Genomics in Precision Medicine
Gulnaz Zaidi, Bioinformatics Scientist, Mibiome Therapeutics
Precision or targeted medicine is an approach that uses a patient’s genetic and lifestyle information to obtain appropriate disease treatment and prevention strategies. Genomics research has facilitated precision medicine by providing improved genetic testing as well as drug target discovery.
Precision medicine can be described as treatment options optimised to take into account the patient’s genetic information and lifestyle. Traditional model of drug development assumes that a drug will show a similar response in all patients afflicted with the same disease. But it is seen that most drugs show a positive response on a subset of the population. Another subset may develop adverse reactions to the drugs while others may not show any response at all. The underlying reason for this variability is the differences in genetic makeup between individuals. Precision medicine aims to understand and utilise these genetic differences to provide more effective therapies, which improve treatment outcome & prevent adverse effects while excluding the need of unnecessary treatments or diagnostic testing.
The first reference for precision medicine can be found in 1892 writing of Dr. William Osler “It is more important to know what kind of a patient the disease has, than to know what kind of a disease the patient has.” In more recent times, the term personalised medicine was first described in 1999 as performing a simple blood test to find out which patients will show positive response to a drug and which ones might show an adverse response. The basis of the diagnostic test would be the minute differences in the genetic makeup of individuals.
Applications of precision medicine
The field of oncology has been the earliest adapter of targeted treatment. The first drug prescribed on the basis of a genetic test was Trastuzumab, approved by USFDA in 1998, which is used for treatment of patients with metastatic breast cancer whose tumours overexpress the HER-2 protein. his was followed by regulatory approval of Imatinib, which inhibits the BCR-ABL protein tyrosine kinase, which is present in BCR-ABL gene fusion positive chronic myeloid leukaemia (CML). Later on, drugs targeting ALK, ROS1, BRAF V600E mutant melanoma and MET-mutant lung cancer were also developed. Multigene panels for diagnostic testing in several types of cancers have been approved by FDA.
In 2017, US Food and Drug Administration (FDA) approved the chimeric antigen receptor T-cell (CAR-T) for treatment of refractory pre-B cell acute lymphoblastic leukaemia and diffuse large B-cell lymphoma. Chimeric antigen receptors (CAR) are patient’s own T-cells that are engineered to express fusion proteins which directs the T-cells to cancer-specific antigens, causing destruction of cancer cells. In 2020, 28 targeted therapies were approved by the FDA in patient populations defined by specific molecular biomarkers.
Cystic fibrosis is caused by one of several defects in the CFTR gene. Majority of cystic fibrosis patients have F508del mutation. Ivacaftor, which works on 5 per cent of patients who carry the G551D mutation, is ineffective for the majority of patients, for whom a combination of Lumicaftor with Ivacaftor is prescribed. In 2021, Evinacumab was approved by USFDA as an add-on treatment with cholesterol lowering agents (for example, Statins) for homozygous familial hypercholesterolemia (HoFH). HoFH patients have two mutations in genes responsible for clearing excess cholesterol from the body. Evinacumab is an angiopoietin-like protein 3 (ANGPTL3) inhibitor. ANGPTL3 slows the function of certain enzymes that break down fats in the body.
Role of genomics in precision medicine
In precision medicine, sequencing and analysis of the patient’s genetic data play a crucial role in disease diagnosis and tailoring treatment. Technologies like Sanger sequencing, real-time PCR and Microarrays were the pioneering techniques utilised for DNA sequencing. Next generation sequencing technologies have enabled fast and efficient DNA sequencing, leading to more biomarkers being identified which can help to classify patients into different subtypes and provide treatment accordingly.
The process of determination of variants from sequencing data is called variant calling. The output of DNA sequencing is recorded in the form of reads, which contain the sequence of a DNA fragment along with the sequence quality scores of each nucleotide base. Any sequencing adaptors or bases with low quality are trimmed before using the reads for analysis. Trimmed reads are aligned to the Human reference genome (GRCh38 is the latest version), and the location in the reference genome where each read aligns is determined. Based on the pileup of all reads overlapping at each nucleotide position, the most likely genotype at that position is determined. Any difference in genotype as compared to the reference sequence is called a variant. There are many classes of variants like Single nucleotide polymorphism (SNP), Insertion/deletion, Structural variation, Copy number variation. All classes of variants can have distinct impacts on disease development and drug metabolism.
Targeted gene panels, which sequence a select set of genes which are known to be associated with the disease being investigated, are the most widely used in genomic testing. However, increas- ing reduction in sequencing costs have generated interest in the use of Whole Exome Sequencing (WES) and whole genome sequencing (WGS) for clinical diagnostic applications.
WES includes enrichment and sequencing of all the protein coding regions of the genome. Since 85 per cent of the disease-causing variants are located in the protein coding regions, WES covers most of the actionable regions of the genome to identify disease causing variants. WES has been used to discover genes associated with Mendelian as well as multigene disorders.
WGS provides the most comprehensive details about the genome. In WGS, a single test can provide the complete genome sequence, including non-coding regions. It also provides better determination of Copy number variations, Structural variants and chromosomal rearrangements as compared to WES. However, WGS has a higher cost of sequencing and data storage.
Whole genome sequence data aids in improved profiling of genetic variants. Genome wide association study (GWAS) is the study of genome-wide genetic variants in a set of individuals to determine if any variant is found to be associated with a trait or disease. In 2005, GWAS found two variants associated with Age-related macular degeneration. Since then, GWAS has identified associated variants for various diseases like coronary heart disease, obesity, type-2 diabetes and schizophrenia. Till date, GWAS have identified more than hundreds of thousand SNP-trait associations, which help to identify newer targets for targeted treatment.
Integration of genomic data with multiomics
Though every cell of an organism has identical DNA, each type of cell performs different functions. This difference in function is resultant of different gene expression patterns in each cell type. Similarly, gene expression can be different between diseased and healthy states. Also, epigenetic modifications (for example, histone modification, DNA methylation) also affect gene expression without changing the DNA sequence. Studying the effect of these changes provides additional layers of understanding about cellular mechanisms.
Apart from genome sequencing, genomics has also facilitated the generation of vast quantities of transcriptomic, epigenomic, proteomic, metabolomic and microbiome data. Integration of these multiomics approaches helps create a deeper understanding of disease mechanisms and opens the door for elucidation of innovative treatments.
Artificial Intelligence and Machine learning in PM
The process of identifying drug targets from large multiomics datasets needs sophisticated computational techniques which can analyse and identify patterns in the data. Artificial intelligence (AI) tools like machine learning, deep learning, natural language processing and network-based approaches are being increasingly used for analysis of genomic datasets. Development of Evinacumab as treatment for HoFH, is an example of AI being utilised to analyse genetic and biochemical information to identify the specific targets for cholesterol regulation. Other examples of AI optimised drugs are Pembrolizumab and Sotorasib, used for cancer treatment.
Gene therapy
The huge amount of available genomic data and advances in genome editing tools (like CRISPR) have also facilitated development of novel gene therapies for diseases with genomic basis, like cystic fibrosis, adenosine deaminase deficiency, familial hypercholesterolemia, cancer, and severe combined immunodeficiency (SCID) syndrome. Gene therapy aims to replace a faulty, mutation-containing gene with a normal, healthy gene within an affected individual’s genome, to cure a disease. Cell and gene therapy can be used for diseases where the causal gene is well known and well-characterized. In 2023, two cell-based gene therapies— Casgevy and Lyfegenia—received FDA approval for treating sickle cell disease.
Future perspectives
Research in precision medicine has huge potential benefits. The insights obtained from multiomics data can have a big impact on prevention, diagnosing and treatment of diseases. Numerous challenges are also evident in attainment of these goals. One of the biggest bottlenecks in implementing precision medicine is lack of genomic data from developing countries. Creation of variant databases catering to all ethnic subpopulations in the world is critical to wider usage of precision medicine. Development of decision support tools which empower physicians to use genetic data for treatment will be an important step. Concerns about data storage, privacy and willingness of individuals to undergo screening tests need to be addressed. Regulatory frameworks that prevent unethical use of medical and genomic data also need to be in place to safeguard patients’ rights. The confluence of AI and genomic data with Precision medicine will increasingly play a significant role in improving disease treatment. The use of advanced computational tools has the potential to generate useful insights from genomic research, which aid in precise and timely clinical decision making, leading to improved patient outcomes.
Author Bio
Gulnaz Zaidi is an experienced researcher in biotechnology industry, and is currently working as Bioinformatics Scientist at Mibiome Therapeutics, Mumbai, India. Her current passion is development of computational analysis workflows for high throughput sequencing genomic, epigenomic and metagenomics datasets.
