nature reviews drug discovery (july 2012)

5/24/2018 Nature Reviews Drug Discovery (July 2012)

1/77

New horizons in pharmaceuticalregulation

Alasdair Breckenridge, Michelle Mello and Bruce M. Psaty

A life cycle approach to pharmaceutical regulation, in which the benefitrisk balanceof new drugs continues to be robustly assessed following market approval, is emergingin both the United States and Europe.

Alasdair Breckenridge is

Chairman of the Medicinesand Healthcare products

Regulatory Agency,

London SW1W 9SZ, UK.

Michelle Mello is Professor

of Law and Public Health

at Harvard University,

Cambridge, Massachusetts,

Massachusetts 02115, USA.

Bruce M. Psaty is Professor

of Medicine, Epidemiology

and Health Services at the

University of Washington,

Seattle, Washington 98101,

USA.

doi:10.1038/nrd3787

When a medicine is first marketed, much is knownabout its quality, pharmacology and efficacy in aselected, carefully screened group of patients involvedin the clinical trial programme. However, these efficacydata may provide an incomplete, and perhaps evenmisleading, indication of the drugs effectiveness in thepopulation overall. Even greater gaps in knowledge existwith regard to its safety in the wider community. Adverseeffects may be too rare to detect in pre-approval clini-cal trials or they may occur in populations that are notrepresented in these trials, such as patients also takingother medicines.

These observations underscore the need for an

ongoing assessment of a new drugs effectiveness andsafety in the post-marketing period. Importantly, the pre-occupation of the press, the public and even politicianswith issues of drug safety is often misplaced, because of afocus on safety alone, even though what determines the

value of a drug is its benefitrisk balance1. Indeed, manydrugs for example, some commonly used anticanceragents or drugs for HIV infection have substantialtoxicities but are sufficiently effective that their benefitrisk profile supports their use in treating these seriousdiseases, often initially under some form of supervisionto maximize the health benefits for patients. Here, wediscuss new directions in pharmaceutical regulation inEurope and the United States that reflect a recognition

of the need for monitoring and recalibrating the benefitrisk profile during the entire life cycle of a drug.

The need for a life cycle approach

The development of innovative drugs has become morechallenging and more costly in recent decades for sev-eral reasons. Regulatory authorities have heightened theirrequirements for evidence of efficacy and safety (in partin response to major drug safety problems), and cash-strapped payers for health care increasingly need strongevidence that new drugs represent value for money com-pared with existing drugs. Indeed, industry may haveshifted the emphasis of its innovative drug development

into areas where the bar for approval is lower owing to alack or absence of effective drugs, but the risk of failureis higher. Nevertheless, in some disease areas there hasbeen considerable success, such as biologics for patientswith rheumatoid arthritis and molecularly targeted drugsfor some cancers (such as chronic myeloid leukaemia).However, the costs of failure can be huge, particularlyfor indications such as Alzheimers disease that requirelengthy clinical trials and that lack validated surrogateend points.

In the United States, and some European countries,an increasingly vocal patient advocacy movement hasdemanded earlier access to new medicines, particularly

for life-threatening diseases. This movement, which hadits origins in AIDS and cancer advocacy in the 1980s, hasput regulators under pressure to relax their evidentiaryrequirements for the approval of potentially life-savingmedicines. Most regulatory agencies have respondedwith accelerated approval tracks and compassionateuse programmes but there are some commentatorswho still press for a broad reversal of a regulatory trendrequiring a higher level of evidence for drug approval.For example, a former commissioner of the US Foodand Drug Administration (FDA) recently called for theUS Congress to allow the approval of new drugs basedsolely on drug safety, with efficacy to be proven in post-approval trials2. The US Congress is currently consider-

ing legislation to speed market access of drugs for seriousdiseases by allowing the expanded use of surrogate endpoints and rapidly measurable clinical end points andthus, presumably, drug approval based on fewer, smallerand shorter trials.

Such an approach, if enacted, means that regulatorsmust move their horizons too, focusing new attentionon the post-approval period of drug development withactive surveillance and post-marketing clinical trials.If standards for initial drug approval are eased andeven if they are not regulatory agencies must havea robust process in place for the continuous evalua-tion of a drugs benefitrisk balance over its entire life

COMMENT

NATURE REVIEWS |DRUG DISCOVERY VOLUME 11 |JULY 2012 |501

2012 Macmillan Publishers Limited. All rights reserved


2/77

cycle. In 2007, an expert committee of the Institute ofMedicine (IOM) in the United States characterized theFDAs approach to benefitrisk analysis as ad hoc, infor-mal and qualitative, and recommended that the agencydevelop a more systematic evaluation as part of a lifecycle approach3.

The need for this life cycle approach again became

apparent as a result of the recent controversy overthe thiazolidinedione antidiabetic drug rosiglitazone(Avandia), which was granted marketing authorizationin 1999 based on its ability to improve glycaemic control,as assessed by reduced levels of blood sugar and haemo-globin A1C surrogate markers that were assumed topredict clinical benefit. However, an increased risk ofcardiovascular events associated with the use of rosigli-tazone highlighted initially in 2007 became appar-ent, emphasizing the need for more effective benefitriskanalysis for recently introduced medicines.

Emerging issues in drug regulation

A life cycle approach has begun to emerge in pharma-

ceutical regulation in Europe and the United States,although much remains to be done. In 2005, the EuropeanCommission proposed that a risk management planbecome part of the required regulatory submission fornew medicines. The plan should include: first, a state-ment of what is known about the safety of the medicinesand also what is not known; second, a plan for how theunknown information should be acquired; third, whatrisk minimization steps, if any, should be followed (forexample, allowing only specialists to prescribe the drugor launching an educational programme for patients);and last, how this information should be communicated.In 2008, the Commission proposed a new set of pharma-

covigilance regulations that will take effect in July 2012.These not only reinforce the importance of risk manage-ment plans but also give regulators the authority to requirepost-marketing studies of effectiveness as well as of safety.

In the United States, the Food and Drug Adminis-tration Amendments Actof 2007 substantially strength-ened the FDAs regulatory authority in the post-marketingperiod, allowing it to require specific studies in certaincircumstances. The FDA was also empowered to requiredrug manufacturers to submit a risk evaluation mitigationstrategy as a condition of drug approval.

Notwithstanding these initiatives, concerns aboutdrug safety and the conduct of post-marketing studiesremained, and were heightened by the case of rosigli-

tazone. Consequently, a second IOM committee wasconvened in 2010 at the request of the FDA and otherfederal agencies to offer further guidance on strategiesfor studying the safety of approved medicines. The finalreport4, released in May 2012, acknowledges the stepsthat the FDA has already taken to improve drug safety

but makes 23 recommendations to implement morefully a life cycle approach to drug regulation. One ofthe most interesting recommendations is for the devel-opment of a benefitrisk assessment management plan(BRAMP) by the sponsor of every new drug. This publicdocument, which would be updated over the life cycleof the drug, would provide an updated assessment of

its benefitrisk profile. Over time, the BRAMP wouldinclude information about any public health ques-tions posed by the drug, a formal benefitrisk assess-ment (which is regularly reviewed) and a rationale forthe type of any post-marketing study of effectivenessand safety. The European Commissions proposals andthose encompassed in a BRAMP converge in their newattention to post-marketing drug safety and an ongoingevaluation of the drugs benefitrisk profile.

Whether the pharmaceutical industry joins in thisconsensus and whether companies will serve as willingor reluctant partners in an invigorated post-marketingdrug regulation scheme remains unclear. Additionalregulatory requirements are never welcome in the

short term, and the post-marketing studies that maybe required under the life cycle approach may imposesubstantial costs on manufacturers. However, industryhas learned the hard way that inadequate responses tosafety signals can have a serious detrimental impact notonly on public health but also in economic and reputa-tional terms. The long-term interests of manufacturersand regulators in well-characterized benefitrisk profilesare essentially the same.

Summary

Experts and regulators in Europe and the United Statesalike now recognize that a robust ongoing assessment of

the benefitrisk balance of new medicines is essential.Regulatory requirements are converging in the directionof the life cycle approach. So too, we hope, will be therecognition that a fully realized vision for drug safetyrequires the commitment of both those who regulatemedicines and those who produce them.

1. Eichler, H.-G. et al.Safe drugs and the cost of good intentions.

N. Engl. J. Med. 360, 13781380 (2009).2. von Eschenbach, A. Toward a 21st century FDA.Wall Street Journal

[online], http://online.wsj.com/article/SB10001424052702303815

404577331673917964962.html(16 Apr 2012).

3. Institute of Medicine. The future of drug safety: promoting and

protecting the health of the public. (National Academies Press,

2007).

4. Institute of Medicine. Ethical and scientific issues in studying the

safety of approved drugs. (National Academies Press, 2012).

Competing interests statementThe authors declare competing financial interests: see Web version for

details.

Disclosure and acknowledgementsThe authors served as members of the Institute of Medicine committees

discussed in this article. They acknowledge important intellectual contri-

butions to the ideas expressed by other members of these committees,

but the views here are their own.

502 |JULY 2012 |VOLUME 11 www.nature.com/reviews/drugdisc

C OMME N T

http://online.wsj.com/article/SB10001424052702303815404577331673917964962.htmlhttp://online.wsj.com/article/SB10001424052702303815404577331673917964962.htmlhttp://www.nature.com/nrd/journal/v11/n7/box/nrd3787_audecl.htmlhttp://www.nature.com/nrd/journal/v11/n7/box/nrd3787_audecl.htmlhttp://online.wsj.com/article/SB10001424052702303815404577331673917964962.htmlhttp://online.wsj.com/article/SB10001424052702303815404577331673917964962.html


3/77

NEWS IN BRIEF

Antibiotic R&D gets a dose of funding

The Innovative Medicines Initiative (IMI) has launched a 223 million (US$280 million) New

Drugs 4 Bad Bugs (ND4BB) programme to spur the development of antibiotics.

The lowdown:Citing an increasing risk of antibacterial resistance and a lack of progress in the

development of new antibiotics, the IMI has launched a call for proposals for its ND4BB programme,

which is aimed at encouraging industry and academic activity in the antibiotic space. The IMI is

contributing 109 million to the programme, and five pharma partners (AstraZeneca, Basilea,

GlaxoSmithKline (GSK), Johnson & Johnson (J&J) and Sanofi) are pitching in a further 114 million

as in-kind contributions. Funders may provide further backing to the project at later stages,

and the IMI anticipates that the programme could use up to 600 million over the next 7 years.

ND4BB marks the first attempt by the IMI to directly fund clinical trial work. The programme

includes a commitment to back the development of GSKs peptide deformylase inhibitor

GSK1322322 in a Phase III trial in community-acquired bacterial pneumonia and in a Phase

IIb trial in acute bacterial skin and skin structure infections in part in the hopes of improving

clinical trial design and infrastructure. The programme could also fund the development of two

AstraZeneca drugs, and back-up GSK compounds, depending on the results of these and other

ongoing trials.

The programme also aims to foster collaborative exchange of data and regulatoryexperiences between companies, and to build and train networks of researchers. A fraction

of the budget is earmarked for preclinical research, including projects that could improve

the communitys understanding of antimicrobial resistance and bacterial cell penetration,

efflux and permeability. Full details of the first call for proposals for ND4BB are available online

(see theIMIwebsite), and submissions are due by 9 July 2012.

made in a plant cell system. Regulatory approval

decisions were pending for several candidates

including Bristol-Myers Squibb/Pfizers

anticoagulant apixaban, Arenas anti-obesity

drug lorcaserin and Sanofis teriflunomide for

multiple sclerosis as Nature ReviewsDrugDiscoverywent to press.

Last year the FDA approved 30 new drugs,

in line with an average of 29.5 per year since

1993, the first full year in which the Prescription

Drug User Fee Act (PDUFA) was in place.

A recent 5-year outlook for the industry forecasts

that the agency will approve, on average, 3035

new drugs per year until 2016 (Nature Rev. Drug

Discov.11, 435436; 2012). An FDA official hassaid that the agency is anticipating that drug

developers will file over 20 approval applications

for cancer drugs this year.

SGLT2, take two?

Interest in antidiabetic sodium-dependent

glucose co-transporter 2 (SGLT2) inhibitors has

picked up again.

The lowdown: Earlier this year, the FDA rejected

AstraZenecas and Bristol-Myers Squibbs SGLT2

inhibitor dapagliflozin, a potential first-in-class

drug for diabetes, probably owing in part to

concerns that it could increase the risk of

cancer. New Phase III data and filing activity

around Johnson & Johnsons SGLT2 inhibitorcanagliflozin are now renewing interest in

the class of drugs, which act by blocking renal

glucose reabsorption and thereby increasing

glucose secretion via urine (Nature Rev. DrugDiscov.10, 645646; 2011). J&J presentedtop-line data from five trials at the American

Diabetes Association (ADA) annual meeting

in Philadelphia, including one trial in which

canagliflozin plus metformin and sulfonylurea

beat sitagliptin plus metformin and sulfonylurea

in terms of reducing haemoglobin A1c (HBA1c)

levels, and another in which it beat glimepiride.

The company did not disclose cancer incidence

data in these trials. J&J filed the drug for USapproval at the end of May.

Dapagliflozins future in the United States,

meanwhile, remains unclear. AstraZeneca

and Bristol-Myers Squibb released new Phase

III data on dapagliflozin at the ADA meeting

showing that the drug in combination with

sitagliptin reduces HbA1c levels versus placebo

plus sitagliptin, but they did not release cancer

incidence data from this trial. The Committee

for Medicinal Products for Human Use (CHMP)

gave the drug the green light for use in the

European Union in April.

First-in-class approvals include Genentechs

Hedgehog inhibitor vismodegib for basal cell

carcinoma, and Vertexs ivacaftor a cystic

fibrosis transmembrane conductance regulator

(CFTR) potentiator that is the first cystic

fibrosis drug to address the mechanism of thedisease rather than the symptoms. Protalixs

taliglucerase alfa, an enzyme replacement

therapy for Gauchers disease, meanwhile

received the first approval for a drug that is

FDA on track with first-halfapprovals

The US Food and Drug Administration (FDA)

approved 12 new drugs and biologics in thefirst half of the year.

The lowdown: The agencys Center for Drug

Evaluation and Research (CDER) has already

approved several novel drugs this year (TABLE 1).

Table 1 | New drugs approved by the FDA in the first half of 2012

Drug name Lead company Indication

Glucarpidase* BTG International Toxic plasma concentrations of methotrexate

Ingenol mebutate Leo Pharma Actinic keratosis

Axitinib Pfizer Advanced renal cell carcinoma

Vismodegib Genentech Metastatic or locally advanced basal cell carcinoma

Ivacaftor Vertex Cystic fibrosisTafluprost Merck Elevated intraocular pressure in patients with

open-angle glaucoma or ocular hypertension

Lucinactant DiscoveryLaboratories

Prevention of respiratory distress syndrome

Peginesatide Affymax Anaemia due to chronic kidney disease

Florbetapir F18 Avid Agent to estimate amyloid-neuritic plaque densityas Alzheimers disease diagnostic

Avanafil Vivus Erectile dysfunction

Taliglucerase alfa Protalix Gauchers disease

Pertuzumab* Genentech HER2-positive metastatic breast cancer

*These drugs were approved as biologics license applications.

N E W S & A N A L Y S I S


http://www.imi.europa.eu/content/stage-1-4http://www.imi.europa.eu/content/stage-1-4http://www.imi.europa.eu/content/stage-1-4http://www.nature.com/nrd/journal/v11/n6/full/nrd3744.htmlhttp://www.nature.com/nrd/journal/v11/n6/full/nrd3744.htmlhttp://www.nature.com/nrd/journal/v11/n6/full/nrd3744.htmlhttp://www.nature.com/nrd/journal/v11/n6/full/nrd3744.htmlhttp://www.nature.com/nrd/journal/v11/n6/full/nrd3744.htmlhttp://www.nature.com/nrd/journal/v10/n9/full/nrd3546.htmlhttp://www.nature.com/nrd/journal/v10/n9/full/nrd3546.htmlhttp://www.nature.com/nrd/journal/v10/n9/full/nrd3546.htmlhttp://www.nature.com/nrd/journal/v10/n9/full/nrd3546.htmlhttp://www.nature.com/nrd/journal/v10/n9/full/nrd3546.htmlhttp://www.nature.com/nrd/journal/v10/n9/full/nrd3546.htmlhttp://www.nature.com/nrd/journal/v10/n9/full/nrd3546.htmlhttp://www.nature.com/nrd/journal/v11/n6/full/nrd3744.htmlhttp://www.nature.com/nrd/journal/v11/n6/full/nrd3744.htmlhttp://www.imi.europa.eu/content/stage-1-4


4/77

Epizyme and Celgene have announced a

partnership on the discovery, development

and commercialization of inhibitors of histone

methyltransferases (HMTs), an emerging family

of epigenetic drug targets, for genetically

defined cancers. Celgene, which markets

two of the four epigenetic drugs currently

approved the histone deacetylase (HDAC)

inhibitor romidepsin (Istodax) and the DNA

methyltransferase inhibitor azacitidine

(Vidaza) gains exclusive rights outside the

United States for Epizymes preclinical DOT1LHMT inhibitor programme against mixed

lineage leukaemia (MLL). The deal, which

involves an upfront payment of US$90 million

to Epizyme, also provides Celgene with

options to license ex-US rights to other

Epizyme HMT inhibitor programmes, through

which Epizyme could earn up to $160 million

in milestone payments.

Aberrant epigenetic changes resulting

in gene dysregulation, such as oncogene

overexpression or tumour suppressor

silencing, can cause cancer. Taiping Chen,

Associate Professor at the MD AndersonCancer Center, Texas, USA, explains that:

Unlike irreversible genetic mutations,

epigenetic mutations are potentially

reversible, which makes epigenetic therapy

an attractive approach for cancer treatment.

Histone-modifying enzymes, including

HDACs and HMTs, have attracted much

interest in the context of anticancer drugs in

recent years and, so far, two HDAC inhibitors

are approved by the US Food and Drug

Administration (FDA).

Compared to HDAC inhibitors, the

development of HMT inhibitors is much less

advanced, but growing evidence implicates

several HMTs as important drug targets

in various cancers, says Yongcheng Song,

Assistant Professor at Baylor College of

Medicine, Texas, USA. One example is

DOT1L, an HMT that specifically methylateshistone 3 lysine 79 (H3K79), which is a target

for acute leukaemias with translocations of

the MLLgene. Several predominant fusion

partners of onco-MLL have been shown to

recruit DOT1L, and H3K79 hypermethylation

was also found to be a hallmark of

MLL-rearranged leukaemias.

Epizymes DOT1L inhibitor EPZ004777

has been shown to selectively inhibit the

proliferation of MLL-rearranged leukaemia cell

lines and prolong survival in a mouse xenograft

model of MLL (Cancer Cell20, 5365;2011).

However, further development is neededbefore DOT1L inhibitors can become clinically

useful drugs, says Song, in particular to

improve on the poor pharmacokinetic

properties of EPZ004777, which has a very

short plasma half-life. Combination therapy

should also be applied to further develop such

compounds he adds. The rationale is that

non-cytotoxic DOT1L inhibitors, which induce

differentiation and apoptosis of leukaemic

stem cells (or leukaemia-initiating cells), work

slowly on this aggressive disease, and so

combination with a cytotoxic drug that

quickly kills non-stem leukaemia cells, thereby

reducing overall cancer burden, should be

useful. Chen also notes that DOT1L is widely

expressed in human tissues and likely to be

involved in several biological processes, raising

questions about the potential toxicity of

DOT1L inhibitors.With regard to targeting HMTs in

general, a major challenge is how to find

selective inhibitors, given that more than

50 human HMTs use S-adenosylmethionine

(SAM) as a common cofactor, says Song.

However, the development of EPZ004777 and

several other selective DOT1L inhibitors that

target the SAM-binding site has indicated

that ligand-based medicinal chemistry,

guided by structure-based design, can be a

viable approach for the discovery of potent

and selective HMT inhibitors, he adds. Given

that multiple companies are now developingsuch inhibitors, Chen concludes that

targeting HMTs, as well as other emerging

epigenetic modifiers such as lysine

demethylases, will be a promising new

frontier for anticancer drug discovery.

Alexandra Flemming

DEAL WATCH

EpizymeCelgene deal signals interest

in new class of epigenetic drugs

BIOBUSINESS BRIEFS



http://www.sciencedirect.com/science/article/pii/S1535610811002273http://www.sciencedirect.com/science/article/pii/S1535610811002273http://www.sciencedirect.com/science/article/pii/S1535610811002273http://www.sciencedirect.com/science/article/pii/S1535610811002273http://www.sciencedirect.com/science/article/pii/S1535610811002273http://www.sciencedirect.com/science/article/pii/S1535610811002273http://www.sciencedirect.com/science/article/pii/S1535610811002273


5/77

EDITORIAL OFFICELondon

[email protected] Macmillan Building, 4 Crinan Street,London N1 9XW, UKTel: +44 (0)20 7843 3620;Fax: +44 (0)20 7843 3629

To subscribe and for more detailed information visitwww.nature.com/reviews/drugdisc

chief editor:Peter Kirkpatrickenior editor:Alexandra Flemming,Charlotte Harrison, Sarah Crunkhorn,Monica Hoyos Flightenior new editor:Asher MullardAitAnt editor:Man Tsuey Tsecopy editor:Mariam Faruqienior copy editor:Catriona Rodwell,Lucie Wootton, Isabel Woodmancopy editing MAnAger:Lewis PackwoodArt controLLer:Susanne Harrisenior Art editor:Vicky Summersby,Patrick Morgan, Kirsten Lee

MAnAging production editor: Judith Shadwellenior production editor:Simon Fenwickproduction controLLer:Natalie Smithenior editoriAL AitAnt:Laura CornseditoriAL AitAnt:Ella Linesweb production MAnAger:Dipti ShahMArketing MAnAger:Tim Redding, Nazly De La RosapubLihing director:Peter Collins

new [email protected] Publishing Group, 75 Varick Street, 9th floor,New York, NY 100131917, USATel: +1 212 726 9200;Fax: +1 212 696 9006

pubLiher (biophArMA):Melanie Brazil

cutoMer ervice: [email protected]

Copyright 2012 Nature Publishing GroupPrinted in Wales by Cambrian Printers on acid-free paper.

peter kirkpAtrick ALexAndrA fLeMMing chArLotte hArrion

MAn tuey teAher MuLLArdArAh crunkhorn

EDITORS

Over the past 25 years, monoclonal antibody-based therapeutics

have become established as effective medicines for several

serious diseases. However, it is only recently that antibodies

have also emerged as valuable tools to inform small-molecule

drug discovery. In a Perspective article, Lawson discusses the application of

antibodies in the validation of targets and design of screening assays, as well as

their potential to aid the identification of modulators of traditionally intractable

targets, particularly proteinprotein interactions. The patents of several

biological drugs, including leading monoclonal antibodies, are due to expire soon,

providing opportunities to develop biosimilar versions. Yet, the complexity of such

biopharmaceuticals raises novel challenges for the development and regulatory

evaluation of biosimilars. In their Review, Jones and colleagues discuss key issues

in assessing the comparability of biosimilars to pioneering biologics and describe

analytical technologies that may be used to measure characteristics such as

post-translational modifications that regulatory authorities have identified as being

important in such comparisons. Monoclonal antibodies have a range of therapeutic

applications in ocular disorders, particularly those targeting angiogenic factors

in the treatment of age-related macular degeneration, diabetic macular oedema

and retinal vein-occlusive diseases. Focusing on retinal diseases and glaucoma,

Zhang and colleagues provide an overview of disease pathogenesis, highlighting

recent innovations in ophthalmic drug discovery and delivery. Finally, Pratt and

colleagues consider the limitations of current schizophrenia therapies and suggest

that the lack of suitable preclinical models may be a key factor hampering

the development of novel treatments. They propose how the use of new rodent

models, in conjunction with translationally relevant end-point assessments,

could increase the chances of success in schizophrenia drug discovery.

Animal models of schizophrenia p560

Analysing biosimilars p527

Getty/LizZador

Getty/RussellThurston


IN THIS ISSUE



6/77

Malini Guha

Although the field of anticancer

immunotherapy has surged inrecent years with the approvals

of Dendreons cancer vaccinesipuleucel-T for advanced prostatecancer and Bristol-Myers Squibbs

ipilimumab for metastatic melanoma,one class of agents that haslanguished is the Toll-like receptor

(TLR) agonists. Once included inthe US National Cancer Instituteslist of immunotherapeutic agents

with the highest potential to treatcancer, TLR agonists have repeatedlydisappointed as systemic drugs.

Most recently, in May, IderaPharmaceuticals top-line Phase IIresults of the oligonucleotide-based

TLR9 agonist IMO-2055 showedthat the drug did not improveprogression-free survival in

second-line recurrent or metastatichead and neck cancer. Merck KGaAhad been partnered with Idera on

this project but stepped out last

year, and the drugs future is nowuncertain. The most high-profilefailure, meanwhile, was Pfizers 2007

discontinuation of its oligonucleo-tide-based TLR9 agonist CPG 7909

(also known as ProMune) after thedrug failed to demonstrate efficacywhen used in combination with

chemotherapy in Phase III trials innon-small-cell lung cancer (NSCLC).The fact that TLRs are expressed by

some cancers, and that TLR agonistshave been found to sometimesdrive tumourigenesis in preclinical

models although this has not beenobserved directly in the clinic hasnt helped matters (Nature Rev.

Cancer9, 5763; 2009).Yet some experts believe that

the many failures should not be

attributed to the agents themselves.They werent used in the right way,says Nina Bhardwaj of the NYU

Langone Medical Center in NewYork, USA. Although investigatorsagree that TLR agonists are unlikely

to prove themselves in most cancers

as monotherapies, they argue thatthe agents still offer promise if theyare better combined with anticancer

drugs that kill cancer cells andthereby release tumour antigens

that stimulate an immune response,or if they are used as therapeuticcancer vaccine adjuvants. The future

is a multipronged approach, saysBhardwaj.

Both the vaccine and

chemotherapy approaches will havean important role, and will depend onthe specific clinical context, says Rob

Hershberg, chief executive of VentiRx.Yet opinions diverge on whichcancer antigens, chemotherapies and

targeted drugs to test the agonistswith. Because members of the TLRfamily each trigger the release of a

different spectrum of inflammatorycytokines, there is also uncertaintyover which targets to agonize to evoke

the most potent antitumour immuneresponses. Other outstandingquestions include whether the

agonists need to be combined with

Anticancer TLR agonists on the ropesToll-like receptor agonists have hit another setback with the Phase II failure of Ideras IMO-2055,but these immunotherapies may still make a comeback if appropriate combinations with vaccineantigens or anticancer drugs can be identified.

NEWS & ANALYSIS


http://www.nature.com/nrc/journal/v9/n1/full/nrc2541.htmlhttp://www.nature.com/nrc/journal/v9/n1/full/nrc2541.htmlhttp://www.nature.com/nrc/journal/v9/n1/full/nrc2541.htmlhttp://www.nature.com/nrc/journal/v9/n1/full/nrc2541.htmlhttp://www.nature.com/nrc/journal/v9/n1/full/nrc2541.htmlhttp://www.nature.com/nrc/journal/v9/n1/full/nrc2541.htmlhttp://www.nature.com/nrc/journal/v9/n1/full/nrc2541.html


7/77

agents that counter the immunosuppressive

tactics of cancer and whether the fast-growingmetastatic tumours that most novel anticancerdrugs are initially investigated in are an

appropriate testing ground. There are manyunresolved questions; we have a lot of work todo, says Bhardwaj.

Vaccine or drug combo therapy?

In the 1890s, surgical oncologist William

Coley noticed spontaneous tumourregressions in some cancer patients who hadcontracted bacterial infections, and speculated

that bacteria somehow activate antitumourimmune responses. TLRs which triggeran immune response upon detection

of pathogens were later linked to theantitumour effect of Coleys toxin, and thebacille CalmetteGurin (BCG) tuberculosis

vaccine, which contains a bacterium thatactivates multiple TLRs, became the first

therapeutic to take advantage of this activitywhen used for localized bladder cancer.

Yet although there have been a couple ofsuccess stories in localized cancer BCG and

the topical TLR7 agonist imiquimod for basalcell skin tumours the class has failed in thetreatment of metastatic disease.

In most of the failures, the TLR agonistswere being tested in combination withchemotherapy or targeted anticancer agents.

Some experts consequently argue that TLRagonists are better suited as therapeutic cancer

vaccine adjuvants, in which the agonists are

used to activate the immune systems dendriticcells (DCs) so that they are better primed torecognize the specific tumour antigens that

make up the core of the vaccines.Indeed, the only experimental TLR agonist

currently in Phase III trials is being usedin a vaccine. GlaxoSmithKlines melanoma-associated antigen 3 (MAGEA3)-specific

cancer immunotherapy (astuprotimut-R)combines a MAGEA3 antigen with anadjuvant system that includes the TLR9

agonist VaxImmune, which is the same TLR9agonist as Pfizers unsuccessful CPG 7909, andthe TLR4 agonist MPL, which is also used

in GlaxoSmithKlines approved Cervarixvaccine. Two large Phase III trials are testingthe vaccine in resected NSCLC and melanoma

in patients whose tumours express MAGEA3,with the aim of reducing disease recurrence.Pivotal data are expected in 2013. MPL is also

being used as an adjuvant in Merck KGaA/Oncothyreons Stimuvax vaccine, whichincludes the MUC-1 antigen and is in a

Phase III trials in unresectable NSCLC.Proponents of the vaccine approach

note that by using a single antigen, such as

MAGEA3, they can develop therapies with a

high specificity for tumour cells. Critics point

to the risk that resistance may arise to a singleantigen vaccine and add that not all patientswith a given cancer usually express the same

antigens (about one-third of NSCLC patientsand two-thirds of melanoma patients havetumours that express MAGEA3, for example).

Chemotherapies and targeted agents,meanwhile, may result in a greater release ofdifferent cancer antigens when they kill cells,

leading others to argue that these should beused to prime DCs that have been activated byTLR agonists. The fact that this combination

approach has not worked so far has notdeterred a few groups from carrying on.They cite other reasons for the failures,

including inappropriate use of chemotherapiesthat were too toxic in combination with theTLR agonist or too immunosuppressive,

or doses of the TLR agonist that were too low.Mologens chief executive Matthias Schroff

says, for example, that its DNA-based TLR9agonist MGN1703 can be given at a dose tentimes higher than the dose of CPG 7909 usedin Pfizers NSCLC trials. In May, Mologen

reported potentially promising top-line resultsof a randomized 55-patient Phase II trial ofMGN1703 in metastatic colorectal cancer.

Bhardwaj also argues that whereas TLRagonists have been delivered subcutaneouslywhen used in combination with standard

anticancer drugs, they would fare better ifdelivered locally. Although local delivery canbe difficult or impossible, systemic delivery

is like trying to stimulate a general immuneresponse throughout the body, hoping thatsomething will happen locally, she says.

Others testing systemic administration say thattheir strategy should work, however, because

activated immune cells circulate throughoutthe bloodstream and will eventually reach thetumour site.

Another possible reason for past failures,says Hershberg, has been the focus on TLR9and TLR7 agonists. TLR9 agonists were

pursued initially because they demonstratedantitumour activity in various preclinicalmodels, whereas TLR7 agonists were pursued

because of imiquimods success (the drugwas found to act through TLR7 after itsclinical efficacy was demonstrated). But, he

says, [TLR9 and TLR7] are not as potent atdriving a T cell response as TLR8 agonists,pointing to cell culture data showing that

VentiRxs small-molecule TLR8 agonistVTX-2337 induces a stronger activation ofthe cytokine interleukin-12 (IL-12) than do

imiquimod and CPG 7909. A 200-patientrandomized Phase II study of VTX-2337plus doxorubicin chemotherapy in metastatic

ovarian cancer is planned for later this year.

Cleveland BioLabs is similarly steering

clear of the pack. Its candidate CBLB502,a recombinant derivative of the bacterialprotein flagellin, acts on TLR5 and is in

Phase I development for solid tumours.The same question of which TLR to

target also plagues the therapeutic vaccine

field. Although it is difficult to compare thedifferent TLR agonists in randomized clinicaltrials, the Cancer Vaccine Collaborative

a joint programme of the Cancer ResearchInstitute and the Ludwig Institute for CancerResearch has conducted a series of

parallel, early-stage trials evaluating TLR3,TLR4, TLR7 and TLR9 agonists as vaccineadjuvants in combination with the cancer

testis antigen NY-ESO-1. It found that theTLR3 agonist poly-ICLC elicited the mostfavourable immune response, as assessed by

antigen-specific antibody levels as well as CD4and CD8 T cell responses. It is now planning

to incorporate the agent into 1015 moreacademic clinical trials.

Tackling immunosuppression

Ruslan Medzhitov, from the Yale School ofMedicine in New Haven, Connecticut, USA,provides a further possible explanation for

the disappointment of TLR agonists: theTLR system evolved primarily as a defenceagainst outside invaders rather than against

cancers, and cancers take advantage of theconserved mechanisms that normally preventautoimmune responses to self-antigens.

We need to figure out how to get around thislimitation with immunotherapy, he says.The immune system is not wired to fight

against cancer, as it is in the case of infection.He and many others believe, therefore,

that drug developers need to tackle tumourimmunosuppression when using TLRagonists. Not only does the tumour secrete

immunosuppressive factors and activateregulatory immune system cells, but TLRagonists themselves seem to induce a negative

feedback mechanism that probably limitsthe uncontrolled inflammation that couldotherwise arise.

Several groups are experimenting alongthese lines. Hailing Lu, from the Universityof Washington, Seattle, USA, showed in

a preclinical study that treatment withtopical imiquimod induced high levels ofthe anti-inflammatory cytokine IL-10, and

blockade of IL-10 enhanced the antitumoureffect of imiquimod (J. Immunol.184,53605367; 2010). She is now hoping to use the

same combination in a Phase I trial in breastcancer patients with chest wall metastases.Similarly, TriMod Therapeutics plans to

initiate a clinical trial of a TLR agonist and a



http://www.jimmunol.org/content/184/9/5360.abstracthttp://www.jimmunol.org/content/184/9/5360.abstracthttp://www.jimmunol.org/content/184/9/5360.abstracthttp://www.jimmunol.org/content/184/9/5360.abstracthttp://www.jimmunol.org/content/184/9/5360.abstracthttp://www.jimmunol.org/content/184/9/5360.abstracthttp://www.jimmunol.org/content/184/9/5360.abstract


8/77

phosphoinositide 3-kinase (PI3K) inhibitor

after preclinical studies showed that thiscombination suppressed anti-inflammatoryIL-10 and regulatory transforming growth

factor (TGF) activity induced by the TLR5agonist flagellin (Cancer Res. 72, 581591;2012). Others are testing TLR agonists

preclinically with cytotoxic T lymphocyteantigen 4 (CTLA)-targeting drugs as well.The potential danger in blocking

immunoregulatory mechanisms, of course,is the risk of excessive immune reactions.Many people are thinking how to make these

therapies more targeted so that we get anoverwhelming inflammatory response only atthe tumour site and not in the whole body, but

so far we dont know how to do this, says Lu.Some in the field of cancer

immunotherapy have also come to believe

that rapidly progressing or bulky metastatic

disease may not be the best setting totest immunotherapies, as numerousimmunosuppressive mechanisms are likely

to be in play in these tumours, and inducedimmune responses take time to becomeeffective. But trials in the adjuvant setting that

are designed to prevent disease recurrence(as is being tested with GlaxoSmithKlinesMAGEA3 vaccine) are long, expensive and out

of reach for many small biotech companies.Within metastatic disease, fortunately, there

are different settings, and some of these might

be more amenable to treatment with TLRagonists than others. Mologens MGN1703,for example, was tested as a maintenance

therapy in patients with colorectal cancerwhose tumours were already stabilized orresponded to standard therapy.



http://cancerres.aacrjournals.org/content/72/3/581.abstracthttp://cancerres.aacrjournals.org/content/72/3/581.abstracthttp://cancerres.aacrjournals.org/content/72/3/581.abstracthttp://cancerres.aacrjournals.org/content/72/3/581.abstracthttp://cancerres.aacrjournals.org/content/72/3/581.abstracthttp://cancerres.aacrjournals.org/content/72/3/581.abstract


9/77

Asher Mullard

US funders are spending US$20 million to put

58 abandoned therapeutics, contributed by

8 drug developers, into the hands of academic

drug hunters. The drug repurposing pilot

scheme which grew out of a meeting last

year between US National Institutes of Health

(NIH) officials, industry and academic

researchers (Nature Rev. Drug Discov.10,

399400; 2011) follows closely on the heels

of a similar venture in the United Kingdom. InDecember last year, the UK Medical Research

Council (MRC) pitched in UK10 million ($15

million) to provide UK researchers with the

means to study 22 compounds that have been

de-prioritized by AstraZeneca.

Through the open-innovation schemes,

drug companies will pull previously shelved

compounds from out of their freezers, and the

US and UK funders will pay for further clinical

research in new indications. For the NIHs

scheme, all compounds have been tested in

humans and so come with extensive safety

and pharmacokinetic data that are nottypically made public but have been

discontinued owing to either lack of efficacy

in their initial indication or strategic

refocusing (the asset list is available at

go.nature.com/5sXoVG). The compounds are

being contributed by Abbott, AstraZeneca,

Bristol-Myers Squibb, Lilly, GlaxoSmithKline,

Johnson & Johnson, Pfizer and Sanofi.

AstraZeneca is sharing some of the same

compounds with the MRC and the NIH,

although the MRCs experiment also includes

a subset of 12 compounds that are currently

only deemed appropriate for preclinical

studies (a list of the MRCAstraZeneca assetsis available at go.nature.com/kNSZYs).

Although the contributed compounds

span therapeutic areas, the US programme

in particular is heavily loaded with central

nervous system drugs(FIG. 1).Several

companies have independently donated

compounds from once-promising psychiatric

drug classes including histamine receptor

antagonists, nicotinic acetylcholine

receptor agonists and 5-hydroxytryptamine

receptor modulators underscoring a

widespread disinvestment in this space

in recent years. And these are not just clear

reject compounds. In the case of the

histamine receptors, Rob Leurs, a medicinal

chemist at the University of Amsterdam,

the Netherlands, and expert in the class,

comments that the contributed drugs could

unlock the enormous clinical promise that

new histaminergic compounds once held.

The US asset list is mostly made up of orally

available small molecules, but also includes

one monoclonal antibody (AstraZenecas

MEDI2338) and one recombinant DNAplasmid (Sanofis XRP0038).

There are plenty of potential uses that

come to mind for most of these that have

at least a reasonable chance of success,

says Bryan Roth, a pharmacologist at the

University of North Carolina, Chapel Hill,

USA.

Under the terms of both crowd-sourcing

programmes, pharma companies will have an

early opportunity to vet incoming ideas that will

be proposed by academics and decide which

theyd like to collaborate on, but the funders will

independently and ultimately decide whichprojects to back. The NIH, via its new National

Center for Advancing Translational Sciences

(NCATS), hopes to fund around eight projects

that have the potential to provide Phase I/II

proof-of-principle data, although it could

support more if individual centres at the NIH

also decide to pitch in. The MRC, which has

already whittled a list of over 100 initial

proposals down to a shortlist of 25 since

announcing their programme last year, plans to

pay for 1015 preclinical and clinical projects.

Responses to the programmes span the

spectrum. I absolutely applaud this pot

of funding, says Aris Persidis, CEO ofrepurposing biotech Biovista. What is good

for repositioning in one arena is good for

everybody in the field, he adds. We might

as well give it a try and see, notes Roth.

$20 million is really cheap as far as these

things go. Others are more sceptical,

with Pfizers former head of research and

development (R&D) John LaMattina among

others arguing that repurposing is already

widely done in-house throughout industry,

and questioning the focus on applied over

basic research (go.nature.com/4TOWv8).

Thinking big

Funders provide several defences to their

critics. For one, argues Kathy Hudson, acting

deputy director of the NCATS, the NIH project

is about more than delivering new drugs to

patients. First and foremost we are pilot

testing a model for multistakeholder

collaboration, she says. Second, we are

hoping that fantastic science will come

out about disease mechanisms.

Chris Watkins, head of translational

research at the MRC, sees the UK project

similarly. Our primary intention is not to

repurpose these molecules; it is to use them to

increase our understanding of mechanisms

of human disease and novel therapeutic

opportunities, he says. As a result, he adds,

individual projects might validate a target for

a specific disease and are therefore more akin

to fundamental discovery science than to

applied research. You can do extremely

innovative clinical research that is discoveryscience, he says.

When confronted with more pointed

questions about why academia is likely to

succeed in its attempts to reposition drugs

after pharma scientists have failed, both point

to the programmes unique ability to tap

into the expertise and knowledge base of

thousands of researchers from across

disciplines, therapeutic areas and biases.

So does Don Frail, a vice president at

AstraZeneca and architect of several

repurposing projects. This concept of crowd

sourcing has worked really well in our earlyexperience with the MRC, he says, adding

that all but one of the 22 MRCAstraZeneca

assets received at least one proposal.

Given the broad goals and early days of the

schemes, its perhaps no surprise that even

some supporters have quibbles about the size,

scope and details of the programmes.

For one, the schemes are focused on on-target

repurposing, whereby drugs are retested on

the basis of the activity they were designed

to possess rather than on the basis of

unanticipated activity. When researchers can

make solid arguments that a specific off-target

activity is likely for example, because of highstructural similarity between the primary target

and a protein of interest they can apply for

funding. But, laments Roth, investigators

who want to hunt further afield for activity

with exploratory target-based, phenotypic

or computational screens are out of luck.

It would have been great to be able to screen

these compounds, says Roth.

Frail defends the screening exclusions

of both programmes by arguing that the

discoveries of therapeutically relevant

off-target activities are low-probability

Drug repurposingprogrammes get lift offWhat will schemes designed to get industrys discarded drug candidates into the

hands of academic researchers deliver?



http://www.nature.com/nrd/journal/v10/n6/full/nrd3473.htmlhttp://www.nature.com/nrd/journal/v10/n6/full/nrd3473.htmlhttp://www.nature.com/nrd/journal/v10/n6/full/nrd3473.htmlhttp://www.nature.com/nrd/journal/v10/n6/full/nrd3473.htmlhttp://www.nature.com/nrd/journal/v10/n6/full/nrd3473.htmlhttp://go.nature.com/5sXoVGhttp://go.nature.com/kNSZYshttp://go.nature.com/4TOWv8http://go.nature.com/4TOWv8http://go.nature.com/4TOWv8http://go.nature.com/kNSZYshttp://go.nature.com/5sXoVGhttp://www.nature.com/nrd/journal/v10/n6/full/nrd3473.htmlhttp://www.nature.com/nrd/journal/v10/n6/full/nrd3473.html


10/77

events in his experience, especially in

the current era of highly optimized drug

candidates. But commercial concerns may

also be at play. Exploration of off-target

pharmacology is more likely to require the

development of new drug doses, formulations

and routes of administration, points out

Persidis, and this can lead to the creation

of new intellectual property (IP) that can

undermine the originators control of a

project. It is essentially impossible to develop

a drug with such extreme specificity that it

will not have some kind of off-target activity,

and the network-pharmacology that renders

each drug unique and also gives all of us

opportunities, he adds.

Another possible sticking point is whether

patent arrangements will get in the way of

delivering new drugs. Part of the need for the

NIH programme, says Hudson, is that it can

take longer to negotiate a deal between

individual investigators and a drug company

than it would take to do the research that

drives the deal. Getting all parties to agree

upfront to boilerplate agreements which

give new use IP to the researchers while

keeping composition of matter IP and a first

chance to license new IP in the hands of the

respective contributing drug firms certainly

saves time. But the chance that successful

findings will eventually make it to patients,

says Persidis, may now hinge on the reversionrights that let the investigators license their

new IP to third parties, and on whether

contributors may block further development

plans if they foresee commercial complications

on the horizon.

Hudson notes that investigators will have

reversion rights, but the speed with which they

will be able to take their new use IP to other

parties remains unclear. Potential licensees

would probably also need to license

the composition of matter IP from the

originators, she adds.

Given the historic success rate of drugdiscovery, adds Hudson, patent issues may be

the least of anyones concerns in terms of

successfully steering the retested compounds

towards approval. We have leapfrogged over

6 or 7 years of preclinical and early-stage

research and $30 million or so of investment

with these compounds thats where the time

saving is. But moving from here to the next

stage will have the same failure rate as

anybody else, she says. With the probability

that a compound will make it to market from

Phase I and Phase II at around 5% and 11%,

respectively (Nature Rev. Drug Discov.10,

328329; 2011), a scenario in which even one

of the rescued compounds will ultimately end

up in a medicine cabinet is not guaranteed.

From pilot to permanent?

A further counterargument to criticisms big and

small is that both the NIH and the MRC see their

ventures as pilot programmes. The scope of

the solicited projects, the number of candidates

made accessible, the language in the

pre-negotiated template agreements and more

can be reassessed once the pilots wrap up, and

amended if everyone is still keen to continue.

With one eye already on such a future,

Watkins says the MRC is already in discussions

with a number of companies that would like

to join in on the UK venture.

Although obvious ways to expand the

project are to get more companies to

contribute a bigger batch of failed drugs andto allow compound screening for potentially

relevant off-target activity, a more ambitious

approach might be to get companies to

contribute their active pipeline candidates as

well. The belief that a particular set of failed

compounds holds promise above and beyond

all manner of other compounds, including

successful compounds, simply fails to pure

logic, says Persidis. In 2010, Frail helped Pfizer

and Washington University to strike a deal that

provides researchers with data and access to

over 500 compounds from across stages of the

pipeline which was essentially a blueprintfor the new NIH and MRC schemes

suggesting that it may eventually be possible

to overcome the legal, regulatory and

commercial barriers that led the NIH and MRC

programmes to focus initially on repurposing

failed drugs.

I agree that the programme has to start

from somewhere, but an over-reliance and

over-representation of failed compounds is

technically limiting and is a disservice to the

possibility of finding new cures, says Persidis.

Figure 1 | Breakdown of the US NationalInstitute of Healths repurposing assets by

therapeutic area. Eight pharmaceuticalcompanies contributed 58 compounds that hadpreviously been tested in 90 indications. Centralnervous system (CNS) includes psychiatricindications and neurodegenerative indications;respiratory includes chronic obstructivepulmonary disease, asthma, allergic rhinitis andcough; pain includes both CNS and other typesof pain; metabolic includes diabetes andobesity; autoimmune includes rheumatoidarthritis, multiple sclerosis and Crohns disease;other includes endometriosis, irritable bowelsyndrome, gastroesophageal reflux disease andgenitourinary indications.

Central nervous systemCardiovascularMetabolicAutoimmune disorders

RespiratoryPainOncologyOther

30%

14%

10%

10%

9%

7%

4%

16%



http://www.nature.com/nrd/journal/v10/n5/full/nrd3439.htmlhttp://www.nature.com/nrd/journal/v10/n5/full/nrd3439.htmlhttp://www.nature.com/nrd/journal/v10/n5/full/nrd3439.htmlhttp://www.nature.com/nrd/journal/v10/n5/full/nrd3439.htmlhttp://www.nature.com/nrd/journal/v10/n5/full/nrd3439.htmlhttp://www.nature.com/nrd/journal/v10/n5/full/nrd3439.htmlhttp://www.nature.com/nrd/journal/v10/n5/full/nrd3439.html


11/77

In April,the Chinese State Intellectual Property (IP) Office published

the National IP Strategy for2012, which aimed to enhance the

quality of IP within the country.

But parts of the plan that called for more effective use of

compulsory licences, combined with an announcement in May

that outlined new measures related to compulsory licence

applications, sparked debate as to whether China was introducing

new laws in an effort to begin making low-cost copies of medicines

that are still under patent protection.

But in reality, China as well as many other countries

already has such compulsory licensing laws in place, which allow

governments to override patents in a national emergency or cases

of extreme urgency.

If you compare existing patent law with the new measureson compulsory licensing, it is apparent that the latter did not add

any material changes, rather the measures just supplemented some

procedural rules, says DannyFriedmann, co-founder at KongHai

Consultancy, Shenzhen, China, and author of the IP Dragon blog.

Compulsory licenses are compliant with international laws (the

World Trade Organizations trade-related aspects of intellectual

property rights; TRIPS). For example, India recently issued a

compulsory licence for the anticancer drug sorafenib, and Brazil and

Thailand have issued licences for the HIV drug efavirenz. Compulsory

licensing is also in the patent legislation of high-income countries

such as Japan, Germany, the United States and Canada.

Chinas National IP Strategy also aims to improve the IP appraisal

and assessment system, and to guide inventors to shift the focus from

patent quantity to quality. But Friedmann notes that the impact of the

plan could be limited by a lack of implementation. In general, Chinas

IP laws have already a lot of strength on paper. The problem andchallenge of most companies that are active in China is that these

laws are insufficiently enforced on the ground.

Two routes to patent challengecan give different results

There are two ways of challenging a patentin the United States: through administrativere-examination proceedings at the US Patent

and Trademark Office (PTO), or throughlitigation in the courts. A dispute relatedto a medical device patent has highlighted

that sometimes these two routes can yield

different results.Prior to the current case, the patent

owned by Baxter International was

found to be invalid when re-examined at thePTO but valid when assessed in concurrentproceedings by the US Court of Appeals for

the Federal Circuit (CAFC).Because of this discrepancy, Baxterappealed to the CAFC, asserting that

the PTOs decision was wrong. Perhapssurprisingly, in the current case the CAFC

said that Baxters patent was indeed invalid seemingly going against its own previousdecision. But it said that the distinction arose

because of differing evidence before the twoappeals; namely that the PTO used certainprior art to hold that the patent was invalid,

yet this prior art had not been raised in courtproceedings.

According to Jake Holdreith, an

attorney at Robins, Kaplan, Miller & Ciresi,Minneapolis, Minnesota, USA: There isno question that ex partere-examinations

[the type of re-examination carried out at thePTO in the current case] create headaches forinventors and patent owners who are trying

to enforce their patents in court. For thatreason, it is a common strategy for an accusedinfringer to bring an ex partere-examination.

However, some comfort for parties involvedin patent validity challenges will come fromthe fact that the CAFC noted that if the same

evidence had been presented to both the courts

and the PTO, then these two bodies would

have arrivedat the same result.But this case might also have detrimental

consequences for patentees. One thing the

In reBaxter decision may do is encourageaccused infringers to draw out the courtproceeding and hang in as long as possible

to try to get to a [validity] result from thePTO, rather than, for example, reaching asettlement, concludes Holdreith.

In ReBaxter: http://www.cafc.uscourts.gov/images/stories/opinions-orders/11-1073.pdf

23andMe patent creates a stir

The announcement that consumer geneticstesting company 23andMe had been issued

with a patent (US 8187811) related tohuman polymorphisms associated withParkinsons disease raised alarm among

several of the companys customers that theirgenetic data was used to create the patents(see The Spittoon website).

According to Pascal Borry, professor ofbioethics at the University of Leuven, Belgium,the use of customers personal data for

developing patented methods is blurring theline between consumers and research subjects.There is nothing wrong [with 23andMe]

using the direct-to-consumer genetic testingmodel for creating business revenues, but thisconflicts with the transparent, science-driven

image 23andMe has been creating of its

PATENT WATCH

China aims to enhance IP laws MACMILLAN

BRANDX



http://www.cafc.uscourts.gov/images/stories/opinions-orders/11-1073.pdfhttp://www.cafc.uscourts.gov/images/stories/opinions-orders/11-1073.pdfhttp://spittoon.23andme.com/news/announcing-23andmes-first-patent/http://spittoon.23andme.com/news/announcing-23andmes-first-patent/http://www.cafc.uscourts.gov/images/stories/opinions-orders/11-1073.pdfhttp://www.cafc.uscourts.gov/images/stories/opinions-orders/11-1073.pdf


12/77

research objectives. Consumers have been

mislead in that regard, and saying that[permission] was in the informed consent[that customers signed] is not sufficient.

The patent which claims methods ofscreening individuals for susceptibility toParkinsons disease was filed before the

high-profile Prometheus case, which castdoubt on the patentability of diagnosticmethods (see Nature Rev. Drug Discov. 11,

344; 2012), so it remains to be seen whetherthe 23andMe patent will survive anypotential subsequent validity challenges.

Charlotte Harrison

PATENT ADVISORS

Daniel M. Becker: Dechert, Mountain View, CA, USA.Luke Kempton: Wragge & Co., London, UK.Leslie Meyer-Leon: IP Legal Strategies, Boston, MA, USA.George W. Schlich: Schlich & Co., London, UK.John A. Tessensohn: Shusaku Yamamoto, Osaka, Japan.Philip Webber: Dehns, London, UK.

Toll-like receptor agonists andcancer

Toll-like receptors (TLRs) which recognizeconserved microbial signature molecules have a central role in the innate immune

response, suggesting that they could betargeted in cancer therapy. In the articleon p503 we highlight that although some

TLR agonists have produced disappointingresults in clinical trials, these therapies couldstill hold promise if they are appropriately

combined with anticancer drugs thatstimulate an immune response, or used asadjuvants for therapeutic vaccines. Here in

TABLE 1we summarize international patentapplications published in the past 18 monthsrelated to TLR agonists and cancer. Data were

researched using the SureChemdatabasefrom Macmillan publishers.

Table 1 | Toll-like receptor agonists and cancer

Patents Assignee Subject

WO 2011022508WO 2011022509

Array BioPharma Substituted benzodiazepines that act as TLR7 and/or TLR8 modulators; useful fortreating cancer, autoimmune diseases or infectious disorders

WO 2012066336 AstraZeneca Benzylamine compounds that act as TLR7 agonists and are selective over TLR8 andHERG; useful for treating allergic diseases, autoimmune diseases, viral diseases andespecially cancer

WO 2012066336 AstraZeneca Novel TLR7 agonists; useful for treating cancer

WO 2012021834 Baylor Research Institute Novel vaccine adjuvants that consist of a dendritic cell-specific antibody conjugated toa TLR agonist; effective in stimulating the immune response

WO 2011044246 Cleveland BioLabs Methods of treating cancer that involve using TLR5 in combination with a TLR agonist

such as flagellin, which recruits cells involved in both the innate and adaptive immuneresponse to kill cancer cells

WO 2011068233 Dainippon Sumitomo Pharma Imidazoquinoline compounds that have immunomodulating properties and act via TLR7;useful for the treatment of viral or allergic diseases and cancer

WO 2011119759 Novartis Cysteine-based lipopeptides that act as TLR2 agonists; useful as vaccine adjuvants

WO 2011031965 Gilead Novel modulators of TLRs; useful for treatingmelanoma, non-small-cell lungcarcinoma, hepatocellular carcinoma, basal cell carcinoma, renal cell carcinomaand myeloma

WO 2011059505 Hemispherx Biopharma Novel double-stranded ribonucleic acids with rugged physicochemical structures thatbind and activate TLR3 with high specificity

WO 2012027017 Idera Pharmaceuticals Oligonucleotide-based compounds that bind to and activate TLR3

WO 2011139769

WO 2011136828

Johns Hopkins University Immunogenic compositions that consist of neoplastic cells expressing GM-CSF

formulated with at least one TLR agonist; useful for treating neoplasiaWO 2011057148WO 2011049677

Novartis Compounds and immunogenic compositions to treat disorders associated with TLR7;useful as vaccine adjuvants

WO 2011128436 Sapienza University of Rome A combination of a HIF1 inhibitor and a TLR3 agonist; useful as an antitumour agentagainst solid tumours, particularly prostate tumours

WO 2011139348 University of California Uses of phospholipid conjugates of synthetic TLR7 agonists (for example, in vaccines);useful for treating inflammation, cancer and pathogen infection

WO 2012037612 University of Melbourne A method of treating or preventing cancer that uses a TLR2 agonist to stimulate aninnate immune response (but not a humoral or cellular immune response)

WO 2012045090 VentiRx Pharmaceuticals A TLR agonist preferably a TLR8 agonist and its use in combination therapies forthe treatment of cancer

GM-CSF, granulocytemacrophage colony-stimulating factor; HERG, ether-a-go-go-related potassium channel protein, HIF1, hypoxia-inducible factor 1; TLR,Toll-like receptor.

Getty



http://www.nature.com/nrd/journal/v11/n5/full/nrd3735.htmlhttp://www.nature.com/nrd/journal/v11/n5/full/nrd3735.htmlhttp://www.nature.com/nrd/journal/v11/n5/full/nrd3735.htmlhttp://www.nature.com/nrd/journal/v11/n5/full/nrd3735.htmlhttp://www.nature.com/nrd/journal/v11/n5/full/nrd3735.htmlhttps://surechem.com/https://surechem.com/http://www.nature.com/nrd/journal/v11/n5/full/nrd3735.htmlhttp://www.nature.com/nrd/journal/v11/n5/full/nrd3735.html


13/77

The success of some targeted anti-cancer drugs, in particular kinaseinhibitors, is known to be due to a

certain amount of off-target activity(polypharmacology). Since kinasesignalling cascades are complex and

interconnected, it is thought thatsimultaneous inhibition of several

signalling nodes is needed to preventthe rapid emergence of resistance totargeted agents. However, to date,the design of anticancer agents with

favourable polypharmacology hasbeen guided by serendipity ratherthan intent. Now, reporting in Nature,

the groups of Shokat and Caganpresent a systems pharmacologyapproach that allows the identifica-

tion of targets and anti-targets for therational design of drugs with an opti-mized polypharmacological profile.

Combining aspects of target andphenotype-based drug discovery,the authors used a Drosophila melano-

gaster model of multiple endocrineneoplasia type 2 (MEN2), kinase

profiling and genetics in a stepwiseapproach. The MEN2 model is basedon an oncogenic RET transgene

(RETMEN2) targeted to developingepithelia such as wing, eye and legusing thepatchedpromotor, with the

transgene dose calibrated to permit50% survival of D. melanogasterlarvaeto pupariation and 0% survival to

adulthood. Oral administration ofclinical kinase inhibitors such as van-detanib, sunitinib or sorafenib resulted

in different levels of suppression ofRET-driven lethality sorafenib wasmost potent with rescue of about 17%

of animals to adulthood. The authorsthen screened a library of compoundsthat target RET in addition to other

classes of kinases, and they identifiedone compound, AD57, that rescued25% of the animals to adulthood.

It suppressed defects in wing and

eye development and hadan improved efficacyand toxicity profile

compared to clinicalkinase inhibitors.

In vitroexperi-

ments showedthat AD57 also

inhibited theviability of patient-derived cell lines withRETMEN2mutations.

In a mouse xenograft modelof RET-driven tumours, AD57potently suppressed tumour growth

and showed no toxicity, validatingthe D. melanogaster model as auseful tool for the discovery of

compounds with improvedefficacy and toxicity profiles.

Using kinase profiling

and further genetic andpharmacological manipulationin D. melanogaster, AD57 and its

derivatives were analysed with regardsto their kinase inhibition profiles

and the relative contributions of theirtargets to the efficacy and toxicityof the compound. Small alterations of

the structure of AD57 were shown tolead to considerable changes in kinaseselectivity. For example, AD58,

a close analogue of AD57, inducedwhole-animal toxicity, providing anopportunity to explore aspects of AD

class drug toxicity. It was found thatERK and SRC inhibition suppressedoncogenic RET signalling, whereas

increased RAS pathway signalling con-tributed to toxicity. AD58 was found toexert its toxic effects through the inhi-

bition of target of rapamycin (TOR),which was thought to be due to the factthat TOR signalling provides feedback

inhibition of the RAS pathway.These insights were used to

generate close AD57 analogues with

optimized polypharmacological

profiles. AD80 and AD81, in whichTOR binding was abolished, were

shown to rescue 7090% of MEN2Drosophilalarvae, and AD80 had animproved efficacy and toxicity profile

in mammalian MEN2 modelscompared to AD57 and vandetanib.

The authors caution that the true

predictive value of this approachawaits further testing. However,they point out that the development

of drugs with optimized polyphar-macology has distinct advantagesover the related approach of using

drug combinations, as the latter ishampered by the cost and complexityof clinical trials, complex target-

profile interactions and differingpharmacokinetics.

Alexandra Flemming

ORIGINAL RESEARCH PAPER Dar, A. C. et al.Chemical genetic discovery of targets and

anti-targets for cancer polypharmacology.

Nature486, 8084 (2012)

C A N C E R

Of targets and anti-targets

R E S E A R C H H I G H L I G H T S

NATURE REVIEWS |DRUG DISCOVERY VOLUME 11 |JULY 2012

Nature Reviews Drug Discovery|AOP, published online 22 June 2012; doi:10.1038/nrd3781



14/77

Previous studies have identified ade novomicrodeletion in SHANK2

(which encodes a molecular scaf-folding protein enriched in neuronalsynapses) in autism spectrum

disorder (ASD), but the role of thismutation in ASD pathogenesis isunclear. Now, a recent paper inNatureshows that Shank2-mutantmice recapitulate many of thebehavioural phenotypes that are

characteristic of ASD. Moreover,the study suggests that hypofunctionof NMDA receptors (NMDARs)

is the underlying molecular causeof impaired social interaction andshows that this impairment could

be pharmacologically reversed.

To elucidate the link betweenthe de novoSHANK2microdeletion

and ASD, the authors generatedtransgenic mice that carried thesame mutation as the human

microdeletion. These mutant mice(Shank2/mice)showed autistic-likeimpairments in social interaction

and in learning and memory. Theyalso exhibited higher levels of anxiety,repetitive behaviours and hyper-

activity compared to wild-type mice.At the cellular level, Shank2/

mice had normal basal synaptic

transmission at hippocampal SchaffercollateralCA1 pyramidal (SCCA1)synapses, and the postsynaptic mor-phology and numbers of excitatorysynapses were unchanged. However,

synaptic plasticity, as assessed bylong-term potentiation (LTP)

and long-term depression (LTD),was severely impaired. Furtherexperiments showed that this

impairment was due to NMDARhypofunction and associatedimpairment of NMDAR signalling.

To examine whether NMDARhypofunction directly contributesto ASD-like behaviours in Shank2/

mice, the authors tested the effectof NMDAR modulators in thesemice. Intraperitoneal injection of

-cycloserine (a partial NMDARagonist) improved social interactionof Shank2/mice. Furthermore,

intraperitoneal injection of CDPPB

a positive allosteric modulator ofmetabotropic glutamate receptor 5

(mGluR5) that ultimately enhancesNMDAR function restored theimpaired LTP and LTD at SCCA1

synapses without affecting basalsynaptic transmission, and fullynormalized NMDAR signalling.

These CDPPB-treated Shank2/micealso had enhanced improvement insocial interaction compared with

mice treated with -cycloserine.However, other behaviours (for

example, anxiety-like behaviours)

were not improved.Together, these results suggest

a causal link between mutations inSHANK2, reduced NMDAR func-tion and impaired social behaviour.

The authors also suggest that

targeting mGluR5 could be a novelstrategy for treating ASD associated

with impaired NMDAR function.

Man Tsuey Tse

ORIGINAL RESEARCH PAPERWon, H. et al.Autistic-like social behaviour in Shank2-mutantmice improved by restoring NMDA receptorfunction. Nature486, 261265 (2012)

N E U R O D E V E L O P M E N T A L D I S O R D E R S

Exploring the links betweenSHANK2 and autism

PHOTOALTO






15/77

Thyroid hormone (TH) confersseveral beneficial metabolic effects,

including lowering of serum choles-terol and reduction of body fat, butthe therapeutic application of this

hormone has so far been hamperedby deleterious effects on the heart,muscle and bone. There is substantial

interest in the design of TH mimeticsthat may avoid such adverse effectsby exhibiting TH receptor isoform-

or tissue-selectivity. Now, Grueteret al.have shown that the microRNAmiR-208a and its target MED13

(mediator of RNA polymerase IItranscription subunit 13) function inthe heart in the control of metabolic

homeostasis through effects onTH signalling, thus presenting new

opportunities for harnessing thebeneficial effects of TH.

miR-208a a heart-specific

microRNA encoded by an intronof theMYH6gene is associatedwith cardiomyocyte hypertrophy and

fibrosis in response to stress, and itsinhibition improves cardiac functionand survival rates in rodent models

of heart failure. Among the strongestpredicted and validated targets ofmiR-208a is MED13, which is a com-

ponent of the multiprotein mediatorcomplex a transcriptional co-activator that acts as a bridge between

DNA-bound transcription factors

and RNA polymerase. However, untilnow, the functions of MED13 in the

heart had not been explored.The authors used pharmacologi-

cal and genetic tools to modulate the

expression of miR-208a and MED13in the heart. They found that micetreated systemically with a locked

nucleic acid anti-miR-208a wereleaner than control mice and wereresistant to high fat diet (HFD)-

induced obesity, displaying reducedfat mass. This phenotype was not due

to a difference in food consumption

or physical activity but was mediatedby an increase in systemic energy

consumption. In addition, suchmiR-208a inhibition significantlyreduced serum triglyceride and

cholesterol levels while improvingsystemic insulin sensitivity and glu-cose tolerance. Mice transgenically

overexpressing MED13 specificallyin cardiomyocytes (Med13-TG mice)

displayed an analogous metabolicphenotype, being similarly resistant

to HFD-induced obesity and insulinresistance. By contrast, cardiac deletionof MED13 enhanced susceptibility of

mice to HFD-induced obesity whileincreasing levels of triglycerides,cholesterol and glucose.

Next, to further explore the role ofMED13 in energy homeostasis, theauthors performed microarray analy-

sis on hearts from Med13-TG miceand controls. More than 60 genes werefound to be downregulated in the

hearts of the Med13-TG mice, mostof which were involved in metabolismand regulated by various nuclear

receptors. Further studies revealed

that many of these genes were respon-sive to TH, with MED13 specifically

acting at the TH receptor -isoform tomodulate TH-dependent responses.Importantly, there were no increases

in heart rate in these mice, which is acommon consequence of enhancedsystemic TH signalling.

Together, these findings reveal anovel role of the heart in systemicmetabolic control and suggest that

miR-208a inhibition may have appli-cations in the treatment of metabolic

disorders, without the risk of adverse

effects associated with previous THmimetics.

Sarah Crunkhorn

ORIGINAL RESEARCH PAPERGrueter, C. E. et al.A cardiac microRNA governs systemic energyhomeostasis by regulation of MED13. Cell149,671683 (2012)

FURTHER READING Baxter, J. D. & Webb, P.

Thyroid hormone mimetics: potential applicationsin atherosclerosis, obesity and type 2 di abetes.

Nature Rev. Drug Discov. 8, 308320 (2009)

M E T A B O L I C D I S O R D E R S

Heart microRNA provides newroute to metabolic control

BRAND X






16/77

Mutant tumour suppressor p53

proteins are expressed in and regulatetumour biology in many humancancers, and so restoration of p53

function could be a viable anticancerstrategy. Writing in Cancer Cell,Yu and colleagues used data from the

US National Cancer institute (NCI)

drug screen (known as the NCI60screen) to identify a compoundthat restored the structure and

function of one of the most commonp53 mutants.

Because the p53 status of the

cell lines used in the NCI60 screen which determined the IC

50(half-

maximal inhibitory concentration)

values of over 48,000 compoundson 60 cell lines was known, theauthors thought these data could be

used to identify compounds thattarget tumours expressing mutantp53. They first used an in silico

methodology that normalized theIC

50values to statistically define

a good response, and then scored

compounds based on whether they

produced a good response againstmutant p53 while not having a good

response against wild-type p53.Three compounds with high scores

were thiosemicarbazone compounds,

of which two (NSC319725 andNSC319726) were shown to inhibitthe growth of mouse fibroblasts

expressing mutant p53. NSC319726was then assayed in tumour celllines expressing different p53 muta-

tions (in amino acid positions 175,248 or 273 of the DNA-binding

domain). These studies showed that

NSC319726 inhibited cellular growththe most in cells expressing the

p53R175mutant. Further work showedthat the NSC319726-mediated reduc-tion in cell growth was mediated by

apoptosis, which was at least partiallydependent on the presence of thep53R175mutant protein.

Next, the authors investigatedthe mechanism of action ofNSC319726. Using conformation-

specific antibodies, they showed

that NSC319726 could restore the

conformation of mutant p53R175

protein to that of the wild-type

protein. Furthermore, this con-

formational change restored theDNA-binding properties of p53R175and increased levels of p53 target

genes, indicating that NSC319726restored the transcriptional func-tions of p53R175.

Thiosemicarbazone com-pounds are metal ion chelatorsand redox modulators. Indeed, the

authors found that chelation ofzinc ions and modulation of cell-ular redox states had a role in the

NSC319726-mediated inhibition ofcellular growth, and hypothesizedthat NSC319726 could act as a source

of zinc to allow the p53R175 mutant

to adopt a correct conformation.When the authors investigated

the effects of NSC319726 in miceexpressing xenograft tumoursderived from human cell lines

expressing different p53 alleles,NSC319726 (delivered byinjection) inhibited the growth

of p53R175Hxenografts but not ofxenografts expressing other mutantp53 alleles.

These data further highlightthat reactivation of mutant p53

using a small molecule is a potential

anticancer strategy. The authorsnote that, unlike previous studies,

NSC319726 was identified froma screen that used many differentcell lines with diverse genetic back-

grounds, which could better reflectthe tumour heterogeneity observedin the clinic.

Charlotte Harrison

ORIGINAL RESEARCH PAPER Yu, X. et al.Allele-specific p53 mutant reactivation. CancerCell21, 614625 (2012)

A N T I C A N C E R D R U G S

Reactivating p53

PHOTODISC





17/77

ERRATUM

Busy panels recommend Pfizers tofacitinib, Arenas lorcaserin and moreAsher Mullard

Nature Reviews Drug Discovery11, 429 (2012)

In the story Busy panels recommend Pfizers tofacitinib, Arenas lorcaserin and more, the la st sentence was incorrect.Johnson & Johnsons rivaroxaban is under review for acute coronary syndrome, not stroke prevention in atrial fibrillation.

This has been corrected online.




18/77

The clinical and commercial success of biologics suchas monoclonal antibodies and recombinant versions ofendogenous proteins is transforming the pharmaceuti-cal industry. In 2010, worldwide sales of all biologicsapproached the US$100 billion mark1, and by 2015 it isexpected that more than 50% of new drug approvals willbe biologics2, rising to more than 70% by 2025 (REF. 3).As these drugs begin to come off patent, substantialopportunities exist for other companies to make copiesor generic versions of these drugs.

For small-molecule drugs, abbreviated regulatorypathways for the development and introduction ofgeneric versions of the drug (following the expirationof patent protection on the original product) have been

established for more than 25 years. Rather than requiringgeneric versions to undergo the same level of evaluationas the original drug, including clinical trials, abbrevi-ated approval for the same purposes is generally basedon demonstrating that the generic drug is pharmaceu-tically equivalent (that is, it contains the same activeingredient in the same purity, strength, dosage form androute of administration) and bioequivalent (that is, it isabsorbed into the body at a similar rate and extent) tothe original drug4. Consequently, abbreviated approval isconsiderably less expensive to achieve, thus dramaticallylowering the costs of generic drugs. This has led to thewidespread use of generic versions and substantial cost

savings for health-care systems; a recent paper noted thatin 2009 almost 75% of small-molecule drug prescriptionsdispensed in the United States were for generics, and theapproval of a generic drug resulted in average savings of77% of the originalproducts cost within 1 year5.

However, for biologics, establishing a regulatory path-way for the introduction of follow-on versions of theoriginal product (once its patent protection has expired)is much more challenging than for small molecules. Somesimple biologics for example, small peptides such asrecombinant insulin and recombinant human growthhormone can be well characterized by establishedanalytical approaches, which has facilitated the regulatoryapproval of follow-on versions under abbreviated path-

ways (based in part on data from the original drug and inpart on analytical data and limited clinical data in somecases)4; however, many biologics such as monoclonalantibodies and other recombinant therapeutic proteinsare much larger and more complex. For such biologics,the extent to which existing analytical technologies canbe used to support the likelihood of clinical comparabilitybetween a follow-on version and the original product ismuch more limited than for small-molecule drugs, and itis not possible to demonstrate that the two products areabsolutely identical.

Consequently, a key question for the developmentand regulation of follow-on biologics also known as

1Analytical Development,

Biogen Idec, 14 Cambridge

Center, Cambridge,Massachusetts 02142, USA.2Department of Chemistry

& Chemical Biology,

The Barnett Institute of

Chemical & Biological

Analysis, Northeastern

University, 360 Huntington

Ave., Boston,

Massachusetts 02115, USA.3Waters Corporation,

34 Maple Street, Milford,

Massachusetts 01757, USA.

Correspondence to G.B.J.

e-mail:[email protected]

doi:10.1038/nrd3746

Analytical tools for characterizingbiopharmaceuticals and theimplications for biosimilarsSteven A. Berkowitz1, John R. Engen2, Jeffrey R. Mazzeo3and Graham B. Jones2

Abstract | Biologics such as monoclonal antibodies are much more complex thansmall-molecule drugs, which raises challenging questions for the development and regulatory

evaluation o

nature reviews drug discovery (july 2012)

Documents

existing drugs

absence of effective

major drug safety problems

new horizons

benefitrisk profile

benefitrisk balance1

entire life cycle

wediscuss new directions