J O U R N A L O F T H E AM E R I C A N C O L L E G E O F C A R D I O L O G Y V O L . 6 5 , N O . 1 5 , 2 0 1 5

ª 2 0 1 5 B Y T H E AM E R I C A N C O L L E G E O F C A R D I O L O G Y F O U N DA T I O N I S S N 0 7 3 5 - 1 0 9 7 / $ 3 6 . 0 0

P U B L I S H E D B Y E L S E V I E R I N C . h t t p : / / d x . d o i . o r g / 1 0 . 1 0 1 6 / j . j a c c . 2 0 1 5 . 0 3 . 0 1 6

THE PRESENT AND FUTURE

STATE-OF-THE-ART REVIEW

Cardiovascular Drug Development

Is it Dead or Just Hibernating?

Christopher B. Fordyce, MD, MSC,* Matthew T. Roe, MD, MHS,* Tariq Ahmad, MD, MPH,* Peter Libby, MD,yJeffrey S. Borer, MD,z William R. Hiatt, MD,x Michael R. Bristow, MD, PHD,x Milton Packer, MD,kScott M. Wasserman, MD,{ Ned Braunstein, MD,# Bertram Pitt, MD,** David L. DeMets, PHD,yyKatharine Cooper-Arnold, MPH,zz Paul W. Armstrong, MD,xx Scott D. Berkowitz, MD,kk Rob Scott, MD,kJayne Prats, PHD,{{ Zorina S. Galis, PHD,zz Norman Stockbridge, MD, PHD,## Eric D. Peterson, MD, MPH,*Robert M. Califf, MD*

ABSTRACT

Fro

Bo

Sch

Te

Sch

zzNUn

Co

Re

Ta

Ph

Pro

Sa

Sq

Pu

Co

tria

lin

ad

Ge

me

the

en

Hi

Co

He

em

Despite the global burden of cardiovascular disease, investment in cardiovascular drug development has stagnated over

the past 2 decades, with relative underinvestment compared with other therapeutic areas. The reasons for this trend are

multifactorial, but of primary concern is the high cost of conducting cardiovascular outcome trials in the current regu-

latory environment that demands a direct assessment of risks and benefits, using clinically-evident cardiovascular end-

points. To work toward consensus on improving the environment for cardiovascular drug development, stakeholders

from academia, industry, regulatory bodies, and government agencies convened for a think tank meeting in July 2014 in

Washington, DC. This paper summarizes the proceedings of the meeting and aims to delineate the current adverse trends

in cardiovascular drug development, understand the key issues that underlie these trends within the context of a

recognized need for a rigorous regulatory review process, and provide potential solutions to the problems identified.

(J Am Coll Cardiol 2015;65:1567–82) © 2015 by the American College of Cardiology Foundation.

m the *Duke Clinical Research Institute, Durham, North Carolina; yBrigham and Women’s Hospital, Harvard Medical School,

ston, Massachusetts; zState University of New York Downstate Medical Center, Brooklyn, New York; xUniversity of Colorado

ool of Medicine, Aurora, Colorado; kDepartment of Clinical Sciences, University of Texas Southwestern Medical Center, Dallas,

xas; {Amgen, Inc., Thousand Oaks, California; #Regeneron Pharmaceuticals, Tarrytown, New York; **University of Michigan

ool of Medicine, Ann Arbor, Michigan; yyUniversity of Wisconsin School of Medicine and Public Health, Madison, Wisconsin;

ational Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Maryland; xxCanadian VIGOUR Centre,

iversity of Alberta, Edmonton, Alberta, Canada; kkBayer HealthCare Pharmaceuticals, Whippany, New Jersey; {{The Medicines

mpany, Parsippany, New Jersey; and the ##Division of Cardiovascular and Renal Products, Center for Drug Evaluation and

search, United States Food and Drug Administration, Silver Spring, Maryland. The Cardiovascular Drug Development Think

nk was supported by Amgen, Inc., Bayer HealthCare Pharmaceuticals, Bayer Pharma AG, Bristol-Myers Squibb, Janssen

armaceuticals, Regeneron, and The Medicines Company. Dr. Fordyce has received support from the Clinician Investigator

gram of the University of British Columbia. Dr. Roe has received research grants from Eli Lilly & Co., Sanofi-Aventis, Daiichi-

nkyo, Amgen, and Familial Hypercholesterolemia Foundation; has received lecture fees from AstraZeneca and Bristol-Myers

uibb; and has received consulting fees from AstraZeneca, Eli Lilly & Co., Merck & Co., Janssen Pharmaceuticals, Elsevier

blishers, and Amgen. Dr. Ahmad has served as a consultant to Roche; has received an educational grant from Thoratec

rporation; and has received a grant from the Daland Foundation. Dr. Libby is an unpaid consultant or involved in clinical

ls for Amgen, AstraZeneca, Boehringer Ingelheim, Bristol-Myers Squibb, Esperion Therapeutics, Genzyme, GlaxoSmithK-

e, Kowa Pharmaceuticals, Merck, Novartis, Pfizer, Sanofi-Regeneron, Takeda Pharmaceuticals; is a member of the scientific

visory board for Athera Biotechnologies and Interleukin Genetics; and his laboratory receives research funding from

neral Electric, GlaxoSmithKline, and Novartis. Dr. Borer has been a paid consultant, clinical trial executive committee

mber, data and safety monitoring committee chair or member, and clinical events adjudication committee member during

past year for Amgen, Boehringer-Ingelheim, JenaValve, Takeda USA, Celgene, Cleveland Biolabs, Somahlution, Cardior-

tis, Celladon, Biotronik, BioMarin, Salix, Novartis, ARMGO, Pfizer, and Laboratoires Servier; and has stock in BioMarin. Dr.

att has received research grants from AstraZeneca and Bayer; and his research organization affiliate of the University of

lorado receives funds from companies including AET Biotech, AstraZeneca, Bayer, Cardioxyl, National Institutes of

alth, CSI, GlaxoSmithKline, Janssen, Kyushu University, Merck, Pluristem, Reneuron, and Takeda. Dr. Bristow is an

ployee, shareholder, and board member of ARCA biopharma. Dr. Packer is a consultant for Actelion, Amgen, CardioKinetix,

ABBR EV I A T I ON S

AND ACRONYMS

AD = adaptive design

NME = new molecular entity

PFS = progression-free survival

ROI = return on investmen

Cardiorenti

employee o

and Merck

peutics, St

pending fo

Johnson &

is a consul

device ind

industry-sp

Sanofi, Boe

Sharp & Do

conjunction

consulting

Axio/Orexi

shareholde

and Drug A

Association

consulting

and Sanofi

and Develo

Medscape

expressed

All other a

Listen to t

You can al

Manuscript

Fordyce et al. J A C C V O L . 6 5 , N O . 1 5 , 2 0 1 5

Cardiovascular Drug Development A P R I L 2 1 , 2 0 1 5 : 1 5 6 7 – 8 2

1568

D espite significant improvements incardiovascular mortality over thelast several decades, cardiovascu-

lar disease remains the leading cause of deathboth in the United States and the rest of theworld (1–3). Heart disease and stroke willresult in an estimated 24 million deaths/year

worldwide by 2030, and will continue to representthe dominant cause of death among themost prevalentchronic diseases (4–6) (Figure 1). Furthermore, mortal-ity and morbidity due to cardiovascular eventscontinue to climb globally as a result of rising cardio-vascular disease rates in low-income countries, result-ing in increasing disparities in outcomes as a functionof wealth and education (7). The burden of cardiovas-cular disease clearly remains both a major publichealth concern and growing global challenge.

Notwithstanding this increase in cardiovasculardisease prevalence worldwide, investment in cardio-vascular drug development has stagnated over the past2 decades, with relative underinvestment comparedwith other therapeutic areas (8–13). This alarmingtrend appears to reflect the business strategy in thepharmaceutical industry of maximizing return on in-vestment (ROI) by focusing on areas currently felt to bemost lucrative (8,9). As discussed in the following text,the reasons for these trends are multifactorial. How-ever, a particularly important factor is the high costof conducting cardiovascular outcome trials in thecurrent regulatory environment that demands a directassessment of risks and benefits using clinically-

t

s, Janssen, Novartis, Pfizer, and St. Jude Medical. Dr. Wasserman

f Regeneron Pharmaceuticals and a retired employee of Merck;

. Dr. Pitt is a consultant for Pfizer, Bayer, AstraZeneca, scPha

ealth Peptides, and Tricida; has stock options in Relypsa, sc

r site-specific delivery of eplerenone to the myocardium; ha

Johnson, Oxygen Biotherapeutics, and Cytopherx; and has serve

tant to the National Institutes of Health, the Food and Drug Ad

ustry on the design, monitoring, and analysis of clinical trials;

onsored data and safety monitoring committees, including Astr

hringer Ingelheim, Teva, and AbbVie. Dr. Armstrong received

hme in conjunction with Duke Clinical Research Institute (DCRI

with DCRI, Merck & Co., Inc., Sanofi-aventis Research and D

fees from AstraZeneca, Boehringer Ingelheim, GlaxoSmithKlin

gen, Merck, Eli Lilly, and Bayer. Dr. Berkowitz is an employee o

r of Amgen, Inc. Dr. Prats is an employee of The Medicines Com

dministration. Dr. Peterson has received research grants from t

, Eli Lilly & Co., Janssen Pharmaceutical Products, and the

fees from AstraZeneca, Bayer AG, Boehringer Ingelheim, Genent

-Aventis. Dr. Califf has received research grants from Amylin, Bri

pment LLC, Merck, and Novartis; and received consulting fees f

LLC/heart.org, Merck, Novartis, Regado NJ, and Roche; and h

are those of the authors and do not necessarily reflect official N

uthors have reported that they have no relationships relevant

his manuscript’s audio summary by JACC Editor-in-Chief Dr. V

so listen to this issue’s audio summary by JACC Editor-in-Chie

received January 28, 2015; accepted March 3, 2015.

evident cardiovascular endpoints for approval ratherthan biomarkers or putative surrogates. These realitiessuggest that although the cardiovascular diseaseburden continues to grow and innovative scientificdiscoveries continue to occur, investors have concernsregardingwhat they describe as regulatory uncertaintyand high development costs, leading to negativeeffects on ROI for novel cardiovascular therapies.

To work toward consensus on improving theenvironment for cardiovascular drug development,stakeholders from academia, industry, and govern-ment convened in July 2014 in Washington, DC. Thispaper summarizes the proceedings of this “thinktank” meeting, the specific aims of which were to:

1. Delineate the current adverse trends in cardiovas-cular drug development;

2. Understand the key issues that underlie thesetrends within the context of a rigorous regulatoryreview process that is a key aspect of drug devel-opment; and

3. Provide potential solutions to the problems identified.

CURRENT TRENDS IN

CARDIOVASCULAR DRUG DEVELOPMENT

Between 2000 and 2009, U.S. Food and Drug Admin-istration (FDA) approvals for new cardiovascular drugtherapies declined by approximately 33% comparedwith the prior decade (11). During the discus-sions, FDA representatives reported parallel adversetrends in investigational new drug applications to the

is an employee of Amgen, Inc. Dr. Braunstein is an

and owns stock in both Regeneron Pharmaceuticals

rmaceuticals, Relypsa, Aurasense, Da Vinci Thera-

Pharmaceuticals, Aurasense, Tricida; has a patent

s served on data monitoring boards of Novartis,

d on the events committee of Juventis. Dr. DeMets

ministration, and the pharmaceutical and medical

and receives compensation for serving on several

aZeneca, Amgen, Actelion, GlaxoSmithKline, Merck,

research grants from Boehringer Ingelheim, Merck

), GlaxoSmithKiline, Amylin Pharmaceutical, Inc. in

evelopment, and Regado Bioscience; and received

e, Merck & Co., Inc., F. Hoffmann-La Roche Ltd.,

f Bayer HealthCare. Dr. Scott is an employee of and

pany. Dr. Stockbridge is an employee of the Food

he American College of Cardiology, American Heart

Society of Thoracic Surgeons; and has received

ech, Janssen Pharmaceutical Products, Merck & Co.,

stol-Myers Squibb, Eli Lilly & Co., Janssen Research

rom Amgen, Bayer Healthcare, BMEB Services LLC,

as equity in N30 Pharma and Portola. The views

ational Heart, Lung, and Blood Institute positions.

to the contents of this paper to disclose.

alentin Fuster.

f Dr. Valentin Fuster.

FIGURE 1 Global Burden of Disease: Chronic and Infectious Diseases 2002 to 2030

OtherOtherHIV/AIDS

Dramatic shift in deaths fromyounger to older and fromcommunicable tononcommunicable diseases.

Infectious excludingHIV/AIDS

PerinatalRespiratory infections

DigestiveRespiratoryCancerCVD

Chronic Diseases Infectious Diseases

40

60

20

Deat

hs (m

illio

ns)

02002 2006 2010 2014

CVD HIV

Year Year2018 2022 2002 2006 2010 2014 2018 2022 2026 20302026

AIDS ¼ acquired immune deficiency syndrome; CVD ¼ cardiovascular disease; HIV ¼ human

immunodeficiency virus. Reprinted with permission from Fuster et al. (4) and adapted with

permission from and Mathers and Loncar (6).

J A C C V O L . 6 5 , N O . 1 5 , 2 0 1 5 Fordyce et al.A P R I L 2 1 , 2 0 1 5 : 1 5 6 7 – 8 2 Cardiovascular Drug Development

1569

Division of Cardiovascular and Renal Drugs fornew molecular entities (NMEs) seeking approval formarketing (with no change in investigational newdrugs and a relative decline in new marketing appli-cations), including NMEs for cardiovascular disease,from 1991 to 2013 (personal communication, N.Stockbridge, December 6, 2014) (Figure 2). Parallelanalyses that evaluated regulatory approvals for car-diovascular therapies demonstrated a similar tempo-ral decline, but showed that approvals in several othertherapeutic areas, including oncology, rose substan-tially during the same time frame (11).CARDIOVASCULAR DRUG DEVELOPMENT COMPARED

WITH OTHER THERAPEUTIC AREAS. The applicationof genomic technologies, as well as systems biologyapproaches, have collectively identified multiplepotential new cardiovascular drug targets, as wellas actual molecules with potential cardiovascularapplications. Unfortunately, these scientific ad-vances have not stimulated an increase in new car-diovascular drug development. Early-phase researchand development appear to have stagnated, withfewer molecules in the cardiovascular research pipe-line compared with other therapeutic areas. In ananalysis comparing the number of drugs undergoingearly-phase development at 2 separate intervals (1990to 1999 vs. 2000 to 2007), the development of anti-neoplastic agents grew from 16.55% to 23.43%(þ6.88%), whereas the development of cardiovascu-lar agents experienced the strongest contraction ofany therapeutic area (–4.57%) (9). Furthermore, thesuccess rate for antineoplastic drug approvalsincreased during this time interval (10). An analysis ofall FDA NMEs from 2000 to 2012 found that oncologydrugs had not only the greatest number of NME ap-plications (n ¼ 61), but also the highest rate of first-cycle FDA approvals of all therapeutic classes (72%)(12). In contrast, cardiovascular drugs had signifi-cantly fewer NME applications (n ¼ 21) and a muchlower rate of first-cycle approvals (32%). The currentportfolio of drugs being evaluated in clinical trialsreflects these trends, with a steep rise in cancer drugsbeing tested over the last 2 decades (Figure 3) (13).RISING DRUG-DEVELOPMENT COSTS AND THE CONCEPT

OF REGULATORY UNCERTAINTY. Notwithstanding theissues that appear to be impeding cardiovascular drugdevelopment, during the last decades, the overallpharmaceutical research and development processeshave become less efficient in converting promisingtherapies into actual approved and marketed agents(Figure 4) (14). In 2005, the average capitalized cost tobring 1 new biopharmaceutical product to market,including the cost of failures, was $1.24 billion (14).These high drug-development costs caused overall

research and development spending to exceed $50billion in the United States in 2008.

The high costs of drug development were consid-ered by meeting participants to include the perceivedcomplexities and obstacles in navigating the regula-tory drug approval process. Although the critical roleof the FDA and other regulatory agencies to balancethe timely approval of effective therapies with theneed to protect the public from harmful drugs wasdiscussed and strongly endorsed, the concept of“regulatory uncertainty” appears to be deep-seated inthe pharmaceutical industry.

The pathway for drug development is becomingmore complex and costly for industry, but given mul-tiple prior instances of drugs receiving regu-latory approval that eventually resulted in patientharm through post-market analyses, there is intensegovernment and public scrutiny on the regulatory drugapproval process (15). Nonetheless, the reality is thatthe pharmaceutical industry will always be the domi-nant funding source for drug discovery—a findingreinforced by recent analyses that found that between40% to 80% (16,17) of randomized controlled trialspublished in top-tier medical journals were funded byindustry sponsors. Pharmaceutical representatives atthe meeting cited regulatory uncertainty as not only acommon concern across industries (18,19), but also akey influence on decisions regarding the specificdevelopment of innovative cardiovascular drugsgiven the high cost of regulatory failures during drugdiscovery. Despite these concerns, the FDA represen-tatives expressed a strong willingness to meet earlywith industry representatives to discuss plans for a

FIGURE 2 Marketing Applications for New Uses and NMEs and Research and Commercial IND Applications Received by the FDA by Year

25 200

180

160

140

120

100

80

60

40

20

0

20

15

10

5

0

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

Research CommercialNew Uses NMEA B

(A) Marketing applications received by the FDA Division of Cardiovascular and Renal Drugs for new uses and new molecular entities (NMEs) by year; (B) investigational

new drug (IND) applications received by the FDA for both research and commercial by year. New use applications are for previously approved drugs that are reviewed for

a new indication in future transactions. NME refers to a new drug application reviewed for its first approved indication. Commercial IND applications are those intended to

be marketed for therapeutic use. Research IND applications are those not intended to be marketed for therapeutic use, but for research purposes.

FIGURE 3 Innovat

900 Anti-inCardio

GastroGenito

EndocCentra

850800750700650600550500450400350300250200150100

5001996 1997 1998

Num

ber o

f Com

poun

ds in

Clin

ical

De

velo

pmen

t

Adapted with permis

Fordyce et al. J A C C V O L . 6 5 , N O . 1 5 , 2 0 1 5

Cardiovascular Drug Development A P R I L 2 1 , 2 0 1 5 : 1 5 6 7 – 8 2

1570

drug-development program to provide guidance andfeedback regarding the potential regulatory pathwaysfor a promising therapy before the launch of pivotalregistration trials.

Studies in economics and finance have found regula-tory uncertainty to be potentially a major source of un-predictable variation in the return on businessinvestment (20). The regulatory state is oftenperceived asa key component of the “external environment” thatpharmaceutical companies must take into account whenmaking capital investments or deciding whether to entercompetitive markets (18). Some therapies may follow a

ive Compounds Between Phase I and III Development Over Time

fectives (including vaccines)vascular

intestinalurinary

HematologyOncologyImmune systemMusculoskeletalRespiratory

rinel nervous system

1999 2000 2001 2002 2003 2004Year

2005 2006 2007 2008 2009 2010 2011

sion from Berggren et al. (13).

known regulatory pathway, whichmay decrease the timeto drug approval. For other therapies, the pathway forregulatory approval has not been defined or has beeninfluenced by prior regulatory decisions for therapies in asimilar class, so the timeline for drug approval may beexpected to be prolonged or difficult to predict. The FDAalso recognizes that although innovative drugs are moredifficult and costly to develop, they may also be moredifficult to review given the lack of experiencewith a newtherapy or indication (21). Thus, although regulatory un-certainty may be a key factor influencing the trends incardiovascular drug development, the path forward mayinvolve early, in-depth discussions with industry repre-sentatives and regulatory agencies that would be ex-pected to guide decisions about how to structure adevelopment program that addresses an unmet need inclinical practice, but also appears viable within thecontext of a rigorous but fair regulatory review process.

REGULATORY ISSUES ASSOCIATED WITH THE

GLOBALIZATION OF CLINICAL TRIALS. From acontemporary United States perspective, theglobalization of cardiovascular clinical trials hasresulted in increased regulatory complexity at boththe FDA and European Medicines Agency (EMA) (22).Although recent efforts have been made to stan-dardize regulatory reporting in Europe (23), the EMAremains a decentralized organization, with each Eu-ropean nation still having its own drug agency andregulations. Because cardiovascular drug registration

FIGURE 4 New Drug Approvals and Pharmaceutical R&D Expenditures in the

United States From 1963 to 2008

60

New

Dru

g Ap

prov

als R&D Expenditures

(in billions [US$, 2008])

New drug approvals

R&D expenditures45

30

15

0 0

13

26

39

52

1963 1968 1973 1978 1983 1988 1993 1998 2003 2008

Dots and orange line indicate new drug approvals and periwinkle shaded area indicates

pharmaceutical research and development (R&D) expenditures. R&D expenditures are

presented in terms of constant 2008 dollar value. The trend line is a 3-year moving

average. The source of drug approval data is the Tufts Center for the Study of Drug

Development. The source of R&D expenditure data is the Pharmaceutical Research and

Manufacturers of America; Industry Profile 2009. Conversion of actual expenses to con-

stant dollars was performed by the Tufts Center for the Study of Drug Development.

Adapted with permission from Kaitin (14).

J A C C V O L . 6 5 , N O . 1 5 , 2 0 1 5 Fordyce et al.A P R I L 2 1 , 2 0 1 5 : 1 5 6 7 – 8 2 Cardiovascular Drug Development

1571

trials typically involve many countries to meetenrollment goals, a large and experienced pool ofcardiovascular site investigators is critical to thesuccess of these trials. However, over the pastdecade, the number of FDA-regulated investigatorsbased outside of the United States has grown by 15%annually, whereas United States–based investigatorshave declined 5.5% annually over the same interval(24). Nonetheless, studies have shown that the FDAapproves drugs faster than the EMA and Health Can-ada (median: 303, 366, and 352 days, respectively)(25), but enrollment in cardiovascular registrationtrials has appeared to shift to non-Western countrieswith perceived greater enrollment potential andlower operational costs. Although difficulty in patientrecruitment has contributed to this shift of trialenrollment to non-Western countries (26), theperceived bureaucratic and expensive regulatory en-vironments in Western countries is another stronginfluence (22).

Although globalization of trial conduct and enroll-ment could theoretically reduce upfront costs forpivotal registration trials, growing concern exists overthe geographical variation in patient selection andresults in cardiovascular outcomes trials (27,28). Therecent TOPCAT (Treatment of Preserved CardiacFunction Heart Failure with an Aldosterone Antago-nist) trial demonstrated clear differences in patientselection for heart failure with preserved ejectionfraction amongst patients in Russia and Georgia,which highly confounded the overall trial results (29).Additionally, low enrollment of patients in the UnitedStates in pivotal registration trials can also introduceconfounding, as was demonstrated in the PLATO(Study of Platelet Inhibition and Patient Outcomes)trial with the antiplatelet agent ticagrelor (30). Thetreatment results were discordant in the United Statescompared with other geographic regions, with <10%of the patients enrolled in the United States. There-fore, the globalization of clinical trials appears tocontribute to regulatory uncertainty and to complicatethe regulatory review process for therapies withpivotal trials that showed geographical differences inpatient selection/features as well as treatment results.

POTENTIAL CAUSES OF

RELATIVE UNDERINVESTMENT IN

CARDIOVASCULAR DRUG DEVELOPMENT

FEW INCENTIVES TO PROMOTE CARDIOVASCULAR

DRUG DEVELOPMENT. Cardiovascular drug devel-opment currently suffers from a lack of “push-pull”incentives relative to other fields, particularly on-cology, which tends to dominate contemporary drug

development. “Push” funding policies aim to incen-tivize the pharmaceutical industry by reducing costsduring the research and development stages, whereas“pull” mechanisms create incentives for the phar-maceutical industry by creating viable market de-mand for novel therapies that address unmet clinicalneeds (31). Push mechanisms may partially offsetresearch and development by underwriting a portionof the costs, whereas pull mechanisms may rewardpositive trial results for novel therapies.

At the federal funding level, the field of oncologyhas several push mechanisms that may accelerate thediscovery of novel therapies by academic researchersthat could potentially be commercialized throughacademic-industry collaborations. First, the 2013National Institutes of Health (NIH) budget wasapproximately $29.3 billion, with $4.8 billion allo-cated to the National Cancer Institute (NCI), thelargest proportion among all institutes and a distinc-tion held by the NCI since at least 1980 (32,33). Incomparison, despite the relative proportional in-crease in the budget of the National Heart, Lung, andBlood Institute (NHLBI) compared with the NCI dur-ing this time period (32), the allocation in the 2013budget to the NHLBI was significantly lower at $2.9billion (34). Second, the FDA has several special-designation programs designed to expedite andfacilitate the authorization and approval of newmedications for unmet medical needs. A detailed

Fordyce et al. J A C C V O L . 6 5 , N O . 1 5 , 2 0 1 5

Cardiovascular Drug Development A P R I L 2 1 , 2 0 1 5 : 1 5 6 7 – 8 2

1572

analysis found that the greatest percentage of appli-cations to each of these programs was for oncologytherapies (35). In comparison with oncology drugs,cardiovascular drug applications were far fewer in allprograms: orphan (9% vs. 38%), priority review (4%vs. 53%), accelerated approval (8% vs. 32%), and fasttrack (1% vs. 56%). Finally, as a result of the FDA’sorphan program and the increasing focus of thepharmaceutical industry on novel “first-in-class”therapies, the number of NMEs targeting orphan in-dications has risen 3-fold over the past 3 decades (36).However, there are few cardiovascular disease con-ditions that would qualify for orphan drug programs,so the interest for novel therapies has naturally shif-ted to the oncology field.

Advocacy and fundraising, major push mecha-nisms, are also dominated by oncology. In 2011, char-itable donations raised during the Susan G. KomenRace for the Cure for breast cancer ($258 million) andMovember for prostate cancer ($147 million) were over7-fold greater than that of the American Heart Asso-ciation (AHA) Jump Rope for Heart campaign($54 million) (37). Furthermore, the AHA raised $509million in 2010, compared with $903 million raisedby the American Cancer Society (38). The oncologyfield has also dominated crowdfunding initiatives—ventures that raise small amounts of money from alarge number of people (typically via the Internet).A recent analysis of 97 crowdfunding campaignsaimed at cancer research found that the averageamount raised per campaign was $45,629 (averagedonation: $186), including 5 rare-disease campaignsraising between $17,217 and $248,734 (39). In contrast,few crowdfunding campaigns have involved cardio-vascular disease, although a new initiative to fundearly-stage products was introduced at the AHA 2014Scientific Sessions (40). Finally, an example of theconsequences of philanthropic discrepancies affectingfuture drug development is reflected in the number ofyoung investigator awards available for fellows ineither subspecialty: the AHA had 0 research grants ofat least 1 year in duration for fellows in 2012, theAmerican College of Cardiology had 4, and the Amer-ican Society of Clinical Oncology had 38 (41).

What drives successful cancer fundraising cam-paigns? It is possible that certain organizations arebetter organized to raise funds. However, fear mayalso explain the documented discrepancy betweenpublic perception of disease severity and reality (42).For example, a 2011 survey conducted by the MetLifeFoundation found that 41% of responders namedcancer as the disease they fearedmost; only 8% namedheart disease, and another 8% feared stroke (43).Surveys of women conducted by the AHA show many

believe cancer to be the number 1 cause of death forfemales, not heart disease, as annual mortality statis-tics consistently confirm (44). Given that the diagnosisof cancer typically elicits a much greater emotionalresponse than that of cardiovascular disease, the dis-parities in the breadth and success of fundraising andadvocacy campaigns for cancer versus cardiovasculardisease likely reflect societal opinions on the “unmetneeds” for medical conditions, despite the aforemen-tioned global burden of cardiovascular disease.DIVERTING INVESTMENT AWAY FROM CARDIO-

VASCULAR THERAPIES. Even though the biotech-nology sector continues to show interest in drugdevelopment, trends for venture capital funding pri-orities demonstrate a shift away from cardiovas-cular disease. Between January 2005 and June 2012,acquisitions of venture-backed life science com-panies with 1 or 2 products in the pipeline remainedstrong (45). However, companies with oncology andinfectious disease products had the greatest acquisi-tions, with approximately double the money spentcompared with acquisitions that focused on cardio-vascular disease products. Similarly, biotechnologycompanies with either an oncology or a neurologyfocus had the most licensing agreements during thesame time period (2005 to 2012)—4-fold higher thanthe cardiovascular arena. The consensus opinionexpressed at the meeting was that although it appearsthat venture capital investor interest in biotech-nology remains strong, the appetite for risk is lowercompared with historical trends, so investment firmstypically are less interested in cardiovascular drug-development programs given the uncertain ROI forthe previously mentioned reasons. Compared withthe blockbuster therapies that were previously de-veloped from biotechnology companies, investors arenow prepared to accept a lower ROI (i.e., an orphandrug for a defined and small population) in exchangefor a lower risk of drug-development failure (46,47).In fact, global sales of oncology drugs in 2013 topped$67 billion, the highest of any class of medications(48). These findings highlight the interest of invest-ment firms in oncology therapies that typically have aclear regulatory pathway and a strong degree ofadvocacy from patient groups.COMMERCIAL VIABILITY OVER EFFICACY OR SAFETY AS

A DRIVER OF CARDIOVASCULAR DRUG-DEVELOPMENT

FAILURE. In addition to a lack of push incentives, alack of pull incentives currently hinders cardiovascu-lar drug development due mainly to industry concernsregarding ROIs. A recent analysis demonstrated thatcardiovascular drug failures following pivotal regis-tration trials most often result from a lack of com-mercial viability rather than from efficacy or safety

FIGURE 5 Drug Failures by Therapeutic Class, 2000 to 2009

60

% o

f Fai

lure

Sha

res

EfficacySafetyCommercialOther

All therapeuticclasses

50

40

30

20

10

0

Cardiovascular

Antineoplastic

Respiratory CNS

Musculoskeletal

GI/Metabolism

Systemic

Anti-infectives

The dashed lines represent the rates of failure among all therapeutic classes. Adapted with

permission from DiMasi (10). CNS ¼ central nervous system; GI ¼ gastrointestinal.

J A C C V O L . 6 5 , N O . 1 5 , 2 0 1 5 Fordyce et al.A P R I L 2 1 , 2 0 1 5 : 1 5 6 7 – 8 2 Cardiovascular Drug Development

1573

issues (10) (Figure 5). Failure from commercialviability for cardiovascular therapies occurred forseveral reasons, including cases where expectedfuture costs rise or expected revenues fall (perhapsdue to increased competition or that testing suggeststhe target profile should be narrowed) to the pointwhere the project is no longer perceived tobe financially viable, or when pharmaceutical firmsabandon a development program for strategic rea-sons (J.A. DiMasi, personal communication, December9, 2014). For cardiovascular therapies, physicians,payers, and regulatory agencies demand large trialsthat are adequately powered to show a difference inclinical outcomes for approval. In general, and apartfrom some conditions such as heart failure, the pro-gression of cardiovascular disease to “hard” naturalhistory outcomes is relatively prolonged comparedwith the more rapid accumulations of such outcomesin oncology. Therefore, cardiovascular outcomes tri-als require large sample sizes and must continue formany years to accumulate enough endpoints to beadequately powered. Additionally, new cardiovascu-lar medications must be tested on a background of amultitude of guideline-recommended therapies thatcomprise the “standard of care.” Because patientsselected for most cardiovascular outcomes trials onbackground medical therapy have relatively low eventrates, trials have become larger, prolonged, and cost-lier with each new therapeutic modality (49).NECESSITY OF MEASURING VALIDATED CLINICAL

OUTCOMES OVER SURROGATE ENDPOINTS. The useof putative surrogate endpoints to bring potentiallybeneficial treatments to patients many years beforethe availability of clinical outcomes at relativelylower cost appeals to investors, scientists, and pa-tients who hope to seek access to beneficial therapies(50). Yet, such shortcuts can have several well-documented unintended consequences, demandingdirect measures of clinical outcomes that matter topatients. In cardiovascular medicine, both bloodpressure and low-density lipoprotein levels haveserved as surrogates of important clinical outcomesunder specific circumstances, but only after valida-tion in multiple randomized clinical trials involvingtens of thousands of patients (51,52). Success of theIMPROVE-IT (Improved Reduction of Outcomes:Vytorin Efficacy International Trial), which demon-strated that low-density lipoprotein lowering with acombination of ezetimibe and simvastatin versussimvastatin alone resulted in improved long-termclinical outcomes, provides further impetus for thedevelopment of novel agents, including PCSK9 in-hibitors (53). Yet, alpha-adrenergic blockers lowerblood pressure but appear to be inferior in reducing

cardiovascular events compared with other bloodpressure-lowering treatments (54). Torcetrapib andlong-acting niacin increased high-density lipoproteincholesterol considerably, but provided no cardiovas-cular outcome benefits in multiple large trials and,conversely, were associated with significant harmdue to off-target drug effects (55,56). Other cardio-vascular clinical trials have exposed harms aftershowing positive results with surrogate outcomes,including morbidity from the utilization of inotropesin cardiogenic shock despite improved hemody-namic parameters or death associated with the use ofclass IC antiarrhythmic agents despite successfulsuppression of premature ventricular contractionspost-myocardial infarction (57–59).

Intermediate endpoints, such as progression-freesurvival (PFS), are more commonly used in oncology.The use of PFS among all systemic-therapy random-ized controlled trials of breast, colorectal, and non–small-cell lung cancers published in the Journal ofClinical Oncology increased from 0% (1975 to 1984) to26%more recently (2005 to 2009), whereas 23% of newFDA drug approval indications from 2005 to 2007 wereon the basis of trials with PFS as the primary endpoint(60–62). Therefore, there has not only been a clearincrease in the use of intermediate endpoints overtime, but the prevalence of intermediate endpoint usehas been consistent across studies. Justification forthe use of PFS as a valid outcome for drug approvalrelates to the extended patient follow-up and to beingconfounded by causes of mortality unrelated to can-cer. Furthermore, as novel therapies demonstrate

Fordyce et al. J A C C V O L . 6 5 , N O . 1 5 , 2 0 1 5

Cardiovascular Drug Development A P R I L 2 1 , 2 0 1 5 : 1 5 6 7 – 8 2

1574

effectiveness along the treatment continuum forcancers such as breast and colorectal, survival may beinfluenced by the use of such therapies when trialshave been completed (63,64). However, the use of PFSas a primary endpoint in many randomized controlledtrials of advanced solid tumors, including breast can-cer, has not been on the basis of evidence of its sur-rogacy for either overall survival or quality of life(60,65). Despite approving many therapies on the ba-sis of PFS, the FDA has long acknowledged that tumorresponses may not necessarily equate with clinicalbenefit, because nonresponding patients may benefitfrom a delay in tumor progression and tumors mayrecur more aggressively if they regress quickly in somecases (66,67). Taken together, despite their appeal tomake trials more efficient, putative surrogate end-points do not fully predict the true balance of risk andbenefit of interventions. This failure usually is a resultof incomplete coupling between the biomarker ofinterest and the array of clinical outcomes and theunknown and unintended pharmacological actions ofan intervention that are independent of the diseaseprocess (68).

Notwithstanding the different types of outcomesthat meet the thresholds for approval in cardiovas-cular disease versus cancer, the quality of clinical trialevidence used for recent approvals of novel thera-peutic agents varies widely across indications. Ananalysis of the pivotal clinical trial evidence requiredfor FDA approval between 2005 and 2012 is striking:cardiovascular disease, diabetes mellitus, and hy-perlipidemia trials were larger than those in othertherapeutic areas (median sample size ¼ 651 sub-jects), were of longer duration (median ¼ 24 weeks),were mostly randomized (98.6%), were typically

TABLE 1 Design of Pivotal Efficacy Trials Providing the Basis for App

2005 and 2012, Stratified by Therapeutic Agent

Therapeutic Agent(Pivotal Trials) Patients Duration, Weeks

CVD, DM, hyperlipidemia(n ¼ 72)

651 (406–926) 24.0 (10.0–26.0)

Cancer(n ¼ 55)

266 (84–610) 18.5 (8.9–29.2)

Infectious disease(n ¼ 57)

585 (319–697) 5.0 (2.5–24.0)

Neurology(n ¼ 38)

358 (234–613) 16.0 (12.0–21.0)

Dermatology(n ¼ 29)

233 (121–491) 4.3 (2.0–13.0)

Autoimmune/MSK(n ¼ 36)

525 (362–749) 24.0 (24.0–28.0)

Psychiatry(n ¼ 43)

432 (275–590) 6.0 (6.0–8.0)

Values are median (interquartile range) or %. Adapted with permission from Downing e

CVD ¼ cardiovascular disease; DM ¼ diabetes mellitus; MSK ¼ musculoskeletal.

conducted in a double-blind fashion (93.2%), andalmost uniformly used a placebo or an active controlin the comparator arm (97.2%) (Table 1) (69). Incontrast, trials used to support cancer drug approvalswere small (median sample size ¼ 266 subjects), wereof shorter duration (median ¼ 18.5 weeks), were lessfrequently randomized (47.3%), were less frequentlyconducted in a double-blind manner (27.3%), and lessthan one-half used a placebo or active control in thecomparator arm (47.3%). The consensus at themeeting was that cardiovascular medicine mustcontinue to rely on proven clinical outcomes in thenew era of drug development, especially in light ofthe previously-mentioned failures of putative surro-gate endpoints (54–59).“MISCLASSIFICATION” OF CARDIOVASCULAR DISEASE:

IMPLICATIONS FOR DRUG DISCOVERY. Traditionally,cardiovascular drug development has focused onsystemic therapies centering on large populations,such as hypertension and heart failure. However,these conditions result from multiple heterogeneousinfluences and pathways and have several potentialtherapeutic biological targets. Relatively few newdrug targets for cardiovascular disease exist that canbe completely separated from potential overlap withexisting targets, because there are often multipleeffective drugs in any given class (i.e., statins, anti-platelet agents, novel anticoagulants), which oftenrenders distinction from existing therapies difficult.This setting presents considerable challenges to thedevelopment of new therapies. Yet, recent scientificadvances, including harnessing genetic and systemsbiology approaches, promise to yield many noveldrug targets in the multiple pathways involved incardiovascular diseases such as thrombosis,

roval of Novel Therapeutic Agents by the FDA Between

Randomized Double-BlindedPlacebo or Active-Control

Comparator

98.6 93.2 97.2

47.3 27.3 47.3

93.0 78.9 91.2

100.0 100.0 94.7

93.1 75.9 86.2

100.0 94.4 100.0

100.0 100.0 100.0

t al. (69).

J A C C V O L . 6 5 , N O . 1 5 , 2 0 1 5 Fordyce et al.A P R I L 2 1 , 2 0 1 5 : 1 5 6 7 – 8 2 Cardiovascular Drug Development

1575

myocardial relaxation and strain, lipid metabolism,and atherosclerosis. The traditional “lumping” ofheterogeneous disease—on the basis of taxonomyestablished by the International Classification ofDiseases that seldom incorporated rapidly-emergingmolecular data, incidental patient characteristics, orsocio-environmental influences on disease (70)—should give way to a subclassification on the basis ofbiological signatures that better represents the un-derlying pathophysiology and permits more precisetargeting of treatments (71). This approach mimics thenow well-established strategy of targeted therapies inoncology (receptor level) and infectious diseases(organism level) and provides enticing drug targets.As opposed to cancers, common cardiovascular dis-eases do not arise from single gene mutations, butrather reflect interactions of risk factors with multiplegenes that each contribute a small portion of risk.This complexity currently furnishes fewer appealingbiological targets in cardiovascular medicine than incancers (9). Nonetheless, in a recent industry survey,most large pharmaceutical companies (14 of 16) havestill expressed an interest in partnering regardingcardiovascular disease (44).

STRATEGIES TO ADVANCE

CARDIOVASCULAR DRUG DEVELOPMENT

REDUCE OPERATING COSTS OF CLINICAL TRIALS.

Well-intentioned efforts to minimize the risk of harmor undesirable outcomes in patients enrolled in clin-ical trials have introduced numerous inefficiencies inthe conduct of trials, particularly in pivotal registra-tion trials that result in rigorous regulatory review(72). Meeting participants uniformly agreed thatreducing the amount of extraneous data collected fora cardiovascular clinical trial offers an important costsavings opportunity. The typical clinical trial in 2012involved 13 endpoints, had 169 case report formpages, and required study volunteers to make 11 visitsover an average of 175 days (49). To advance discus-sions about reducing the amount of extraneous datacollected for a large cardiovascular outcomes trial,the participants recommended that pharmaceuticalindustry sponsors meet early with key stakeholders(regulators, academic collaborators) to omit collec-tion of unessential data and to simplify case reportforms. Furthermore, there was a clear recommenda-tion from FDA representatives to the pharmaceuticalindustry to meet with the FDA to gain a commitmentearly during the drug-development process to limitthe amount of data to be collected. Academiccollaborators should be encouraged to support thescientific legitimacy of this approach, with the goal

of emphasizing quality by design (discussed in thefollowing text). Adopting FDA guidance with respectto centralized monitoring practices and risk-basedmonitoring is a strategy that should continue toensure subject protection and overall study qualitywhile increasing efficiency (73). Other strategies toreduce costs include the use of centralized institu-tional review boards (74,75) and improving the sys-tem of reporting and interpreting unexpected seriousadverse events through decreasing the review of un-interpretable case reports (76). The Clinical TrialsTransformation Initiative (CTTI)—a public-privatepartnership founded by the FDA and Duke Univer-sity to identify barriers to the conduct of large, simpletrials—has recognized other important impedimentsto clinical research involving regulatory bodies,sponsor-imposed delays, academic impediments, andhealth system and clinical practice site impediments(72). Finally, harnessing the massive National Patient-Centered Clinical Research Network (PCORnet) pro-gram to centralize and organize clinical data will notonly augment our understanding of risk and out-comes for people with specific diseases, but shouldalso provide answers that are vital to patients morequickly, efficiently, and at a lower cost than previ-ously possible (77).INCREASE FOCUS ON PRACTICAL, STREAMLINED

TRIALS. Despite a temptation to use surrogate end-points to decrease sample sizes and shorten theduration of clinical trials (ostensibly to reduce thelikelihood of drug-development failures), it wasagreed that we must continue to promote large,pragmatic trials to measure clinical outcomes whenevaluating new cardiovascular therapies. Larger trialsalso help resolve conflicting data. After post-approvalanalyses of data from small, randomized trials sug-gested that use of nesiritide was associated with arate of worsening renal function and death, theASCEND-HF (Acute Study of Clinical Effectiveness ofNesiritide in Decompensated Heart Failure) demon-strated the safety of nesiritide but demonstrated noefficacy benefits with this agent (78). Finally, if con-ducted properly, large, simple trials can prove tobe successful despite high rates of background,guidelines-recommended therapies that are known toimprove outcomes when used widely. A prime ex-ample of the contemporary success of a novel agentstudied in a large trial is PARADIGM-HF (ProspectiveComparison of ARNI [Angiotensin Receptor–NeprilysinInhibitor] with ACEI [Angiotensin-Converting–EnzymeInhibitor] to Determine Impact on Global Mortality andMorbidity in Heart Failure Trial), which comparedLCZ696 versus enalapril in heart failure with reducedejection fraction (79). This trial was stopped early due

Fordyce et al. J A C C V O L . 6 5 , N O . 1 5 , 2 0 1 5

Cardiovascular Drug Development A P R I L 2 1 , 2 0 1 5 : 1 5 6 7 – 8 2

1576

to efficacy in the LCZ696 arm. Finally, integratingquality by design principles during the planning oflarge, simple trials is a key ingredient for trialstreamlining. CTTI promotes quality by designwith the goal that clinical trials will be more stream-lined, fit for purpose, and quality driven (80). Trialswith adequate sample sizes to assess benefits andrisks using cardiovascular events should, therefore,remain the reference standard for cardiovasculardrug development.

LEVERAGE PHASE II TRIALS TO INFORM PHASE III

TRIALS. Many participants in the meeting felt that theresearch and development community could moreeffectively leverage data collected from phase II trialsto better plan for phase III trials. Although there is aneed to use phase II trial data to predict phase IIIoutcomes and to adequately design phase III trials forregulatory approval, it is equally important to use thedata gathered to understand the impact of a noveltherapy on biology and disease pathways to planfuture development. Phase II studies should, thus,focus on informing and refining important vari-ables such as dosing, population pharmacokineticmodeling, side effects, and the treatment effect onbiomarkers before commencing pivotal phase IIIregistration trials. In fact, biomarkers are most ap-propriately used in phase II trials to screen for prom-ising new therapies through evaluation of biologicalactivity (68). Embedding adaptive design (AD) withina single clinical trial may furnish a novel strategy aswell. This approach allows a review of accumulatinginformation during a trial that may suggest modifica-tion of trial characteristics, thus addressing uncer-tainty about choices made during planning (81). ADallows pre-specified updates to the maximum samplesize, study duration, treatment group allocation,dosing, number of treatment arms, or study end-points. AD could translate into more efficient therapydevelopment by reducing trial size. In turn, this couldlead to more viable studies, with less risk for sponsors.However, although some academic and industry rep-resentatives may be eager to use AD strategies, regu-lators have been rightfully cautious to accept all ADtrials for approval due to the broad spectrum ofpotentially problematic adaptations involved (81,82).

USE NOVEL NIH PROGRAMS TO SUPPORT EARLY-

PHASE DEVELOPMENT. Despite pre-clinical testingand selection, many drugs that enter human studiesnever make it to market because of failure betweenphases I and III (83). The NIH has initiated new pro-grams in search of an effective approach to activelysupport early clinical development, hoping to providea bridge to facilitate successful innovation. Although

the majority of NIH funding goes to support non-human research, it is also committed to fundingapplied sciences (84). The newest NIH institute, theNational Center for Advancing Translational Sciences,was created in 2010 to accelerate translation ofideas into treatments via different programs, in-cluding those that enable repurposing of existingdrugs for new indications (85).

The NHLBI has also developed several new ap-proaches to specifically facilitate the translation ofbasic cardiovascular discoveries into clinical applica-tions. The goal is to continue to develop a variety ofprograms that create teams of academic investigatorsand industry partners (86) (Figure 6). The NIH Centersfor Accelerated Innovations (NCAIs) aim to identifyand advance the development of promising emergingtechnologies toward new commercial products for theprevention and management of medical conditionsaffecting the cardiovascular, pulmonary, and hema-tologic systems. The NCAIs are developing a central-ized institutional approach to move basic sciencediscoveries through the early stages of technologydevelopment to enable their subsequent commer-cialization. Inventors, especially those who are newto the product-development process, have access torelevant personal training and mentoring opportu-nities. The new Vascular Interventions/Innovationsand Therapeutic Advances program (87) aims toaddress the need for new or better therapeutic in-terventions (drugs or devices) and diagnostic modal-ities in several medical conditions that have beentraditionally neglected by industry. Strategiesinclude: 1) no restrictions on the applicant’s type ofinstitutional affiliation or on geographic location(within the United States); 2) assistance with projectmanagement, regulatory issues, and expert industryadvice; and 3) opportunities to leverage other exist-ing NIH translational programs. Areas of recentfunding include vascular disorders, thrombotic dis-eases, and pulmonary hypertension. The Small Busi-ness Innovation Research funding mechanismprovides seed funding to support the development ofa broad array of commercial products to detect, di-agnose, treat, and prevent disease. Other currentprograms include the Science Moving TowardsResearch Translation and Therapy program and theGene Therapy Research Program (GTRP). These newprograms are unproven, but they do represent aneffort by the NHLBI to encourage and enable cardio-vascular drug research even in the risk-adverse capi-tal environment that currently exists.LEVERAGE ACADEMIC EXPERTS AS AN INTERFACE

BETWEEN INDUSTRY AND REGULATORY BODIES.

Optimal cardiovascular drug development depends on

FIGURE 6 NHLBI Programs That Exemplify the Opportunities Created to Support

Translation Using Funding or Assistance-Type Mechanisms

Adapted with permission from Galis et al. (87). GTRP ¼ Gene Therapy Resource Program;

NCAIs ¼ National Institutes of Health Centers for Accelerated Innovations; NHLBI ¼ Na-

tional Heart, Lung, and Blood Institute; SBIR ¼ Small Business Innovation Research;

SMARTT ¼ Science Moving Towards Research Translation and Therapy; VITA ¼ Vascular

Interventions/Innovations and Therapeutic Advances.

J A C C V O L . 6 5 , N O . 1 5 , 2 0 1 5 Fordyce et al.A P R I L 2 1 , 2 0 1 5 : 1 5 6 7 – 8 2 Cardiovascular Drug Development

1577

true partnerships amongst academic representatives,regulatory officials, industry scientists, and clinicalresearchers, as well as practicing clinicians—all ofwhom represent the needs of patients with cardio-vascular disease (Central Illustration). The use of aca-demic clinical trialists is important in the design andconduct of pivotal cardiovascular registration trialswhere expertise on the cardiovascular disease condi-tion, clinical trial operations, and established rela-tionship with regulatory agencies are integratedattributes that are unique to academic researchers(88). Academic researchers can work with the phar-maceutical industry to help identify and define unmetclinical needs and to translate mechanisms of drugeffect into clinical scenarios where the benefits mayexceed the risks. Those qualifying as academic re-searchers are typically university-affiliated, haveexpertise in the field, and have prior clinical trialexperience. Selected academic research organizations,especially in cardiovascular medicine, foster the “sci-ence of clinical operations” and stand well-positionedto enact and conduct efficient and streamlined clinicaltrials. However, academic clinical trials do not have tobe affiliated with an academic research organization.Above all, with expertise and understanding ofclinical practice, academic researchers can facilitatediscussions between the regulatory bodies and thepharmaceutical industry from the perspective ofexpertise in disease therapeutics and patient-centeredclinical care.CONTINUE TO STRENGTHEN THE SCIENCE. Manycardiovascular drugs have been developed on thebasis of proven biological activity in modifying abiomarker, hoping that the biomarker directly reflectsdisease outcomes. As discussed previously, manysuccessful therapies have targeted biomarkers (spe-cific lipid fractions and hypertension), resulting inimproved clinical outcomes, but the definitiverecommendation for therapy occurred only aftervigorous testing in the form of multiple clinical trialswith thousands of patients and valid clinical end-points. In this setting, the effect sizes may be modest,and therefore, interventions require large trials toshow benefit. At the meeting, there was generalagreement that to improve efficiency from a scientificstandpoint, development strategy should move awayfrom targeting a directional change in biomarkers to:1) use biomarkers to form enriched populations andfurther classify disease; and 2) learn more about thebiologic targets and causal mechanisms that drivemost cardiovascular diseases.

Using biomarkers to enrich study populations mayhelp refine our definitions of disease and lead tosuccessful drug development. Several examples

already exist in the published data. In the COMPAN-ION (Comparison of Medical Therapy, Pacing, andDefibrillation in Heart Failure) trial, which demon-strated a 20% reduction in all-cause mortalityor hospitalization with treatment with chronic re-synchronization therapy (with or without an intra-cardiac defibrillator), the patient population wasenriched by only including those with a QRS intervalof $120 ms. The investigators widely acknowledgethat if this therapy had been applied to the entireheart failure population, it almost certainly wouldhave been negative (89). The JUPITER (Justificationfor the Use of Statins in Prevention: an InterventionTrial Evaluating Rosuvastatin) primary preventiontrial, which demonstrated a reduction of major car-diovascular events with rosuvastatin, enriched an ap-parently healthy population by including only thosewith an elevation of an inflammatory marker, C-reac-tive protein (90). Finally, although the TOPCAT trial ofspironolactone in patients with preserved ejectionfractionwas negative overall, a pre-specified subgroupof patients with elevated brain natriuretic peptidemay have benefitted—a finding that must be regardedas exploratory and requires replication (91). Enrich-ment with biomarkers could help fuel studies totackle several unmet cardiovascular research needs,

CENTRAL ILLUSTRATION Goals of Stakeholders Involved in New Cardiovascular Drug Development and Their Influence on Each Other

Regulatory AgenciesApproval of effectiveand safe therapies Protecting public safetyRigorous processes to support regulatory approvals

Government FundingAgenciesEffective and safe therapies Scientific discovery and translation

IndustryEffective and safe therapies Investment in R&DReturn on investment for investors and shareholders

PractitionersEffective and safe therapiesAddress unmet clinical needs

PatientsEffective and safe therapiesTransparencyAccess to careImproved quality of life and outcomes

Academic ResearchersEffective and safe therapiesMore efficient and timelystudies to support drug development

Fordyce, C.B. et al. J Am Coll Cardiol. 2015; 65(15):1567–82.

Optimal drug development for patients with cardiovascular disease requires true partnerships among academic researchers, regulatory officials, industry representa-

tives, government-funded researchers, and practicing clinicians.

Fordyce et al. J A C C V O L . 6 5 , N O . 1 5 , 2 0 1 5

Cardiovascular Drug Development A P R I L 2 1 , 2 0 1 5 : 1 5 6 7 – 8 2

1578

including pulmonary hypertension, hypertrophic car-diomyopathy, and congenital heart disease. Therefore,although the use of biomarkers as surrogate endpointscan prove treacherous, they may be useful to enrichpopulations that may benefit from specific therapies.

To improve understanding of the causal mecha-nisms of disease, several novel approaches havebeen leveraged. A genome-wide association studyidentified 13 new susceptibility loci for coronary ar-tery disease (92). Of the 13 loci, only 3 were associatedwith conventional, established cardiac risk factors,and all were found to be of greater significance thanPCSK9, which missed the genome-wide significancelevel by a small margin. This analysis shows thatthere are many potential nonconventional targetsassociated with the pathogenesis of coronary arterydisease available for mining. Next-generation sequ-encing found a strong association between titin anddilated cardiomyopathy (93), and the importanceof systems biology approaches for drug discoveryhas now been recognized for over a decade (94).Other strategies include Mendelian randomization,

proteomics, and metabolomics. New basic sciencetechnologies have helped further our understanding ofsome of the causal mechanisms of cardiovascular dis-ease and have the potential to stimulate futuredrug development. Increasing transdisciplinary ex-changes of knowledge between the basic and clinicalresearch fields through training and collaborationswillenhance efforts to identify new targets, pathologicmechanisms, and drugs to address the unmet need incardiovascular disease.STRATEGIES TO IMPROVE INDUSTRY-SPONSORED

CARDIOVASCULAR DRUG DEVELOPMENT. The phar-maceutical industry is generally composed of 2 mainindustry groups: those from the top 15 largest phar-maceutical companies (“Big Pharma”) and othercompanies considered small to medium-sized enter-prises (95). From a drug-development process, thisdistinction may be important when evaluating re-search and development strategies. Small to medium-sized enterprises typically focus on the biology andpotentially large effect sizes in serious cardiovascularillnesses. As discussed, leveraging human genetics or

TABLE 2 Key Issues Underlying Adverse Trends in Cardiovascular

Drug Development and Potential Solutions

Issue Solution

Rising costs � Reduce extraneous data collectedB Meet early with key stakeholders

(regulators, academic collaborators)to set limits

B Adopt quality by designB Implement centralized monitoring

practices/risk-based monitoringB Prioritize the collection of suspected

unexpected serious adverse reactions;limit collection of other adverseevents

� Centralize institutional review boards� Harness PCORnet to help centralize and

organize clinical data

Regulatory uncertainty � Ensure frequent and early communica-tion with regulators

� Leverage academic experts to interfacebetween industry and regulatory bodies

� Use NIH programs to mitigate financialrisk in early phases

� Consider novel trial designB Adaptive designB Pre- and post-marketing approval

studies

Necessity of validated clinicaloutcomes over surrogateendpoints

� Increase focus on practical, streamlinedtrialsB Adopt quality by designB Use phase II to inform phase III:

dosing, pharmacokinetic modeling,side effects, and biomarkers

� Use biomarkers to enrich populations,but continue to demand clinicaloutcomes

Apparent decline in newcardiovascular drug targets

� Strengthen novel scientific methods tofurther define the pathophysiology andcreate new “biological signature”B Genome-wide association studiesB Next-generation sequencingB Mendelian randomizationB Proteomics and metabolomics

� Leverage NIH programs to facilitatetranslation of basic cardiovascular dis-coveries into clinical applications

Discord between cardiovascularburden and public perception

� Channel advocacy through the AmericanHeart Association and American Collegeof Cardiology

NIH ¼ National Institutes of Health; PCORnet ¼ National Patient-Centered Clinical ResearchNetwork.

J A C C V O L . 6 5 , N O . 1 5 , 2 0 1 5 Fordyce et al.A P R I L 2 1 , 2 0 1 5 : 1 5 6 7 – 8 2 Cardiovascular Drug Development

1579

known human pharmacology will be extremely valu-able to identify drug targets with a higher likelihoodof success, instead of investing in a multitude oftargets and accepting that some will fail. An exampleof this was the identification of the PCSK9 protein as acompelling target to treat hypercholesterolemia(96,97). The convergence of deoxyribonucleic acidsequencing, computational biology, and statisticalmethods is uncovering new compelling targets thatshould lead to the next wave of therapies aimed atreducing cardiovascular morbidity and mortality. Inaddition, this new paradigm should mitigate invest-ment risk and facilitate the acceleration of high-potential programs. As a corollary, some meetingparticipants felt that a move toward reduced invest-ment in targets supported only by nonclinical geneticsand disease models would be beneficial. Programswith the greatest risk of failure should be reserved forthe most grievous illness, novel science, and potentialbiggest incremental advance for patients. Last, tech-nological advances in small and large molecule gen-eration andmanufacturing as well as new insights intobiology that allow for different therapeutic modal-ities have positioned the industry to address cardio-vascular disease in ways previously not achievable.

From a “Big Pharma” industry perspective, simplerandomized trial design could prove to be key. Thispathway emphasizes clinical outcomes over putativesurrogate endpoints. Trials must be sufficiently largeto detect small but clinically-meaningful differencesin mortality and other major outcomes important toboth the individual and to society. Priority should begiven to suspected unexpected serious adverse re-actions and to limiting the recording of less impactfuladverse and serious adverse events. Streamlined,risk-based, and remote monitoring would be benefi-cial. Big firms should continue to strive to improvetrial efficiency and emphasize meaningful clinicaloutcomes.PROMISING TRIAL DESIGNS TO IMPROVE DRUG

DEVELOPMENT. A potential novel strategy to helpstreamline clinical trials in the future would be toperform a smaller pre-marketing study but agree to alarger post-marketing comparative effectiveness trial.Currently, post-marketing studies cannot be used tosupport approval and are primarily used to refine thelabeling of an approved drug therapy. These aregenerally studies required of or agreed to by asponsor that are conducted after the FDA hasapproved a product for marketing, and they typicallyserve to gather additional information about a prod-uct’s safety, efficacy, or optimal use (98). By func-tioning as a more comprehensive extension of atypical post-marketing study, a “continuum” study

design would pre-specify integrated, systematic,monitored data from pooled pre-market randomizedand post-market registry (99). Gathering pre- andpost-market “real world” safety data might bothimprove safety assessment and support the assuranceof safety in smaller pre-market studies. From an in-dustry perspective, smaller studies resulting inreduction of pre-market cost and time could provideincentive for a serious commitment to complete post-market registries that are larger and more rigorousthan current post-market data collections. In 2012,the FDA released its vision for post-market surveil-lance for medical devices (100). Although this effortspecifically covered medical devices, this system

Fordyce et al. J A C C V O L . 6 5 , N O . 1 5 , 2 0 1 5

Cardiovascular Drug Development A P R I L 2 1 , 2 0 1 5 : 1 5 6 7 – 8 2

1580

could be adapted and implemented as a form of post-market enforcement to study novel drugs. Thesesystems can and should be used to also assess efficacy(101). Furthermore, such post-marketing comparativeeffectiveness trials could align with 1 of severalPatient-Centered Outcomes Research Institute initia-tives, including PCORnet (discussed previously),whose goal is to build clinical research into the healthcare process by creating a national network for con-ducting clinical outcomes research (102,103). An ex-ample of a future clinical trial integrating an industrysponsor, the FDA, and patient-centered outcomesmight be a post-marketing comparative effectivenesstrial supported by the Patient-Centered OutcomesResearch Institute, the data from which are thencaptured, integrated, and studied using PCORnet.

CONCLUSIONS: A CALL TO END

THE HIBERNATION

Despite the growing global burden of cardiovasculardisease, metrics indicate that cardiovascular drugdevelopment has steadily declined relative to thegrowth of scientific discovery and the expansion ofdevelopment in other therapeutic areas. This trend isparticularly concerning given the unprecedented ef-fect of cardiovascular drugs on outcomes, particularlyin high-income countries (104). Investment in car-diovascular drug development has shifted elsewhere.This trend seems driven primarily by economic fac-tors, including ROI with increased reimbursementand the perceived reduction in investment risk inbringing noncardiovascular drugs to market.

Nevertheless, the science of clinical therapeuticsdrives the continued requirement for the field of car-diovascular medicine to have large cardiovascularoutcomes trials to demonstrate the balance of benefitsand harms of potential novel therapies. Fortunately,

despite the current perceived hibernation of cardio-vascular drug development, most large pharmaceu-tical firms still have an interest in pursuing novelcardiovascular therapies should the right opportunityexist (45). The key issues and potential solutions aresummarized in Table 2. Leveraging academic collabo-rations and novel governmental programs to identify,derisk, and develop potential therapeutic targetsthrough pre-clinical and early-phase developmentappears to provide a positive path forward. Bystrengthening and refining the scientific questions andearly-phase discoveries, cardiovascular drug devel-opment can rise once again by targeting enrichedpopulations, using novel operational approaches tostudy design and conduct afforded by advances in thescience of clinical trials, and emphasizing the pursuitof unmet clinical needs. Outcomes research and pop-ulation health are coalescing around streamlining andsimplification. As such, the “hibernation period” forcardiovascular drug development could draw to a closeby application of the principles recommended in thispaper.

ACKNOWLEDGMENTS The authors would like tothank Drs. Kenneth Kaitin, Joseph DiMasi, and JoshuaCohen from the Tufts Center for the Study of DrugDevelopment as well as Dr. Zubin Eapen from theDuke Clinical Research Institute for their insightfulcomments during the synthesis of the manuscript.The authors would also like to thank Morgan deBle-court for her editorial contributions, which wereprovided as part of her regular duties as an employeeof the Duke Clinical Research Institute.

REPRINT REQUESTS AND CORRESPONDENCE: Dr.Christopher B. Fordyce, Duke Clinical ResearchInstitute, 2400 Pratt Street, Durham, North Car-olina 27705. E-mail: christopher.fordyce@duke.edu.

RE F E RENCE S

1. Go AS, Mozaffarian D, Roger VL, et al. Heartdisease and stroke statistics—2014 update: areport from the American Heart Association. Cir-culation 2014;129:e28–292.

2. Gaziano TA. Cardiovascular disease in thedeveloping world and its cost-effective manage-ment. Circulation 2005;112:3547–53.

3. Nabel EG, Braunwald E. A tale of coronary ar-tery disease and myocardial infarction. N Engl JMed 2012;366:54–63.

4. Fuster V. Global burden of cardiovascular dis-ease: time to implement feasible strategies and tomonitor results. J Am Coll Cardiol 2014;64:520–2.

5. Fuster V, Kelly BB, Vedanthan R. Promotingglobal cardiovascular health: moving forward.Circulation 2011;123:1671–8.

6. Mathers CD, Loncar D. Projections of globalmortality and burden of disease from 2002 to2030. PLoS Med 2006;3:e442.

7. Reddy KS. Cardiovascular disease in non-Western countries. N Engl J Med 2004;350:2438–40.

8. DiMasi JA, Feldman L, Seckler A, Wilson A.Trends in risks associated with new drug devel-opment: success rates for investigational drugs.Clin Pharmacol Ther 2010;87:272–7.

9. Pammolli F, Magazzini L, Riccaboni M. Theproductivity crisis in pharmaceutical R&D. Nat RevDrug Discov 2011;10:428–38.

10. DiMasi JA. Causes of clinical failures varywidely by therapeutic class, phase of study. TuftsCSDD Impact Report 2013;15(5):1–4.

11. Kaitin KI, DiMasi JA. Pharmaceutical innovationin the 21st century: new drug approvals in the firstdecade, 2000–2009. Clin Pharmacol Ther 2011;89:183–8.

12. Sacks LV, Shamsuddin HH, Yasinskaya YI,Bouri K, Lanthier ML, Sherman RE. Scientific andregulatory reasons for delay and denial of FDAapproval of initial applications for new drugs,2000–2012. JAMA 2014;311:378–84.

13. Berggren R, Moller M, Moss R, Poda P,Smietana K. Outlook for the next 5 years indrug innovation. Nat Rev Drug Discov 2012;11:435–6.

14. Kaitin KI. Deconstructing the drug develop-ment process: the new face of innovation. ClinPharmacol Ther 2010;87:356–61.

J A C C V O L . 6 5 , N O . 1 5 , 2 0 1 5 Fordyce et al.A P R I L 2 1 , 2 0 1 5 : 1 5 6 7 – 8 2 Cardiovascular Drug Development

1581

15. Eichler HG, Pignatti F, Flamion B, Leufkens H,Breckenridge A. Balancing early market access tonew drugs with the need for benefit/risk data: amounting dilemma. Nat Rev Drug Discov 2008;7:818–26.

16. Lundh A, Krogsbøll LT, Gøtzsche PC. Sponsors’participation in conduct and reporting of industrytrials: a descriptive study. Trials 2012;13:146.

17. Lundh A, Barbateskovic M, Hróbjartsson A,Gøtzsche PC. Conflicts of interest at medicaljournals: the influence of industry-supportedrandomised trials on journal impact factorsand revenue–cohort study. PLoS Med 2010;7:e1000354.

18. Whitford AB. The reduction of regulatory un-certainty: evidence from transfer pricing policy.Saint Louis University Law Journal 2010;55:269–306.

19. Hoffmann VH, Trautmann T, Schneider M.A taxonomy for regulatory uncertainty—applica-tion to the European Emission Trading Scheme.Environmental Science & Policy 2008;11:712–22.

20. Morgan DP. Rating banks: risk and uncertaintyin an opaque industry. Am Econ Rev 2002;92:874–88.

21. Milne CP, Kaitin KI. FDA review divisions: per-formance levels and the impact on drug sponsors.Clin Pharmacol Ther 2012;91:393–404.

22. Glickman SW, McHutchison JG, Peterson ED,et al. Ethical and scientific implications of theglobalization of clinical research. N Engl J Med2009;360:816–23.

23. Berntgen M, Gourvil A, Pavlovic M,Goettsch W, Eichler HG, Kristensen FB. Improvingthe contribution of regulatory assessment reportsto health technology assessments—a collaborationbetween the European Medicines Agency and theEuropean Network for Health TechnologyAssessment. Value Health 2014;17:634–41.

24. Getz KA. Global clinical trials activity in thedetails: individual countries hold the key tofinding hot spots in growth regions like Centraland Eastern Europe. Appl Clin Trials 2009. Avail-able at: http://www.appliedclinicaltrialsonline.com/appliedclinicaltrials/article/articleDetail.jsp?id¼453243. Accessed October 14, 2014.

25. Downing NS, Aminawung JA, Shah ND,Braunstein JB, Krumholz HM, Ross JS. Regulatoryreview of novel therapeutics–comparison of threeregulatory agencies. N Engl J Med 2012;366:2284–93.

26. Thiers FA, Sinskey AJ, Berndt ER. Trends in theglobalization of clinical trials. Nat Rev Drug Discov2008;7:13–4.

27. Mentz RJ, Kaski JC, Dan GA, et al. Implicationsof geographical variation on clinical outcomes ofcardiovascular trials. Am Heart J 2012;164:303–12.

28. Fiuzat M, Califf RM. Conduct of clinical trialsin acute heart failure: regional differences inheart failure clinical trials. Heart Fail Clin 2011;7:539–44.

29. Pfeffer MA, Claggett B, Assmann SF, et al.Regional variation in patients and outcomes in theTreatment of Preserved Cardiac Function HeartFailure With an Aldosterone Antagonist (TOPCAT)trial. Circulation 2015;131:34–42.

30. Wallentin L, Becker RC, Budaj A, et al. Tica-grelor versus clopidogrel in patients with acutecoronary syndromes. N Engl J Med 2009;361:1045–57.

31. Grace C, Kyle M. Comparative advantages ofpush and pull incentives for technology develop-ment: lessons for neglected disease technologydevelopment. Global Forum Update on Researchfor Health 2009;6:147–51.

32. Sampat BN. Mission-oriented biomedicalresearch at the NIH. Res Policy 2012;41:1729–41.

33. Mervis J. U.S. spending. Final 2014 budget helpsscience agencies rebound. Science 2014;343:237.

34. National Institutes of Health website. Appro-priations history by institute/center (1938 to pre-sent). Available at: http://officeofbudget.od.nih.gov/approp_hist.html. Accessed October 6, 2014.

35. Milne CP. Prospects for rapid advances in thedevelopment of new medicines for special medicalneeds. Clin Pharmacol Ther 2014;95:98–109.

36. Kinch MS, Merkel J, Umlauf S. Trends inpharmaceutical targeting of clinical indications:1930–2013. Drug Discov Today 2014;19:1682–5.

37. Belluz J. The truth about the Ice Bucket Chal-lenge: viral memes shouldn’t dictate our charitablegiving. Available at: http://www.vox.com/2014/8/20/6040435/als-ice-bucket-challenge-and-why-we-give-to-charity-donate. Accessed August 29, 2014.

38. Wood S. Money—is cancer beating cardiovas-cular disease? Medscape. August 16, 2011. Avail-able at: http://www.medscape.com/viewarticle/790999. Accessed October 16, 2014.

39. Dragojlovic N, Lynd LD. Crowdfunding drugdevelopment: the state of play in oncology andrare diseases. Drug Discov Today 2014;19:1775–80.

40. Heart Innovation Forum. Available at:http://www.heartinnovationforum.org/. AccessedNovember 14, 2014.

41. Ahmad T, Becker RC. The unmet need forphilanthropic funding of early career cardiovas-cular investigators. J Thromb Thrombolysis 2014;37:527–31.

42. Siegel KR, Feigl AB, Kishore SP, Stuckler D.Misalignment between perceptions and actualglobal burden of disease: evidence from the USpopulation. Global Health Action 2011;4.

43. Wood S. Glory—is cancer beating cardiovas-cular disease? Medscape. August 18, 2011. Avail-able at: http://www.medscape.com/viewarticle/791006#vp_2. Accessed October 14, 2014.

44. Mosca L, Mochari-Greenberger H, Dolor RJ,Newby LK, Robb KJ. Twelve-year follow-up ofAmerican women’s awareness of cardiovasculardisease risk and barriers to heart health. Circ Car-diovasc Qual Outcomes 2010;3:120–7.

45. Giniatullina A, Boorsma M, Mulder GJ, vanDeventer S. Building for big pharma. Nature Bio-technol 2013;31:284–7.

46. Mathieu MP. PAREXEL BiopharmaceuticalR&D Statistical Sourcebook 2013/2014. Waltham,MA: Parexel International Corp, 2013.

47. Bioassociate Innovative Consulting website.2013 Biotech & pharma IPO review–most populartherapeutic trends, trending clinical stages

and post-IPO winners & losers. December 14,2013. Available at: http://www.bioassociate.com/2013-biotech-pharma-ipo-review-most-popular-therapeutic-trends-trending-clinical-stages-and-post-ipo-winners-losers/. Accessed October 8,2014.

48. IMS Health website. Top-line market data.December 2013. Available at: http://www.imshealth.com/portal/site/imshealth/menuitem.18c196991f79283fddc0ddc01ad8c22a/?vgnextoid¼6521e590cb4dc310VgnVCM100000a48d2ca2RCRD&vgnextfmt¼default. Accessed October 29, 2014.

49. Eapen ZJ, Lauer MS, Temple RJ. The impera-tive of overcoming barriers to the conduct oflarge, simple trials. JAMA 2014;311:1397–8.

50. Temple R. Are surrogate markers adequate toassess cardiovascular disease drugs? JAMA 1999;282:790–5.

51. Turnbull F, Neal B, Ninomiya T, et al., for theBlood Pressure Lowering Treatment Trialists’Collaboration. Effects of different regimens tolower blood pressure on major cardiovascularevents in older and younger adults: meta-analysisof randomised trials. BMJ 2008;336:1121–3.

52. Mihaylova B, Emberson J, Blackwell L, et al.,for the Cholesterol Treatment Trialists’ (CTT)Collaborators. The effects of lowering LDLcholesterol with statin therapy in people at lowrisk of vascular disease: meta-analysis of individ-ual data from 27 randomised trials. Lancet 2012;380:581–90.

53. Cannon CP. IMPROVE-IT trial: a comparison ofezetimibe/simvastatin versus simvastatin mono-therapy on cardiovascular outcomes after acutecoronary syndromes. Paper presented at: AmericanHeart Association Scientific Sessions; November 14,2014; Chicago, IL. Available at: http://my.americanheart.org/idc/groups/ahamah-public/@wcm/@sop/@scon/documents/downloadable/ucm_469669.pdf. Accessed December 3, 2014.

54. ALLHAT Collaborative Research Group. Majorcardiovascular events in hypertensive patientsrandomized to doxazosin vs chlorthalidone: theantihypertensive and lipid-lowering treatment toprevent heart attack trial (ALLHAT). JAMA 2000;283:1967–75.

55. Barter PJ, Caulfield M, Eriksson M, et al. Ef-fects of torcetrapib in patients at high risk forcoronary events. N Engl J Med 2007;357:2109–22.

56. AIM-HIGH terol levels receiving intensivestatin therapy. N Engl J Med 2011;365:2255–67.

57. O’Connor CM, Gattis WA, Uretsky BF, et al.Continuous intravenous dobutamine is associatedwith an increased risk of death in patients withadvanced heart failure: insights from the FlolanInternational Randomized Survival Trial (FIRST).Am Heart J 1999;138 Pt 1:78–86.

58. Felker GM, Benza RL, Chandler AB, et al. Heartfailure etiology and response to milrinone indecompensated heart failure: results from theOPTIME-CHF study. J Am Coll Cardiol 2003;41:997–1003.

59. Echt DS, Liebson PR, Mitchell LB, et al. Mor-tality and morbidity in patients receiving encai-nide, flecainide, or placebo. The CardiacArrhythmia Suppression Trial. N Engl J Med 1991;324:781–8.

Fordyce et al. J A C C V O L . 6 5 , N O . 1 5 , 2 0 1 5

Cardiovascular Drug Development A P R I L 2 1 , 2 0 1 5 : 1 5 6 7 – 8 2

1582

60. Booth CM, Eisenhauer EA. Progression-freesurvival: meaningful or simply measurable? J ClinOncol 2012;30:1030–3.

61. Kay A, Higgins J, Day AG, Meyer RM,Booth CM. Randomized controlled trials in the eraof molecular oncology: methodology, biomarkers,and end points. Ann Oncol 2012;23:1646–51.

62. Sridhara R, Johnson JR, Justice R, Keegan P,Chakravarty A, Pazdur R. Review of oncology andhematology drug product approvals at the USFood and Drug Administration between July 2005and December 2007. J Natl Cancer Inst 2010;102:230–43.

63. Hackshaw A, Knight A, Barrett-Lee P,Leonard R. Surrogate markers and survival inwomen receiving first-line combination anthracy-cline chemotherapy for advanced breast cancer. BrJ Cancer 2005;93:1215–21.

64. Louvet C, de Gramont A, Tournigand C,Artru P, Maindrault-Goebel F, Krulik M. Correla-tion between progression free survival andresponse rate in patients with metastatic colo-rectal carcinoma. Cancer 2001;91:2033–8.

65. SaadED,KatzA,Hoff PM,BuyseM.Progression-free survival as surrogate and as true end point:insights from the breast and colorectal cancerliterature. Ann Oncol 2010;21:7–12.

66. JohnsonJR,WilliamsG,PazdurR.Endpoints andUnitedStatesFoodandDrugAdministrationapprovalof oncology drugs. J Clin Oncol 2003;21:1404–11.

67. Johnson JR, Ning YM, Farrell A, Justice R,Keegan P, Pazdur R. Accelerated approval ofoncology products: the food and drug administra-tionexperience. JNatl Cancer Inst 2011;103:636–44.

68. Fleming TR, DeMets DL. Surrogate end pointsin clinical trials: are we being misled? Ann InternMed 1996;125:605–13.

69. Downing NS, Aminawung JA, Shah ND,Krumholz HM, Ross JS. Clinical trial evidencesupporting FDA approval of novel therapeuticagents, 2005–2012. JAMA 2014;311:368–77.

70. Committee on a Framework for Developmenta New Taxonomy of Disease; National ResearchCouncil. Toward Precision Medicine: Building aKnowledge Network for Biomedical Research anda New Taxonomy of Disease. Washington, DC:National Academies Press, 2011.

71. Ahmad T, Fiuzat M, Pencina MJ, et al. Chartinga roadmap for heart failure biomarker studies.J Am Coll Cardiol HF 2014;2:477–88.

72. Kramer JM, Smith PB, Califf RM. Impedimentsto clinical research in the United States. ClinPharmacol Ther 2012;91:535–41.

73. U.S. Department of Health and Human Ser-vices, Food and Drug Administration. Guidance forIndustry: Oversight of Clinical Investigations—ARisk-Based Approach to Monitoring [draft guid-ance]. Silver Spring, MD: U.S. Department ofHealth and Human Services, August 24, 2011.

74. Menikoff J. The paradoxical problem withmultiple-IRB review. N Engl J Med 2010;363:1591–3.

75. Flynn KE, Hahn CL, Kramer JM, et al. Usingcentral IRBs for multicenter clinical trials in theUnited States. PloS One 2013;8:e54999.

76. Kramer JM, Vock D, Greenberg HE, et al.Investigators’ experience with expedited safetyreports prior to the FDA’s final IND safetyreporting rule. Ther Innov Regul Sci 2014;48:413–9.

77. Califf RM. The Patient-Centered OutcomesResearch Network: a national infrastructure forcomparative effectiveness research. N C Med J2014;75:204–10.

78. O’Connor CM, Starling RC, Hernandez AF,et al. Effect of nesiritide in patients with acutedecompensated heart failure. N Engl J Med 2011;365:32–43.

79. McMurray JJ, Packer M, Desai AS, et al.Angiotensin-neprilysin inhibition versus enalaprilin heart failure. N Engl J Med 2014;371:993–1004.

80. Clinical Trials Transformation Initiative. QBD:project summary. Workshops on quality by design(QbD) and quality risk management (QRM) in clinicaltrials. Available at: http://www.ctti-clinicaltrials.org/what-we-do/investigational-plan/qbd-qrm. AccessedOctober 15, 2014.

81. Kairalla JA, Coffey CS, Thomann MA,Muller KE. Adaptive trial designs: a review ofbarriers and opportunities. Trials 2012;13:145.

82. U.S. Department of Health and Human Services,Food and Drug Administration, Center for DrugEvaluation and Research, Center for Biologics Eval-uation and Research. Guidance for Industry: AdaptiveDesign Clinical Trials for Drugs and Biologics [draftguidance]. Rockville, MD: U.S. Department of Healthand Human Services, February 2010.

83. Kola I, Landis J. Can the pharmaceutical in-dustry reduce attrition rates? Nat Rev Drug Discov2004;3:711–5.

84. Galis ZS, Hoots WK, Kiley JP, Lauer MS. On thevalue of portfolio diversity in heart, lung, andblood research. Circ Res 2012;111:833–6.

85. National Center for Advancing TranslationalSciences. Available at: http://www.ncats.nih.gov/index.html. Accessed March 24, 2015.

86. Galis ZS, Black JB, Skarlatos SI. NationalHeart, Lung, and Blood Institute and the trans-lation of cardiovascular discoveries into thera-peutic approaches. Circ Res 2013;112:1212–8.

87. National Heart, Lung, and Blood Institute. VITAProgram. Available at: http://www.nhlbi.nih.gov/research/resources/vita. AccessedMarch 24, 2015.

88. Brass EP, Hiatt WR. Improving the FDA’sadvisory committee process. J Clin Pharmacol2012;52:1277–83.

89. Bristow MR, Saxon LA, Boehmer J, et al.Cardiac-resynchronization therapy with or withoutan implantable defibrillator in advanced chronicheart failure. N Engl J Med 2004;350:2140–50.

90. Ridker PM, Danielson E, Fonseca FA, et al.Rosuvastatin to prevent vascular events in menand women with elevated C-reactive protein.N Engl J Med 2008;359:2195–207.

91. Pitt B, Pfeffer MA, Assmann SF, et al. Spi-ronolactone for heart failure with preserved ejec-tion fraction. N Engl J Med 2014;370:1383–92.

92. Schunkert H, Konig IR, Kathiresan S, et al.Large-scale association analysis identifies 13 new

susceptibility loci for coronary artery disease. NatGenet 2011;43:333–8.

93. Herman DS, Lam L, Taylor MR, et al. Trunca-tions of titin causing dilated cardiomyopathy.N Engl J Med 2012;366:619–28.

94. Butcher EC, Berg EL, Kunkel EJ. Systemsbiology in drug discovery. Nat Biotechnol 2004;22:1253–9.

95. O’Connell KE, Frei P, Dev KK. The premium ofa big pharma license deal. Nat Biotechnol 2014;32:617–9.

96. Blom DJ, Hala T, Bolognese M, et al.A 52-week placebo-controlled trial of evolocu-mab in hyperlipidemia. N Engl J Med 2014;370:1809–19.

97. Schwartz GG, Bessac L, Berdan LG, et al. Effectof alirocumab, a monoclonal antibody to PCSK9,on long-term cardiovascular outcomes followingacute coronary syndromes: Rationale and designof the ODYSSEY Outcomes trial. Am Heart J 2014;168:682–689.e1.

98. U.S. Food and Drug Administration. Post-marketing requirements and commitments: intro-duction. Available at: http://www.fda.gov/Drugs/GuidanceComplianceRegulatoryInformation/Post-marketingPhaseIVCommitments/. AccessedOctober 16, 2014.

99. Rao SV, Califf RM, Kramer JM, et al. Post-market evaluation of breakthrough technologies.Am Heart J 2008;156:201–8.

100. Center for Devices and Radiological Health,U.S. Food and Drug Administration. Strengtheningour national system for medical device postmarketsurveillance. September 2012. Available at: http://www.fda.gov/downloads/AboutFDA/CentersOffices/OfficeofMedicalProductsandTobacco/CDRH/CDRHReports/UCM301924.pdf. Accessed October 16, 2014.

101. Vavalle JP, Ohman EM. Left ventricularsupport systems for high-risk percutaneouscoronary interventions: how can we improveoutcomes for rare procedures? Circulation 2013;127:162–4.

102. Frank L, Basch E, Selby JV. The PCORIperspective on patient-centered outcomesresearch. JAMA 2014;312:1513–4.

103. Fleurence RL, Curtis LH, Califf RM, Platt R,Selby JV, Brown JS. Launching PCORnet, a na-tional patient-centered clinical research network.J Am Med Inform Assoc 2014;21:578–82.

104. Ford ES, Ajani UA, Croft JB, et al.Explaining the decrease in U.S. deaths fromcoronary disease, 1980–2000. N Engl J Med2007;356:2388–98.

KEY WORDS biomarkers, cardiovascularagents, clinical trials, drug costs,pharmacological, policy

APPENDIX For a list of think tank partici-pants, please see the online version of thisarticle.