The value of large cohort studies in discerning the subtle etiology of disease is contingent on a variety of factors, perhaps most importantly, the study design. Participant sample numbers need to be high enough across all of the multiplicity of variables that will be examined. In addition, if these studies are to be used to to discover new biomarkers that could predict disease development, prognosis and response to specified treatment regimen, they must also incorporate collection of biospecimens from their participants. To move beyond discovery of disease associated biomarkers into early diagnosis and treatment, the new biomarkers must be validated both in the laboratory and in the clinic, and further studies requiring biospecimen collection must be initiated. Unfortunately, however, pre-diagnostic biospecimens of certain cancers and other rare diseases are especially difficult to find, thus finding very early predictors of disease frequently go undiscovered.
For this reason, all longitudinal cohort studies, ideally, should include collection of biospecimens from at least a subset of participants. However, as we all know, the issue is that large cohort studies have become prohibitively expensive, especially in the US, where there is no centralized national entity that can maximize efficiency. Is there a better way to conduct cohort studies and speed up the discovery of new biomarkers and development of therapies?
In 2010—four years ago—the NIH convened a symposium entitled New Models for Large Prospective Studies, to explore ways of increasing efficiency and maximizing value of these efforts, especially by looking at the UK Biobank and its collaboration with the UK’s National Health Service. The symposia highlighted the benefits of a national health system for conducting longitudinal epidemiological studies: recruitment of hundreds of thousands of participants in a short time is easier, as is access to health records and outcome data. Ensuring diversity in the study population is easier as well. Similarly, NCI's Epidemiology and Genomics Research Program (EGRP) sponsored a workshop entitled "Trends in 21st Century Epidemiology: From Scientific Discoveries to Population Health Impact" in December of 2012. The Workshop generated 12 sets of recommendations for Epidemiology over the next 12 years. The first set focused on Study Design, including developing research designs in the context of existing health care systems. [The second set of recommendations focused on the sharing of data, which will be discussed in part 2 of this blog post.]
There are numerous examples of various health care systems in the US conducting cohort studies, including the Marshfield Clinic Personalized Medicine Research Project (the Marshfield Epidemiology Study Area, or MESA) the Kaiser Permanente Research Program on Genes, Environment and Health (RPGEH),the California Teacher's Study, and the Golden Retriever Lifetime Study, all of which are collecting medical, environmental, and genetic information from participating members. Given that massive amounts of data (and biological samples) are already being collected within so many health care settings, is there a way in which those samples/patient visits/clinical information can be mined for discovery of biomarkers or validation?
Blood and other specimens drawn in the course of ordinary health care are generally discarded once a test is conducted. Researchers know that people can be eager to participate in research resulting in the healthy patient effect and thus we are missing an opportunity to collect pre-diagnostic specimens – the most valuable of all, perhaps, for the discovery of biomarkers.(These are not valuable UNLESS you have follow up samples from the same patient).
The Nation Cancer Institute’s Biorepositories and Biospecimen Research Branch (BBRB) is encouraging both clinical sites and patients to donate specimens. Acknowledging that clinical sites may not know how to approach patients about donating specimens, they created a resource to help them get started: a booklet for potential donors, in English and Spanish. The booklet can be used by healthcare providers in any setting to answer patient questions about donating specimens and what the specimens will be used for. The brochure also includes the broadest terms of consent in order to facilitate availability of resources across studies. The brochure can be downloaded and printed from the website and also be distributed by email. Up to 50 print copies (in English only) can also be ordered from the BBRB online. Please click here to download this brochure.
What does a clinical site do with these specimens? If you are reading this blog, then you already know that sample storage is easily outsourced. However, the BBRB invites clinical site staff to contact them for information about how specimens can be collected and used to further research.
The journey toward precision medicine, as well as a new taxonomy of disease depends heavily on data generated by genomic, other “omics” analyses, and molecular epidemiology, and this depends on biospecimens and the patient data associated with them. Let’s quit wasting opportunities!
A summary of the 2010 NIH symposium is available at https://nihroadmap.nih.gov/pdf/NewModels_Meeting Report.pdf; some of the overarching considerations that resulted from the symposium were reported by Manolio et al in the American Journal of Epidemiology (Manolio, T.; Weis, B.; Cowle, C.C.; Hoover, R.N. & Hudson, K. et al. (2012). New models for large prospective studies: Is there a better way? American Journal of Epidemiology, 175, pp. 859–66.) A summary of the 2012 Workshop on the future of Epidemiology is available at http://epi.grants.cancer.gov/workshops/century-trends/summary.pdf.
If you want to learn more about US conducted cohort studies, check our eBook on the California Teacher's Study, Next Generation Cohort Studies and Biobanking: How Cloud Technology is Accelerating Translational Research below.