Genomics is the subject of a new Anticipatory Report “Predicting disease risk in populations in the era of personalized precision medicine” in the framework of the Observatory of Trends in the Medicine of the Future promoted by the Roche Institute Foundation.
The development of most common diseases, such as cancer or coronary pathologies, is usually due to multiple factors conditioning health or disease, which may be biological, environmental or exposure factors, lifestyle or socioeconomic factors.
In fact, these factors can be used to predict the risk of developing a certain disease through the creation of risk prediction models.
These are statistical analyzes that, in addition to being used to predict each person’s specific risk of suffering from a certain disease, allow the integration of knowledge about individual factors and characteristics with public health information to identify factors that condition the development of diseases at the population level.
According to the report, advances in the field of personalized precision medicine and the incorporation of new technologies, such as the omics sciences, have enabled the development and improvement of preventive medicine and precision public health.
The genomic and technological revolution of recent years favors having more information on the genetic basis of diseases, which is incorporated into the analyzes and thus being able to quantify the differences in risk between different people through the so-called Polygenic Risk Estimates or PRS, for the acronyms in English (Polygenic Risk Scores).
Contributions of genomics to risk prediction models
From a genetic point of view, diseases can be “mendelian” (those that are determined by an alteration in genes of high penetrance, i.e. alterations that confer a high probability of suffering from the disease), or complex (whose development is conditioned by the presence of multiple genetic variations and/or other non-genetic factors).
The detection of Mendelian diseases can be performed in routine clinical practice through genetic diagnostic tests for the identification of mutations that cause pathologies or that increase the risk of suffering from a disease.
However, in most common diseases the “Mendelian” component explains only part of the heredity.
These cases are mainly due to genetic variations that are relatively frequent in the population, of low or medium penetrance and that confer a low risk individually, but that together with the effect of other genes can be considerable, explains the report on genomics
In fact, most common diseases have a Mendelian component, with genetic variations of high penetrance, and a more complex component with genes of low penetrance that, together and due to the interaction with the environment, confer a variable risk of disease so, for the detection of complex diseases, other factors must be taken into account.
In this sense, family genetic studies make it possible to establish the degree to which a specific disease is genetically determined (that is, its heritability).
These studies have made it possible to identify specific mutations associated with pathologies such as, for example, mutations in the BCRA1 and BCRA2 genes associated with breast cancer and other tumors.
Mutation of these genes in a person’s genome has been shown to significantly increase their predisposition to breast cancer, thus constituting a risk factor for developing the disease.
In fact, routine genetic testing is used in clinical practice for the detection of these mutations in families for the early detection and clinical management of high-risk individuals.
However, these mutations are very rare and highly penetrant, and most breast cancer patients do not have these types of mutations, but the development of breast cancer is influenced by a combination of other factors.
On the other hand, to learn about common variants with low penetrance, genome-wide association studies or GWAS, for short in English (Genome Wide Association Studies), are used, which allow the identification of multiple variants such as, for example, polymorphisms in a single nucleotide or SNPsm, for the acronym in English (Single Nucleotide Polymorphism), associated with the risk of developing diseases.
In fact, this type of study in which the genomes of individuals are compared has made it possible to identify variants present more frequently in people with a given disease.
For example, in the case of coronary heart disease, more than 300 genomic variants have been identified that are present more in people with this pathology.
This information on genetic variants can be incorporated into the development of risk prediction models and contribute to improving their effectiveness in estimating the risk of polygenic diseases.
The recommendations of the experts
The Anticipant report of the Roche Institute Foundation on genomics includes a series of recommendations relating to disease prediction models:
• Promote study in omic sciences. In this sense, the exposome and other modifiable conditioning factors of health or disease, among others omics, can contribute to improving the development of prediction models.
• Create integrated databases of information on health and disease determinants: Sharing genetic information and other determinants of health would be useful for the development of prediction models.
• Generate validated models in different populations: Essential, not only for the effectiveness of the models, but also to promote equity in use and application in different populations.
• Establish standards for the generation of risk prediction models: It is advisable to promote initiatives aimed at establishing well-defined criteria in order to reduce variability and facilitate their implementation.
• Evaluate the efficiency of the models in the relevant health context. With the aim of incorporating risk prediction models into real clinical practice in Spain and contributing to the sustainability of the system, it is important to generate economic data that will allow evidence of the advantages of its use from the point of view of ‘efficiency in the current health environment and its potential incorporation into the common portfolio of services of the National Health System.
• Increase training in the field of risk prediction. Training of healthcare professionals in the operation of risk prediction models, their limitations and implications.
• Raise public awareness of the implications of using genetic information. It is important not to generate excessive expectations in the population about the predictive capacity of predictive models and to convey their results correctly.