NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.
FDA-NIH Biomarker Working Group. BEST (Biomarkers, EndpointS, and other Tools) Resource [Internet]. Silver Spring (MD): Food and Drug Administration (US); 2016-. Co-published by National Institutes of Health (US), Bethesda (MD).

BEST (Biomarkers, EndpointS, and other Tools) Resource [Internet].
Show detailsDefinition
A biomarker used to identify individuals who are more likely than similar individuals without the biomarker to experience a favorable or unfavorable effect from exposure to a medical product or an environmental agent.
Examples
- Squamous differentiation in non-small cell lung cancer may be used as a predictive biomarker to identify patients who should avoid treatment with pemetrexed, on which they are likely to have worse survival or progression-free survival outcome compared to treatment with other standard chemotherapies such as docetaxel or cisplatin in combination with gemcitabine (Scagliotti et al. 2009).
- Certain cystic fibrosis transmembrane conductance regulator (CFTR) mutations may be used as predictive biomarkers in clinical trials evaluating treatment for cystic fibrosis, to select patients more likely to respond to particular treatments (Davies et al. 2013).
- BReast CAncer genes 1 and 2 (BRCA1/2) mutations may be used as predictive biomarkers when evaluating women with platinum-sensitive ovarian cancer, to identify patients likely to respond to Poly (ADP-ribose) polymerase (PARP) inhibitors (Ledermann et al. 2012).
- Human leukocyte antigen allele (HLA)–B*5701 genotype may be used as a predictive biomarker to evaluate human immunodeficiency virus (HIV) patients before abacavir treatment, to identify patients at risk for severe skin reactions (AIDSinfo 2007).
- Thiopurine methyltransferase (TPMT) genotype or activity may be used as a predictive biomarker when evaluating patients who may be treated with 6-mercaptopurine or azathioprine, to identify those at risk for severe toxicity because of high drug concentrations (PharmGKB 2016; Relling et al. 2011).
- Mutations in the BRCA1/2 genes may be used as predictive biomarkers for sensitivity to ionizing radiation because they may impair the function of the genes’ protein products in the repair of double stranded DNA breaks, which are one type of damage induced by ionizing radiation (Pijpe et al. 2012).
Explanation
A predictive biomarker is used to identify individuals who are more likely to respond to exposure to a particular medical product or environmental agent. The response could be a symptomatic benefit, improved survival, or an adverse effect.
A familiar example of use of a predictive biomarker in medical product development is predictive enrichment of the study population for a randomized controlled clinical trial of an investigational therapy, in which the biomarker is used either to select patients for participation or to stratify patients into biomarker positive and biomarker negative groups, with the primary endpoint being the effect in the biomarker positive group (U.S. Food and Drug Administration 2012). If the biomarker is in fact predictive of a favorable outcome, then the effect of the investigational therapy compared to a control therapy (including no therapy) will be greater (or present at all) in patients with the biomarker or some level of the biomarker. When only a small fraction of the patients who receive the investigational therapy are expected to show a meaningful effect, identification of that small group using a predictive biomarker is critical to the feasibility of demonstrating the intervention’s effectiveness (Betensky et al. 2002; Maitournam and Simon 2005). The notion of a predictive biomarker applies to a wide variety of interventions, including drugs, biologics, medical devices or procedures, and behavioral or dietary modifications for treatment or prevention of diseases or conditions.
The utility of predictive biomarkers is not limited to a clinical trial setting, as these biomarkers can also assist in informing patient care decisions, such as determining who might benefit from a particular treatment or selecting among multiple interventions. In the latter situation, evidence that a biomarker predicts the comparative effectiveness of an intervention should be accompanied by specification of the alternative interventions involved in the comparison.
Predictive biomarkers for effects of interventions may be characteristics of the individual’s biological constitution (“host characteristics”) or characteristics of the disease process or other medical condition. Biomarkers representing host characteristics are present irrespective of the individual’s disease or medical condition status, such as germline DNA, HLA type or cytochrome P450 enzyme phenotype, renal or hepatic function, or metabolic characteristics. Examples of biomarkers characterizing a disease process or medical condition include protein levels in diseased tissues, mutations in tumors, low or preserved ejection fraction in heart failure, and serum protein levels in pregnancy. Predictive biomarkers for drugs are often chosen initially based on the mechanism of action of the drug and understanding of pathophysiology, but they could also be identified empirically, e.g., based on previous studies. Understanding the impact on outcome of both host and disease or condition characteristics is important for efficient development and optimal application of interventions.
Establishing that a biomarker is predictive for an intervention’s effect generally requires a comparison of the intervention to a control treatment in individuals with and without the biomarker, usually in randomized trials. Although studying only biomarker positive patients would establish effectiveness of a particular intervention it does not specifically demonstrate the role of the biomarker. It is therefore generally appropriate to stratify patients in the randomized trial by presence or absence of the biomarker (if dichotomous). Randomization to treatment and control groups is usually important because demonstrating that individuals who are positive for a biomarker and receive an investigational therapy experience a better outcome than those who receive the same therapy but are negative for the biomarker does not establish that the biomarker is predictive. Differences in outcome associated with the biomarker could be due to prognostic abilities of the biomarker and may be present irrespective of the therapy received. The greater differences between treatment and control in the biomarker positive compared to biomarker negative groups are what establish the biomarker as predictive. (See also discussion in Understanding Prognostic versus Predictive Biomarkers).
Studies designed to evaluate a predictive biomarker should usually include patients with a range of biomarker values (or positive and negative for binary biomarkers). Sometimes there is sufficient prior evidence to strongly suggest that an investigational therapy will not be effective (or could even be harmful) in a certain subgroup of individuals defined by a biomarker; these circumstances may require excluding patients who are negative for the biomarker from trials of the investigational therapy. When a biomarker identifies a subgroup of patients who will benefit most from an investigational therapy, enrichment of a trial with individuals from that subgroup will provide increased statistical power for detection of the (larger) effect of that therapy; use of such an enrichment strategy will also affect the intended population to receive the therapy after its regulatory approval (U.S. Food and Drug Administration 2012).
The predictive biomarker concept can be extended beyond interventional trials to studies of exposures to environmental toxins, tobacco smoke, nicotine, alcohol, food additives, environmental or occupational radiation, or infectious agents or to studies of the unintended ancillary effects of interventions. In this document, an exposure is distinguished from an intervention in that it may occur passively (e.g., second- hand smoke or exposure to ultraviolet radiation from the sun during outside activities) or without intent to influence an affected system (e.g., kidney toxicity from exposure to aminoglycosides used to treat an infection in another organ). In the exposure setting, a predictive biomarker is one that is associated with increased or decreased likelihood of experiencing a particular outcome of interest when an individual is subjected to the exposure. The predictive biomarker can be used to assess degree of vulnerability to an exposure and can be viewed as an effect modifier.
References
- AIDSinfo. Guidelines for the Use of Antiretroviral Agents in HIV-1-Infected Adults and Adolescents. December 1, 2007. Accessed October 2016. https://aidsinfo
.nih .gov/guidelines/html /1/adult-and-adolescent-arv-guidelines /7 /hla-b--5701-screening. - Betensky RA, Louis DN, Cairncross JG. Influence of unrecognized molecular heterogeneity on randomized clinical trials. J Clin Oncol. 2002 May 15;20(10):2495–9. [PubMed: 12011127]
- Davies JC, Wainwright CE, Canny GJ, Chilvers MA, Howenstine MS, Munck A, Mainz JG, Rodriguez S, Li H, Yen K, Ordoñez CL, Ahrens R. VX08-770-103 (ENVISION) Study Group. Efficacy and safety of ivacaftor in patients aged 6 to 11 years with cystic fibrosis with a G551D mutation. Am J Respir Crit Care Med. 2013 Jun 1;187(11):1219–25. [PMC free article: PMC3734608] [PubMed: 23590265] [CrossRef]
- Ledermann J, Harter P, Gourley C, Friedlander M, Vergote I, Rustin G, Scott C, Meier W, Shapira-Frommer R, Safra T, Matei D, Macpherson E, Watkins C, Carmichael J, Matulonis U. Olaparib maintenance therapy in platinum-sensitive relapsed ovarian cancer. N Engl J Med. 2012 Apr 12;366(15):1382–92. [PubMed: 22452356] [CrossRef]
- Maitournam A, Simon R. On the efficiency of targeted clinical trials. Stat Med. 2005 Feb 15;24(3):329–39. [PubMed: 15551403] [CrossRef]
- PharmGKB. Clinical Pharmacogenetics Implementation Consortium (CPIC) guideline information for azathioprine and TPMT. May 2016. Accessed October 2016. https://www
.pharmgkb .org/guideline/PA166104933. - Pijpe A, Andrieu N, Easton DF, Kesminiene A, Cardis E, Noguès C, Gauthier-Villars M, Lasset C, Fricker JP, Peock S, Frost D, Evans DG, Eeles RA, Paterson J, Manders P, van Asperen CJ, Ausems MG, Meijers-Heijboer H, Thierry-Chef I, Hauptmann M, Goldgar D, Rookus MA, van Leeuwen FE. GENEPSO; EMBRACE; HEBON. Exposure to diagnostic radiation and risk of breast cancer among carriers of BRCA1/2 mutations: retrospective cohort study (GENE-RAD-RISK). BMJ. 2012 Sep 6;345:e5660. [PMC free article: PMC3435441] [PubMed: 22956590] [CrossRef]
- Relling MV, Gardner EE, Sandborn WJ, Schmiegelow K, Pui C-H, Yee SW, Stein CM, Carrillo M, Evans WE, Klein TE. Clinical Pharmacogenetics Implementation Consortium Guidelines for Thiopurine Methyltransferase Genotype and Thiopurine Dosing. Clin Pharmacol Ther. 2011 Mar;89(3):387–391. [PMC free article: PMC3098761] [PubMed: 21270794] [CrossRef]
- Scagliotti G, Hanna N, Fossella F, Sugarman K, Blatter J, Peterson P, Simms L, Shepherd FA. The differential efficacy of pemetrexed according to NSCLC histology: a review of two Phase III studies. Oncologist. 2009 Mar;14(3):253–63. [PubMed: 19221167] [CrossRef]
- U.S. Food and Drug Administration. Draft Guidance for Industry: Enrichment Strategies for Clinical Trials to Support Approval of Human Drugs and Biological Products. December 2012. Accessed March 2016. http://www
.fda.gov/downloads /drugs/guidancecomplianceregulatoryinformation /guidances/ucm332181.pdf.
- Predictive Biomarker - BEST (Biomarkers, EndpointS, and other Tools) ResourcePredictive Biomarker - BEST (Biomarkers, EndpointS, and other Tools) Resource
Your browsing activity is empty.
Activity recording is turned off.
See more...