GIENS WORKSHOPS 2014/CLINICAL

PHARMACOLOGY

Thérapie 2015 Janvier-Février; 70 (1): 11-–19 DOI: 10.2515/therapie/2014231 © 2015 Société Française de Pharmacologie et de Thérapeutique

Translational Research: Precision Medicine, Personalized Medicine, Targeted Therapies: Marketing or Science? Pierre Marquet1, Pierre-Henry Longeray2, Fabrice Barlesi3 and participants of round table N°1 of Giens XXX: Véronique Ameye4, Pascale Augé5, Béatrice Cazeneuve6, Etienne Chatelut7, Isabelle Diaz8,9, Marine Diviné10, Philippe Froguel11,12, Sylvia Goni13, François Gueyffier14, Natalie Hoog-Labouret15, Samia Mourah16,17,18, Michèle Morin-Surroca19, Olivier Perche20, Florent Perin-Dureau21, Martine Pigeon22, Anne Tisseau2 and Céline Verstuyft16,23† 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

UMR 850 INSERM, CHU Limoges, Université de Limoges, Limoges, France Laboratoire Merck Serono SAS, Lyon, France Aix Marseille Université; Assistance Publique – Hôpitaux de Marseille, Service d’Oncologie Multidisciplinaire et Innovations Thérapeutiques, Marseille, France Laboratore Novartis Pharma SAS, Rueil Malmaison, France Inserm-Transfert, Paris, France Laboratoire Lilly France, Neuilly-sur-Seine, France Institut Universitaire de Cancérologie, Toulouse, France LEEM, Paris, France ARIIS, Paris, France Laboratoire Amgen, Neuilly-sur-Seine, France Imperial College, London, Royaume-Uni Institut Pasteur, Lille, France Laboratoire Lundbeck SAS, Issy-les-Moulineaux, France Université Lyon 1, HCL, Lyon, France Institut National du Cancer, Boulogne Billancourt, France Assistance publique – Hôpitaux de Paris, Paris, France Université Paris 7, Paris, France Inserm, Paris, France HAS, Saint Denis la Plaine, France Laboratoire Roche Diagnostics, Meylan, France ANSM, Saint Denis, France CNAMTS, Paris, France Faculté de Médecine Paris-Sud, Le Kremlin Bicêtre, France

Text received December 17th, 2014 ; accepted December 18th, 2014

Keywords: personalized medicine; stratified medicine; biomarkers; companion diagnostic; therapeutic drug monitoring; pharmacogenetics †

Abstract – Personalized medicine is based on: 1) improved clinical or non-clinical methods (including biomarkers) for a more discriminating and precise diagnosis of diseases; 2) targeted therapies of the choice or the best drug for each patient among those available; 3) dose adjustment methods to optimize the benefit-risk ratio of the drugs chosen; 4) biomarkers of efficacy, toxicity, treatment discontinuation, relapse, etc. Unfortunately, it is still too often a theoretical concept because of the lack of convenient diagnostic methods or treatments, particularly of drugs corresponding to each subtype of pathology, hence to each patient. Stratified medicine is a component of personalized medicine employing biomarkers and companion diagnostics to target the patients likely to present the best benefit-risk balance for a given active compound. The concept of targeted therapy, mostly used in cancer treatment, relies on the existence of a defined molecular target, involved or not in the pathological

Articles, analyzes and proposals from the Giens workshops are those of the authors and do not prejudice the proposition of their parent organization.

Article publié par EDP Sciences

12

Marquet et al.

process, and/or on the existence of a biomarker able to identify the target population, which should logically be small as compared to the population presenting the disease considered. Targeted therapies and biomarkers represent important stakes for the pharmaceutical industry, in terms of market access, of return on investment and of image among the prescribers. At the same time, they probably represent only the first generation of products resulting from the combination of clinical, pathophysiological and molecular research, i.e. of translational research. Abbreviations: see end of article.

1. Introduction The participants in the round table N°1 of the Giens meeting 2014 chose as objectives: to define personalized medicine, its avatars (stratified medicine and precision medicine) and its composing parts (targeted therapies, biomarkers and companion diagnostics); to evaluate the contribution of translational research to personalized medicine; to evaluate its status in current and future medicine; to clarify its modalities of registration, reimbursement, post-marketing benefit-risk and medico-economical evaluation; and to evaluate the scientific justifications and/or marketing motivations behind the claims for this label.

2. Definitions Personalized medicine, following the commonly accepted definition taken over by the European Medicines Agency (EMA),[1] consists in “giving the right treatment to the right patient, each drug being given at the right dose and at the right time”, to which can be added “and for the right duration”. Following the Food and Drug Administration (FDA),[2] it is the “tailoring of medical treatment to the individual characteristics, needs, and preferences of a patient during all stages of care, including prevention, diagnosis, treatment, and follow-up”. Personalized medicine is often presented as “tailored” medicine, in contrast to a medicine that would be “ready to wear” or even “one size fits all”. It concerns all stages of the medical act, from molecular diagnosis using biomarkers to detailed therapeutic modalities. Stratified medicine, often abusively mistaken for personalized medicine, consists in identifying subgroups of patients in whom a “targeted” drug will have the best benefit-risk ratio (hence will be indicated), i.e. a particular case of the first step “the right drug to the right patient”. The wording “precision medicine” has been proposed to comply with the current line of thought in the USA, as a replacement to personalized medicine, owing to two contradictory reasons: one being that all good medicine is by definition personalized, and the other that a true personalized medicine is an illusion in the sense that the clinical situations where it is really possible to individualize the choice of the drug are extremely rare.[3] Indeed, the “targeted therapies” improve the benefit-risk ratio of only a part of the population

© Société Française de Pharmacologie et de Thérapeutique

presenting the pathology, while the other patients can only receive “conventional” drugs. Following its proponents, precision medicine is based on molecular information which improves the diagnosis precision, hence the way patients are treated.[3] It therefore only deals, once again, with a population stratification in order to prescribe the drug correctly and in this respect, precision medicine can be considered as a synonym of stratified medicine, and not of personalized medicine that relies on very diverse modalities, beyond simple population stratification.

3. The right drug to the right patient 3.1. Translational research To improve the benefit-risk ratio of existing drugs, it is necessary to delineate the diseases more precisely, i.e. to make nosology progress by evolving from a classification based on syndromes to a classification based on cellular or molecular biology. Such translational research, relying on genomic and genetic analysis of particular or extreme forms of the disease or deriving from basic research, also allows better understanding of the pathophysiological mechanisms that will be the basis of new therapeutic targets. The biomarkers employed to select the patients likely to be responders, whether being genetic variants or abnormal, under- or over-expressed proteins, must then be integrated into the “environmental” factors category, pertaining to the patient (pathophysiological status), associated pathologies, concurrent treatments (drugdrug interactions), etc. Taking into account all these factors simultaneously will increasingly require the setting up of personal, comprehensive health databases. If the drugs targeting these genes or proteins do not exist already, they can be new research objectives. 3.2. Targeted therapies If the concept of a drug specific to a target linked with the pathology has been known and used for many years in many therapeutic fields (e.g., hypertension, antibiotherapy…), the term “targeted therapy” is more recent and seems to be preferentially employed in cancer and chronic inflammatory diseases, where the

Thérapie 2015 Janvier-Février; 70 (1)

Precision Medicine, Personalized Medicine, Targeted Therapies

13

Table I. Targeted therapies on the market in 2013, in the cancer area. (French Cancer Institute. Annual Report on the situation of cancer treatments. With courtesy INCa [N. Hoog Labouret]).[4]

Tyrosine kinase and related inhibitors

Monoclonal antibodies

INN

Brand name speciality

Year of marketing authorization

Drugs reimbursed on top of INN hospitalization costs

Brand name speciality

List of drugs Year of reimbursed on top marketing of hospitalization authorization costs

imatinib erlotinib

GLEEVEC® TARCEVA®

2001 2005

No No

rituximab trastuzumab

RITUXAN® HERCEPTIN®

1998 2000

sorafenib

NEXAVAR®

2006

alemtuzumab

MABCAMPATH® 2001

sunitinib dasatinib nilotinib temsirolimus lapatinib gefitinib everolimus pazopanib vandetanib vemurafenib axitinib crizotinib bosutinib ponatinib afatinib regorafenib dabrafenib

SUTENT® SPRYCEL® TASIGNA® TORISEL® TYVERB® IRESSA® AFINITOR® VOTRIENT® CAPRELSA® ZELBORAF® INLYTA® XALKORI® BOSULIF® ICLUSIG® GIOTRIF® STIVARGA® TAFINLAR®

2006 2006 2007 2007 2008 2009 2009 2010 2012 2012 2012 2012 2013 2013 2013 2013 2013

cetuximab ibritumomab tiutexan bevacizumab panitumumab catumaxomab ofatumumab ipilimumab brentuximab vedotin pertuzumab ado-trastuzumab

ERBITUX® ZEVALIN® AVASTIN® VECTIBIX® REMOVAB® ARZERRA® YERVOY® ADCETRIS® PERJETA® KADCYLA®

2004 2004 2005 2007 2009 2010 2011 2012 2013 2013

Yes Yes Withdrawn from the market Yes Yes Yes Yes No Yes Yes No No No

aflibercept

ZALTRAP®

2013

No

No No No No Yes No No No No No No No No No No No No No

INN: international non proprietary names.

previous active compounds targeted ubiquitous proteins or signaling pathways, poorly specific of the pathological process. The drugs claiming to be targeted therapies in oncology are of two types, biodrugs and small, so-called “chemical” molecules (table I). Biodrugs, derived for genetic engineering and biological production, are mostly monoclonal antibodies, whose international common denomination bears the suffix “-mab”, as well as fusion proteins incorporating the extracellular fraction of a membrane receptor for instance. They are administered intravenously or subcutaneously, have a high molecular weight, a long elimination halflife and mostly an extracellular target. The small molecules are mainly signaling pathway inhibitors, in particular tyrosine kinase inhibitors (“-nib”), as well as mammalian target of rapamycin (mTOR) inhibitors or immune modulators. They are low molecular weight, orally active molecules undergoing hepatic metabolism, which are prone to drug-drug interactions. Their pharmacokinetic properties are very variable and their target can be intracellular.

© Société Française de Pharmacologie et de Thérapeutique

Out of 34 targeted therapies in cancer in 2013,[4] 16 are accompanied by a biomarker that obviates their target and conditions their prescription. The study of these 34 drugs and the search of the common features shows that they present one or more of the following 4 characteristics: they are specific to a molecular target; this target is a protein most often directly involved in the pathogenesis; there is a biomarker allowing the selection of the patients who can benefit from the treatment; the target population is most often much smaller that the population presenting the disease. However, certain “targeted therapies” fulfill only one of these 4 criteria, such as rituximab which depletes B lymphocytes, whether normal or blastic, or everolimus which is only considered to be a targeted therapy in the treatment of cancer but not in its other two indications, organ transplantation (the main one) and tuberous sclerosis. In contrast, many monoclonal antibodies or specific signaling pathway inhibitors exist in other therapeutic areas, without them being called targeted therapies.

Thérapie 2015 Janvier-Février; 70 (1)

14

Marquet et al.

The motivations of the pharmaceutical industry to present their products as targeted therapies in the cancer area rely on the willingness of a rapid market access (after phase II, or even phase I trial in certain cases) and of an image of greater efficacy, excellent benefitrisk ratio among the prescribers. The round table allowed highlighting the different understandings of what targeted therapies are for researchers, the industry, regulatory agencies and prescribers. France is at the forefront of the use of targeted therapies in cancer treatment, owing to the national cancer plans. The French National Institute of Cancer (Institut National du Cancer, INCa) ran a national programme to ensure equity and high quality access to a targeted therapy if available, for all the patients with malignant hemopathies and solid tumours, by performing high-quality molecular tests (biomarker screening). This programme helped to set up, on the basis of a regional organization, 28 molecular genomic platforms sponsored by the Ministry of Health and INCa, all functioning in a multidisciplinary cooperative scheme involving clinicians, pathologists and molecular biologists. However, it seems important that these targeted therapies be more precisely defined as this status influences their registration procedure (in the USA) as well as their market access, marketing and prescription. The share of targeted therapies in the therapeutic strategies against cancer (indications, drug associations, etc.) is rapidly growing and its economic consequences must be anticipated. A proposal put forward during the round table, which did not reach unanimity among the participants, was that a targeted therapy should fulfill 3 out of the 4 abovementioned criteria. As previously mentioned, a few anticancer drugs currently considered as targeted therapies would not pass through such a filter.

4. Biomarkers and associated diagnostics 4.1. Definition and categories of biomarkers In 1998, the American National Institute of Health Biomarkers Definitions Working Group defined a biomarker as follows: “A characteristic that is objectively measured and evaluated as an indicator of normal biological processes, pathogenic processes, or pharmacologic responses to a therapeutic intervention.”[5] Biomarkers are tools used to select a patient group with the highest probability of a favourable response to a drug and/or to exclude those who may be susceptible to adverse effects. They can be classified into the three following categories: prognostic (biomarkers used to predict the natural outcome of a confirmed disease), predictive (biomarkers used to predict a patient’s potential response to a drug) or pharmacodynamic (biomarkers used to determine the effectiveness and modulate the dose of a drug). Predictive biomarkers are mainly used in hematology and oncology, which count for 38% of the total number of biomarkers in the different therapeutic areas.[6]

© Société Française de Pharmacologie et de Thérapeutique

4.2. Biomarkers and clinical development stakes There are currently 13 915 clinical trials using biomarkers[7] while 94% of pharmaceutical companies use biomarkers in their research programmes and 12-50% of company pipelines involve “personalized medicine”.[8] The critical steps in the development of biomarkers and associated diagnostics include the development of a well characterized and validated assay and the demonstration of the clinical validity of the biomarker in well-conducted studies. Ideally the predictive biomarker and the therapeutic agent should be co-developed prospectively. The randomized and controlled trial (RCT) design with predefined endpoints and measures remains the gold standard. However, biomarker identification and validation may occur at any time during the drug development process and even post-marketing. Indeed, it is sometimes not before the phase III trial results that clinical phenotypes suggesting a genetic study to identify subpopulations of increased benefit are discovered. In certain circumstances, conducting a new RCT is therefore not feasible for timing and ethical reasons. Furthermore, the length of time such a study would take might delay patient access to a potentially life-saving treatment, as well as withholding access to a treatment considered as the standard of care during this period. Additionally, since the optimal biomarker-driven RCT design may require the demonstration of a lack of effect (or even of a detrimental effect) in biomarkernegative patients, this raises additional ethical issues. Therefore, the evaluation of targeted therapies together with biomarkers requires rethinking the RCT design and analysis. Because patient populations in RCTs are heterogeneous, often due to genetically-based differences in therapeutic responsiveness,[9,10] a detailed assessment of an RCT cannot be restricted to an overall test of significance[9,10] and an analysis of subgroups is critical for identifying patients who respond to a given therapy or present with adverse effects.[9,11] Additionally, when biomarkers are discovered after the clinical utility of a new therapeutic has been demonstrated, the use of archived deoxyribonucleic acid (DNA) specimens from previous prospective RCTs could enable their validation.[12,13] This method is referred to as prospective-retrospective analyses (PRA) because it is based on retrospective analyses of prospectively collected samples.[13]

4.3. Biomarkers evaluation and value assignment stakes Over the past two years, the EMA, the FDA, the National Institute for Health and Clinical Excellence (NICE) in the UK, the Institut für Qualität und Wirtschaftlichkeit (IQWIG) in Germany and the French High Authority for Health (Haute Autorité de Santé, HAS) in France have issued guidelines on the planning of the phase III clinical trials undertaken by pharmaceutical companies that intend to claim enhanced safety or efficacy for a drug associated to biomarker-

Thérapie 2015 Janvier-Février; 70 (1)

Precision Medicine, Personalized Medicine, Targeted Therapies

based population stratification. Prospective biomarker validation still represents the gold standard for regulatory submission to the agencies, drug approval, access to reimbursement and value- and improvement-based pricing. However, most of these decision bodies also recognize that such prospective RCTs are not always feasible or even ethical. Therefore they have mentioned in their guidelines the methodological criteria that could enhance PRA validity and value assignment in their decisions. Amongst others, they request: 1) Consistent evidence from multiple trials and analyses, with a replication of the results in one or more independent sample sets; 2) Adequate sample sizes and statistical methods to avoid a selection bias; 3) Documentation of the hypothesis and statistical analysis plan prior to retrospective evaluation, with pre-specification of a particular subgroup mentioning the evidence supporting its biological or clinical plausibility. Statistical procedures exist that can address the methodological problems associated with subgroup analysis, i.e., multiplicity.[9,11] Also, the biological plausibility of the conclusions drawn from subgroup analyses and replication of evidence are both key considerations.[10] Involving clinicians and methodologists in the development of such guidance may be of great benefit. An interesting example of such an approach was published in 2011,[13] where a systematic decision tree was developed to identify the value of the PRA of a clinical trial that had been stratified using a predictive biomarker. Nine key questions were raised enabling the decision maker to conclude whether the considered PRA actually brings strong evidence value or not: 1) Well-controlled trial; 2) Biomarker with high biological plausibility; 3) Independent confirmation of association of a biomarker; 4) Biomarker hypothesis and statistical plan prior to trial initiation; 5) Hypothesis established prior to analysis of samples; 6) Statistical analysis plan; 7) Collection of “adequate” samples; 8) Well-defined assay for assessment; 9) Results meet statistical significance as per plan? The authors considered that if the nine answers were positive, then the PRA would be appropriate for regulatory and health technology assessment submissions. One approach to provide cost-effective healthcare is to use stratified medicine. Every effort to clarify the value of PRA upstream will have a very positive impact on the confidence that the health care systems may have in such analyses. This would help to clarify the public health benefit and economic values that targeted treatments may provide to society. Indeed, when the efficiency of new stratified medicines has to be demonstrated, the validity of clinical inputs used in the pharmacoeconomic models is logically put under high scrutiny and has to be analyzed and valued following precise and systematic methodology. 4.4. Quality stakes Delivering a reliable and relevant result is a daily concern for all those involved in the diagnosis. The parties involved are actually

© Société Française de Pharmacologie et de Thérapeutique

15

very much implicated on both sides, that is to say the users (clinical biologists and pathologists) and the industrial suppliers of in vitro diagnostic devices (IVD). For many years now, quality mechanisms have been reinforced by legal or regulatory obligations. For manufacturers, the European directive 98/79/EC has defined regulatory requirements related to IVD CE marking. The CE marking certifies the conformity of the IVDs’ to essential requirements of health and safety of the directive, explained in harmonized European standards. Harmonized standards provide methods to demonstrate that the IVD is safe and that its performance meets the claims of the manufacturer and the specified usage. The directive requires that the manufacturer is dealing with a quality management system in light of the ISO 13485. User compliance with manufacturer’s recommendations will ensure relevant results. Professionals who want to implement a homemade technique (LDT) to perform a diagnosis must demonstrate and validate their protocols, or in other words have to implement a CE marking process like the EU directive in the lab (figure 1). An evolution of current European regulation will soon also reinforce the obligations of manufacturers when producing CE products. Since 1994, a good practice guide (guide de bonne exécution des analyses, GBEA) has been published for medical laboratories. Since 1998 with the publication of an equivalent guide for pathology laboratories (recommandations de bonnes pratiques en anatomie et cytologie pathologiques, RBPACP), users have been permanently seizing the opportunity to demonstrate compliance to a regulation. This environment has been continuously reinforced. In 1999, a decree defined the GBEA as a mandatory process. As a next step and following 2 reports, one from the general inspection of social affairs (2006) and the other one from Dr. Ballereau in 2008, other obligations have been enforced by the issuance of a ruling in 2010 and a new law in 2013.[14] Since 2013, the accreditation of medical laboratories according to the standard ISO 15189 has to be implemented with finalization between 2016 and 2020. This has still not been decreed for pathology and cytology laboratories. In biology, the level of requirements has therefore significantly increased and compliance checks have been organized with regards to these obligations. These recent rules are aimed at a reorganization of lab activities on the French territory while framing implementation terms of this discipline. Their aim is also to medicalize this discipline. The clinical biologist is accountably moving from the obligation of using the appropriate means to the obligation of providing good results. Accreditation will be a better guarantee of quality results for the patient and will improve the integration of the clinical biologist in the course of care. Nevertheless, a number of challenges remain, especially for companion tests. For manufacturers, a major challenge is to be able to obtain CE marking on diagnostic products put on the market through a reliable process ensuring consistent use of products and standardized results. This approach would minimize the variability of results regarding different technologies performed at various

Thérapie 2015 Janvier-Février; 70 (1)

16

Marquet et al.

EU Direcve 98/79/CE CE Products CE or not CE

CE Products Laboratories

IQC

Home brew technics Variabilies, no standardizaon

Product usage following manufacturer package insert recommendaons

Home brew technics

IQC

Protocol validaon with essenal requirements 98/79/CE

Results Fig. 1. Regulatory environment to produce diagnostic in vitro results. IGC: internal quality control.

locations. It would provide clinicians with reliable and homogenized results to prescribe targeted therapies. For clinical biologists and pathologists, it remains necessary to harmonize the regulatory environment and to define a national quality control for companion tests.

5. The right dosage for each patient and the right moment for treatment Beyond choosing the best drug for each patient, the choice of the best initial dose and the possible adjustment of this dose during the course of treatments are other examples of personalized medicine. As well as taking into account classic environmental factors (medical history, associated drugs, etc.), the choice of the initial dose can depend on a pharmacogenetic test, making it possible to identify patients with a “pharmacokinetic” risk of over or under dosage, or those with a “pharmacodynamic” risk of undesirable effects or a weak therapeutic response. Table II presents examples of pharmacogenetic biomarkers which make it possible to improve the use of the corresponding drugs and prevent iatrogenesis. Indeed, the national pharmacogenetic network (Réseau National de Pharmacogénétique, RNPGx) would like these and other tests to be registered on the list of biological medical acts and be reimbursable, which is not the case at present. A population pharmacokinetic approach can also be used for the choice of the initial dose, allowing for the estimation of the

© Société Française de Pharmacologie et de Thérapeutique

patient’s most probable pharmacokinetic parameters depending on biometric factors (height, weight), demographic factors (age, gender), pathological factors and biological biomarkers (such as serum creatinine for drugs with renal elimination, for example). Such models have been developed for aminoglycosides, because dosing based only on serum creatinine is insufficient to encompass the very large variability of their elimination clearance depending on the situation; in adults (badly burnt patients, patients in resuscitation, very old patients) as in children (in particular, newborn and extremely premature babies). Therapeutic drug monitoring (TDM), aiming at adapting the dose of drugs with a narrow therapeutic margin for each patient, on the base of a blood concentration or of a systemic exposure index, is one of the oldest examples of personalized medicine based on a biomarker. Today, around a hundred active compounds are concerned, in the whole of the treated population or sometimes only in particular sub-populations at risk of under or over dosage. TDM can use pharmacokinetic modelling, when the pertinent exposure index is difficult to measure directly, such as the average concentration (or area under the curve of the concentrations). Even though TDM has been validated by experience and, more rarely, by random comparative trials, it is often not used to its full capacities or not used wisely, and is rarely the object of official recommendations, beyond those provided by national and international scientific societies of TDM or medical specialities. This example shows that one of the forecasted obstacles for the generalization of new methods of personalized medicine is the slowness and difficulty in attaining the medical practice, due to a lack of understanding of the

Thérapie 2015 Janvier-Février; 70 (1)

Precision Medicine, Personalized Medicine, Targeted Therapies

17

Table II. Pharmacogenetic tests and personalized medicine.

Tests

Indications

HLA*B57:01

Risk of allergy with abacavir

UGT1A1*28

Risk of irinotecan toxicity

CYP2D6/CYP2C19 (several polymorphisms)

Antidepressant selection in case of adverse effects or resistance

CYP3A5*1

Initial dose adjustment of tacrolimus

TPMT

Risk of azathioprine and 6-mercaptopurine toxicity

DPYP (several polymorphisms)

Risk of 5FU and capecitabine toxicity

5-FU: 5-fluorouracile; CYP2C19: cytochrome P450 2C19; CYP2D6: cytochrome P-450 CYP2D6 ; CYP3A5: cytochrome P450 3A5; DPYP: dihydropyrimidine dehydrogenase; HLA: human leukocyte antigen; TPMT: thiopurine S methyltransferase; UGT1A1: UDP-glycosyltransferase 1 polypeptide A1.

procedures by the prescribers or the lack of time to put them into place and to use them. Defining the right moment to administer a treatment can correspond to different objectives: early prevention, detection and treatment of a disease;[15] a strategy for life-long treatment of chronic diseases, making it necessary to choose the sequence of treatment lines; or the best administration time of an active compound, the pharmacokinetics or pharmacodynamics of which is the object of circadian variations.[16] Aside from anticancer drugs, only few studies and clinical applications have been carried out on these last two situations, probably because translational research in this field is only academic. Indeed, the pharmaceutical industry is less interested in therapeutic strategies involving their competitors’ active compounds, as well as by chronotherapy, considered as a complication which can hold up the prescription.

6. Discussion and recommendations

basic research to the clinics. The pharmaceutical industry increasingly uses the results of phase III studies to identify specific clinical phenotypes (in particular, response profiles), suggesting a complementary genetic study (PRA). The same strategy can and must be used with the post-marketing cohorts and registers where the patients’ variability is even larger, for the research of clinical and biological characteristics (biomarkers) influencing drugs benefitrisk ratio, which can open up new pathways for research. However, clinical research results would require an average of 17 years to be routinely implemented in clinical practice.[15] Recommendations should be delivered by a computer to the physician as specific courses of action, at the time and location of decision making, so as to improve the use of evidence-based personalized interventions. It will therefore be necessary to develop computerized decision-making algorithms (health information technologies) as well as comprehensive personal medical databases, which can also be considered as translational research but which pose problems with regards to both technology and data protection.

6.1. The role of translational research Translational research, which should allow the accession to new ways of personalized medicine, involves a long bidirectional process: from the patient to the laboratory (pathophysiological research and the search for biomarkers) and from the laboratory to the patient (evaluation of new therapeutic procedures). Close relations are necessary between health practitioners and research laboratories in both the academic field and the industry. In France, the French Alliance of Life and Health Sciences (Alliance des Sciences de la Vie et de la Santé, AVIESAN) and the French National Institute of Health and Medical Research (Institut National de la Santé et de la Recherche Médicale, INSERM) encourage such translational research by promoting interactions between clinicians and researchers and by valorising biological collections, patient cohort studies, industrial collaborations and the transfer of the results of

© Société Française de Pharmacologie et de Thérapeutique

6.2. Personalized medicine: a scientific reality Several types of personalized medicine rely on experimental and clinical evidence. For example, some targeted therapies for cancer have provided evident and in some cases major therapeutic benefits, in comparison with classical, cytotoxic chemotherapies. The introduction into the market of new active compounds results in regular progress for the patients, in terms of recovery but above all of survival. This is the same for the clinical application of pharmacogenetic tests or TDM, resulting from academic research. Personalized medicine necessitates training courses, which means that it also participates in the progression of medical skills. It goes together with a constant progression in basic and translational knowledge (for example, the study of the relapse factors to targeted

Thérapie 2015 Janvier-Février; 70 (1)

18

Marquet et al.

therapies which could make it possible to discover new targets). It therefore represents the beginning of a reply to the complexity. The scientific stakes which are open involve new approaches in the search for biomarkers such as epigenetic and all the omics, as well as new technological tools (new generation sequencing, new imaging techniques, mass spectrometry, etc.) and in particular the treatment and storage of the enormous mass of data they generate, which will require consequent financial, technological and human investments. Personalized medicine, notably in cancer, is also an organisational stake and this is a reality in certain countries such as France. Indeed, it is not sufficient to show that such and such drug whose use is based on the presence of a biomarker, improves the efficacy of the management of such or such pathology, but the determination of this biomarker must be technically feasible, for all the patients, with a good reproducibility and at an acceptable cost. Moreover, the multiplication of biomarkers in different pathologies (89 000 analyses in cancer pathologies in 2013 for example) will impose the setting up of new analytical techniques (high throughput analyses) and sophisticated methods for the collection and interpretation of data (“big data”). 6.3. Personalized medicine: a marketing reality The creation and the use of the term “targeted therapy” meets not only with scientific aims but also with the objectives of the industry and in particular marketing in a restrictive economical context. “Targeted therapies” are not all well targeted or innovative. Even if the regulations of industrial marketing are clearly set out by the agencies, the concepts of targeted therapy and even more, of personalized medicine, are positive driving forces for the promotion and distribution of a drug because they benefit from a positive preconception by the prescribers (based on the model of orphan drugs). This allows them a more rapid appropriation of the proven results and therefore a more rapid prescription. The scientific community is also concerned by the interest of marketing in the concept of personalized medicine. It may be in their interest to oversell their field of research in order to make it easier to obtain credits and financial support from the institutions and the industry. Personalized medicine is also a stake in communication destined for the general public, for example to justify the contributions by the French cancer programme and INCa. Finally, the concept of personalized medicine has become an internal marketing tool for certain drug companies, to promote a new corporate strategy: research for targets and biomarkers to reinforce efficiency in the development of candidate drugs. 6.4. Recommendations One of the first recommendations of the round table concerns the standardization of clinical applications of personalized medicine

© Société Française de Pharmacologie et de Thérapeutique

(even if these terms seem to be contradictory), on the basis of consensual criteria/biomarkers, of sensitive and validated analytical methods showing identical results between them, and algorithms in the choice of the treatment or adaptation of the dose also validated by the scientific and medical community and standardized. The other recommendations concern essentially translational research, which will be at the basis of progress in personalized medicine. It is indeed eminently important that the brut data of clinical trials be made available to the scientific community to allow them to carry out retrospective analyses of phenotypes and/or particular responses. Moreover, the setting up of an individual, comprehensive and protected medical database will represent a major study tool for individual response to treatment, on condition that such tools and bioinformatics skills are available.

7. Conclusion Ideal personalized medicine depends on the precise diagnostic of diseases, the existence of a suitable treatment for each type of patient, of methods in the choice and adjustment of the dose and of biomarkers of treatment efficacy, toxicity, discontinuation or relapse. Stratified or precision medicine uses biomarkers and companion diagnostic tests to target patients susceptible to present the best benefit-risk balance for a treatment. The concept of targeted therapy is above all used in oncology and deserves to be better defined. Targeted therapies and biomarkers represent important stakes for the pharmaceutical industry, in terms of access to the market, return on investment and image perceived by the prescribers. Those currently available probably only represent the first generation of products stemming from the combination of clinical research and physiopathological and molecular research, that is to say translational research. Conflicts of interests. None. Abbreviations. ACP: pathology and cytology; AVIESAN: French Alliance of Life and Health Sciences (Alliance des Sciences de la Vie et de la Santé); CE: European community; DNA: deoxyribonucleic acid; EMA: European Medicine Agency; FDA: Food and Drug Administration; GBEA: good analytical practice guide (guide de bonne exécution des analyses); HAS: French High Authority of Health (Haute Autorité de Santé); HTA: health technology assessment; INCa: French National Insitute of Cancer (Institut National du Cancer); INSERM: French National Insitute of Health and Medical Research (Institut National de la Santé et de la Recherche Médicale); IQWIG: Institut für Qualität und Wirtschaftlichkeit im Gesundheitswesen; IVD: in vitro diagnotic devices; mTOR: mammalian target of rapamycin; NICE: National Institute for Health and Clinical Excellence;

Thérapie 2015 Janvier-Février; 70 (1)

Precision Medicine, Personalized Medicine, Targeted Therapies

NIH: National Institute of Health; PRA: prospective retrospective analysis; RBACP: good practice recommendations in pathology and cytology (recommandations de bonnes pratiques en anatomie et cytologie pathologiques); RNPGx: National Pharmacogenetic Network (Réseau National de Pharmacogénétique); RCT: randomized and controlled trial; TDM: therapeutic drug monitoring.

References 1.

EMA. Personalised medicine towards the market and patients: the approval process. European Medicines Agency’s perspective. Mai 2011 http://ec.europa.eu/research/health/pdf/event06/13052011/marisapapaluca-amati_en.pdf Accessed December 18th, 2014 (17 pages)

2.

FDA. Personalized medicine. FDA’s unique role and responsabilities in personalized medicine. Octobre 2013 http://www.fda.gov/scienceresearch/specialtopics/personalizedmedicine/default.htm Accessed December 18th, 2014

3.

Katsnelson A. Momentum grows to make ‘personalized’ medicine more ‘precise’. Nature Med 2013; 19(3): 249

4.

Institut national du cancer. Rapport annuel sur la situation de la chimiothérapie des cancers. Juillet 2013 http://www.e-cancer.fr/rss-soins/7763-rapport-annuel-sur-la-situationde-la-chimiotherapie-des-cancers Accessed December 18th, 2014

5.

Strimbu K, Tavel JA. What are biomarkers? Curr Opin HIV AIDS 2010, 5(6): 463-6

6.

FDA. Table of pharmacogenomic biomarkers in drug labeling. Aout 2014 http://www.fda.gov/Drugs/ScienceResearch/ResearchAreas/Pharmacogenetics/ucm083378.htm Accessed December 18th, 2014

© Société Française de Pharmacologie et de Thérapeutique

19

7.

http://www.clinicaltrial.gov Accessed December 18th, 2014

8.

Tufts center for the study of drug development impact report. Tufts study shows drug development for personalize medicine on the rise. Decembre 2010 http://www.policymed.com/2010/12/tufts-study-shows-drug-development-for-personalize-medicine-on-the-rise.html Accessed December 18th, 2014

9.

Koch A, Framke T. Reliably basing conclusions on subgroups of randomized clinical trials. J Biopharm Stat 2014; 24: 42-57

10. Hemmings R. An overview of statistical and regulatory issues in the planning, analysis, and interpretation of subgroup analyses in confirmatory clinical trials. J Biopharm Stat 2014; 24: 4-18 11. Koch G, Schwartz T. An overview of statistical planning to address subgroups in confirmatory clinical trials. J Biopharm Stat 2014; 24: 72-93 12. Mandrekar SJ, Sargent DJ. Predictive biomarker validation in practice: lessons from real trials. Clin Trials 2010; 7(5): 567-73 13. Patterson SD, Cohen N, KarnoubM, et al. Prospective–retrospective biomarker analysis for regulatory consideration: white paper from the industry pharmacogenomics working group. Pharmacogenomics 2011; 12(7): 93951 14. Loi n° 2013-442 du 30 mai 2013 portant réforme de la biologie médicale. http://www.legifrance.gouv.fr/ affichTexte.do?cidTexte=JORFTEXT000027478077 Accessed December 18th, 2014 15. Ginsburg GS, Willard HF. Genomic and personalized medicine: foundations and applications. Transl Res 2009 Dec; 154(6): 277-87 16. Fu L, Kettner NM. The circadian clock in cancer development and therapy. Prog Mol Biol Transl Sci 2013; 119: 221-82 Correspondence and offprints: Pierre Marquet, UMR-S850 INSERM, Université de Limoges, Service de Pharmacologie, Toxicologie et Pharmacovigilance, CHU de Limoges, 87042 Limoges, France. E-mail: [email protected]

Thérapie 2015 Janvier-Février; 70 (1)