Personalized medicine refers to the use of a patient's unique genetic signature to aid in making diagnoses, selecting medications, stratifying the risk of developing complications from interventions, choosing an appropriate preventative measure, or assisting with decisions regarding reproduction. This is of particular relevance in neurology.
Personalized medicine refers to the use of a patient’s unique genetic signature to aid in making diagnoses, selecting medications, stratifying the risk of developing complications from interventions, choosing an appropriate preventative measure, or assisting with decisions regarding reproduction. This is of particular relevance in neurology, as disease expression and response to therapy is invariably tied in some way to each person’s genome. Many neurologic disorders are being reclassified based on their underlying molecular biology. This emerging knowledge is providing neurologists with additional tools that allow for more accurate diagnosis, and in the near future, treatment options customized for an individual patient. Neurologists will need access to tools that recommend care based on frequently updated knowledge resources integrated with patient-specific clinical and genomic information. The electronic health record (EHR) will likely play an expanding and central role in the “genomic era.”
Recent advances in genomic medicine
The field of medical genomics (along with the other “-omics” such as proteomics, metabolomics, cytomics, etc) attempts to understand biological diversity at the level of the individual patient. The human genome and the less well-known HapMap projects have led to techniques that have greatly accelerated our ability to identify genetic factors related to disease. The human genome contains roughly three billion base pairs. Not counting mutations, individuals differ by about three million base pairs, and these variances are referred to as single nucleotide polymorphisms (SNPs). This genetic diversity is what makes us unique as individuals. However, specific patterns of SNPs aff ecting one or more genes are associated with tendencies (eg, height, intelligence), specific disorders, and responses to treatment. Many common disorders, such as hypertension and diabetes, are associated with polymorphisms in multiple genes. These are felt to act in concert, causing dysfunction through complex interactions between protein products and changes to how genes are expressed. It is hoped that the recognition of SNP patterns associated with diseases or responses to therapies will lead to treatments that can be customized at the level of the individual patient. For example, subtle changes to proteins caused by SNPs that code for proteins involved in receptor complexes could affect the efficacy of an anticonvulsant medication. Although mapping of an individual’s entire genome is not currently cost-eff ective or widely available, tests that target the most active coding regions called “genome wide association studies” allow for the relatively inexpensive identifi cation of SNP patterns in an individual’s DNA.
Application of genomic knowledge in neurology
The neurological literature is replete with reports of single gene mutations that are being associated with a rapidly growing number of disorders. This is forcing a reclassification of many disorders that, under the scrutiny of molecular analysis, are found to represent distinct diseases. For example, in the last 11 years, numerous breakthroughs in the genetics of the Parkinsonian syndromes have led to significant changes in our understanding of what was formerly felt to be a sporadic disease. Several gene mutations (eg, SNCA, Parkin, PINK1, DJ-1, LRRK2, and ATP13A2) have been causally linked to Parkinsonism symptoms in certain families.
Genetic tests will increasingly play a role in this disease and are already worthy of consideration for patients with the early onset of Parkinsonian symptoms or a strong family history. While this information is enriching our understanding of this condition, these mutations are only found in a small percentage of patients with Parkinsonism. Less dramatic changes (eg, SNPs) to the Parkinsonian genes listed above, particularly in concert, appear to play a role in the majority of “sporadic” forms of this disease. These findings may allow for early detection, early treatment, and potentially preventative strategies customized to an individual patient.
Although the significance of this finding is not clear, SNPs have been found in multiple sclerosis patients in genes associated with IL3 and IL7 production. As these interleukins are felt to play a role in the pathogenesis of this condition, this finding is being closely examined by researchers. Numerous other common neurologic conditions not typically felt to have a genetic origin are now being looked at closely for common SNP patterns. This is a promising direction that will lead to an improved understanding of the underlying pathophysiology of these disorders, and over time new methods of prevention and treatment.
Epigenetics: The role of environment on genetic expression
Polymorphisms appear to play a significant role in disease, response to therapy, and recovery from neuronal injury. These genetic patterns are also believed to make certain patients more susceptible to environmental factors. This has given rise to the field of epigenetics, which studies the influence of factors, such as chemicals and radiation, on gene structure and function. A process as basic as methylation of a single nucleotide at a critical locus within an intron can have major effects on the gene’s transcribed protein.
Epigenetic influences may have a significant role in the pathogenesis of multiple sclerosis (MS). For monozygotic twins, the risk that the second twin will develop MS once the first twin is affected is approximately 36%. For dizygotic twins the incidence is only about 4%. This suggests that there is a genetic predisposition for this disorder, but as 64% of monozygotic twins avoid developing MS, it would appear that exposure to specific environmental infl uences must play a significant role in the pathogenesis of this condition. MS may result from a combination of “nature and nurture,” caused by the misfortune of having a genetic susceptibility coupled with exposure to specific environmental factors. Molecular biologists and clinical geneticists are beginning to explore epigenetic influences in more detail, with the hope that some epigenetic triggers will be reversible.
The role of the consumer
Information on the risk of developing diseases based on an individual’s DNA has been made directly available to consumers through proprietary and unregulated sources. Th e accuracy of the data used to make these determinations has been highly suspect. Nonetheless, physicians are now being asked to confi rm or refute information presented to them by patients
concerned about their risk of developing certain diseases obtained from these sources. Scheuner, et al. (JAMA, 2008 Mar 19;299(11):1320-34) recently reviewed the literature to assess the state of clinician readiness for the era of personalized healthcare and found that clinicians are poorly prepared. Reliable genetic information that could be used to infl uence treatment decisions is appearing in scientific literature on a daily basis. One challenge will be determining how and when this information should be used to guide clinical management. Another will be determining the ways in which reliable clinical information specific to a given patient’s genome can be effectively presented to healthcare providers at the point of care.
A concern that has been raised is how genetic information can be protected once associations between specific genome patterns and disorders/tendencies become better established. For example, a genetic predisposition to developing amyotrophic lateral sclerosis could be used against a patient for employment, healthcare insurance, and even socially (eg, premarital spouse
“screenings” based on DNA profi les). While steps are being taken to prevent this information from being used against individuals (eg, the recently passed Genetic Information Nondiscrimination Act), ethical issues associated with the control of this knowledge are just beginning to surface.
Linking DNA databases together in evolving healthcare information exchanges will also lead to privacy considerations for related individuals in a community. For example, DNA testing for conditions not related to paternity yield an unsuspected (at least to the father and child) incidence of non-paternity of over 5% of tested samples. As one can imagine, this could be highly
disruptive to the structure of aff ected families. The genomic era will lead to an abundance of new information that may require a paradigm shift in how society deals with issues of privacy and the control of medical information.
The role of electronic health records in managing genomic information
The use of EHRs in the US has remained fairly limited, with less than 20% of clinicians using systems to assist them with patient care in ambulatory settings. EHRs have been widely promoted as tools that benefit patient safety through automated alerts and reminders (eg, when a patient on Coumadin is due for an INR), medication error prevention, tracking the status of all ordered tests, suggesting alternative diagnoses, providing context-specific and immediate access to authoritative references, immediate access to information from any location, analysis of the quality of care provided to patients, improved workflow efficiency, coding accuracy, and numerous other activities. Physicians in countries where EHR usage approaches 100% have difficulty imagining the practice of medicine without the use of these tools, but in the US, clinicians have thus far been reluctant to embrace this technology. The American Academy of Neurology has attempted to encourage EHR adoption through a formative evaluation of EHR products suitable for neurologists, and physician adoption has slowly been increasing (see www.aan.com/go/practice/electronic).
Employing an EHR may become a desideratum of healthcare even in the US, however, as genomics and its attendant exponential increase in clinical information enters the day-to-day practice of medicine. To illustrate this challenge. I present a hypothetical neurology consultation from the year 2020: A 62-year-old male presents to a neurologist with tremor and gait instability. The patient has previously provided informed consent to have his entire genome mapped, and this information is stored in a secure registry with access controlled by the patient via his personal health record. The patient grants the clinic access to his electronically stored genome and other health information (with the exception of his mental health history, which he elects to not share at that time). This information is processed by the EHR, which then runs rules-based and context-specifi c
queries that respond to new information obtained during the visit. None of the mutations associated with his symptoms are present; however, a pattern of single nucleotide polymorphisms known to be associated with Parkinsonian symptoms is identified.
The physician is also presented with the patient’s risk of developing numerous related and unrelated conditions (eg, NPH, glioblastoma multiforme, dementia, peripheral neuropathy, etc) based on his genome. Recommendations for diagnostic procedures, therapy, and preventive measures based on evidence-based medicine-supported guidelines are presented to the clinician for each condition. The EHR may also assess potential epigenetic influences, such as exposure to chemicals (eg, medications), underlying conditions, exposure to light, previous viral infections (eg, EBV), and other information. The system will recommend medications for the patient’s Parkinsonian symptoms that take into account the likelihood of a desired therapeutic response (eg, based on his unique receptor morphology) and avoid medications that may result in an adverse reaction—again based on the patient’s unique genetic profile.
The ability to provide care tailored to a patient’s unique genetic attributes holds significant promise toward improving patient care. Personalized medicine will allow for improved diagnostic accuracy, selection of treatments that will have greater efficacy and lower risk, and patient-specifi c preventive measures. The amount of information that physicians will be required to manage is likely to overwhelm the capacity of traditional paper-based practices. EHRs that can process, synthesize, and present clinically useful genomic information to clinicians will become essential tools in the era of personalized medicine.
Michael Stearns, MD, CPC, is president of e-MDs, Inc.