Stephen C. Peiper, M.D., Peter A. Herbut Professor & Chair, Thomas Jefferson University
Erica S. Johnson, Ph.D,. Associate Professor, Thomas Jefferson University
Healthcare in the United States is currently in the throes of radical change after years of organizational stasis and disheartening metrics. Our healthcare system was ranked 37th among 191 member states for health outcomes, responsiveness, and fairness of distribution by the World Health Organization in 2000, and 31st in a group of 51 for efficiency and life expectancy by Bloomberg in 2014. Recently, however, diverse forces started to converge, leading to forecasts of a “perfect storm” of healthcare reinvention.
These forces include 1) an emphasis on value and outcomes for quality of care and patient safety; 2) a shift from reactive disease management to health promotion; 3) a changing compensation model; 4) access to universal health insurance coverage; and 5) availability of state-of-the-art diagnostic technologies and cutting-edge therapies.
Many models have been put forward for improving U.S. healthcare, but the most elegant conceptual framework is P4 medicine, or predictive, preventive, personalized, and participatory medicine. The term was coined by Leroy Hood, M.D., Ph.D., the creator of the technological foundation for modern genomics and an advocate for the application of systems biology to medicine.
The predictive and personalized aspects of P4 medicine are fundamentally linked to diagnostic analyses that employ technologies with the highest power of resolution to expose mechanisms driving diseases.
Support for Development Efforts
To enhance the impact of our current technological armamentarium on the health of our population, President Obama has committed to fund the development of the following:
- A cohort of at least one million volunteers to advance insight into health and disease, and to promote research on active participants that includes data sharing ($130 million to the National Institutes of Health)
- Increased research to identify genomic mechanisms that drive malignant diseases and generate therapies to these molecular targets ($70 million to the National Cancer Institute)
- Highly annotated databases that will transform and enhance personalized/precision medicine and public health ($10 million to the Food and Drug Administration)
- Measures to address standards and requirements for the secure transmission of data ($5 million to the Office of the National Coordinator for Health Information Technology)
The Role of NGS
The application of genomic technologies in clinical diagnostics, most notably next-generation sequencing (NGS), has made the greatest impact on the diagnosis and treatment of cancer patients. The efficacy of drugs designed against targets known to function as drivers of the malignant process has been made possible by companion diagnostic tests that determine whether the molecular target has been activated, and therefore whether the patient is eligible for this therapy.
Since NGS can simultaneously analyze thousands of regions distributed in hundreds of genes, it represents an ideal technique to interrogate tumors for candidate mutations that can serve as a driver of malignant behavior. For example, NGS has been used to screen patients for participation in a new type of clinical trial, a “basket trial.” Here, eligibility for a new drug is based on the detection of activation of the corresponding molecular target instead of the traditional criteria of primary site and pathologic stage.
This sort of “genomically informed medicine,” however, still faces multiple challenges. Some of these challenges are logistical; others are more basic. At this time, there are no NGS oncology diagnostic tests that have secured premarket approval (PMA) or clearance (510(k)) from the Food and Drug Administration (FDA), and there are efforts to more closely regulate the performance of laboratory-developed NGS tests in oncology.
Indeed, one of the goals of the Precision Medicine Initiative is to modernize the regulatory landscape for NGS testing. The current menu of oncology tests approved by the FDA includes testing for HER2 amplification; detection of mutations in KRAS, EGFR, ALK, and BRAF genes; and analysis of prognostic gene-expression profiles in breast cancer. It seems unavoidable that the individual companion diagnostic (gene-by-gene approach) will be replaced by NGS analysis of extended gene panels. In addition to the absence of FDA-approved/cleared tests to determine the mutational signature of tumors, there are limited reagents available to perform proficiency testing that would satisfy the Clinical Laboratory Improvement Amendments, that is, the CLIA regulations.
The detection of somatic driver mutations as molecular targets for therapy makes multiple assumptions that could have an impact on the efficacy of individual targeted therapies. For example, it is not clear that the biology of a tumor is dominated by a single driver mutation or whether driver mutations behave the same way in different tumor types.
It is clear from early work on carcinogenesis in colonic polyps that malignant transformation is a multistep process resulting from the accumulation of somatic mutations, and that resistance mechanisms which develop following targeted therapy may utilize collateral signaling pathways or result from additional somatic mutations in the driver gene. Thus, the biology of tumors is not binary. A systems medicine approach may be an important consideration for a process that may include multiple abnormalities that contribute to tumor biology, such as the effects of modifier mutations on recognized driver mutations.
Operational challenges include optimizing features of “good genomic practice.” These features include the analysis of small, noninvasive specimens; the preanalytical handling of specimens; the identification of proper areas of tumor tissue for analysis; the validation of instruments and software; the use of public databases; the inspection of data for “medical grade” sequencing; and data storage.
In addition, it is critical to consider ethical and legal issues including the discovery of unexpected findings and the reporting of unanticipated germline abnormalities to patients and family members. The importance of the confidentiality of genomic information is illustrated by a recent incident in Idaho where police obtained DNA data from multiple individuals (without a search warrant) from a website that analyzes ancestry. Forensic analysis revealed a match at 35 of 36 loci with evidential DNA from a murder case. While the implicated individual was exculpated early in the process, this incident raises awareness of the power of genomic analysis and the importance of providing appropriate procedures for confidentiality.
Moving forward, while it is clear the vanguard of clinical genomics will be to detect somatic mutations in oncologic disorders using NGS, some mutation-specific immunohistochemical tests—such as those using monoclonal antibodies to BRAF (V600E), EGFR (L858R), EGFR with deletions in exon 19, and ALK—will be invaluable for the analysis of small tissue samples from minimally invasive procedures. In the event that sufficient DNA cannot be derived from the tissue specimen, it will still be possible to morphologically identify malignant cells and determine whether a mutation is present. In addition, expression assays that measure the composite phenotype of malignant tumors, such as Prosigna Breast Cancer Prognostic Gene Signature Assay (PAM50), will provide prognostic evidence that guides the selection of the best possible therapy.
In addition to cancer, multiple medical specialties will be fertile ground for genomic analysis. Whole-genome sequencing was successfully used at the Medical College of Wisconsin to modify a child’s therapy in 2010, and now numerous pediatric hospitals have implemented whole-exome sequencing or whole-genome sequencing for dysmorphic neonates. This analysis may provide important information to the family regarding the presence, or absence, of inherited germline mutations.
Collection of ever larger sets of family exome and genome sequences, as proposed in the Personalized Medicine Initiative, will also allow the elucidation of multigenic traits, including those associated with important health threats such as heart disease and diabetes. Since there is evidence that an individual’s microbiome may have a significant impact on multiple diseases, including obesity, metagenomic analysis may also provide indications for appropriate therapies, including fecal transplantation.
Genomic information will play a key, but not exclusive, role in personalized and predictive medicine. In addition to the healthcare team, it is likely that stakeholders such as health and life insurance companies will investigate the ability to make predictions based on genomic analyses. The architecture of genomically informed medical diagnostics and therapeutics has not yet been established, and appropriate reimbursement mechanisms are unclear.
Thus, not unlike the Spanish explorer Alvar Nunez Cabeza de Vaca—“The Gasser” in the quotation referenced at the beginning of this article—we do not know exactly where we stand. It is certain that the field will advance from the current state with genomic data central to the evolution of our understanding of many diseases, either as a primary factor in the individual or as a susceptibility to their environment.
Stephen C. Peiper, M.D. ([email protected]), is Peter A. Herbut Professor & Chair and Erica S. Johnson, Ph.D., is Associate Professor in the Department of Pathology, Anatomy & Cell Biology at Thomas Jefferson University.
This article was originally published in the July 2015 issue of Clinical OMICs. For more content like this and details on how to get a free subscription to this digital publication, go to www.clinicalomics.com.