The new genetics in clinical practiceBMJ 1998; 316 doi: https://doi.org/10.1136/bmj.316.7131.618 (Published 14 February 1998) Cite this as: BMJ 1998;316:618
- John Bell (), Nuffield professor of clinical medicine
- Nuffield Department of Clinical Medicine, University of Oxford, John Radcliffe Hospital, Oxford OX3 9DU
Common diseases are currently defined by their clinical appearance, with little reference to mechanism. Molecular genetics may provide the tools necessary to define diseases by their mechanisms. This is likely to have profound effects on clinical decisions such as choice of treatment and on our ability to characterise more clearly the course of disease and contributory environmental factors. This information also raises the possibility that new therapeutic interventions can be obtained rationally, based on a clear understanding of pathogenesis. Most of these genetic factors will act as “risk factors” and should be managed ethically and practically, as would other risk factors (in hypertension or hypercholesterolaemia, for example). The rapid advances in human molecular genetics seen over the past five years indicate that within the next decade genetic testing will be used widely for predictive testing in healthy people and for diagnosis and management of patients.
Molecular genetics was originally used in medicine to map and identify the major single gene disorders, such as cystic fibrosis1 and polycystic kidney disease.2 The excitement in the field has shifted to the elucidation of the genetic basis of the common diseases. With the help of very large, well characterised family collections, genetic linkages for many of the major causes of morbidity and mortality in Western populations have been identified. The genes and DNA variants responsible for these disorders are now being cloned at an ever increasing pace. Large scale genotyping, increasingly integrated genetic and expressed sequence maps,3 and large scale sequencing programmes4 have all contributed to this remarkable evolution in our understanding of how genes might modify our susceptibility to disease.
Considering the current rapid acquisition of genetic information relating to common disease and the dramatic technological developments that continue to fuel the field, it would be surprising if most of the major genetic factors involved in human disease were not defined over the next 5-10 years. This information will form an important template for redefining disease, clarifying biological mechanisms responsible for disease, and developing new treatment for most disorders.
Genetic information is likely to transform the practice of clinical medicine
Genetics will provide a taxonomy of disease that is based on biochemical mechanisms rather than phenotype
Genetic information will be used to identify individuals who are likely to respond to or suffer toxicity from drugs
Genetic variation will be another form of “risk factor” and will permit early treatment and directed screening
The rapid developments in human molecular genetics have often been underestimated, largely due to a failure to recognise the power of new technologies being applied to the problem. The use of information encoded within the genome for clinical practice has previously been limited by problems of scaling up accurate detection of DNA variation for rapid and inexpensive analysis. The problem will soon be resolved, perhaps by the use of oligonucleotide array technology or “chips.” 5 The ease with which this can be accomplished will determine how widespread DNA diagnostics will become, but there is little doubt that the problem is likely to be solved, technologically, in the near future.
The role of genes for susceptibility to disease has been emphasised in clinical medicine; it is now clear that this represents too narrow a perspective for the genetics of the future. Although such genes will be critical for redefining diseases and understanding their pathogenesis, equally important will be loci that determine disease progression, disease complications, and response to treatment.
A new taxonomy of disease
Perhaps the most important single contribution of the new genetics to health care is that it will create a biological rather than a phenotypic framework with which to categorise diseases. Clinical physiology and biochemistry have provided many insights into the biological disturbances that accompany disease, but it is genetics that is able to identify the pathways that are unambiguously involved in pathogenesis. Such genetic information will eventually lead to the redefinition of disease on the basis of biochemical events rather than phenotype; on molecular events driving biological processes rather than a correlation of clinical syndromes and outcomes.
The ability to redefine common human disease, using genetics to define the biochemical processes responsible for disease, will allow the subdivision of heterogeneous diseases such as hypertension or diabetes into discrete entities. Such subdivision is likely to help explain the wide variation of these diseases, including apparent differences in physiology, clinical course, and response to treatment, and it might also provide a basis for identifying environmental factors that contribute only to certain subtypes of disease. This has already begun in diabetes, where definition of the involvement of HLA genes suggested an immune mechanism in a subset of patients, leading to the subdivision into type I and type II diabetes.6 More recently, type II diabetes has been subdivided further on the basis of distinct mechanisms involving glucose phosphorylation and insulin (glucokinase) secretion, transcriptional regulation (HNF), and insulin receptor dysfunction.
A new taxonomy of disease: diabetes
Type I Autoimmune (HLA, INS) Type II Insulin resistant (INS receptor) Insulin secretion (glucokinase) Insulin transcription (?) (HNF1α, HNF4α, IPF)
Disease mechanisms have led to clear definitions of infectious diseases. For example, our understanding of hepatitis has progressed: it used to be viewed as a clinical syndrome with a wide variety of outcomes, and is now seen as a set of quite specific diseases defined by the aetiological agent, each with its own clinical course, prognosis, and (perhaps) response to treatment. An understanding of the biological process underlying the clinical phenotype has been of unquestionable benefit in defining and managing disease, and doctors are unlikely to attempt to manage a jaundiced patient with hepatitis without attempting to define the specific viral agent involved. Similarly, pharmaceutical companies are unlikely to attempt to develop novel vaccines or therapies without precise information about the disease type. Even in a well defined disease such as viral hepatitis, aspects of disease progression such as viral persistence will need to await genetic clarification.
Understanding the biological events and pathways identified by genetics as contributing to disease will lead to clear definition of disease. Such information may become the starting point in the management of most patients.
A new taxonomy of disease based on genetics is already being developed. The first examples of disease definition have come from the loci in common disease that seem to resemble autosomally inherited traits in families. Although these contribute to disease in only a small proportion of affected people, they provide considerable insights into disease mechanisms. Breast cancer (BRCA1, BRCA2), 7 8 colon cancer (FAP, HNPCC),9 and diabetes (MODY 1, 2 and 3)10-12 all have such highly penetrant loci, and their elucidation has provided some of the first insights into disease pathogenesis. The controversy over the potential role of impaired insulin secretion versus insulin resistance has been clarified by our understanding of mechanisms that result in each type of pathophysiology (glucokinase mutations versus insulin receptor mutations). Disease genes that contribute a component of susceptibility but require other genetic and environmental factors for disease to occur are also now available for disease definition. Apo E4 involvement in Alzheimer's disease is leading to revelations about its pathogenesis,13 while angiotensin converting enzyme14 and angiotensinogen15 probably contribute to different forms of cardiovascular disease in more predictable ways. The result of these developments is that we are beginning to move toward a refined taxonomy in medicine that is based on biochemical mechanisms and driven by genetics.
Genetic information in clinical practice
Early diagnosis, patient stratification, and improved management
With an increasing trend to focus healthcare resources so that they are most efficiently used, to develop accurate definition of disease to predict its clinical course, to target other forms of screening, and to choose optimal treatment, it is likely that genetic information will be an essential part of future clinical practice. Already it is possible to identify people at high risk of breast or colon cancer and to focus screening (such as mammography or colonoscopy) or early interventional treatment on these groups. In both breast and colon cancer we understand the genetic basis for about 5% of cases, a sufficiently large number of patients to overwhelm the already stretched genetic screening capacity in the United Kingdom. As we learn more about the effect of individual mutations on phenotype and as we identify more high frequency, low penetrance genes in both of these diseases, the pressure for screening in populations with and without symptoms will increase. Similarly, in diabetes, both the aetiological mutations (HNF, glucokinase) 10 11 and other loci (ACE)16 contribute to the course of the disease or the frequency of complications; hence these and other genes will be important prognostic indicators for those managing the disease and will need to be tested for. Even relatively simple management decisions regarding individuals at risk of deep venous thrombosis (patients with total hip replacement, or those taking the oral contraceptive pill) may benefit from evaluation of their factor V Leiden status.17 Decisions about the best treatment (CETP alleles and statins, 5'-lipooxygenase in asthma, or tacrine in Alzheimer's disease) or the side effects of drugs (cytochrome P-450 and flecanide) may rely on genetic stratification.18 These and many other indications for the use of genetic screening in patients with disease are likely to emerge in the coming years, and the pressure from patients and doctors for such services is likely to increase steadily.
Clinical benefit accruing from genetic studies of disease
A new taxonomy of disease based on mechanisms, not phenotype
New drugs developed rationally from our understanding of pathogenesis
Drug development and utilisation focused on disease subtypes likely to respond to treatment
Adverse effects of drugs avoided by genetic screening
“Risk factor” analysis will facilitate environmental modification, screening, and therapeutic management of people before they develop symptoms
Discovery and development of drugs
One of the earliest applications of this genetic information will be in the discovery and development of new drugs. Genetics is now widely used to identify new targets for drug designs, and it is increasingly recognised that defining disease populations by genotype will probably correlate with response to drug treatment. The variety of mechanisms that underlie complex disease may account for the wide variations in response seen in clinical practice and the difficulty often encountered in drug development of showing consistent large benefits in trial populations. Wise pharmaceutical companies are already introducing genotyping in their trials to predict response, and eventually this information will be needed to protect individuals from receiving a drug if they are unlikely to respond to it. Effort will also be focused on defining more clearly the basis for severe side effects of drugs and not giving them to people likely to experience side effects. Disease definition and drug response will go hand in hand, and lifelong treatment is unlikely unless an accurate genetic diagnosis provides an indication of response. Development of drugs along genetic guidelines will be a major force driving implementation of genetic screening by healthcare providers, as both response to treatment and complications will have been defined genetically for many new therapeutic agents.
Genes as risk factors
An indication of how important genetic information will be in defining disease and predicting outcomes in complex diseases can be gained from our knowledge of Apo E4 and Alzheimer's disease. Homozygosity for this allele is associated with a shift of about 20 years in the average age of onset of Alzheimer's disease.13 These effects are at least as great as other more conventional risk factors in common disease (such as hypertension in hypercholesterolaemia). Although current recommendations suggest that Apo E genotyping be used as an adjunct to diagnosis in cognitively impaired people, it is likely that genetic stratification by Apo E genotype will define drug response, and hence such genotyping may soon be applied in clinical trials and eventually will be more relevant to daily clinical practice.
Examples such as Apo E4 raise the question of whether a genetic susceptibility factor might best be treated as another “risk factor.” Other risk factors (blood pressure or cholesterol concentrations) show similar patterns of incomplete penetrance and have been considered for population screening. There is little reason that risk factors based on DNA should not be treated in the same way. Genetic factors that can be used to predict the risk of a population rather than an individual should be viewed in the same way as other risk factors, particularly if safe treatment or environmental modification were available.
This raises the possibility of population screening to detect important susceptibility loci when intervention becomes available. The obvious requirement for such screening would be validation by large scale trials on the benefits of such early detection and treatment. A combination of conventional and genetic risk factors may be optimal for identifying populations at risk. In hypertension or hypercholesterolaemia, risks vary greatly. Treating the extremes of variation has the most favourable cost benefit ratio, but most “at risk” patients fall within the normal range. Genetics could be used to identify those who have additional genetic risks and in whom reduction of these variables might be beneficial, even where such variables might be in the “normal” range. There are some trial data to support such an approach.19
The widespread redefinition of disease through genetics will be accompanied by the use of genetics for prediction and diagnosis and to optimise treatment in most common diseases. This is likely to occur within the next decade. Testing for genetic “risk factors,” even in people without symptoms, may develop (as it has for other risk factors), and this information may be used to identify people at increased risk, for early intervention. There is a possibility, however, that DNA diagnostics and pharmacogenomics will be used without proper evaluation—especially as few resources are available for rigorous evaluation and pressure continues to introduce this information in routine clinical practice.
Funding: Wellcome Fund.
Conflict of interest: JB sits as a non-executive member on the board of Oxagen, a genomic biotechnology company, but holds no equity.