Brussels Broods over Biomarkers

Bloggers, Barroso, and biomarkers have more in common than their initial letter. They are all terms that have started to gain currency only very recently. And while Brussels has had no difficulty in digesting the latest fad in public diary-keeping on the Internet, and only a little more difficulty in getting used to the obscure Portuguese politician who was thrust into the limelight when he became European Commission President a year or so ago, it is taking rather longer to come to terms with biomarkers.

Peter O'Donnell

But it is not for want of trying. In a bid to move the debate along, the European Medicines Agency has been assiduously promoting exchanges of information on the subject. It held its first wide-ranging meeting on the subject in December, bringing together investigators and researchers, learned societies, health professionals, and regulators. It published a report on the subject in February. And it is promising a further major meeting, to include industry participants too, during the course of 2006.

According to the agency, the prospects overall for biomarkers are "very positive," although it admits that generally the development is still "at a very early stage," as scientists continue searching for relevant biomarkers and as efforts continue to define and validate them. But the agency's view is clear: "Intellectual pressure should be put on the implementation of translational research early in clinical trials."

Agency workshop

As a sign of the importance attributed to the subject by the agency, the workshop was co-chaired by Dr. Daniel Brasseur, chair of the agency's Committee for Human Medicinal Products, and by Professor Bruno Flamion, chair of the Scientific Advice Working Party. It paid particular attention to current views on validation of biomarkers. Presenting experience with the validation of surrogate endpoints in HIV, Professor Michael Hughes of the Harvard School of Public Health stressed the importance of having a reasonable biological model, sensitive assays, potent treatments, large databases (in particular when treatments are not so potent), and developing methodologies for linking information from shorter-term studies when endpoints are distant in time.

It also discussed the influence of pharmacogenomics on new drug therapies, and examples of successful biomarker programs. Much of the discussion focused on cancer. Dr. Francesco Pignatti presented the agency's own experience with biomarkers in oncology drug approvals and the revision of the CHMP oncology guideline to include considerations on the use of biomarkers throughout the development of noncytotoxic compounds. In his remarks on statistical validation of surrogate endpoints in oncology, Professor Thomasz Burzykowski of Hasselt University in Belgium underlined the importance of meta-analysis techniques to look at data from several randomized trials and study the association between the surrogate and the true endpoint at the trial level. Professor Helgi Helgason of the Netherlands Cancer Institute noted the clinical utility of proteomics in oncology drug development, including potential applications in malignant disease for screening, monitoring progression or relapse of disease, treatment monitoring, and possible new leads for drug targeting.

The role of biomarkers in translational research—including successful and so far unsuccessful developments—was discussed by Professor Heinz Zwierzina of the European Society for Medical Oncology, who underlined the importance of identifying subgroups of patients with a high chance to respond to targeted therapy in order to speed up drug development and avoid turning down potentially active drugs, and to define optimal combinations. Single biomarker approaches have not been shown to have a strong potential in lung cancer so far, observed Dr. Alexandre Passioukov of the European Organisation for Research into Treatment of Cancer in his presentation of developments in biomarker identification and validation for lung cancer, including detection/diagnostic biomarkers and prognostic biomarkers. He suggested that the future might rather depend on the successful use of a set of biomarkers.

Cardiovascular care

Dr. Pieter de Graeff of the Medicines Evaluation Board in the Netherlands discussed regulatory implications for the use of biomarkers in cardiovascular research. His cautious view was that biomarkers are useful for drug development and risk estimation, but currently not as endpoints for pivotal studies in the cardiovascular field. But Professor Carlo di Mario of the Royal Brompton Hospital in London claimed that biomarkers could become very useful in coronary vascular imaging interventions in atherosclerosis, because this condition is clinically silent until it reaches advanced stages.

The complexity of the issue was demonstrated by the remarks of Dr. Francois Alhenc-Gelas of INSERM in Paris, whose discussion of genetic variations and genomic biomarkers in cardiovascular imaging ranged across monogenic forms of complex arterial diseases as well as genomic wise scans versus candidate gene approach for cardiovascular diseases such as the ACE gene polymorphism and disease outcome.

Participants at the workshop took the view that one bottleneck to development is the pre-analytical phase, and that prospective sampling of fresh frozen material required for most techniques should be promoted, with a focus on methods that improve reproducibility. The importance of sharing data among companies as required for validation of surrogate endpoints, in particular from the statistical point of view, was also underlined.

In many areas the trend is to use a combination of several biomarkers, but this raises issues of multiplicity that need addressing from a statistical point of view. At the same time, it will be essential to establish the role and weight of the different biomarkers within a composite score, and to establish the link to the true endpoint.

And while application of biomarkers may, in the best of cases, result in an overall decrease in the size of the trials, or the increased use of prespecified subgroups, this will entail study design with what the agency politely refers to as "increased plausibility." As a clear indication of the agenda still to be covered in this rapidly developing field, the role of biomarkers for safety was mentioned only briefly at the workshop, "and will require more discussions," according to the agency.

Innovative medicines agenda

Development of biomarkers also plays a prominent role in the plans being pushed by the European pharmaceutical industry for a new European initiative to support innovative medicines. The draft strategic research agenda² being developed for this project suggests that "thoughtful and proactive use of biomarkers can improve the mechanistic information generated in drug development, allowing a better understanding of the sources of variation and the correlation between discovery, preclinical, and clinical information." The result should, it says, be better early decision-making, thus reducing late-stage—and more costly—attrition of projects.

The Innovative Medicines Initiative defines biomarkers as "quantitative measures of biological effects that provide informative links between mechanism of action and clinical effectiveness," and claims they can provide "new insights into a drug's mechanism of action, metabolism, efficacy and/or safety and into disease mechanisms and disease course." They can be used as tools to understand the biology of a disease, to understand the effects of a new drug, and to provide information on patient subpopulations that might respond to a new drug or be susceptible to side effects. "The value of biomarkers is that they hold enormous potential to point us in the direction of critical information for developing better diagnostics and drugs, and for helping the industry to manage the innovation process in a more cost-effective manner."

In fact, such is the enthusiasm of this manifesto that biomarkers are depicted almost as the Holy Grail of future drug design. "The application of biomarkers in the drug development process will translate into such benefits as: increasing the probability of program success and reduced cycle times; matching patients with therapy; faster optimization of therapy; improved compliance with therapy; reduced complications of therapy and disease; more efficient drug development; more efficient health care delivery; and ultimately reduced societal health care burden," it says.

But—and there are still plenty of "buts"—"the vision will only be achieved if there is the right approach to optimization of biomarker investment, performance, and application," admits the IMI research strategy. Validation of biomarkers is a lengthy and expensive exercise involving many patients and years, and "there is no real consensus among all partners." The strategy paper recognizes the need for further refinement to the technologies such as improvements in resolution, sensitivity, and comparability, and a pressing need for greater access of patients to centers of excellence in imaging methods. "For this to happen, standards and registries of biomarker and clinical data will need to be agreed upon, and existing European-wide national networks will need to be coordinated."

EFGCP urges clarification

Biomarkers also reared their heads at the European Forum for Good Clinical Practice meeting in Brussels in late January. Fritz Buhler from the European Center of Pharmaceutical Medicine in Switzerland insisted on the need to construct mechanistic validity of biomarkers and model-based linkage with disease and pharmacological mechanisms and/or clinical endpoints. Ole Bjerrum of the European Federation of Pharmaceutical Sciences called for research into new biomarkers for characterizing chronic disease. Ian Ragan, in charge of European scientific affairs at Eli Lilly, urged greater acceptance of efficacy biomarkers in regulatory approvals and the establishment of a framework for biomarker development. And summing up a workshop session on reducing the bottleneck between preclinical and clinical development, Denis Lacombe of EORTC reported that one of the key topics to emerge was the use of biomarkers to allow more individualized treatment. However, as he pointed out, "biomarker" still has multiple meanings.

There's growing awareness that wider and sharper use of biomarkers may be crucial in constructing shorter and smaller, but more precise, clinical trials. Unfortunately, there's still not enough agreement about how to use them. And that may have to wait until Barroso is writing about biomarkers in his blog.

Agency's View of Biomarkers

"Biomarkers play an increasingly important role in the development of new drugs. It is expected that they will contribute to increase the rate of success of new developments and to expedite the development of drugs. Also, biomarkers are key in the shift away from the 'one size fits all' to 'the right drug at the right dose in the right patient' approach. Hence, biomarkers play an important role for scientists and industry in drug development and for regulators in the approval process."¹ —PO

Peter O'Donnell is a freelance journalist who specializes in European health affairs and is based in Brussels, Belgium.

References

1. Report on the 16 December 2005 EMEA/CHMP Biomarkers Workshop, 16 February 2006, European Medicines Agency, www.emea.eu.int.

2. The Innovative Medicines Initiative Strategic Research Agenda, European Commission, http://europa.eu.int/comm/research/fp6/index_en.cfm?p=1_innomed.