’t need any. All in all, then, theincremental spending required to develop a mar-ketable diagnostic test is, relatively speaking, paltry.
Second, rewards are prompt. Diagnostics, in bypass-ing most of the traditional steps of pharmaceuticalR&D, can be brought to market not only far morecheaply than drugs, but far more quickly too. Drugcompanies are thereby able to realize some unusu-ally fast payback on their R&D spending.
Third, the market outlook is favorable. As new thera-pies proliferate, more diagnostics will be demanded;and as technologies advance, new types of diagnos-tics will become available. The signs are good.
These opportunities are to some extent offset, how-ever, if not by risks, then at least by challenges.
There is the challenge of novelty, for instance: formany traditional companies, diagnostics wouldinvolve manufacturing an unfamiliar kind of product—a kit—and that in turn would involve developing newcapabilities, or else partnering with a dedicated diag-nostics company. Companies that have an in-housediagnostics division, such as Hoffmann-La Roche,Abbott, and Bayer, will have an advantage here.
Then there is the challenge of intellectual-propertyrights: companies might find it more difficult toassert those rights over diagnostics than over theirfindings in pharmaceutical research.
Perhaps the most daunting challenge is timing: diag-nostic tests will tend to emerge too speedily, becom-ing available sooner than the therapies they indicate.So the chief appeal of investing in diagnostics—itsprompt availability—may be undercut. Drug compa-nies may have to delay marketing their diagnostics(and thus delay capitalizing on the opportunities)until their drug R&D pipelines catch up.
genetics-based form of pharmacogenomics (seesidebar, “Pharmacogenomics—Some Definitions”),and comes into play later, in the developmentphase: it involves predicting the efficacy and sideeffects of candidate drugs.
The data explosion detonated by genomics tech-nology has created vast amounts of genetic infor-mation, ready for sifting. The findings of theHuman Genome Project and related endeavors aremerely the starting point. The ultimate goal is toelucidate the genetic basis of human disease anddrug response. In the short term, genetics researchwill enable scientists to predict disease susceptibilityand likely drug response in individuals; in thelonger term, it should help to improve the qualityof pharmaceuticals and medical diagnoses.
Attaining the short-term goal is, conceptually, sim-ple enough. The genetic codes of individuals differin tiny, but sometimes decisive, details. By compar-ing an individual’s genetic variations against the“standard” genome, scientists should be able to pre-dict whether that individual is at risk for a specificdisease, and, if so, how well suited he or she is to aparticular drug therapy—the work respectively ofdisease genetics and pharmacogenetics.
The two approaches benefit R&D economics in dif-ferent ways. Disease genetics will improve efficiencyin target discovery and, by leading to the discoveryof particularly high-quality targets, will bring aboutimproved success rates in validation and down-stream. Pharmacogenetics, by enabling scientists toselect patients more appropriately for clinical trials,
26
P H A R M A C O G E N O M I C S — S O M E D E F I N I T I O N S
Pharmacogenomics is the use of genomicsapproaches to elucidate drug response. There arethree relevant approaches: via DNA, via RNA, andvia proteins, and three corresponding forms of phar-macogenomics: pharmacogenomics using geneticapproaches (or pharmacogenetics), expression pro-filing (or expression pharmacogenomics), and pro-teomics (or proteomic pharmacogenomics).
Pharmacogenetics predicts patients’ drug responseby analyzing the genetic variations in their DNA. It isthe form of pharmacogenomics discussed in themain text here.
Expression pharmacogenomics predicts patients’drug response by analyzing their RNA levels—specif-ically, by comparing the amounts of RNA found indifferent samples to determine which genes areexpressed at different levels. An example: a researchgroup at The Whitehead Institute studying two verysimilar leukemias (AML and ALL) has observed adistinct difference in expression levels of specificgenes, and thereby provided a quick and reliablemethod for differentiating them. Patients are nowless at risk of being misdiagnosed and being given
an incorrect, and possibly lethal, drug treatment: ineffect, the test screens for adverse drug response.Expression pharmacogenomics seems to be movingfrom academic studies and biotechs into more main-stream pharmaceutical R&D. Witness the recent pur-chase by Merck and Co. of Rosetta Inpharmatics, abiotech founded specifically to develop expressionpharmacogenomics.
Finally, proteomic pharmacogenomics predictspatients’ drug response by analyzing their proteinlevels—specifically, by comparing protein readingsin different tissue samples to identify proteins thatdiffer either in structure or in expression levels.Consider the example of an aberrant fusion of twoproteins called Bcr and Abl, which occurs in morethan 95 percent of patients with CML (chronicmyeloid leukemia, which accounts for about 20 per-cent of all cases of adult leukemia). This aberrantfusion protein is present only in cancer cells. It dis-tinguishes itself from its normal counterparts by itsincreased size. It can be used not only to monitor theprogression of the disease but also to test whetherGleevec, a revolutionary new drug, would provide aneffective therapy.
will not only help to make the trials faster andcheaper, but also allow some drugs to pass thatwould otherwise have failed owing to poor efficacyor side effects.
High-Risk, High-Reward Research In the near-to-medium term, genetics promises toreduce R&D costs dramatically. Or does it? Ourmodel shows that the application of disease geneticsand pharmacogenetics together could, in the verybest case, save as much as two-thirds of the currentcost to develop a drug and nearly two years. Ofthese potential savings, the vast majority comesfrom disease genetics. Of the remainder, some can-not be clearly apportioned to either disease genet-ics or pharmacogenetics but must be credited totheir joint efforts, being the product of synergiesrealized when disease genetics information is incor-porated into pharmacogenetics-driven clinical tri-als. (Although genetics can be combined with thegenomics technologies described in the previouschapter, the potential savings are not additive. Inthe next chapter, we will assess the total savings real-istically achievable through the application of geno-mics and genetics.)
Our model also shows that realizing this potentialwill be far from straightforward and is far fromguaranteed. The high-end savings estimate assumesa positive resolution of several scientific, technical,and market risks. With every setback, the savingsdiminish. Too many setbacks, and the savings fall tozero. Implementing genetics could even turn out tohave a negative impact on value, owing to adversemarket dynamics.
Such a double-edged sword is awkward to wield. Ifcompanies fail to grasp it at all, they forfeit theopportunity to reap enormous rewards—nearlydouble the savings possible with genomics tech-nologies. If they do grasp it, they put themselves insome danger, and will need to develop a geneticsstrategy aligned with their risk profile and with spe-cific market conditions. As an approach to researchand development, genetics remains risky, but full ofpromise, too. This chapter assesses the promise andthe risk alike.
Disease Genetics
Underlying many diseases are genetic variants, orpolymorphisms—alterations in the DNA sequenceof a given gene that influence individual risk of dis-ease. (The term polymorphism usually refers to acommon variant—one found in more than 1 per-cent of the population.) Often the genetic differ-ence consists of just a single altered letter in thegenetic code (the variant is then known as a singlenucleotide polymorphism, or SNP). Such a tiny alter-ation can have fatal consequences; for example, thesubstitution of an A for a T in the hemoglobin ßgene is responsible for sickle cell anemia. The goalof disease genetics is to identify such DNA alter-ations; so far, this goal has proved elusive for all butthe most strongly inherited conditions. (See side-bar, “Disease Genetics—Various Approaches toVariants.”) But researchers persevere, since findingthe causal variants can be a major step to finding atreatment—and potentially a cure.
Uncertainty persists: the technologies have yet toprove their full worth by producing the necessaryquota of practical results. But if all goes accordingto plan, the savings realized by pharmaceuticalcompanies will be enormous.
The PotentialThe savings would come partially from improvedefficiency in discovery, but principally fromimproved success in target validation and clinicaltrials. The improvements in efficiency result fromthe consolidation of target discovery into a singlestep; the improved success rates result from therefining of target identification, making it possibleto pinpoint the targets associated with disease sus-ceptibility in humans.
Once the targets have been located, they can boasta collective superiority in three particular respects.
First, their relevance to human disease is certain.They have, after all, been validated in humans,showing directly that modulation of gene activityleads to alteration in the intensity or duration ofdisease. (By contrast, other targets are usually iden-
27
28
D I S E A S E G E N E T I C S — V A R I O U S A P P R O A C H E S T O V A R I A N T S
When disease genetics is used to identify or testvariant genes, the investigation can take a variety offorms. There are three key dimensions in the designof such an investigation: narrow/broad, linkage/asso-ciation, and direct/indirect. These dimensions allhave a bearing on cost and the chances of success.
First, researchers can examine some of the genes (ina narrow study, or candidate-gene study) or all ofthe genes (in a broad study, or genome-wide scan).
Second, these genes can be examined through inher-itance patterns in families prone to the disease (in alinkage study) or by comparing individual patientswith healthy individuals in the population at large (inan association study).
Finally, within association studies, each variant genecan be studied directly or indirectly: researchers cantest each variant individually for any involvement inthe disease (in a direct study) or test clusters ofclosely positioned variants for the presence of a cul-prit among them (in an indirect study).
The earliest disease genetics investigations, con-ducted prior to the 1980s, were association studies,and direct, and as narrow as could be—studying justa single gene, which had been selected on the basisof biological knowledge about disease mechanisms.The researcher would seek polymorphisms in thegene, and compare their frequencies in patients andcontrols. This early approach did achieve somenotable successes, including, in 1956, the first dis-covery of an inherited genetic variation found tocause disease—the variant underlying sickle cellanemia. The approach had a serious limitation, how-ever: it allowed very few genes to be examined, andit required a prior hypothesis.
In the 1980s and 1990s, attention turned to fami-lies showing an inherited pattern of disease. Theinvestigations took the form of broad linkage studies,and were tremendously successful in identifyingsome genes responsible for single-gene disorders,
notably the cystic fibrosis gene in 1989. Such stud-ies, being genome-wide, were now unbiased andcomprehensive. But the actual identification ofgenes remained a slow, painstaking laboratoryprocess. So the early versions of such studies werereally suitable only for monogenic diseases. Hopeswere raised in the early 1990s, when companiessuch as Millennium, Sequana, and Myriad were setup to develop and exploit these techniques in thequest to identify the genes implicated in commonpolygenic diseases. Their initiative seems to bestalled at the moment: although the localizing ofdisease-related genes has become more efficient,the actual locating of them remains discouraginglydifficult. That task would be better suited to associ-ation studies.
Given the limitations in genome-wide linkage stud-ies, association studies have recently come back intofashion, fortified by the efforts of the HumanGenome Project, Celera, and the SNP Consortium.These studies have high-throughput technologies toundergird them, as well as comprehensive databasesof gene sequences and SNPs.
Of the two possible approaches here, direct and indi-rect, the former looks like a very formidable task.The researcher conducting a direct association study,and aiming to find the actual polymorphism underly-ing the specified disease, is confronted by the entireset of common variants in the genome—expected tonumber some ten million. To examine such a hordeof variants with current technology would be inordi-nately time-consuming and expensive.
Hence the hopes—and funds—now being invested inindirect studies. Since variants in close proximitytend to form clusters (known as haplotypes), it maybe possible to track down the disease-related poly-morphisms using only a small proportion of allSNPs. In light of recent research findings, indirectassociation studies of this kind do look practicable,if not quite imminent.
tified through the use of animals and tissue cul-tures, and so their relevance to human disease islargely speculative.) In other words, there is no pos-sibility of failure in target validation, because iden-tified targets are ipso facto validated. (This is, ofcourse, no guarantee of their drugability—theirresponsiveness to small-molecule intervention.)
It would not be possible to overstate the value of invivo human validation. Most of what passes fortarget validation today is largely conjectural inrelation to the disease in question.
—Diabetes researcher, Harvard Medical School
Second, the frequency of the causal polymorphismsis known at the outset. If a study identifies multiplegenes associated with a particular disease, it willalso reveal their relative culpability. Consider theexample of Alzheimer’s disease, a heritable butgenetically complex disorder. On the one hand,there are variants in three genes—PS1, PS2, andAPP—that are very rare but very potent: if a personhas any of them, he or she is almost certain todevelop Alzheimer’s. On the other hand, there isthe ApoE4 polymorphism of the ApoE gene, whichhas a more modest effect on disease susceptibilitybut is much more common in the population atlarge, and among patients with Alzheimer’s.Information of this kind can be useful for predict-ing a drug’s potential marketability: although itmight be equally feasible to develop a drug thatinfluences the rarer variants, a drug targetingApoE4 might expect broader effectiveness, andthus a larger market, and so might take precedencein further research. (The rarer variants may still beworth pursuing, using pathway analysis.)
Finally, the nature of the relevant polymorphisms isknown—different disease-inducing mechanismsamong variant forms, for instance. Such informa-tion may help to streamline clinical trials if used byefficacy-based pharmacogenetics to identify “nonre-sponders”—patients who lack the crucial DNAalteration, and hence are unlikely to experience theintended effect of a candidate drug—and excludethem from the trials. (In modeling the potential of
disease genetics, we have included this effect ofefficacy-based pharmacogenetics. See the sectionon pharmacogenetics below for further details.)
Depending on the approach taken, cost savings perdrug could be as great as $420 million, with thepotential time savings ranging from 0.7 to about 1.6years (producing an added $290 million of valueper drug). (See Exhibit 6.) Of the cost savings, thevast majority would be yielded by the improvementsin success rates: $390 million, consisting of $110million in validation and $280 million in the clinic.
29
EXHIBIT 6DISEASE GENETICS OFFERS GREAT SAVINGS POTENTIAL
Cost to drug
$M
Time to drug
880Pre-genomics
460Candidate gene study
485Genome-wide scan
455Genome-wide scan plus pathway analysis
Pre-genomics
Candidate gene study
Genome-wide scan
Genome-wide scan plus pathway analysis
Years
13.1
14
16.5
ID
14.7
1,0008006004002000
20151050
Development
Preclinical Clinical
Chemistry
Screening Optimization
Biology
Target ID Target Validation
SOURCES: Industry interviews; scientific literature; public financial data;
BCG analysis.
Efficiency improvements in target discovery ac-count for the remaining savings.
The UncertaintyFor these vast savings to materialize, two require-ments will have to be met. First, disease geneticsmust prove scientifically feasible for the relevantcommon diseases. Second, not only must studies inhumans work; in addition, the targets they identifymust be drugable; failing that, identifying the dis-ease genes is pointless, and all the effort that hasgone into finding them will be wasted. (See sidebar,“Drug-Resistant?—Are Disease Genes DrugableTargets?”)
Feasibility—the Limitations of Technology Fundamental technological concerns still hover overdisease genetics. Can it actually be done? The resultsso far have been very modest. The bonanza ofclearly documented disease-susceptibility genes forcommon multigenic diseases has yet to materialize.Candidate gene studies, for instance, are by defini-tion limiting: they focus on a subset of genes definedby a prior hypothesis, and therefore risk excludingsome crucial culprits. And genome-wide linkagestudies, although highly successful in addressing single-gene diseases, have proved disappointing for
the more common multigenic kind of disease: a dis-ease-related gene might be accurately pinpointed inaffected families (such as the BRCA1 breast cancersusceptibility gene), only for it then to show very lowprevalence outside the families used in identifyingit. True, these two approaches might become moretractable now, in the wake of the sequencing of thehuman genome and the development of compre-hensive SNP maps (catalogs of the characteristicsand locations of SNPs in the genome).
As for genome-wide association studies, consideredby many experts to be the most promising of all,they have only recently became practicable: all therequisite tools (a full genome sequence with a SNPmap to match, genotyping technologies, and so on)appear to be in place. But the approach remains vir-tually untested, owing to the still exorbitant cost ofgenotyping. The preferable form of genome-wideassociation studies would clearly be the indirectkind—still covering the entire genome, but geno-typing far fewer SNPs—yet even here the currentcost is a prohibitive $400 million or so for each dis-ease investigated. Within five years, however, geno-typing costs are expected to fall to $20 million orless, and the essential proof-of-concept tests canthen take place more routinely. (See sidebar,
30
D R U G - R E S I S T A N T ? — A R E D I S E A S E G E N E S D R U G A B L E T A R G E T S ?
The skeptics pose an awkward question: Will dis-ease-related genes ever prove to be drugable in sig-nificant numbers? The record so far is hardly encour-aging. Some disease genes, such as CFTR in cysticfibrosis, were identified long ago, yet have failed togenerate therapeutics. The infrequency of successstories, such as Ceredase—a drug for type IGaucher’s disease that was essentially a creation ofdisease genetics—only highlights the general trendof failure.
These long-identified disease genes tend to be forsingle-gene disorders, however. And such disordersare by their nature difficult to cure. They are binaryphenomena: the gene is broken, you get the disease.
Finding a small-molecule therapeutic to repair a com-pletely defective protein is an extremely difficult chal-lenge. (Indeed, Ceredase is a protein therapeutic.)
Most disorders, by contrast, are attributable not to asingle gene but to multiple genes, and perhaps otherfactors too. This means that the system as a wholecan still function, just hampered to a greater orlesser degree. Such cases benefit from patching up,so drugs can be beneficial without actually consti-tuting a cure. There is little reason to doubt thatsuch palliative drugs will soon emerge in abun-dance, as disease genetics becomes ever faster atidentifying some of the genes implicated in multi-genic disorders.
“Genotyping—Hopes Rise as Prices Fall.”) Andonce that happens, the two major questions out-standing can be resolved: the number of patientsneeded to attain statistically significant results, andthe utility of SNP maps.
Regarding patient populations, first: the morepatients, of course, the easier it is to discern a truedifference above randomly occurring fluctuations(noise)—and the more expensive. Size your sampletoo sparingly, and you risk emerging empty-handed;too generously, and you overspend. Striking theright balance is tricky. It depends on how commonthe sought-for SNPs are, and that is very difficult toestimate. It also depends on the strength of theassociation between the disease and the suspectpolymorphism. For common multigenic diseases,susceptibility depends on a specific combination ofseveral genetic changes, and is influenced by envi-ronmental factors as well, which weakens the asso-ciation. The weaker the association, the larger thesample needed to detect the influence of a specificgene. Perhaps the issue will dissolve before beingresolved. If prices plummet as expected, and ifdatabases become as comprehensive as hoped, suf-ficient sample sizes will become easily affordable.
SNP maps can help in the quest to identify a disease-susceptibility gene, but only if two condi-
tions prevail. First, for any association studies(whether narrow or broad) to work, the genetic vari-ants associated with the disease need to be fairlycommon—prevalent in more than 1 percent of thepopulation at large—and that is far from guaran-teed. Then, these polymorphisms need to be eitherrecognizable (that is, they must produce discerniblechanges in a protein) or at least detectable by anindirect measure (called linkage disequilibrium;that is, the presence of a particular SNP cluster inindividuals with a given disease). Assuming reason-able costs, indirect genome-wide association stud-ies—of all the approaches, the one most likely toprevail—would, according to our model, result in atotal savings of $395 million in cost per drug on aver-age, with 0.7 years of time saved to market (produc-ing an additional $260 million of value per drug).
Of course, for that to happen, you need to do morethan find disease genes—you have to turn them togood effect by ultimately producing a drug. Andhere disease genetics presents a further challenge:given the long odds involved in pharmaceuticalR&D, will the number of targets yielded by diseasegenetics be sufficient?
Practicability—the Limitations of Human StudiesThe problem is that human studies—identifyingpolymorphic genes in humans—will tend to pro-
31
G E N O T Y P I N G — H O P E S R I S E A S P R I C E S F A L L
For companies wishing to pursue disease genetics,one of the major stumbling blocks has been the pro-hibitive cost of genotyping. The current averagegenotyping cost in the industry is about 50 cents perSNP. At that price, even narrow candidate-gene stud-ies could cost as much as $15 million; a directgenome-wide association study could cost upwardsof $5 billion.
Matters are about to change, however. Costs aredeclining, and are expected to continue to fall dra-matically—as much as 50-fold—over the next fewyears, thanks to competition and customer demandon the one hand, and improved automation and
miniaturization (which allows companies to reducetheir consumption of expensive reagents) on theother.
Our calculations are, accordingly, based on a cost ofone cent per SNP genotype—a likely price across theindustry, according to expert consensus, within thenext five years. (Some companies must already bebenefiting from genotyping costs considerably lowerthan the current industry average.) That said, thecost of conducting disease genetics studies willremain far from negligible, and companies will needto continue to take it into account when assessingrisk.
duce targets high in quality but low in quantity, per-haps just one to five for an average disease. Thatwould produce at best a 25–30 percent chance ofyielding a drug, given that chemistry and develop-ment remain far from fail-safe. For disease geneticsto live up to its promise, it will need to improvethose odds considerably. And to do that, it will haveto call on a supplementary technique: pathwayanalysis.
Disease-susceptibility genes, if identified throughdisease genetics, serve not only as targets them-selves, but also as guides to additional targets. Thegenes form part of broader disease pathways, andthese pathways contain other molecules that mayserve as targets (perhaps 10 to 15 targets per path-way, according to experts). These new targets areidentified by pathway analysis, often taking the formof the study of “simple” experimental systems, suchas those of Drosophila melanogaster or C. elegans (a.k.a.fruitflies and worms). When studied in the labora-tory, these systems disclose the genetic componentsof a given pathway. (Other approaches to pathwayanalysis include expression profiling of tissue sam-ples and studies of protein-protein interaction.)
Implementing pathway analysis will result in a lowercost per drug on average: it improves efficiency.Costs are reduced because pathway studies are rela-tively inexpensive and fewer human-derived targetsare required—pathway analysis expands the pool ofpotential targets tenfold or more. As targets, theyprove to be of high quality, moreover (in keepingwith the disease-susceptibility gene that inspiredtheir discovery), achieving good success rates inclinical trials. For although they themselves are notyet validated in humans or clearly implicated in thedisease, they participate in a pathway that is.
So pathway analysis should give human studies therequisite boost, with enough targets emerging toyield a drug more often than not. On the downside, there are the time and cost of additional ani-mal research and the loss of some advantages inclinical trials. Adding pathway analysis, via geneticstudies of Drosophila melanogaster, to indirectgenome-wide association studies would result in a
total savings of $425 million for each drug on aver-age, regardless of the original human studyapproach, though it would add nearly two years ofadditional work (producing an additional $255 mil-lion of value per drug).
The first one to do genetic studies takes a huge risk.If it works, you’ll see everyone running to join thecrowd.
—R&D executive, leading pharmaceutical company
Implementing Disease GeneticsThe savings promised by disease genetics are enor-mous, and companies cannot ignore them. But theycannot ignore the risks either, and will need to exer-cise rigorous selectivity and discipline when it comesto pursuing specific disease genetics studies—whichapproaches to adopt, for instance, and which dis-eases to explore. Placing bets in this way is going tobe nerve-racking enough. But companies faceanother difficult set of choices as well, in the variousoperational issues that need to be addressed.
Placing BetsSome diseases have clear appeal as objects ofgenetic research: asthma, Alzheimer’s disease, anddiabetes, for example, being complex diseases thatafflict large populations and obviously contain heri-table factors. They have already become competi-tive areas of study. Although opinion is still dividedover the likely impact of disease genetics, substan-tial bets are being placed by various companies—emerging biotechs and established pharmaceuticalcompanies alike. A few claim they are already see-ing benefits from their investments: following itsalliance with Roche, deCODE genetics claims tohave succeeded in identifying a gene contributingto cerebrovascular disease; GlaxoSmithKline hasannounced finding genes associated with migraine,Type II diabetes, and psoriasis; and Genset, aFrench biotech company, is reported to have iden-tified genes implicated in prostate cancer and schizophrenia.
But all companies embarked on, or intent on, pur-suing disease genetics must acknowledge that it is
32
still a scientifically risky endeavor. As they constructtheir customized portfolio of bets, they will have tokeep reviewing not just their internal capabilities butalso the degree of risk they are prepared to accept.
Putting Disease Genetics into Operation If companies do opt to participate in disease genet-ics, they may still choose to maintain some distanceby outsourcing the activity or licensing in theresults. Those that decide to launch a major diseasegenetics program in-house will confront a numberof significant challenges as they put the programinto operation.
Take the problem of obtaining the required sam-ples, for instance. From whom should samples becollected? And how are samples to be stored? Theuse of human tissue raises ethical considerations aswell. What constitutes consent? How can privacy beprotected? Who “owns” the tissue material? Andwho should profit from it? (Several companies,such as Genomics Collaborative and the not-for-profit First Genetic Trust, are emerging to addressthese conundrums.)
And, of course, there are major organizationalquestions. What are the implications for humanresources and labor relations? For big pharmaceuti-cal companies, new capabilities will be demanded,and new skills will have to be acquired—statisticalgeneticists, for instance, and experts on pathwaygenetic studies. And other capabilities may sud-denly be less in demand—perhaps even obsolete.
We will discuss implementation issues such as thesein more detail in the final chapter of this report.
Pharmacogenetics
Just as some genetic variations among individualsmay influence their susceptibility to diseases, soothers may influence their responsiveness to drugtreatments for those diseases. It is the goal of phar-macogenetics to seek out and characterize some ofthese latter variations.
The savings that pharmaceutical companies mighthope to harvest are considerable, though nothing
like as high as those that disease genetics stands toachieve in ideal circumstances.
The PotentialThe impact of pharmacogenetics on R&D produc-tivity will derive from the increased flexibility itintroduces into clinical development. Currently,drug-development policy is dominated by a binaryscenario in its later stages: either shepherd a com-pound through its clinical trials and out to market,or abandon it as unmarketable if it stumbles in thetrials. Pharmacogenetics provides a more nuancedscenario, with an expanded range of possible out-comes, by allowing the exclusion of patients geneti-cally predisposed to respond poorly to the drug.Two particular benefits emerge: “standard” clinicaltrials can now be streamlined; and “failing” com-pounds can now be salvaged. (See Exhibit 7.)
The streamlining of trials would apply only to com-pounds destined to proceed successfully throughthe clinical trial process anyway. That path can nowbe made smoother. The trials can be designed moresubtly. They can be smaller and quicker thanbefore, now that it is possible to preselect promis-ing patients—that is, patients whose geneticmakeup is likely to maximize the drug’s efficacyand minimize its side effects. So, patients lackingthe drug-susceptible variation of the target gene
33
EXHIBIT 7PHARMACOGENETICS EXPANDS DEVELOPMENT CHOICES
High
Prop
ortio
n of
pat
ient
s sho
wing
poor
or n
o re
spon
se
Low
Current options
Abandon drugbefore market
Continue clinicaltrials to market
Options available with pharmacogenetics
Continue trials safely byexcluding at-risk patients
Optimize clinical trials,making them
smaller and shorter
SOURCE: BCG analysis.
would be excluded from the trial, in order to showhigh efficacy levels for the subset of patients whowould eventually use the drug. Also excluded wouldbe patients having a specific genetic variation asso-ciated with side effects.
To see the streamlining effect of such exclusions,consider the case of Herceptin, a treatment foradvanced breast cancer. It is effective only in a sub-set of patients—the 25–30 percent whose tumorsoverexpress the HER2/neu oncogene. It is thisgene that serves as the drug target. By screening forHER2/neu expression, Genentech was able toexclude nonresponders—some two-thirds of thesubjects originally tested—early in the clinical trial.Without this prescreening, Genentech would haveneeded nine times as many patients in phase III toachieve significant results. The cost of such a trialwould have made Herceptin economically unviable.
Turning to the second benefit, for failing com-pounds pharmacogenetics lowers the hurdle by eas-ing the conditions for market viability. Considerspecifically those candidate drugs that reveal seriousside effects in a significant proportion of the sub-jects (such as the 7 percent of Caucasians with lowlevels of CYP2D6, an enzyme that helps metabolizesome 25 percent of all drugs). Traditionally, anysuch drug would be perceived as too risky to market,and would be abandoned in preclinical studies.Today, however, pharmacogenetics makes it possibleto identify the at-risk patients, so the drug would notbe disqualified right away, and could go on to provemarketable—patients would just need to be testedfor vulnerability before being given prescriptions.
These are the potential benefits to R&D, the pri-mary focus of this report. Even greater potential,some observers believe, may lie in market advan-tages. Three such advantages are possible: pricepremium, share shift, and new patients. In otherwords, if a perception emerges that the pharmaco-genetics-assisted drug is distinctly less risky or dis-tinctly more efficacious for the (now restricted) tar-get patient population, payers may tolerate a higher
price for the drug, physicians may favor it whenoffering new patients a prescription, and patientswho have shunned previous medications (owing toside effects, typically) may now choose to try it.Although it seems reasonable for pharmaceuticalcompanies to expect some market upside frommore efficacious, better tolerated therapies, itremains to be seen to what extent they will be ableto reap these market rewards.
Putting figures on the cost savings pharmacogenet-
ics benefits might achieve, our model estimates an
average of $335 million in the cost to drug—if
pharmacogenetics were to work every time it were
applied. But pharmacogenetics won’t work every
time. Given the set of cases where it is applied and
succeeds, the expected savings would average about
$80 million, as discussed below. And of course, cor-
responding to the potential market upside, there is
the counterpart scenario—potential destruction of
value in the market. Why the uncertainty? (See
Exhibit 8.)
34
EXHIBIT 8PHARMACOGENETICS’ POTENTIAL IS CONTINGENT
Cost to drug
$M
880Pre-genomics
545Pharmacogenetics: the promise1
800
ID
1,0008006004002000
Pharmacogenetics: expected savings2
Development
Preclinical Clinical
Chemistry
Screening Optimization
Biology
Target ID Target Validation
SOURCES: Technical literature; industry interviews; publicly available infor-
mation; BCG analysis.
1Savings per drug assuming pharmacogenetics can be applied across the
R&D pipeline.
2Average savings across R&D pipeline, given scientific and market limitations.
The UncertaintyThe sizable potential savings are mirrored by siz-able risks—market and regulatory risks this time, aswell as scientific and technical risks. In some cir-cumstances, market dynamics might be so unfavor-able that companies would be well advised to stepback and forgo the potential savings altogether.Once again, given the range of possible outcomes,the economic impact could ultimately be a negativeone—far from resulting in savings, applying phar-macogenetics could result in a net loss for R&D. Toassess pharmacogenetics realistically, companiesneed to ask two questions: How feasible is it? Andhow desirable is it?
Feasibility—the Technological LimitationsPharmacogenetics will not apply to all drugs. It willapply only where differing drug response is dueentirely to genetic variation, and where that rela-tionship can be elucidated.
It is fairly rare for both of these conditions to pre-vail. For one thing, biology-based variation in drugresponse can be due in part to environmental fac-tors: grapefruit juice, for example, is known tomodify the effect of certain drugs in certain indi-viduals, sometimes raising the uptake to dangerouslevels. For another, drug-response variation willoften be the work of multiple genes, acting severallyor jointly, and that compounds the statistical andtechnological difficulties of the search.
The business guys hear about this stuff, and arelike, “Great! Make it happen!” We’re left scratchingour heads, looking like poor sports, because a lot ofit just isn’t possible.
—Senior scientist, major biotech company
Even if the two conditions are fulfilled, a furtherchallenge lies in wait—to find the relevant genes intime to streamline the trial. For pharmacogeneticsto effect this streamlining, you need to be able toscreen out nonresponders. And that means findingthose variants, or identifying the nonrespondergenotype before designing any streamlined trial.
Now, developing a robust pharmacogenetic test
would generally require more than 1,000 patients.
A phase I trial is far too small for that purpose, so
no streamlining would be possible for a phase II
trial, despite the excitement surrounding that
prospect. Streamlining could be possible in time for
phase III trials, but even that is far from assured.
(See Exhibit 9.) (The other kind of screening test,
for side effects, is less problematic, since the associ-
ated metabolic variations are often determined in
preclinical trials. But that kind of test does not help
to streamline clinical trials, which have to be sized
to test for efficacy rather than for side effects.)
It is hard to see how these phase II trials will beused for pharmacogenetics, because most of thevariants are expected to be relatively infrequent.Certainly, 30 percent prevalence [which fallswithin standard clinical trial sizes] would be rela-tively rare.
—Geneticist,
The Whitehead Institute
35
EXHIBIT 9STATISTICAL REQUIREMENTS LIMIT PHARMACOGENETICS’POTENTIAL TO STREAMLINE TRIALS
Frequency of responder genotype
6050403020100
Number of patients required to develop a test
Prevalence >30% needed todevelop a test prior to phase III trials
Typical phase II trial size
5,000
4,000
3,000
2,000
1,000
Typical phase I trial size
SOURCES: Technical literature; industry interviews; publicly available infor-
mation; BCG analysis.
NOTE: Graph analyzes a scenario from Nature Biotechnology, vol. 18, May
2000. Scenario uses efficacy pharmacogenetics to identify the ApoE
responder genotype, a predictor of efficacy of Cognex (tacrine) for
Alzheimer’s. Strength of effect, a different variable, is not considered here.
Given these technological limitations, we estimatethat less than 15 percent of drugs will be amenableto the application of pharmacogenetics.
Desirability—Market EconomicsAs already mentioned, there are circumstances inwhich a company might have an incentive to shunpharmacogenetics entirely. After all, by excludingpatients from trials, you are in effect giving thedrug a restricted label when trying to market it.
Gauging the likely effect of a restricted labelinvolves some complex analysis. For a start, youneed to consider two distinct groups of patients:those who take a prescription for the full course oftreatment (which could last many years, or even theremaining lifetime for those suffering from chronicdiseases), and those who embark on a prescriptionbut then discontinue it for reasons of inefficacy orside effects.
The pharmacogenetics test would shrink these twopotential patient groups in different ways. From theformer, it would eliminate the “placebo respon-ders.” From the latter, it would eliminate some ofthe nonresponders and negative responders.
Market fragmentation has happened in manyindustries—the marketing group can’t put theirheads in the sand. We have to figure out what to doabout pharmacogenetics.
—Genetics director, leading biotech company
Since pharmacogenetics seems to be chipping awayat a drug’s market base, why pursue it in the firstplace? The answer may lie, in part, in competitivedynamics and game theory: companies may have toembrace pharmacogenetics because their competi-tors are doing so. Merck & Co., for example,according to a recent Wall Street Journal article, isbusy developing capabilities to reproduce pharma-cogenetic analyses conducted by its competitors, ifonly to disprove any claims that a rival drug mightbe superior to its own.
But the compensatory advantages can be more pos-
itive, too—the potential for market upside, once
again: price premium, share shift, and new patients.
What a company has to judge, before adopting
pharmacogenetics for any drug in development, is
the likely breakeven point—the point at which a
price premium or increased market share begins to
offset the volume loss. Our model assumes a modest
market premium of 20 percent, and calculates the
breakeven point in various scenarios, based on four
different approaches to pharmacogenetics. (See
sidebar, “Pharmacogenetics—Four Applications,”
and Exhibit 10.)
Efficacy-based pharmacogenetics can reduce trial costs
considerably. But the market dynamics could then
cast a cloud over that economic picture. If the
restricted label, by disqualifying placebo respon-
ders and some nonresponders and negative respon-
ders, translates into an overall revenue loss of just 2
percent, that cancels out the savings achieved in the
clinical trials.
36
EXHIBIT 10PHARMACOGENETICS’ VALUE DEPENDS ON MARKETDYNAMICS
Revenue increase required from market premium (%)
3002001000
Patients lacking good response (%)1
Pharmacogenetics can optimize clinical trials
Abandon drug
Conduct normal trialsPharmacogenetics
trials make drug viable
100
80
60
40
20
SOURCES: Industry interviews; BCG analysis.
1Example based on a scenario in Nature Biotechnology, vol. 18, May 2000
of ApoE4 efficacy in tacrine response; assumes response rate of 41 percent
among patients with the SNP versus 20 percent among those without it;
also assumes 50 percent of nonresponders discontinue.
37
P H A R M A C O G E N E T I C S — F O U R A P P L I C A T I O N S
To evaluate pharmacogenetics properly, companiesneed to take an especially close look at the marketdynamics. These dynamics vary according to howpharmacogenetics is used. We have identified foursuch applications (each associated with a differentcategory of patient responding to a given drug). Thefirst three are used to exclude patients from trials;the fourth is used to expand the potential market forthe drug.
First, efficacy prediction identifies patients who willshow no real or significant response to the drug—perhaps because they metabolize the drug in anunusual way, or have an unusual form or combina-tion of susceptibility genes. A typical drug producesthis negligible response in about a third of patients,but sometimes the proportion is far higher. Forexample, Cognex (tacrine), the first drug forAlzheimer’s, is inefficacious in more than 50 percentof patients. The varying response is associated withdiffering versions of the ApoE gene, and is thereforereadily predictable by a pharmacogenetic test.
Second, common-side-effect prediction identifiespatients likely to experience familiar side effects, asa result of metabolic difficulties caused by well-known enzymes. A test can screen out negativeresponders—“slow acetylators,” for instance. Theacetylation polymorphism in the NAT2 gene is one ofthe commonest genetic variations in drug metabo-lism; it has the effect of reducing the enzyme’s life-span and thus reducing the effective amount of theenzyme in cells at any one time. This polymorphismis present in more than 50 percent of Caucasians,who are thus at greater risk of drug toxicity.Knowledge of this polymorphism could save a drugin clinical trials that would otherwise be abandoned.
Third, very-rare-side-effect prediction identifies pa-tients at risk for unconventional side effects, butcomes into play only after the drug is on the market.Unlike most of the common side effects, which areassociated with metabolic pathways and usually
emerge in preclinical studies, these rare side effectstend to be provoked by nonmetabolic genes, and tobe overlooked at first. They cannot easily be pre-dicted, since there are too many possible sources(modifications of the target or of the disease path-way, or unrelated pathways), and they may occur toorarely to show up in clinical trials.
A case in point is Lotronex, a drug for irritable bowelsyndrome, now withdrawn from the market. Onlyafter its market launch, and 450,000 prescriptions,did its severe side effect (bowel impaction) becomeapparent. About one in 6,500 patients wasaffected—a frequency far too rare for a standardclinical trial to detect beforehand. (A typical trialinvolves about 5,000 patients: for this side effect tohave been manifest in a statistically significant way,a trial of nearly 100,000 patients would have beenneeded.) Pharmacogenetics could in certain casescome to the rescue of such compromised drugs, bybelatedly devising a screening test.
Finally, market expansion identifies patients who arecurrently unsuited to the drug but potentially respon-sive to it. Since fine-tuning of dosages or formulationcan often reduce side effects and occasionallyimprove efficacy, pharmacogenetics could reassessand upgrade many of the supposedly ineligiblepatients. The market for the drug might expand con-siderably as a result.
Take the case of cyclophosphamide, a chemotherapydrug, which works only when metabolized by theenzymes CYP3A4 and CYP3A5. Some patientsappear underresponsive to it: a genetic variationsuppresses the activity of the enzymes, therebydecreasing the amount of active drug in the blood-stream. The best course is not to discontinue thedrug, but to compensate by taking a higher dosage.A pharmacogenetic test could identify the appropri-ate patients prior to treatment, and their consump-tion of the drug, instead of declining to zero, wouldactually increase.
Side-effect-based pharmacogenetics for common side effectscan save a candidate drug that would otherwise fail.This form of pharmacogenetics does not streamlinetrials; in fact, it imposes a moderate increase in costs,since more patients have to be recruited initially forthe vetting process. It requires a smaller upside tobreak even than efficacy-based pharmacogenetics,however, since its powers of exclusion apply only tothe second type of patient (those who would ordi-narily try the drug and then discontinue it). They donot apply to the first group (those who would take afull course of the drug), since placebo respondersdo not suffer from side effects.
Side-effect-based pharmacogenetics for very rare side effectsis the type that would be applied for a drug alreadyon the market. Once the number of adverse events(instances of severe side effects) reaches a criticalmass, the drug’s reputation suffers, and its contin-ued marketability is jeopardized. (In severe cases,regulatory agencies require the drug to be removedfrom the market.) To salvage it would involve imple-menting screening tests for all potential patients—a kind of postmarket surveillance. This type ofpharmacogenetics would increase costs fairlysteeply, yet it could still make economic sense if thedrug were saved.
The economics hinge on a paradox: the fewer theadverse events, the harder it might be to save thedrug. To identify the culprit genetic marker for usein the screening test, you need a certain minimumnumber of patients who have experienced the sideeffect. That could be as low as 20 (assuming youachieved a 100 percent association with a singleSNP marker), and that would carry the modestprice tag of $100,000 (assuming the expected geno-typing cost of one cent per SNP). But the requirednumber could be 2,000 (assuming you achievedonly a 10 percent association), and the likelihood isthat such a number of side-effect sufferers wouldsimply never emerge.
It’s a crime that a very small percentage of patientscan sometimes eliminate an otherwise highly bene-ficial drug from the market. Pharmacogenetics ben-efits everyone here.
—Research executive,leading pharmaceutical company
Finally, market-expansion pharmacogenetics for themost part has highly favorable economics. Since itseffect is to expand rather than contract the market,all it needs to ensure is that the expansion be largeenough to cover the cost of the required trial.
The prospects hinge to some extent on the inci-dence of the genetic variant. Consider two drugs,one producing side effects in poor CYP2D19metabolizers (including 20 percent of Asians) andthe other in poor CYP2D6 metabolizers (including7 percent of Caucasians); and suppose that dosageadjustments could resolve the side effects. It mightturn out that the former case warrants the invest-ment and the latter does not, given the differencein their potential market expansions.
All in all, these limitations reduce the expected sav-ings that pharmacogenetics would bestow on anaverage drug to about $75 million. But of coursethere is no such thing as a true “average” drug. Insome cases, pharmacogenetics could yield potentialsavings as high as $335 million and potentially cap-ture additional upside through price premiums oran increase in market share. In other cases, it mightsave nothing, or even destroy value in the market.The key to success is to be selective.
Implementing PharmacogeneticsAlthough the savings attainable through pharmaco-genetics appear less dramatic than those attainablethrough disease genetics, they are in the right cir-cumstances quite substantial. And the total valueadded would be enormous if the hoped-for marketadvantages were realized.
38
Realizing this value is a matter not only of marketdynamics, but also of various more speculative fac-tors: how acceptable pharmacogenetics will proveto payers, patients, physicians, and regulatory agen-cies; how readily physicians and patients willembrace the screening tests to generate share shift;and so on. So the different types of pharmacoge-netics will probably come into effect at differenttimes. Detection of rare side effects will probably beintroduced first, as pharmaceutical companies arehighly motivated to save drugs from failure. Efficacypharmacogenetics will probably progress on aslower timetable, owing to concerns about marketfragmentation. It might even take FDA action toturn efficacy testing into a routine procedure.
When contemplating their pharmacogenetics pol-icy, companies will need to scrupulously analyzespecific drugs and markets. Deciding shrewdly justwhere and when to apply pharmacogenetics, forinstance, will mean assessing market dynamics ear-lier than ever before in the clinical trials phase. Andthat in turn will demand new decision-makingprocesses and communication channels, includingstronger ties between research and development,and between R&D and commercial activities. It ison operational and organizational issues of thiskind that the spotlight will fall in the next chapterof this report.
A Final Word
If the new genetics can realize its full potential, theeconomics of pharmaceutical R&D will undergo ametamorphosis. Efficiency will improve hand-somely and success rates will surge. The sums savedcould exceed a half billion dollars per drug, morethan halving the current cost.
That prospect is far from assured. There areenough risks and uncertainties to temper excite-
ment. The range of possible outcomes is wide, andcompanies will have to examine minutely and applyselectively the various genetics opportunities.
Contrast genomics technology: the productivityimprovements promised by its implementation maybe more modest, but they are clearly achievable,despite the operational challenges. With genetics,the operational challenges are formidable too, butthey are compounded by less distinct and possiblymore intractable challenges: technological limita-tions, scientific unknowns, and (in the case of phar-macogenetics) the vagaries of the marketplace.
So, companies determined to acquire and exploitgenetic information need to know what they are let-ting themselves in for. They need to consider howapplicable genetics is to their current researchstrategy. They need to spell out the level of risk theyare prepared to take on, and then plan how to man-age that risk. In short, they need a genetics strategy.
In the case of disease genetics, risk managementwould best begin by contemplating the sheer mag-nitude of the undertaking. Companies will beprompted to ask themselves questions such as these:How feasible is it for us to establish an extensivedisease genetics program in-house? On which dis-eases should our program focus? Are there oppor-tunities to share the risk, perhaps by joining a “pre-competitive” industry consortium, along the linesof the SNP Consortium? Or, should we adopt a wait-and-see stance, and then hope to license in thefruits of others’ labor?
In the case of pharmacogenetics, risk managementbegins by reevaluating the pipeline. On that basis,companies will try to determine the drugs to whichpharmacogenetics applications would add mostvalue. Companies will also want to study intently themarket context and competitor landscape, thinkingthrough potential competitor moves and counter-
39
moves, along with the relative scientific feasibilityfor each drug or therapeutic class. Such assess-ments will need to be revised continuously, as dif-ferent drugs present themselves for considerationand perhaps suggest different approaches, and asthe market and the regulatory environment con-tinue to evolve.
A genetics strategy would encompass all of theseissues, and would optimize any potential synergiesamong genetics approaches. If a company decidesto implement both disease genetics and pharmaco-genetics, it will need to decide how to integrate andharmonize them. Which diseases, for example,might be amenable to disease genetics on the onehand, and be likely to provide a market premium
on the other? If genetic redefinition of diseasesmakes it possible to develop suites of drugs andthereby address several smaller markets, how canresearch best collaborate with marketing to maxi-mize the impact? The answers to these questionswill generate still more questions: Do we have therequisite skills and capabilities to pursue the strat-egy? Do we have the right alliances in place, or theright alliance strategy?
Genetics is a risky endeavor. Companies cannotavoid the risk—but they cannot lightly ignore thepotential jackpot either. They need to be selectiveand smart in deciding how and where to place theirbets. With such vast winnings at stake, it seemsappropriate that the odds should be fairly long.
40
41
Chapter 3: Managerial Challenges
Preface: Looking Back and Looking Forward
The Story So FarThe genomics revolution is poised to sweep asidethe old economics of pharmaceutical R&D. Thebiotechnology and pharmaceutical industries—andperhaps health care delivery in general—are on thebrink of transformation, and companies thatembrace the revolution in the right way stand toreap enormous benefits. Developing a new drugshould become considerably less unpredictable andmuch less expensive. Companies will recordimprovements both in efficiency and in successrates all along the value chain, and the average costand time needed to bring a new drug to market willfall correspondingly. (See sidebar, “PotentialSavings—From Theoretical to Practical.”)
But this benign prospect is clouded by some warn-ings: great rewards will require comparably greatefforts; a new paradigm in R&D economics maynecessitate paradigm shifts in R&D management;above all, the great promise is offset by great risks—though, as in any revolution, the risks of standingaside may be greater than those of getting involved.
Ensuring Your FutureAll biopharmaceutical companies are, or should be,actively deciding how best to engage in the revolu-tion. Making such decisions is no easy matter. Thefamiliar bearings are no longer there, since thecompetitive and regulatory landscapes havechanged so much—and continue to change—in re-sponse to the promise that genomics offers. Compa-
nies have been rushing to claim intellectual prop-erty rights (in the so-called IP land grab), now thatthe sequencing of the human genome has beencompleted. Statutes and court decisions regulatingthose IP rights keep emerging and modifying thepicture. (See sidebar, “Intellectual Property—Lostand Found.”) And the corporate map is being re-drawn: the major mergers of recent years have cre-ated industry superpowers, and the pace of acquisi-tions and alliances is set to quicken, if anything.(See sidebar, “Industry Changes.”)
With so much change occurring, there are boundto be winners and losers. Although the decisionswill be unfamiliar and difficult, success will in theend be determined by traditional criteria. The win-ners will be those who make optimal strategicchoices and then implement them in an optimalway. The two components of the winning combina-tion will differ from company to company, accord-ing to each company’s size, aspirations, financialpower, capabilities, and so on. In this final chapterof our report, we identify the strategic and opera-tional issues and examine the various options thatdifferent companies might exercise.
To begin with the strategic issues, then—specifi-cally, the challenge of defining a strategy in thegenomics era.
Strategy—Searching for Genomic Competitive Advantage
Before genomics, biopharmaceutical companiesused two basic tools—chemistry and molecular biol-
42
In the first two chapters of this report, we assessedthe potential savings for the two waves of thegenomics revolution: first, the substantial savingsattainable through genomics technologies; second, thegreater but far less certain savings attainable throughgenetics approaches, notably disease genetics andpharmacogenetics. Those assessments show the highend of the achievable range, and they view the twowaves singly rather than jointly; that is, they indicatediscrete and best-case scenarios, which will be diffi-cult for companies to realize in practice and impossi-ble to combine.
A more integrated assessment needs to average outthe achievable range—to take account of worst-casescenarios too. And it has to analyze the various likelycombinations of approaches from the two waves,rather than treating genomics and geneticsapproaches in isolation.
According to the combination selected, the R&Dvalue chain as a whole will assume a particular newform and favor a particular subset of potential drugs.That is a crucial consideration for a company engagedin building a portfolio of technologies: the more com-
binations of approaches, the greater the company’sversatility in pursuing different drug subsets.
Drawing once again on our economic model, wehave examined the impact of each feasiblegenomics-based approach to target discovery, aug-mented by pharmacogenetics and genomics technol-ogy whenever possible. And we have estimated therealizable value in each case: first, by calculating thecost, time, and added value for each possible com-bination of approaches, and then—adjusting for thepercentage of targets each approach is able toprocess—by calculating a weighted average perdrug. The result is three broad scenarios:
• A genomics-based approach: industrialized targetidentification, supplemented where applicable bydownstream genomics technologies (in silicochemistry, in silico toxicology, in vitro ADME, sur-rogate markers, and pharmacogenetics)
• Chemical genomics: industrialized target identifi-cation, followed by chemistry and traditional vali-dation conducted in parallel, and supplementedwhere possible by the downstream technologiesjust listed
P O T E N T I A L S A V I N G S — F R O M T H E O R E T I C A L T O P R A C T I C A L
Development
Preclinical Clinical
Chemistry
Screening OptimizationOptimization
Genomics: In silico/in vitro tests, surrogate markers
Pharmacogenetics
Genomics and pharmacogenetics
Traditional
73%
25%
<1%
2%
30%In silico
70%Traditional
Genomic target discovery
Chemical genomics target discovery
Genetic target discovery with pathway analysis
Biology
Target ID Target Validation
DRUG R&D VALUE CHAIN ACTIVITIES
43
• Disease genetics supported by pathway analysis,combined where possible with the various down-stream technologies
The chart on page 42 shows the three scenarios,and the weighting applied to calculate the averagecost per drug. Here, by way of example, are realisticsavings estimates generated by our economic modelfor these three approaches.
First, a genomics-based approach in a traditionalvalue chain structure. This provides an ideal founda-tion for a portfolio perspective on drug discovery. Itapplies to both known and unknown target classes,and offers impressive potential savings of $200 mil-lion and 1.5 years per drug.
Next, chemical genomics is probably the most com-petitive approach for targets where there is an estab-lished high-throughput chemical screening assay.The time savings are particularly important—nearly3.5 years—boosting drug revenues by means ofintellectual property rights and first-to-marketadvantages. When combined with other genomicsapproaches, chemical genomics offers potential sav-ings of about $100 million per drug. The approachlends itself particularly well to certain targets, suchas GPCRs, so a company electing not to pursuechemical genomics would be at a disadvantage if itretained such targets on its wish list.
Finally, disease genetics supported by pathwayanalysis is, theoretically, the most direct route tounderstanding human disease. This approach isapplicable to known and unknown target classes andto targets overlooked in animal-based studies. Andon the face of it, it is the most attractive approachfinancially, with cost savings of $400 million. Thereis an extended time to drug, however, amounting toabout one year, though that drawback would often
be offset by the early securing of intellectual prop-erty rights. Unfortunately, this inviting approach isstill not affordable, and in fact its scientific feasibil-ity remains unproven. (See the graphs below for asummary of expected savings.)
REALIZABLE RESULTS BASED ON DISCOVERY APPROACH
Cost to drug
Cost ($M)
Weighted average of approaches across value chain
Time to drug
880Pre-genomics
675Genomics-basedapproach
780Chemical genomics-based approach
475Genetics-based approach1
Pre-genomics
Genomics-basedapproach
Chemical genomics-based approach
Genetics-based approach1
Time (years)
13.3
11.3
15.8
14.7
1,0008006004002000
20151050
1With pathway analysis, 50% through genetic simple systems, 50% through
genomics expression profiling. Assumes resolution of scientific and techno-
logical questions.
44
ogy—to discover new drugs. Broadly speaking, thedrugs that emerged were much indebted toserendipity. Research strategy consisted mainly ofchoosing which therapeutic areas to investigate, anddiscovery efforts focused on individual drug targets.Development provided even fewer strategic choices:a promising compound emerging from chemistrywould be tested on animals and humans in largeand inefficient trials (inefficient because there wasno means of identifying in advance likely respon-ders or nonresponders). With the rise of genomics,there have come new technologies, new approaches,new information, and new ways of thinking about
research and development. These have broughtwith them a new opportunity, or imperative, to turnresearch to competitive advantage.
So companies now have weighty strategic issues to
address. At the corporate level, the question is how
much to invest, given the current environment. For
R&D leadership, the question tends to be where to
focus those investments—in what therapy areas, on
what target classes, and so on—as well as which
technologies to adopt and how to adopt them (in-
house or externally, for example), and how to miti-
gate the associated risks.
I N T E L L E C T U A L P R O P E R T Y — L O S T A N D F O U N D
Gene patent applications are flourishing: in 2000alone, more than 20,000 were submitted to theUnited States Patent and Trademark Office. Despitethe large number of applications, two central ques-tions have yet to be answered. What exactly can bepatented? And what rights does a patent actuallyconfer on its holder?
For a gene or gene fragment (an expressed sequencetag, or EST) to secure a patent, its “utility” has to beestablished. In January 2001, the United StatesPatent and Trademark Office issued Utility Exami-nation Guidelines to clarify the standard used. (Thatin itself was encouraging to those in favor of genepatenting, reinforcing the view that genetic materialcan indeed be patented.) Following on the interimguidelines released in 1999, the new guidelinesadvise patent seekers to provide at least one “spe-cific, substantial, and credible” use for the gene orgene fragment in question. This restatement effec-tively fixes the height of the hurdle for applicantsand disqualifies undersubstantiated applicationsfrom the outset. Some uncertainty remains, how-ever: whether it is necessary when presenting evi-dence of utility to understand the actual biologicalfunction of the genetic material, and whether gene
fragments, as distinct from full-length genes, are eli-gible for a patent.
If getting a patent approved seems daunting, all themore so is enforcing it, or sheltering confidently inits protective embrace. The strength of protectionafforded by a gene patent is still a developing legalissue. One recent court decision, though, clearlymarks a setback for patent holders—Festo Corp. vShoketsu Kinzoku Kogyo Kabushiki, decided by thefederal circuit court in November 2000. It appearsto weaken many patents by precluding a broad inter-pretation of most patent claims; it does so by virtu-ally excluding the “doctrine of equivalents.” This is adoctrine that generally allows extension of a patent’sclaim beyond its literal language, so that a would-beinfringer who makes trivial changes to the patentedproduct is not thereby exempt from the patent’s con-straints. According to the Festo ruling, if the lan-guage of the legal claim diverges from that of thepatent itself, the doctrine no longer applies. Manypatents now look very narrow and vulnerable, andcompanies will have to plan their patent submis-sions even more carefully in the future. Or hope fora change of fortune: the U.S. Supreme Court isexpected to review Festo in the 2001–2002 term.
45
The Starting PositionAlthough these same broad questions will applyequally to all companies, there can be no standardanswers. The actual options available to any com-pany will depend on its starting position.
Company SizeA key constraint on a company’s strategic options issize. The largest pharmaceutical companies boastcapabilities and finances on a scale that allows fullparticipation in the new technologies, even whenthe risk is high. Not that this exempts them fromhaving to make choices. In fact, since scale givesthem so many options, they arguably carry a greaterburden of strategic decision making. How to selectfrom such an embarrassment of riches? In addition,they face the challenge of managing complexity. Ifthey are not selective enough, and embrace toomany options, the operational problems couldprove overwhelming.
The narrower capabilities and lesser scale of small-to-medium-sized pharmaceutical companies and thelarger biotech companies could represent either asevere drawback or a distinct advantage. On the onehand, there are reduced opportunities and even theprospect of being locked out by the big pharmaceu-tical firms: with disease genetics, for instance, a com-pany with insufficient scale to build an in-housecapability would risk forfeiting potentially lucrativeintellectual property rights. On the other hand,since lesser scale often means lesser complexity,these modest-sized companies can compete moreflexibly, changing their tactics quickly in response totechnological advances or competitor moves.
To see how scale can affect a company’s options,consider the differing ways in which large and mid-size companies approach the target land grab. Thelarger companies have been able to take veryaggressive approaches—scaling up or pursuing bigdeals to secure intellectual property rights to tar-gets. The smaller companies, lacking in resources,have been unable to follow suit, but some of themhave compensated by choosing very focused strate-gies, concentrating on their special competenciesand imposing higher quality standards.
Building the Fact BaseApart from company size, the two most importantfacets of a company’s starting point are the beliefsand hypotheses held by its leadership team (roughly,its corporate culture) and its current R&D capabili-ties. Companies need to scrutinize both.
It is crucial to understand and shape the beliefs andhypotheses of leaders throughout the organization,especially since, with genomics and genetics, thecontributions and effects are cross-functional—thatis, the managers or sections that contribute mostare not necessarily those that benefit most. Allthose affected need to articulate their perceptionsof the value and applicability of genomics andgenetics to the company. Once tested, these per-ceptions should be given considerable weight whenit comes to defining company strategy.
An equally thorough assessment needs to be madeof the company’s relevant R&D capabilities—itstechnologies, skills, specific knowledge of diseasesand disease mechanisms, and so on. Ideally, this willinclude an audit of current R&D productivity atevery step in the value chain, identifying bottle-necks and other constraints. The more accurateand detailed the assessment, the more effectivelythe company can address the strategic questions asthey pertain to its specific situation.
Corporate Decisions: How Much to Invest and WhereAs suggested above, even the largest pharmaceuti-cal companies will have to make choices. Considersome of the huge deals of recent times: the $500million deal between Bayer and Millennium for tar-gets, the $800 million deal between Novartis andVertex for in silico chemistry, the $500 million dealbetween Roche and deCODE for disease genes.Note that these deals concern discrete steps of thevalue chain: in each case, it appears likely that thecompanies concerned were acting on an explicitpreference—an established strategic preference.After all, given the magnitude of these deals, itseems unlikely that any one company would haveplaced all three bets. More to the point, such largedeals, although essentially R&D ventures, are not
46
I N D U S T R Y C H A N G E S
The genomics landscape features many small start-ups amid the larger genomics companies and thegenomics divisions of big pharmaceutical corpora-tions. But that landscape is changing. The numberof deals—of genomics companies combining witheach other or being taken over by big pharmaceuti-cal firms—has been growing steadily. What is driv-ing this tendency toward consolidation?
The Pressure to Extend ScopeIncreasingly, genomics companies are aspiring tobecome full-fledged drug companies. No specializedcompany, it seems, has yet succeeded in building atruly stable competitive position as a drug-industrysupplier; intellectual-property statutes do notappear to be enough to guarantee long-term protec-tion; and the chances of proprietary advantage arebeing nullified by the trend toward public-privatepartnerships or consortia, underwritten by big phar-maceutical companies.
Wall Street appears to place a far higher value onintegrated drug producers than on pure technologycompanies (if only because the drug sector has tra-ditionally enjoyed such high profits and such highregard among investors). According to a recent USBWarburg study, the average integrated drug com-pany has been able to raise $870 million, as againsta mere $330 million for the average technologycompany. (The study noted a further interestingdivergence among technology companies them-selves: biology companies—those focused on targetidentification and validation—raised $480 millionon average, whereas companies in the chemistryarea—those focused on screening and lead opti-mization—raised on average only $170 million.)
In keeping with this expansionist aspiration, most ofthe recent deals have consisted of acquisitions ofdownstream drug-development capabilities. Witness
LION’s acquisition of Trega (for $35 million),Celera’s acquisition of AxyS (for $173 million), andLexicon’s acquisition of Coelacanth (for $32 million).
The Pressure to Achieve ScaleAs sections of the value chain have become indus-trialized, the value of scale in R&D has gainedprominence. And for genomics platform companiesand pharmaceutical companies alike, it may appearquicker and neater to achieve scale through a mergerthan through painstaking in-house upscaling. (Ofcourse, pharmaceutical companies might have otherreasons to acquire genomics companies: to jump-start their genomics efforts, for instance, or toacquire otherwise rare capabilities.)
Sure enough, most of the recent mergers and acqui-sitions have clearly been initiated for the sake ofincreasing scale: Sequenom’s acquisition of GeminiGenomics (for $238 million), for example, orSangamo’s acquisition of Gendaq (for $40 million).These scale deals have been primarily in target iden-tification and validation rather than in chemistry—areflection of the urgency of the land grab.
The Pressure to SpendWhatever the inducements to merge, there is a tra-ditional impediment—lack of wherewithal. The spiritis willing but the purse is weak. That is certainly nota constraint, however, on some of the largegenomics companies at the moment. In mid-2001,three top companies—Human Genome Sciences,Celera, and Millennium—boasted over $4 billion incash between them, representing about 25 percentof their combined market capitalizations. Idle moneycries out to be spent—probably, for these compa-nies, on diversification more than on scaling.
Market expectations, a sense of urgency, an abun-dance of funds: all signs pointing to continued con-solidation in the near term.
47
R&D decisions alone. Almost certainly, the deci-sions were thrashed out at the corporate level.
For more modest-sized companies, strategic choicesoften go beyond matters of preference or emphasis.The question might be whether to concentrate alltheir efforts on some value chain steps and forgoothers altogether. Certainly it no longer makessense for even midsized pharmaceutical companiesto compete in target identification. And at thesmaller end of the scale, companies with less than$400 million in R&D, say, may find themselves ask-ing even more radical questions: Can we affordresearch at all? Should we not focus exclusively onlicensing instead? Again, it is at the corporate level,rather than within R&D alone, that such questionswill eventually be settled.
It is not just through major partnerships andinvestment decisions, however, that the corporatelevel is impinging on R&D strategy. More and more,specific R&D activities are having ramificationsbeyond R&D itself, and invoking corporate-levelparticipation. Pharmacogenetics, for instance,often touches on corporate strategy as much as onR&D strategy. Should the company continue to pur-sue a promising compound, say, when the risk ofmarket fragmentation might outweigh the positivemarket effects? Should the company attempt to res-urrect candidate drugs previously killed because ofrare side effects? And so on.
R&D Leadership Decisions: Where and How to CompeteWith genomics and genetics now part of the land-scape, R&D decision making has become morecomplex. The options are far more numerous:there are more ways of gaining access to capabili-ties, more technologies to choose among, and evennew dimensions in which to compete. R&D execu-tives must select a combination of options that notonly dovetail with the company’s starting positionand aspirations but can also be integrated smoothlywith one another.
Choosing a Research FocusThe dimensions of competition include:
• Disease states. Some disease states have becomemore tractable, thanks to genomics approaches,and any company continuing to investigate themwill have to deploy genomics if it is to remaincompetitive. Just which therapeutic areas or dis-ease states are most amenable to genomics isdetermined by several factors: the degree towhich the disease is genetic in nature, the currentunderstanding of disease processes at a molecularor genetic level, and so on.
• Target class. Some genomics approaches are atodds with traditional therapeutic-area borders,and favor a broader deployment—around targetclass—rather than the old focus on disease state.(The targets within a class are usually similar instructure and biochemical function.)
• Therapeutic modalities. Small-molecule drugsstill dominate the market, but they no longermonopolize it. Some new therapeutic modalitieshave already established a foothold—injectibleprotein therapeutics, for instance, based onsecreted factors and antibodies. Others remainvery much in the experimental stage—gene ther-apy and anti-sense techology, for example—thoughadventurous companies are pursuing themundaunted (as exemplified by Lilly’s recent $200million deal with Isis to gain access to anti-sensecapabilities).
These dimensions are interconnected, of course,and even interdependent. Take Novartis’s interestin oncology, for example—a broad disease state.Given that interest, it made sense for the companyto focus on kinases, a key target class in oncology.Kinases constitute one of the few target classes thatare amenable to a particular genomics approach, insilico drug design. Novartis has duly set about aug-menting its expertise with the appropriategenomics technology, forming an alliance withVertex to that end.
48
Selecting Technologies According to the research focus adopted by thecompany, certain technologies will press theirclaims immediately. An oncology program, forinstance, would certainly argue for the incorpora-tion of a transcription profiling approach, as moreand more cancers are being redefined at the levelof RNA expression. But each claim would have to beassessed by reference to the company’s aspirationsand current capabilities. How comfortably would acandidate technology fit in with the company’s riskprofile or existing skills mix, for example?
In addition, companies will need to consider thecurrent stage of development of genomics tech-nologies. When is the best time to buy into thefavored technology? As noted throughout thisreport, although some genomics approaches arepracticable today—in the early steps of the valuechain, notably—others remain speculative:genome-wide association studies, for instance. Acompany’s risk profile will determine whether itwishes to be on the “bleeding edge” or to be a tech-nology follower. Either way, the company will wantto chart the evolution of genomics technologiesand approaches, and adjust its own strategy accord-ingly. A technology scouting function is indispensa-ble, now more than ever.
Whether the technology is proven or unproven,companies will need to decide not just whether andwhen to invest, but also how—how to keep a sharpfocus and mitigate the risks involved. The optionsvary from company to company, again according tocompany size. With disease genetics, say, a largepharmaceutical company that chose to pursue thetechnology in-house would face the question of howto apply it—to which therapeutic areas, for exam-ple. A smaller company, by contrast, unable to builda program in-house, and obliged to take a differentapproach, would face such questions as what kindof joint ventures to pursue and what focus to apply.
Deciding How to Acquire or Gain Access to CapabilitiesIn general, there are several ways to attain a desiredcapability, but in some cases the options are limited.
When the item is a proprietary database or tool, forinstance, the company will have to license it in (orpay a provider for service) rather than buy it out-right; or when a company views its own informationas too confidential to outsource, it will be forced toimplement the related technology in-house. Inmany cases, though, a company will face the choicebetween building in-house capabilities and out-sourcing. The in-house option, to justify itself,would have to confer some significant strategic orcost advantage. A company could have a cost advan-tage if it had developed a proprietary method, forexample, or if it could boast greater scale or expe-rience in a given approach.
Some though not all of the new technologies showclear scale benefits, thanks to industrialized proc-esses and informatics. (Among the most obligingtechnologies in this regard are expression profiling,traditional HTS and µHTS, and exploitation ofinformatics-based analysis. The least obliging aremedicinal chemistry and animal models, and some-where in between are compound synthesis andmanagement, proteomic expression analysis, struc-tural biology, and in silico chemistry.) Unfortu-nately, building scale in-house could be dispropor-tionately costly for small-to-midsized pharma-ceutical companies, even for the most scale-friendlytechnologies. These companies are unlikely to real-ize cost advantages; they risk spreading their tech-nology dollars too thin. The wiser option would bepartnering or licensing.
If a company decides to develop a given technologyin-house, it should review that decision regularly.What is today a strategically advantageous capabilitymay be commoditized tomorrow. The perception ofsequencing, for instance, seems to be shifting, froma need-to-have technology to something that canreadily be outsourced.
If a company decides to outsource a given technol-ogy, it will have to decide further on a prospectivepartner or partners. It might even opt to join forceswith competitors. A model partnership of this kindhas been the SNP Consortium. A group of pharma-ceutical companies, helped by various academic
49
institutions, banded together to identify 300,000
SNPs (in the end, the total was about one million)
and put them into the public domain. This joint
effort had two very beneficial effects for its partici-
pants. First, it enabled the companies to concen-
trate more on their core interest, finding drugs;
second, it forestalled the efforts of genomics com-
panies, which would have sought to patent and
extract rents from these SNPs. Other candidates for
“coopetition” of this kind include protein structure
modeling and broad-scale sample collection for dis-
ease association studies.
Putting the Strategy into Operation
Defining a genomics strategy is a good start, but
even the most brilliant strategy is futile if it remains
defined on paper only. The point is to put it into
operation. Putting a strategy into operation consists
essentially of making changes and managing them
effectively. In the case of genomics and genetics,
the changes that need to be made are profound,
affecting all aspects of the R&D organization and,
by extension, the corporation as a whole—core
processes, organizational structure, job descrip-
A midsized pharmaceutical company initiated anR&D strategy overhaul. With the broad goal of in-creasing productivity, the effort was directed at re-ducing the number of and concentrating the areas ofinvestigation and boosting the company’s access torelevant technologies.
The CEO and R&D director commissioned a reviewof the company’s research and development capabil-ities. A cross-functional project team then set aboutdefining those capabilities precisely at all steps ofthe value chain, rating their quality, aligning themwith the diseases and markets of interest, and iden-tifying gaps and synergies.
With specific therapeutic areas in mind, the com-pany next turned to optimizing its genomics technol-ogy portfolio. The project team embarked on a three-stage assessment of the various strategic optionsand their corresponding technologies.
First, having audited the company’s capabilities andresearch interests, the team identified relevantindustry trends, and from these generated a list ofthe strategic options. (Genetics was listed, for exam-ple, as a major target-discovery trend in a favoredresearch area—a therapeutic area with many herita-ble diseases.) The team then compiled a list of
matching technologies, whether currently owned ornot—those that would enhance the options’ chanceof success. (Against the genetics option, forinstance, were listed such matches as SNP mapsand genotyping technologies.) Finally, the team eval-uated each technology’s likely impact on productiv-ity, using a sophisticated productivity model thecompany had established.
What emerged was a ranking of various strategicoptions and their required technologies—in effect,the basis for a new, integrated technology strategyand the blueprint for an optimized technology port-folio. Company executives were now in a position toponder access arrangements—whether upgrading,licensing, or partnering.
More broadly, the result of this effort has been a dra-matic focusing and aligning of research activities,directed by a well-articulated strategy. The R&Dmanagers have been able to commit to specific pro-ductivity metrics and time frames. They have beenable to agree on a distinct research focus: three keytherapeutic areas and limited modalities. And theynow have a specific technology strategy, with a clearplan for acquiring the necessary components and aninvestment program to make it happen.
C A S E S T U D Y : B R I N G I N G R E S E A R C H I N T O F O C U S
50
tions, interfaces, and so on. The necessary work canbe divided into three broad areas:
• Rebalancing the value chain
• Establishing the new organization and its governance
• Managing organizational change
Rebalancing the Value ChainThe old ways of conducting R&D are often unsuitedto the new era. As the first chapter showed, the tra-ditional R&D value chain no longer works. For onething, its smooth flow quickly becomes disrupted bya series of bottlenecks, induced by the differentproductivity gains at different phases. For another,its sequence is unsustainable, since the new tech-nologies dance to a different schedule: much of thechemistry phase might now take place simultane-ously with target validation, for example. To rein-state a smooth flow (while enjoying the new, muchaccelerated rate of throughput), R&D needs to easethe bottlenecks and adjust to a reconfigured valuechain. And that means redistributing resourcesand, more importantly, redesigning processes, aswell as keeping the new value chain in balance.
Restoring Balance: Reallocation versus Redesign At first sight, the bottleneck problem would seemfairly simple to resolve: scale up downstream stepsto meet the increased demand. But how feasiblewould that be? The number of targets identifiedcould increase sixfold or more. To scale up to meetthat increase, a company accustomed to spending$1 billion on all of R&D would now have to spendmore than $1.5 billion on target validation alone.
Another simple approach suggests itself: adjustresources along the value chain in order to bringthe uneven phases back into balance, shifting fundsfrom more efficient phases (notably target discov-ery) to less efficient downstream phases (such aspreclinical). But such reallocation of resources is,on its own, an overcautious measure, and will nothave a really dramatic impact on R&D economics. It
neglects, or even distracts from, the central oppor-tunity that genomics offers: the opportunity to“raise the game” by changing fundamentally theway R&D is conducted.
This transformation of R&D will derive above allfrom the bold reconfiguring of processes, for thesake of both physical process flow and informationflow. For the former, new technology platformsneed to be integrated and optimized, both withinvalue chain steps and across the value chain. (Insome cases, this may require a discipline and arearrangement comparable to the moving assemblyline introduced by Henry Ford in 1913.) As forinformation flow, the tremendous amount of datagenerated by the new technologies remains worth-less unless translated into functional informationand supplied punctually—that is, in time to influ-ence the decisions being made. (See sidebar,“Establishing a Unified Informatics Structure.”)
The extent of the redesign, and the particularshape the new flows take, depend very much on thecompany’s strategy choices. Processes that are newlyindustrialized, but that still follow a traditionalR&D sequence, need to be systematized. In somecases, however, the traditional R&D value chain willneed to be disrupted. To integrate chemical geno-mics and genetics, for instance, would necessitate amajor restructuring of the value chain. Chemicalgenomics introduces a new parallelism, as targetvalidation and chemistry activities are conductedsimultaneously; the two processes now interactrather than just interface. And genetics introducesfeedback loops, where late-phase findings (such asgenetic information from the clinic) feed back intoearlier steps of the value chain (such as disease-genetics-based target discovery).
In anticipation of any process redesign, individualfunction heads should be pondering the contin-gencies: how and when genomics might affectthem, and what actions to take when it does. As onebottleneck is relieved, another is created: When willthe bottleneck reach their step in the chain, andwhat will its impact be? What new technologies and
51
E S T A B L I S H I N G A U N I F I E D I N F O R M A T I C S I N F R A S T R U C T U R E
For a company to extract full value from the newtechnologies and the copious data that will emergefrom them—that is, to transform data into knowl-edge—it will have to devise a comprehensive infor-matics vision and architecture. The vision will artic-ulate the role of informatics as a potential source ofcompetitive advantage. The architecture will havethree essential components
First, there must be an optimized information flowacross the newly industrialized research and devel-opment value chain. Standards will need to beestablished to ensure that data are formatted, or-ganized, and defined consistently. Hardware andapplications will have to be linked and networkedappropriately so that information can move where itneeds to, feeding subsequent steps in the process.Second, a centralized knowledge management sys-tem is required, to capture and store the data, inte-grate it with external data, and make it availablethroughout the company. Finally, powerful analyticaltools will be required, to mine and make sense of thedata—sophisticated algorithms, visualization tools,and so on.
To develop and integrate these components, mostcompanies will have to invest heavily. The costs maylook particularly high in relation to traditional costs,but that is partly because the industry has generallyunderinvested in IT. (One large biotech company, fol-lowing its informatics upgrade, reports a threefoldincrease in its annual IT budget.) To focus the invest-ment accurately, a coherent plan is once againessential. A critical decision is whether to developthe capabilities in-house, outsource to solutionsproviders, or purchase and integrate informaticspackages. The choice or choices made will dependon such factors as available internal expertise, theamount of integration with legacy information sys-
tems required, and the availability of reliable inte-gration vendors, package suppliers, or solutionsproviders.
That last factor may prove particularly difficult toassess. Who would provide the most reliable andsuitable assistance? Today, no single provider ofapplication software provides all the functionalityneeded. The informatics industry is crowded withsmall start-ups offering niche products; the cumula-tive market capitalization of all publicly listed bioin-formatics companies scarcely amounts to one-eighthof GlaxoSmithKline’s annual R&D budget. And whilelarger IT solutions providers are gearing up to servethe burgeoning needs of this market, they are still inthe process of developing internal life-sciences capa-bilities. Given that each prospective solutionsprovider is likely to try to make its offering the cen-terpiece of the company's informatics architecture,and given that multiple solutions will need to beknitted together, companies would be wise to solicitindependent, unbiased advice before deciding onparticular vendors.
In addition to managing all this informatics com-plexity, companies will have to deal with a furtherchallenge if they are to implement the new informa-tion regime successfully. They will have to find a wayto resolve the human-resources and organizationalissues that are bound to arise. Talented, experiencedinformatics personnel are difficult to come by: howto find and keep the right people and how to fit theminto the organizational structure are questions thatcompanies will need to address more actively andimaginatively than ever. So too the question of howto change processes and behaviors generally,throughout the organization, to ensure fullest use ofthe new informatics tools—a question examined insome detail elsewhere in this chapter.
52
approaches will be available, and how effectivelywill they relieve the bottleneck? The head of devel-opment, for example, should already be contem-plating the inevitable increase in demand for clini-cal trial capacity and weighing the various options,such as pharmacogenetics, for meeting it.
Retaining Balance: Capacity Planning and Management (CPM)After the initial jolt of genomics, supply and demandshould get back into alignment, thanks to the com-bined forces of resource reallocation and processredesign. But this restored balance is a precariousone, and needs careful and regular maintenance.That is where capacity planning and management,or CPM, can play an invaluable role. By enabling anorganization to keep supply and demand aligned,CPM also enables it to make rational plans, linked tocapacities and resources, and thereby to manageprojects with optimal efficiency.
Though well established in high-profile corpora-tions such as General Electric, Hewlett-Packard,and Cisco, CPM is conspicuously rare in biotechand pharmaceutical companies. For the genomicsrevolution to realize anything like its full productiv-ity potential, efficient CPM will be immensely bene-ficial if not imperative.
Establishing the New Organization and itsGovernanceImplementing the process changes just mentionedwill entail a thorough review of a company’s exist-ing hierarchies and procedures. For the processchanges to yield optimal value, changes also needto be made in traditional decision-making methodsand in organizational structures.
New Linkages and InterfacesTo begin with organizational changes. With thevalue chain so much altered in appearance, and
In a global pharmaceutical company, the discoverydivision was slowly undergoing a change of charac-ter, from a pre-genomics one of small, independentefforts to an up-to-date one of highly defined andsequenced processes. The company was anxious tospeed up and optimize this inevitable transition. Ithad recently created several centers of excellence inresearch, each containing a particular combinationof technologies, resources, and expertise, but thesenew groupings remained in need of improvedprocess flow, both within and between them.
An “industrializing” approach was proposed: whynot treat each center of excellence as a factory, andin that way rethink or refashion the discovery pro-gram as a whole?
A factory-based structure imposes a strict discipline.Typically, factories have clear, measurable objec-tives, with defined inputs and outputs, and specifiedresources and roles. Efficiency is closely regulated:internal processes and interfaces are optimized for
scale, quality, and productivity; external interactionsare monitored regularly for compatibility and costeffectiveness.
In keeping with this ethos, the company set aboutdefining processes within each potential factory asclearly as possible. The project team set targets forinputs, outputs, and quality standards; it identifiedactivities that could be completed inside the factory,as distinct from those supplied as support from out-side; and it itemized links between factories them-selves, between factory and nonfactory research,and between research units and units outside re-search and development.
The result has been a subtly redesigned discoverydivision. Processes and functions are now clearlyassigned to specific factories, expectations andachievements are more transparent than before, andthe interactions throughout discovery research arenow easily tracked from factory to factory, with themap being constantly refined.
C A S E S T U D Y : R E D I S C O V E R I N G D I S C O V E R Y R E S E A R C H
53
processes now so different from before, many disparities and stresses will inevitably develop in an unaltered managerial system. The old struc-tures will creak and strain under the unfamiliarnew pressures. To restore congruence, a com-pany may need to undertake some bold organiza-tional reshaping—shifting or removing divisionalborders, reassigning personnel, redistributingareas of responsibility, and so on—not just withinR&D, but also within the company as a whole andeven beyond, in the alliances the company mightenter into.
The R&D Department. Incorporating the requisitenew capabilities, it goes without saying, represents aformidable organizational challenge: not only dothey have to mesh with existing capabilities, theyneed to coordinate with one another as well. Toimplement in silico drug design, for example, itwould be almost essential to provide an informaticsinterface between structural biology and chemistrydata. Meanwhile, a comparable reorganization ofpersonnel has to be undertaken. Biologists andchemists, for example, can no longer proceed inisolation, but must now work alongside each
A large pharmaceutical company was facing a capac-ity crisis. Both development and staffing levels wereunder pressure, mainly as a result of productivityimprovements in basic research and competition forscientific talent. As a key part of the remedy, the com-pany undertook worldwide implementation of capac-ity planning and management—a considerable chal-lenge for such a complex organization, where demandwas uncertain and resources were not fungible.
Development of CPM had four main components:
• Quantifying capacity and demand. Appropriateunits of capacity and demand were defined foreach function. In clinical departments, forinstance, the typical unit of capacity was definedas a team of monitors, coordinators, and supportpersonnel, and the unit of demand was a study.
• Business processes. To exploit CPM fully and fos-ter cooperation among departments and projectteams, various new business processes were initi-ated—most importantly, the tracking and interpre-tation of demand and capacity information, andthe consequent adjustment of timelines andresource allocation. Appropriate linkages neededto be made to related functions such as facilitiesplanning and human resources.
• Change management. Since CPM tends to affectdeeply the way an organization operates—publi-
cizing the relative productivity and workload ofdifferent departments, for example—some man-agers react more negatively than others. The com-pany took steps, both before and during theimplementation of CPM, to ease the transition.The message was constantly reinforced—that thechanged regimen was beneficial, essential, andpermanent.
• IT support. With CPM quickly generating a wealthof information, some centralized and some requir-ing broad dissemination, the company recognizedthat its CPM initiative needed extra IT support. Itidentified suitable vendors with the requisite flexi-bility and pharmaceutical experience.
The CPM endeavor has been widely hailed. Nolonger is the question “Do we have the capacity todo these projects?” met with silence. Nowadays, theCPM team can provide a detailed, graphical depic-tion of capacity and demand in each department andoverall, and an analysis of the capacity impact ofeach project.
Looking ahead, the company expects CPM to con-tribute to enhanced revenues, by speeding time tomarket and increasing the number of indications percompound. It also expects to use CPM to improveportfolio management, and to save costs throughmore rational investment in hiring and facilities.
C A S E S T U D Y : M A N A G I N G C A P A C I T Y — E M P O W E R M E N T T H R O U G H C P M
other—often literally—on collaborative projects orin formal discovery partnerships. And geneticsrequires far closer collaboration between basicresearch and development than ever before.
One excellent example of rethinking traditionalorganizational structure and boundaries isGlaxoSmithKline. Alert to the impact of scale, thecompany has on the one hand consolidated func-tions where scale and coordination provide a clearadvantage, and on the other, engaged in decentral-izing where size and complexity could prove a draw-back. Specifically, prompted by the scale benefits,the company decided to organize centrally both thefront end and the back end of R&D (that is, targetdiscovery and full development). For the steps inbetween, conversely, where the company’s enor-mous scale would risk encumbering innovation, ithas established smaller, more autonomous centersof excellence (based on different therapeuticareas), which attempt to simulate the feel of smallerbiotech companies.
The Entire Corporation. So, enhanced control of dataand increased cross-functionality of personnel areset to change the structure and tone of the R&Ddepartment. But their sphere of operation isbroader than that. As with the strategic issues dis-cussed earlier, the company as a whole is impli-cated. New lines of communication, and possiblynew chains of command, will need to be extendedbetween R&D and other units. In particular, therelationship between R&D and marketing will befundamentally transformed: with R&D facinggreater choice and placing bigger bets earlier thanever, commercial input will be crucial. And phar-macogenetics will require new ways of thinkingabout markets, competitors, and customers. (Phar-macogenetics may also inspire new linkagesbetween pharmaceutical and diagnostic units forcorporations that have both).
Coordinating the commercialization processbetween R&D and marketing has always been a del-icate balancing act. Most biopharmaceutical com-panies have established product development proj-ect teams to drive the process. These cross-
functional teams are charged with developing prod-uct strategy and coordinating the various functionsas products progress from R&D into the market.The job has now become even more complex andtricky, owing to larger global efforts, greater infor-mation flow, more specialized functions, andincreased liaison with global strategic marketing(especially when companies consider the optionsfor applying pharmacogenetics to molecules indevelopment).
Beyond the Corporation. Finally, new partnershipmodels need to be considered. Although tradition-ally organized partnerships are still appropriate inmany cases, new and more flexible forms of alliancewill sometimes be required, notably when it comesto collaborating with academic or not-for-profitinstitutions and to joining horizontal networks orconsortia.
R&D Governance One potential source of gain in R&D is improveddecision making. Consider again the example at thestart of the value chain—the glut of identified tar-gets and the need to decide which ones should pro-ceed to the next phase. Genomics technologieshave created this quandary, but they have also pro-vided the means for solving it. Using new genomicmethods of “aptitude-testing,” decision makers canconfidently preselect the most promising targetsand forward them downstream.
Even decisions unrelated to genomics technologiesstand to improve, since the new genomics regimenfosters a culture of rigorous selection criteria. Infact, one of the most important, though perhapsleast noted, benefits of genomics is the way itencourages a thorough rethinking of decision-making processes. New kinds of data now presentthemselves for interpretation, and they enter thecalculations earlier and in greater abundance thanthe old kinds did. And R&D decision makers haveto take into account a new set of factors too, be-yond the confines of R&D, in order to maximizevalue—factors such as marketing and IP implica-tions, for example.
54
Managing Organizational Change With the advent of genomics, R&D personnel sud-denly find themselves in alien territory. As the sci-entific methods change, the old instinctiveapproaches and behaviors need to change as well.Among the greatest challenges facing R&D execu-tives is managing the human side of change.
How the Scientist’s Job is ChangingR&D science is shifting from an arena of experi-mentation to one increasingly concerned with the-oretical biology. The challenge is now less how toget the data than what to do with the data collected.Scientists who formerly could do their jobs virtuallyon their own—conduct their own experiments, andgenerate and analyze the data themselves—nowfind they need to collaborate with others who havemore specialized technological skills, in areas suchas informatics, robotics, or microfabrication.Indeed, the scientists of the pre-genomics era aredestined to evolve into two kinds of successors:
those who interpret the data and devise plans forexploiting it, and those who continue to developand optimize the technologies required for gener-ating the data. (Companies should be sure to rec-ognize and reward the latter group for its contribu-tions, and not relegate it to second-class status.)
All scientists will need to become comfortable withnew ways of working together—more sharing or col-lectivist now, less conducive to solitary initiative.The scientists of the future will still take responsi-bility for their own work, but perhaps will no longertake the credit for it: that will be ascribed to teameffort.
Managing the TransitionChanging from bench-based to information-basedwork in this way, and from favoring fairly independ-ent endeavors to promoting a more collaborativeethos, is bound to be awkward or even painful formost of those involved, scientists and managers alike.
55
A large drug company recently completed an inten-sive three-month project to redesign discovery gov-ernance and is already reaping the benefits. Thecompany had always placed a high value on thequality of its scientists and their entrepreneurialdrive. Now, however, it was growing increasingly dis-satisfied with its existing system of allocatingresources: the decision-making procedures wereproving very troublesome to navigate, decision mak-ers were difficult to identify, communication waspoor, and the decisions themselves often seemedpolitically motivated rather than guided by scientificand commercial promise.
In redesigning the decision-making procedures, thecompany began with a thorough review of its currentgovernance process, both as espoused and as prac-ticed (the two were remarkably distinct in certaininstances). Various root causes of undesirable out-comes were identified: these included perverseincentives (that is, incentives encouraging behavior
at odds with company strategy); unclear criteria,which project champions were disinclined to clarify,let alone follow; and inadequate allocation of deci-sion rights (that is, too vague a definition of who wasentitled to make which decisions), which oftenmeant that no decision was made at all.
From the lessons learned, a new governance processwas devised. Not just devised, but activated: bymodifying incentives, the company ensured thatpractice was now properly aligned with espousal.
The new process is working well: R&D managersnavigate it easily, and decisions are being made andcommunicated clearly and consistently. It allows sci-entists more time to focus on their projects, and itgives those projects appropriate funding and man-agement involvement. It has accordingly won theconfidence of those affected by it, and can claim aconsiderable contribution to the marked improve-ment in productivity that has followed its adoption.
C A S E S T U D Y : R E D E S I G N I N G R & D G O V E R N A N C E
The formidable operational and organizationalchanges will entail cultural changes too: in fact, thenew processes and structures may prove far less diffi-cult to establish than new habits and attitudes.
Consider informatics. It is not enough simply tointroduce powerful new IT tools within traditionalsilos—within chemistry, for example, where in silicoapproaches would boost the efficiency of screeningand optimization. To achieve their full impact,these IT tools need to be deployed across functions:to bring biologists and chemists together, to incor-porate data from the clinic into discovery, and soon. And that will require not just new software, oreven new managerial positions, but new ways ofthinking and of relating to colleagues.
Some idea of what lies in store can be gleaned fromthe history of another transformational technol-ogy—CAD/CAM for airplane design. Likegenomics, it promised to transform a costly andlabor-intensive R&D process into a highly auto-mated and efficient one. After languishing in nicheapplications in the 1970s and ’80s, it finally provedits worth in the 1990s, when Boeing used it indesigning the first “paperless” airplane, the Boeing777. To exploit the technology fully, the companyhad to break down departmental barriers andencourage collaboration across the full range offunctions. Jobs and job responsibilities had tochange. Cherished traditions were called into ques-tion. The company held quarterly meetings atwhich employees could ask questions and voicetheir concerns. The transformation was a struggle,but ultimately a great success: Boeing continues topush the envelope in “in silico” airplane design.
When pharmaceutical companies convert to geno-mics, they will have to temper the discomforts oftransition in their turn. And that means engagingthe emotional and behavioral issues—the human
issues—as deeply as the operational ones.4 Attentivemanagement of the human issues, which has playedsuch a prominent role in so many industries in thethroes of reform, is going to be particularly crucialwhen it comes to the massive institutional changesdemanded by the genomics revolution.
A Final Word
To stake a claim in the changing biopharmaceuticallandscape, let alone feature prominently within it, acompany will have to make itself radically amenableto change. Defining a strategy is certainly a step inthat direction, and initiating that strategy is cer-tainly a gesture of commitment. But wholeheartedcommitment is evidenced not by initiating the strat-egy but rather by maintaining it—that is, monitor-ing the new structures and procedures constantly,responding to shifts in external and internal cir-cumstances, and introducing further changesrepeatedly, aggressive or defensive, as new opportu-nities or new challenges arise, though always in linewith the controlling wisdom of the strategy itself.
If the unfamiliar outer landscape provokes feelingsof unease, so too will a company’s inner landscape,once all the requisite operational and organiza-tional changes are in place. In particular, the in-crease in cross-functional activity may be disorient-ing for some executives of the old school. Many ofthe ancient landmarks, tidy borders, and familiarcategories will no longer be there to give them theirbearings. Short of attempting a counterrevolutionor withdrawing into obscurity, they will need tofamiliarize themselves with the new terrain fairlypromptly—and accept it affirmatively, not grudg-ingly. Changes in attitude will perhaps prove themost difficult changes of all to bring about, and acompany’s prosperity could be in jeopardy if theyfail to take effect.
56
4. For a fuller discussion of the emotional aspects of change, read The Change Monster, by Jeanie Daniel Duck, published by Crown in 2001.
57
Conclusion
The international pharmaceutical industry is press-ing ahead in an unexpectedly difficult environ-ment. Drug companies face unfamiliar frustrations.On one side, pricing policies are coming increas-ingly under threat (witness the recent moves in var-ious U.S. states to restrict access to costlier drugsfor Medicaid patients). From the other side, thepressure of expectation increases too, with financialanalysts continuing to count on triumphant prod-uct launches and enormous growth. In such anenvironment, corporate well-being, or even sur-vival, depends on boosting productivity.
It is against this background that the genomics rev-olution is unfolding. In their quest for improvedproductivity, companies should welcome the newtechnologies and approaches. Genomics promisesprodigious benefits: it will unlock storehouses ofinformation about the workings of human disease,and greatly refine—perhaps even personalize—health care. More to the point, it promises to trans-form how pharmaceutical research is conducted.The paradigm will shift from small-scale andserendipitous to global, industrialized, and system-atic; and from methodical and compartmentalizedto fluid and cross-functional. The impact on R&Deconomics is likely to be tremendous: in the bestcase, productivity could as much as double.
Looking beyond R&D, genomics and genetics alsopromise to transform the way pharmaceutical com-panies conduct their business in the coming years.
If genetics realizes its potential, for example, treat-ments will become more sophisticated, markets mayfragment, and the shape and value of marketingand sales organizations will change dramatically.The entire system of health care delivery, already influx, will complete its metamorphosis.
The offer that genomics and genetics are holdingout is really an offer that companies cannot refuse.Companies that fail to accept the offer adequatelywill find themselves not simply uncompetitive butpossibly right out of contention. There is nowhereto hide, and certainly no safety in inaction. To shunthe promise of pharmacogenomics out of a fear ofmarket fragmentation, for instance, is not to avertthe fragmentation but simply to cede the market toone’s rivals.
Embracing the revolution appropriately will requireboth boldness and finesse: managers will have tomake major strategic decisions, and to implementthem will have to radically reconfigure operations.The decisions take careful analysis to get right, andthe operational hurdles need nimble negotiation tosurmount. It all adds up to a formidable but by nomeans impossible task. And for companies that doit well, the rewards will be handsome.
The opportunities are unprecedented. So are thechallenges. The shrewd company will be one thatremains responsive to both, as it tries to keep itshead and to prosper in these revolutionary times.
59
Methodology
The many diverse studies on drug developmenthave reached diverse conclusions. They have putthe current price tag at anywhere between $350 mil-lion and $800 million per drug (all underestimates,in our view: our own calculation is $880 million).Not surprisingly, when it comes to the likely impactof genomics and genetics on the economics of drugdevelopment, opinions diverge again.
For our study, we conducted an extensive programof discussions in an effort to compile accurate fig-ures for all the main activities in the R&D process,both pre- and post-genomics. The result is a robustbottom-up model of R&D, based on the time, cost,and likely success rate for each step of the valuechain.
Our model goes beyond existing models of R&D inthree important ways:
• It is the product of primary research. Other esti-mates have tended to build on the findings of ear-lier work; our model also draws on more than 100discussions at nearly 50 companies and academicinstitutions.
• It analyzes the discovery phase more closely thanhas previously been possible. Earlier models typi-cally assigned a conjectural figure to representthe sunk cost of discovery, but the art of discoveryis becoming industrialized, and we have duly
been able to model its activities more scientifi-cally. A more detailed understanding of the eco-nomics of discovery has resulted, and that in turnhas allowed us to more accurately quantify theimpact of genomics on R&D. (See the chart onpage 60 for technologies modeled.)
• It is activity-based and flexible. The numberscited in this report represent an average drug,unless noted otherwise, but in our research weranged far wider than that, and assessed each stepof the value chain under a range of circum-stances. So the model allows for scenario buildingand sensitivity analysis, as well as enables us to tailor inputs to match the unique circumstancesof individual companies.
As already mentioned, all numbers cited in the textare for an average drug. In any individual case, costand time will vary according to factors such as ther-apeutic area and target class.
All numbers cited in the text are for a relevantdrug, that is, one to which the technology underdiscussion could be applied. Various technologiesmay not apply to all targets or drugs. Where specificlimitations are likely to be a significant factor, thatis pointed out.
When we discuss the “value added” to a drug, weare referring to its net present value: the current
value of expected profits, discounted by a represen-tative hurdle rate, less the cost of developing thedrug. For the average drug, we assume peak annualsales of $500 million and 11 years to patent expira-tion. Also, our numbers reflect the fact that R&D
dollars saved are pretax dollars; these “saved” dol-lars are subject to taxation, which explains in partwhy the expressed NPV calculations can be lowerthan the R&D savings.
60
GENOMICS TECHNOLOGY ASSUMPTIONS
Development
Pre-clinical Clinical
Chemistry
Screening Optimization
Biology
Target ID Target Validation
Target identification
• Limited numbers of genes
• Molecular biology and biochemistry techniques
Target validation
• Cell and tissue studies
• Mouse knockouts
Target identification
• Large numbers of genes
• Industrialized techniques (e.g., gene chip expression)
• Bioinformatics (e.g., database searches for homologies)
Target validation
• Cell and tissue studies
• Mouse knockouts
Screening
• Parallel synthesis for library design
• Assay development for high-throughputscreening (HTS)
• HTS
Chemical optimization
• Bench synthesis
• Parallel synthesis
Screening
• Structural biology (target structure)
• SAR profiling of library
• Assay development for LTS1
• Virtual screening and LTS1
Chemical optimization
• In silico-supported bench synthesis
• In silico early ADME/tox
Preclinical (ADME/tox)
• Animal testing
Clinical
• Patient trials
Preclinical (ADME/tox)
• Animal testing
• In silico ADME/tox
• In vitro toxicology
• Surrogate markers
Clinical
• Patient trials
• Surrogate markers
Post
-gen
omic
sPr
e-ge
nom
ics
SOURCES: BCG analysis; industry interviews; scientific literature.
1LTS = LOw-throughput screening; generally more information-rich, but less standardized, assays that cannot be used in HTS.
The Boston Consulting Group is a general management consulting firmthat is a global leader in business strategy. BCG has helped companiesin every major industry and market achieve a competitive advantage bydeveloping and implementing unique strategies. Founded in 1963, thefirm now operates 51 offices in 34 countries. For further information,please visit our Web site at www.bcg.com.
The Boston Consulting Group publishes other reports
that may be of interest to senior health care executives.
Recent examples include:
The Pharmaceutical Industry into Its Second Century:
From Serendipity to Strategy
A report by The Boston Consulting Group, January 1999
Ensuring Cost-Effective Access to Innovative Pharmaceuticals:
Do Market Interventions Work?
A report by The Boston Consulting Group and Warner-Lambert,
April 1999
Patients, Physicians, and the Internet: Myth, Reality, and Implications
A report by The Boston Consulting Group, January 2001
Vital Signs: The Impact of E-Health on Patients and Physicians
A report by The Boston Consulting Group, February 2001
Vital Signs Update: The E-Health Patient Paradox
A BCG Focus by The Boston Consulting Group, May 2001
Vital Signs Update: Doctors Say E-Health Delivers
A BCG Focus by The Boston Consulting Group, September 2001
In addition, BCG’s Health Care practice publishes Opportunities for Action
in Health Care, essays on topical issues for senior executives.
For a complete list of BCG publications and information about
how to obtain copies, please visit our Web site at www.bcg.com.
AmsterdamAthensAtlantaAucklandBangkokBerlinBostonBrusselsBudapestBuenos AiresChicagoCologneCopenhagenDallasDüsseldorfFrankfurt HamburgHelsinki
Hong KongIstanbulJakartaKuala LumpurLisbonLondonLos AngelesMadridMelbourneMexico CityMilan MonterreyMoscowMumbaiMunichNew DelhiNew YorkOslo
ParisSan FranciscoSão PauloSeoulShanghaiSingaporeStockholmStuttgartSydneyTokyoTorontoViennaWarsawWashingtonZürich
www.bcg.com
BCG
A Revolution in R&D
H O W G E N O M I C S A N D G E N E T I C S A R E T R A N S F O R M I N G T H E B I O P H A R M A C E U T I C A L I N D U S T R Y
BCG REPORT