National Cancer Institute Public Health Genomics Priorities: Translating New Genomic Discoveries into Practice and Assessing their Impact on Providers, Patients and the Public's Health

Public health genomics is "a multidisciplinary field concerned with the effective and responsible applications of genome-based research for the benefit of population health." The NCI's Division of Cancer Control and Population Sciences, of which the Epidemiology and Genomics Research Program (EGRP) is a part, in collaboration with many partners from within NIH, and the CDC Office of Public Health GenomicsExternal Web Site Policy currently has a special emphasis on interdisciplinary translation research that assesses the impact of emerging genomic tools and tests on providers, patients and the public’s health.

For genomic applications to impact cancer care and prevention, we urgently need a coordinated effort involving multiple scientific disciplines (e.g., epidemiology, biostatistics, behavioral and social sciences, communication, clinical trials, health services research, economics). A number of research activities and products are needed for an evidence–based implementation of genomic and personalized medicine. These research priorities also will provide near term opportunities for employment in academia, public health and the private sector. The following is a list of near term public health genomics-related research priorities and a brief description of each.

The following areas represent scientific priorities for the consideration of investigators when thinking about proposals that would advance the field of Public Health Genomics (PHG). These descriptions are examples of research to further PHG efforts, and are in no way intended to be limited. Original investigator-initiated ideas are also welcomed.

  1. Development of prototype knowledge base for cancer genetic associations and gene-environment interaction
    Show More

    The support of genetic epidemiology consortia has been a priority activity for the Epidemiology and Genetic Research Program for the last decade. Recently, disease-based consortia have emerged as a powerful tool to support genomic epidemiology research. In most cases, large scale genome-wide association studies (GWAS) would not have been possible without the existence of consortia. This has held true for breast and prostate cancers (Cancer Genetic Markers of Susceptibility (CGEMS), Breast and Prostate Cancer Cohort Consortium (BPC3), and NCI Cohort Consortium), colon cancer (Colon Cancer Family Registry (C-CFR), Multiethnic Cohort (MEC)), lung Cancer (International Lung Cancer Consortium (ILCCO)), pancreas cancer (Pancreatic Cancer Genetic Epidemiology Consortium (PACGENE)), ovarian cancer (Ovarian Cancer Consortium (OCAC)) and melanoma (Genetic Epidemiology of Melanoma (GEM)). Moreover, genetic association in past studies of candidate genes or pathways and their interaction with environmental factors in cancer and other diseases have been systematically reviewed through meta-analyses, often supported through consortia, (Network of Networks). The integration and systematic review of the evidence generated by GWAS and candidate genes/pathway studies through meta-analyses can provide the essential underpinning for the design of the next generation of genomic epidemiology studies and for the prompt translation of such results into prevention and clinical practice. A concerted approach to conducting systematic reviews of genetic association and gene-gene and gene-environment interactions has been advocated by the global collaboration Human Genome Epidemiology Network (HuGENet).External Web Site Policy

    Such effort would leverage on the collaborative power of consortia and provide both systematic reviews of published and (if possible) unpublished evidence, but also continuously updated data synthesis. Pilot efforts are needed to focus on evidence of association for breast, prostate, lung and colon cancers to longitudinally integrate and periodically revise results from GWAS and candidate genes association studies. The complete set of field synopses will be maintained online by the HuGE navigatorExternal Web Site Policy and continuously updated, providing a critical resource for scientists and clinicians alike. This information will eventually be merged with the one derived from analogous knowledge synthesis efforts in related disciplinary domains, such as biology and clinical sciences.

  2. Evidence synthesis and decision support dissemination for promising candidate genomic applications for cancer prevention, management and prognosis
    Show More

    With the rapid growth in the field of genetic testing for cancer care and prevention, there are few genomic tools that have been evaluated regarding validity and utility for improving health outcomes. This information deficit is a significant barrier to the effective integration of emerging genomic tools into cancer practice. Evidence-based information on genetic tests informs payer and provider decisions regarding appropriate use of genetic tests in practice, and protects the public from unnecessary spending on emerging genetic tests of unproven validity and utility, an ever-increasing concern as the field of direct-to-consumer marketing of genetic testing takes hold. Of the 1400 available genetic tests, almost 100 have been identified as having potential for broad public health impact, yet evidence-based recommendation statements have been developed for only a handful. More than half of these tests are related to cancer care and prevention. Moreover, the integration into practice of the few evidence-based recommendations that do exist has been uneven at best. For example, the 2005 U.S. Preventive Services BRCA1/2 recommendations have not been well integrated into cancer control programs. Currently, technology assessments of genetic tests are conducted by a variety of entities, including academic institutions, health plans, and fee-for-service providers using varied evaluative methods. The Evaluation of Genomic Applications in Practice and PreventionExternal Web Site Policy (EGAPP) initiative was developed by the CDC in collaboration with NCI and other federal partners to establish an evidence-based process for the evaluation of genetic tests.

    The EGAPP initiative has already evaluated 4 genetic tests, 3 of them related to cancer control and prevention. EGAPP methods can now be disseminated and integrated into practice by providing short-term funding to the existing technology assessment infrastructure to adopt these methods. The resulting evidence-based assessments will summarize the state of the science for emerging genetic tests and provide decision support for providers and consumers as well as inform existing NCI, CDC, AHRQ and other NIH institutes genomics translation research and programs. There is currently a great need to conduct evidence-based reviews to assess the validity and utility of genetic tests related to cancer care and prevention. This can be achieved by integrating the established methods of CDC's EGAPP initiative into the existing technology assessment infrastructure.

  3. Population health genomics modeling
    Show More

    Various individual and consortia based projects have used simulation modeling to: (1) understand the impact of cancer control interventions (screening, prevention, treatment) on current and future trends in incidence and mortality, (2) extrapolate evidence from RCT's, epidemiologic, and observational studies to determine the most efficient and cost-effective strategies for implementing technologies in the population, and (3) be responsive to challenges due to the increased pace of technology, by helping to determine which new technologies are the most promising when scaled up to the population level. There is a crucial need to incorporate family history and genomic information into population modeling models to investigate the population impact of using family history and genetic testing to stratify the population to more appropriately target prevention, early detection and treatment strategies. For example prevention strategies could target a range of activities, increasing in costs, invasiveness and potential harms, from programs to modify risk factor behavior, screening programs customized to individual risk, chemo prevention to prophylactic procedures for those at highest risk. Similarly, treatment targeted to tumor characteristics and risk of progression can potentially provide the most appropriate treatment for each individual while minimizing side effects. Examples of this type of research include Ramsey et al. (Cancer Epid. Prev. Biomarkers, 2005), who evaluated the potential cost effectiveness of implementing different colorectal cancer screening guidelines for those with a family history of colorectal cancer, and Gorlova O. et al. (Human Heredity, 2003) who investigated the potential impact of using DNA repair capacity and bleomycin sensitivity in addition to smoking history to stratify the population for lung cancer screening regimens.

    To perform this type of analysis, models would incorporate family history/genomics information into existing models. This information may take several different forms depending on the current state of knowledge in each cancer site and the questions of interest (e.g. genetic mutations, risk assessment models that include family history of disease, characteristics of the tumor cells after disease is present). Model outputs could include mortality reductions and the cost-effectiveness of competing strategies. For situations where key information is missing, the analyses could focus on minimum characteristics (cost, population prevalence and penetrance of the allele), which would make a strategy for genetic testing cost effective, or could perform sensitivity analyses on unknown parameters. The outcomes of these efforts would include 1) Model revisions to include the genomic information most relevant for the cancer site and policy question being considered. 2) Model outputs that consider various prevention, screening or treatment options and predicts the benefits, harms and costs of the various approaches. 3) Publication of study results, including the implications for public policy and critical gaps in knowledge and studies that could be conducted to fill those gaps.

  4. Trans-disciplinary "teams of future" approaches to considering the role of genomics in disparities in cancer outcomes
    Show More

    New genomics discovery adds to the inherent complexity of cancer prevention and early detection research. Moreover, the intractability of some health disparities suggests that disciplinary-centric "research as usual" may be not meet the complex demands of these problems. Accordingly, NIH leaders have called for a teams-of-the-future (TOF) approach in which scientists from a broad array of disciplines (e.g., genetics, social and behavioral science, clinicians and epidemiologists) come together in trans-disciplinary teams that cross multiple levels of influence to conduct research to address specific health disparities. The TOF approach can be distinguished from other collaborative science by a process that includes: 1) demonstration of opportunities for structured and unstructured cross talk related to cancer disparities amongst scientists from a broad array of disciplines, 2) capability to develop trans-disciplinary heuristic models for characterizing a research approach to address the disparity, and 3) demonstrated potential of, short and long term, practical outcomes amenable to translation.

    For example, it has been suggested that vitamin D deficiency may be an important factor in health disparities, a problem that potentially could benefit from a TOF approach. Vitamin D deficiency has been implicated in a broad array of negative health outcomes including most common cancers (breast, colon, prostate). UV-B radiation is the major source of vitamin D. However skin pigmentation strongly influences vitamin D production because melanin is a highly effective filter against UV-B. Blacks have much higher rates of vitamin D deficiency than whites (e.g., among women 42.4% vs. 4.4%, respectively). Moreover, genetic factors (e.g., variants in vitamin D receptor genes) also play a role in vitamin D levels. These factors in combination with lower uptake of, and/or access to screening for early detection that has been reported among African Americans may contribute conjointly to negative cancer outcomes.

  5. Translation research and programs for genomic applications in cancer care and prevention
    Show More

    NCI, the CDC Office of Public Health Genomics and several NIH Institutes have recently initiated a new collaborative- the Genomic Applications in Practice and Prevention NetworkExternal Web Site Policy (GAPPNet), a coordinated collaboration to close the large chasm between gene discoveries and their successful applications in healthcare and disease prevention. Many of these applications are related to cancer care and prevention. GAPPNet features a collaboration of individuals and organizations funded by NCI, CDC and other NIH institutes. The types of projects include 1) translation research to evaluate validity, utility and impact of genomic applications in the real world and how to disseminate and implement recommended genomic applications; Translation research includes comparative effectiveness research (CER) that assesses effectiveness and cost-effectiveness of genomic and personalized cancer care and prevention compared to non-genomic approaches; 2) translation programs, clinical and community genomics activities that enhance practice, education, surveillance and policy development. These projects will accelerate and streamline the effective integration of validated genomic knowledge into the practice of medicine and public health.

  6. Using epidemiology consortia to translate genomic epidemiology information into promising applications
    Show More

    Substantial investments have been recently made in association studies to identify the genetic determinants of cancer. In addition, a new initiative currently in the implementation phase will support interdisciplinary team following up on GWAS studies through the continuum of post-GWAS research. Because of its character and scope, the almost totality of this research will be conducted by large interdisciplinary cancer consortia, including basic scientists, epidemiologists, clinician, behavioral scientists, geneticists etc. This approach will certainly provide a vast body of evidence on the genetic and environmental determinants of cancer and lay the pre-translational foundation to the first phase of translational research. However, recent analyses show that less than 2% of the grants currently supported by NCI address translational research topics. Now more than ever it is essential that basic genome-based discovery be promptly translates into candidate health application. Recent analyses also show that trans-disciplinary consortia flourish within the framework of population sciences, and that epidemiology consortia built infrastructure capable of supporting the rapid implementation of clinical and translational research and to effectively attract scientists in the relevant disciplines. Epidemiology consortia are needed to conduct translational projects including the full span of T1 to T4 research, from discovery to candidate health applications to the development of evidence-based guidelines to the implementation in health practice and evaluation of health impact. This research will build on current genomic epidemiology results and exploit the interdisciplinary teams and population infrastructure of the approximately numerous existing epidemiology consortia.

    Learn more about EGRP-supported cancer research consortia.

  7. Comparative effectiveness Research in Genomic Applications in Cancer Care and Prevention
    Show More

    Advances in human genomics and the recent progress in identifying susceptibility genes for a wide variety of cancers, are ushering in a new era of personalized cancer care and prevention. Using pharmacogenomic testing, we expect that cancer drugs could become tailored by genetic backgrounds minimize adverse effects and increase treatment effectiveness. Moreover, genetic stratification of cancer risk using biological markers such as genetic variants and protein markers are expected to increase early detection and primary prevention efforts. Several examples of genetic risk stratification for treatment and prevention are already available in breast cancer, colorectal cancer, prostate cancer and leukemias. Nevertheless, to date, there has been no systematic research conducted to compare the effectiveness and cost-effectiveness of cancer care and prevention based on genomic tools and markers compared to existing standards of care and prevention that do not use genetic stratification tools. Without such research, the promise of personalized medicine may not be fulfilled. We propose that NCI funds projects in comparative effectiveness research in genomics. Priorities should be given to clinical and population studies, both observational studies and embedded into clinical trials in the following areas: 1) comparing the clinical validity and utility of risk stratification tools and algorithms for cancer prediction based on genetic markers to existing algorithms that currently do not use genetic tools and markers; 2) comparing the clinical validity and utility of pharmacogenomic testing in cancer treatment and prophylaxis to existing treatments that do not use pharmacogenomic tests; 3) conducting cost-effectiveness analyses and other decision models to assess the added value of genetic stratification in cancer care and prevention, both from an individual and population perspectives. Comparative effectiveness research in genomics is a key first step in the translation of promising genomic applications to cancer care and prevention.

  8. Pharmacogenomics projects to identify and characterize validity and utility of genetic markers for cancer care and prevention
    Show More

    The rapid expansion in the use and development of new targeted cancer therapies, and the advances in the field of molecular biology and pharmacogenomics have recently motivated the development of a new research agenda at NCI. This research agenda focuses primarily on rapidly connecting discoveries in basic sciences with clinical and population sciences to enhance dissemination and impact of new discoveries in cancer therapies to the general population. Various research networks and database infrastructures now exist which contain population-based electronic health care delivery information. This information could be accessed and used to conduct targeted studies on the epidemiology and pharmacogenomics of anticancer drugs to assess responsiveness and toxicity in clinical practice after they have been evaluated in randomized clinical trials. However, certain components of these health information systems may be inadequate and may limit their utility for epidemiologic studies of cancer drugs aimed at identifying informative clinical, epidemiologic, and pharmacogenomic profiles.

    We are initially focusing on building research opportunities in pharmacogenomic epidemiology to address questions that cannot easily be answered using clinical trial data and investigating research infrastructure's feasibility and capacity to conduct rapid, cost-effective, and targeted observational studies necessary for translation of discoveries to the population. Examples of such questions include the genetic studies of effectiveness and adverse effects of oncology drugs when RCT samples may be inadequate; long-term response and/or long-term adverse events that occur beyond the follow up for clinical trials; or issues pertaining to the need to study more diverse populations (according to age at diagnosis, gender differences, race-ethnicity, co-morbidities, concomitant medications).

    Moreover, an ongoing and considerable challenge for the effectiveness of drug therapies is patient noncompliance. Addressing questions about whether and when matching drug therapies to patients' genetic profiles influences adherence to drug regimens also will be needed to establish the clinical utility of pharmacogenomics.

    Learn more about pharmacogenomics and pharmacoepidemiology.

  9. Training in Public Health Genomics
    Show More

    Training opportunities in public health genomics are needed in several schools of medicine and public health, as well as cancer control programs around the country. These opportunities will provide interns, fellows, and professionals with research experience as well as development of tools in genetic risk assessment and accurate interpretation of results from genetic and genomic tests and identify needs for referrals. Training involves analysis and interpretation of the evidentiary standards for genomic applications in cancer control and prevention, as well as designing and evaluating communication strategies for conveying genetic/genomic information. These opportunities would assure an essential pipeline for professionals in the emerging field of public health genomics to help translate genomic applications into cancer care and prevention in the next decade.

    Learn more about CDC training opportunities for genomics.External Web Site Policy

Return to Top