Biostorage.com

The Three ‘Cs' of Biomarker Development
Dr Elizabeth Foot, CEO of London
Genetics, argues that biomarker-driven
drug development is the only way for
biopharmaceutical companies to survive the
changing face of healthcare, and outlines
the three essentials for the successful
development of targeted drugs. Comments
are based on a paper given at the BioStorage
Technologies first annual Global Sample
Management Benchmarking Symposium,
Germany, September 15 2010.
The idea that pharma will lose revenues as it turns away from the blockbuster model towards the development of nichebusters has undoubtedly caused fear for the industry, but this change should be seen for what it truly is – an advance in the science of drug development that will not only significantly improve patient care, but also bring economic rewards.
The goal of clinical development has always been to identify those patients likely to derive most benefit from a given drug with a particular mode of action, and with the development of biomarkers, the tools are now available to predict more effectively who these patients will be.
The drivers to identify biomarkers and develop targeted medicines are getting stronger. Regulatory approval may still be the key milestone for any new medicine, but the shown by the Voluntary Exploratory Data No Samples, No Science
final hurdle to market increasingly lies with Submissions for the FDA (VXDS), and Scientific Pharma should take these opportunities and the payors and reimbursement authorities. Advice Working Party and Pharmacogenomic work with regulators early in development In a move away from free market pricing, Working Party for the European Medical to discuss and gain advice on issues such as flexible pricing structures are now being put Agency. The VXDS and PGxWP further offer biomarker strategy, clinical design, what kinds in place or are under discussion in many the opportunity for a common forum across of tissue samples they should be collecting, countries, with reimbursement only being agencies including the Pharmaceuticals and and which analyses should be conducted. given for those patients in whom medicines Medical Devices Agency in Japan.
The big questions under discussion are how are shown to be cost effective. The result for Indeed there are increasing numbers of to identify robust biomarkers efficiently, how the biopharmaceutical industry is that financial guidances and white papers now available or to build accurate biomarker hypotheses, and reward is ever more closely linked with the currently out for consultation (see Table 1), the how to do so early enough in development to clinical value obtained by the patient.
most recent being from the FDA, which earlier incorporate into the development plan. And Regulators have also been pushing this this year released for comment guidelines all this needs to be considered alongside the approach and encouraging and guiding for the development of targeted drugs and development of a companion diagnostic, ready pharma companies on how to incorporate diagnostics, further supporting the central for approval at the same time as the medicine.
biomarker approaches into clinical role of stratified approaches into the clinical While the details are debated, it is clear at development plans. Their willingness to development process right from the earliest least that in the meantime, the consideration discuss outside the main regulatory path is days of clinical planning.
of, and planning for, the use of biomarkers The Three ‘Cs' of Biomarker DevelopmenT
Table 1: Regulatory white papers and reflection papers: European Medicines Agency and US Food and Drug Administration
Regulatory Reflection paper/Guidance
Publication date/ Effective date
European Medicines Agency*
Guideline on the use of pharmacogenomic methodologies in the pharmacokinetic evaluation of medicinal products Finalisation expected 2011.
Reflection paper on methodological issues associated with pharmacogenomic biomarkers in relation to clinical Draft reflection paper Released for consultation Q4 2010. Finalisation development and patient selection expected 2011.
Reflection paper on co-development of pharmacogenomic biomarkers and assays in the context of drug development Draft reflection paper Released for consultation July 2011. Deadline for comments November 2011. Finalisation expected 2011.
Reflection paper on genomics and personalised medicines Release for consultation Q3 2011. Finalisation expected 2012.
Food and Drug Administration**
Draft Guidance for Industry and Food and Drug Administration Staff: In Vitro Companion Diagnostic Devices July 2011. Comments by November 2011.
Clinical Pharmacogenomics: Premarketing Evaluation in Early Phase Clinical Studies February 2011. Comments by May 2011.
International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use concept paper on pharmacogenomic biomarker qualification: Format and data standards International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use Topic E15 definitions for genomic biomarkers, pharmacogenomics, pharmacogenetics, genomic data and sample coding categories but in others, particularly when predicting Table 2: Example of pharmacogenetic data in drug labels
adverse safety events, data may be limited or Drug name
Genetics Biomarker Label Update
Type of Evidence
completely absent.
Developing technologies now afford greater opportunities to start exploration absence of benefit and build hypotheses, and play an important Stevens-Johnson syndrome/ role at this hypothesis generation stage by toxic epidermal necrolysis investigating expression changes across tissue types (Bilello JA, 2005 The agony and ecstasy of "omic" technologies in drug development. Current Molecular Medicine, 5: 39-52.). Taking both tissue and biological samples from the preclinical stage and onwards and conducting exploratory analysis is a first step to start building these hypotheses.
must start very early in the development path development programmes and be more One success in this area – from the of the drug.
confident that they are on the right road for Predictive Safety Testing Consortium, a Most fundamentally, it is vital that the future success.
collaboration between pharmaceutical appropriate samples are collected, handled and For every clinical development programme companies, biotech and academia as well as stored in a robust manner to ensure the conclusions it is essential to: regulatory authorities (FDA and EMA) – has made based upon the derived data are reliable. If been the identification and subsequent there are no samples, there is no science – no 1) Consider the impact of emerging data qualification (establishment of utility) of seven way to identify the responding populations and and build a robust biomarker hypothesis biomarkers for the detection of renal injury a lost opportunity to investigate unexpected and strategy to better define the patient in preclinical studies. These markers have and otherwise unexplained responses and population and evaluate the drug response subsequently been accepted by the Japanese adverse events arising during development or regulatory authority, the PMDA, as useful in beyond (see Mills FJ (2009). The Need for Good This should be based upon prior assessing the safety of new drugs. To guide Storage Practice. Biopreservation and Biobanking, 7 knowledge of the drug target, signalling thinking in this area of biomarker discovery (2), pp 115-117).
pathways and disease. For example, is and validation, the EMA has developed a For the best chance of success in there a genetic hypothesis? Are there gene Biomarker Qualification process and issued developing targeted drugs with proven variants with high likelihood of modulating guidance to help guide industry in the use of biomarkers, companies must use as drug response within the drug target, drug biomarkers within drug development. In July their starting point in all their clinical signalling pathways or genes involved in of this year it issued for comment (deadline 25 programmes the three ‘Cs' of biomarker how the drug is metabolised? Are there November 2011) a draft reflection paper on sample management – Consider, Collect data from a drug in a similar class already the use of pharmacogenomic biomarkers as and Conduct. Using this formula, pharma in development? In some cases candidate patient selection and treatment stratification can apply the emerging science to its genes will be known that can be investigated, tools in drug development.
scripintelligence.com Informa UK Ltd 2011 The Three ‘Cs' of Biomarker DevelopmenT
2) Collect samples either for prospective et al presented recently simulation data the HLA-B*5701 allele in predicting patients analysis or for storage for retrospective quantifying the loss in statistical power to most at risk. This led to a drug label change, analysis in response to data arising during detect true biomarkers owing to diminishing and genotype testing for this risk allele being clinical development concentrations of measured analytes in the incorporated into clinical guidelines with a As part of any risk management strategy, it is samples, together with the impact of poor resultant decrease in the incidence of abacavir imperative to ensure appropriate samples are sample handling and storage conditions hypersensitivity, and for GSK, an increase in collected as part of clinical development. These (Balasubramanian R, Muller L, Kugler K, Hackl Ziagen's sales.
can range from a single DNA sample to a broad W, Pleyer L, Dehmer M, Graber A (2010) The marketed neurological drug spectrum of samples to look at RNA expression The impact of storage effects in biobanks on carbamazepine had its US label changed and/or biomarkers from a range of different biomarker discovery in systems biology studies. in 2007 to include a recommendation that, tissues that may hold the key to understanding Biomarkers, 15: 677-783).
before starting therapy, patients with Asian the effects of investigational drugs in different More recently, Kugler et al provided ancestry get a genetic blood test that can populations. At the very minimum samples further data, presenting estimates of bias for identify a significantly increased risk of must be collected and appropriately stored to association of biomarkers in discovery studies developing rare, but serious, skin disorders allow analysis either for an individual study or (Kugler KG, Hackl WO, Mueller LAJ, Fiegl H, including Stevens-Johnson syndrome and collectively across studies as appropriate in the Graber A and Pfeiffer RM (2011) Journal of toxic epidermal necrolysis. It was found light of arising data.
Clinical Bioinformatics, 8: 1 doi:10.1186/2043- that patients who have an inherited variant With the increasing number of samples 9113-1-9). Ensuring the highest quality sample of the immune system gene, HLA-B*1502, being collected within a clinical programme management processes are adopted as part of seen almost exclusively in people of Asian it is also imperative to ensure samples are a clinical study is of vital importance to ensure ancestry, run a greater risk of developing the handled appropriately, shipped under the right the robustness of the data, decision-making rare skin diseases.
conditions and stored to ensure their integrity and conclusions from these studies.
Abacavir's labelling was based upon and quality together with a fully documented prospective testing, whereas the label change trail of custody, all factors that will be required if 3) Conduct appropriate studies and analysis and black-box warning for carbamazepine samples are used for analyses included within a The need for prospective versus retrospective was based upon retrospective analysis of a analysis of samples collected in order to achieve relatively small sample set.
Sample integrity is of paramount regulatory approval is not clearly defined Factors such as strength of the association importance in order to make robust and data obtained from both retrospective and seriousness of the adverse effect will conclusions from the data – from the point of and prospective studies have been approved all impact upon the conclusions made. collection, through handling and shipment, by the FDA for inclusion on drug labels. For Table 2 provides examples of genetic data storage and sample management (to ensure example, there is the case of GlaxoSmithKline's in labels and whether they are based upon efficient retrieval of the correct sample and antiretroviral Ziagen (abacavir, now marketed retrospective or prospective data. Study and ability to link with relevant clinical data), by the GSK/Pfizer joint venture Viiv healthcare), analysis design are issues that may be, and together with a detailed and complete which was initially launched without a test to should be, discussed as part of a VXDS or audit trail. A number of recent studies predict the incidence of serious hypersensitivity PGxWP discussion and will depend upon have demonstrated the adverse impact (approximately 8%), instead relying on clinical factors including strength of association, of poor storage on the quality of samples characteristics to determine those most at risk. clinical impact and utility.
and ultimately on the reliability of the data However, later extensive pharmacogenetic derived from their analysis. Balasubramanian studies identified and validated the utility of Dr Elizabeth Foot Informa UK Ltd 2011 scripintelligence.com

Source: http://www.biostorage.com/wp-content/uploads/2011/12/The-Three-Cs-of-Biomarker-Development.pdf