Assessing the current proteomics field

There has been much interest in the promise of proteomics to deliver biomarkers with utility for disease diagnosis and classification, and for assessing therapeutic efficacy and monitoring disease progression. However recently, particularly in the past year, expressions of concern have started to emerge regarding the paucity of protein biomarkers that have reached the stage of FDA approval or at least that have been sufficiently validated in independent studies. Therefore, a clear understanding of the current situation with respect to biomarkers and proteomics would be useful in assessing whether the field does hold promise and is ready to ‘deliver’ or whether effort should be focused elsewhere. The observations and derived conclusions presented here are intended to assess the current status of the field.

It is clear that there is currently no road map or agreed multi-step process to move biomarkers from discovery to validation analogous to the process of drug development. This situation is not limited to proteomics biomarkers but to the entire field of biomarkers. The process of biomarker clinical trials is not currently established like therapeutic clinical trials. The discovery process leading to those thousands of biomarker publications is often undertaken in an isolated laboratory and is loosely structured in a manner that does not lend itself to a seamless progression from discovery to validation. Furthermore, the discovery process is undertaken using platforms that are customised, as appropriate to address a particular objective. Additional effort that falls outside of the original scope of the research is required to develop the tools necessary to allow further testing of the biomarkers at the necessary scale to demonstrate utility.

Another issue of significance is that the proteomics field may be the victim of its own success given that there are too many reported biomarkers. The sheer number of promising biomarkers discovered through proteomics or other means creates an impediment to their further development as it is not easy to recognise those markers that have the greatest potential. Companies interested in further developing biomarkers are faced with a real dilemma as to which markers to invest in and therefore often require that markers have reached an advanced stage of validation and assay optimisation before they get involved. Thus, there is a gap between discovery and deployment that the private sector has become reluctant to fill.

An important related issue is that when discovered markers are scrutinised individually, they often do not exhibit sufficient sensitivity and specificity on their own to merit their further development. Transforming single markers into panels of markers is quite challenging. A marker with “modest credentials” may not be appealing to include in a panel, but such a marker may be an excellent complement if it is particularly informative for a subtype of disease or for predicting response to a particular therapy. However, this particular feature may not be easily ascertained. Issues pertaining to assay development and Intellectual Property may also impede assembly of a broad set of markers for their validation as a panel.

Proteomics technologies are evolving constantly. The field is quite dynamic and highly innovative with the technologies available being constantly under improvement. What is driving innovation is the need to increase throughput and sensitivity, which requires increased automation and miniaturisation. This trend is reflected, in part, in the development of mass spectrometry instrumentation with increased capabilities for protein identification and quantitative analysis, and for studies of post translational modifications. Therefore, given the likelihood that additional ‘better’ biomarkers will be discovered, the question must be asked; why invest limited available resources in the ones in hand? It follows that there may be hesitation in investing in the current generation of markers given the likely possibility that better markers and marker panels may come along.

From a pharmaceutical industry point of view, it is not clear what strategy to adopt. There are several options including one or more of the following:

• Actively engage in an internal biomarker development program to benefit drug development and clinical trials
• Farm out biomarker development to smaller biotech with the required expertise
• Do little for the time being and keep all options open

An open dialogue regarding the role of biomarkers and their acceptability as surrogates for monitoring response to therapeutic intervention among other, engaging private, public and government organisations should be helpful in guiding future progress. One such dialogue has already begun engaging the FDA, the NIH and the pharmaceutical and biotech industry as well as academia. Identifying approaches to fast track the process of discovery and validation, for example by including both discovery and validation objectives into clinical trials, would be highly beneficial given the rigour with which clinical trials are undertaken. Such a concept is gaining momentum. Opening up the vast specimen collections from past trials and from large cohort studies to discovery and validation studies which have been frozen away would also contribute to the fast tracking of biomarker development. Such specimens are largely kept untouched until biomarkers have reached an advanced stage of development, yet making them available for discovery would help ensure reduce biases and other limitations associated with the use of ‘ad hoc-’ type specimens. Demonstrations of success in the months to come are likely to emerge which would contribute to a substantial acceleration of progress in the field.

Sam Hanash MD, PhD

Molecular Diagnostics Program Head, Fred Hutchinson Cancer Research Center

Dr. Hanash is program head for Molecular Diagnostics at the Fred Hutchinson Cancer Research Center in Seattle, Washington. Dr. Hanash’s interests and expertise focus on the development and application of integrated approaches to the molecular profiling of cancer, with particular emphasis on proteomics. He has been a program principal investigator for several multi-investigator aimed at biomarker discovery and validation using proteomics. Dr. Hanash has organised and participated in several workshops related to diagnostics and proteomics. He is the inaugural president of the Human Proteome Organization which is involved in a number of scholarly activities, notably developing standards for the field of proteomics.