Any drug-based treatment, be it prophylactic or therapeutic, is bound to suffer from two limitations. The first one is that the drug will never be active in 100% of individuals. In fact, even in classes of drugs such as prophylactic vaccines, which exhibit very high levels of efficacy, a full efficacy rate is very rarely obtained. In the case of small chemical drugs, the range of efficacy is approximately 50–70% and overall, 30% of patients do not benefit from current medicines. The second limitation of any drug is its safety profile. An ‘active‘ drug is likely to induce side effects that will vary in severity from one individual to another. It is noteworthy in this context that the number of reported serious adverse events by the biopharmaceutical industries is clearly increasing every year.
Therefore, these two limitations led to the explosion of a field of science coined pharmacogenomics, which is the investigation of variations of DNA and RNA characteristics as related to drug response.
As a consequence, there is a need to give the right drug to the right patient and at the right time, and to predict its safety profile in a given patient. The search for biomarkers has been the hope to resolve those critical issues and is currently the subject of intense R&D activities, both in academic and industrial laboratories.
Moreover, the life science industry is facing several other challenges, including drug pipelines failures and delays, safety-based drug withdrawals, blockbuster patent expiries and generic erosion. The innovation gap is also getting wider with a tendency for fewer drug approvals and for more R&D spending.
In this context, the availability of biomarkers will indeed greatly impact all of the aforementioned issues. The positive added value of biomarkers will be at the R&D, registration, pricing and marketing phases. Obviously, selecting the right patients will speed up the time of enrolment during the clinical trials and will improve the chance of ultimately achieving clinical success, of registration by the regulatory authorities, the chance for good pricing and the chance for successful marketing by also providing a competitive advantage.
Nevertheless, one should not underestimate the difficulty of finding and validating the right biomarkers. It is a very complex endeavor involving multiple disciplines and heterogeneous populations, with evolving methods and regulatory environments particularly where standards have to be established. In addition, R&D of biomarkers is expensive, implicating many diverse staff competences and multiple technologies.
What is a biomarker?
A biomarker is a characteristic that is objectively measured and evaluated as an indicator of normal biological processes, pathogenic processes or pharmacological responses to a therapeutic intervention Citation[1]. It is also a molecular indicator of a specific biological property; a biochemical feature or facet that can be used to measure the progress of the disease or the effects of treatment [NIH, Internal Commun.]. In other words, a biomarker is any parameter that can be used to measure an interaction between a biological system and an environmental agent, which may be chemical, physical or biological [WHO, Internal Commun.].
An ideal biomarker should be amenable to easy, reproducible, cost-effective testing, in easily accessible body fluids, such as serum or urine, and should persist (half-life). Another key point is that a biomarker test must be ethically acceptable.
A biomarker test has to be sensitive, in other words, the proportion of case subjects (individuals with confirmed disease) who test positive for the biomarker and specific (i.e., the proportion of control subjects; individuals without the disease) who test negative for the biomarker. In this context an ideal biomarker will have 100% sensitivity and specificity, that is to say everyone with the disease would have a positive test, and everyone without the disease would have a negative test.
Ultimate confirmation of the validity of a biomarker has to be proven in a prospective clinical study. Nevertheless, well designed retrospective studies using well characterized samples in repositories can be performed and frequently yield viable candidates.
Currently, there are already many biomarkers, for example, cholesterol is one of the most well-known biomarkers of cardiovascular disease, body temperature is used as a marker of fever, blood pressure as a marker of stroke risk, blood sugar level for diabetes, viral antigens for hepatitis, proteins for heart attack and genetic variations for Huntington‘s disease.
Behind the word biomarker, one may distinguish several types:
| ▪ | A translation biomarker: a biomarker that can be applied in both a preclinical and clinical setting; | ||||||||||||||||||||||||||||
| ▪ | An efficacy biomarker: a biomarker that reflects positive effect of a given treatment; | ||||||||||||||||||||||||||||
| ▪ | A staging biomarker: a biomarker that distinguishes different stages of a disease; | ||||||||||||||||||||||||||||
| ▪ | A surrogate biomarker: a biomarker regarded as substitute to a clinical outcome; | ||||||||||||||||||||||||||||
| ▪ | A toxicity biomarker: a biomarker that reflects a toxicological effect in vitro and/or in vivo; | ||||||||||||||||||||||||||||
| ▪ | A mechanism biomarker: a biomarker that reports a downstream effect of a drug; | ||||||||||||||||||||||||||||
| ▪ | A target biomarker: a biomarker that reports interaction of the drug with its target; | ||||||||||||||||||||||||||||
| ▪ | A disease biomarker:
| ||||||||||||||||||||||||||||
A surrogate end point is a biomarker intended to substitute for a clinical end point being a characteristic or variable that measures how a patient feels, functions or survives.
A surrogate end point is always expected to predict clinical benefit (or harm or lack of benefit) based on epidemiologic, therapeutic, pathophysiologic or other scientific evidence Citation[2]. So, the ideal surrogate end point is a substitute measurement of benefit that is biologically relevant and scientifically sound, which directly relates to disease activity, with a rapid time course and which is easy and cheap to measure in a noninvasive and reproducible manner. The potential of surrogate end points like all the other types of biomarkers are numerous. They can lower the costs of drug development, reduce the time to filing, and increase go/no go milestone decisions. In addition, they may help clinical adaptive designs, increase response rate by stratifying patients, reduce the development risks and revive some dead drugs.
A biomarker classification has also been proposed on the following criteria:
| ▪ | Type 0 – measures natural history of disease correlating over time with known clinical indicators (e.g., prostate-specific antigen in prostate cancer); | ||||
| ▪ | Type 1 – indicates the intervention effect of a therapeutic drug (optimal drug dosing – Phase I); | ||||
| ▪ | Type 2 – surrogate end point markers (e.g., serum low-density lipoprotein). | ||||
Identifying biomarkers could be hypothesis driven, based on the drug mechanism of action, the target of the drug and the disease, but could or should also be data-driven, with no a priori whatsoever on the type of biomarkers that could be found. In fact, this is particularly relevant when the preclinical models are not predictable, when the mechanism of action is not fully understood and when this is a ‘first-in-class‘ new drug.
Actually, biomarkers can be identified via a panel of technologies. Those could be part of classical standard laboratory analysis (e.g., common blood tests, immunohistochemistry, flow cytometry, serum proteins) or via more recent approaches known under ‘omics‘, such as genomics, transcriptomics, proteomics, peptidomics, glycomics, lipidomics and metabolomics.
Genomics is the study of an organism‘s entire genome. The field includes intensive efforts to determine the entire DNA sequence of organisms and fine-scale genetic mapping efforts.
Transcriptomics is the study of the transcriptome, the complete set of RNA transcripts produced by the genome at any one time.
Proteomics and peptidomics are the large-scale study of proteins or peptides, respectively, particularly their structures and functions.
Glycomics is the study of carbohydrates.
Lipidomics is the full characterization of lipid molecular species and their biological roles with respect to expression of proteins involved in lipid metabolism and function, including gene regulation.
Metabolomics is the study of cell metabolism: the measurement of the metabolites of low molecular weight in an organism‘s cells at a specific time under specific environmental conditions. In other words, biomarkers are a child of the omics technologies, which will ultimately reduce the risk of drug development and improve patients‘ outcomes.
It is very likely that more than one biomarker will be necessary to generate a signature predictive of efficacy or toxicity for a given disease and for a given patient. This set of biomarkers may even come from different technologies described above.
Therefore, biomarkers are the foundation of evidence-based medicine: who should be treated, how and with what? Most evidently, biomarkers will be beneficial to all types of diseases, but cancer stands out as one of the most pertinent.
Actually cancer is a collection of complex, heterogeneous chronic diseases that will benefit enormously from biomarkers. Moreover, oncologists are specialists who are very much aware of new technologies (e.g., genotyping) and can follow the explosion of omics technologies. The oncology field can also yield clear quality of life measurements and survival end points. If there is a disease category that needs to benefit from safer and more efficacious medicines, cancer certainly is the one. Besides those clinical justifications, market-wise, cancer also comes top of the list. This is a very competitive environment for which biomarkers can offer a competitive edge. A better pricing could be obtained for drugs showing a clear cost–benefit ratio. There is also a clear patient and public awareness that cancer is an increasing issue. Finally, cancer-related biomarkers may offer the opportunity for ‘niche-busters‘ by segmenting the right patients to treat.
There are already several tumor-associated biomarkers with some predictive value, such as β-human chorionic gonadotropin in testicular tumors and choriocarcinoma, α-foeto-protein in hepatocellular carcinoma and testicular tumors, calcitonin in medullary thyroid carcinoma and prostate-specific antigen in prostate cancer.
In spite of the technology drive of those omics approaches, the identification of biomarkers is difficult and risky. Obviously, there are numerous confounding factors and bias explaining too many failures, such as:
| ▪ | Bad quality control for sample adequacy, including validity of disease categorization, sample integrity and degradation, and contamination (microorganisms and extraneous material); | ||||
| ▪ | Absence of standardization of methods (planning phase), centralization of measurements or normalization of measurements; | ||||
| ▪ | Accuracy of disease phenotype is indeed critical (i.e., patients must all have the same disease and several causes can lead to the same disease); | ||||
| ▪ | Inappropriate diagnostics used; | ||||
| ▪ | Inappropriate statistics used; | ||||
| ▪ | The same biomarker signature can result in different diseases due to the effect of environment, age, sex, comedication, concomitant pathologies and so on. | ||||
However, the identification of biomarkers paves the way for diagnostics that will help patient stratification.
The combination of a therapy and a diagnostic for personalized medicine led to the appearance of the so-called theranostics field. Some current applications in the field of oncology, for instance, include Genentech‘s Herceptin® (Roche, Basel, Switzerland) and Her-2/Neu biomarker in breast cancer, Novartis‘s Gleevec® (Basel, Switzerland) and Bcr/Abl biomarker in chronic myeloid leukemia, and Genomic Health‘s Oncotype DX® (CA, USA).
No doubt those biomarkers will shape the healthcare business in the future. Biomarkers will ultimately help to move from disease definition by symptoms to mechanism, uniformity of disease to heterogeneity, uniformity of patients to variability, universal treatment to individualized therapy and sickness to predictive/preventive care.
The time is now for personalized medicine biomarkers, the companion clinical biomarker that links the disease and the patient to a drug and that defines the disease and/or predicts response and risk and/or determines dose of the drug. The right drug to the right patient and at the right time, this is what personalized medicine is all about.
What is driving personalized medicine?
There is a convergence in scientific advances and enabling technologies. In addition, there is an increase in patient care and in rising consumerism. Nowadays, patients are more and more empowered and really want to stay well. Finally, there is a clear economic and financial incentive from politicians, payers, insurances, physicians and providers, patients and consumers.
In conclusion, nothing could be less certain that biomarkers will revolutionize personalized medicines and are becoming increasingly important in human health. Many studies reflect their development and there is considerable interest in their use and application to better treat human diseases.
Financial & competing interests disclosure
The author is a full-time employee of Transgene SA, France. The author has no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.
No writing assistance was utilized in the production of this manuscript.
Bibliography
- Biomarker Definitions Working Group: Biomarkers and surrogate endpoints: preferred definitions and conceptual framework. Clin. Pharmacol. Ther.69(3) , 89–95 (2001).
- De Gruttola VG , ClaxP, DeMetsDL et al.: Considerations in the evaluation of surrogate endpoints in clinical trials. summary of a National Institutes of Health workshop.Control. Clin. Trials22(5) , 485–502 (2001).