Intrapatient Multi-omics: Defining the New Standards in Medicine
by Wadie D. Mahauad-Fernandez, PhD | August 6, 2020
In 400 BC, Hippocrates promoted the medical notion of “treating the patient who has a disease rather than the disease who has the patient.” Centuries later, in the late 1800s, Sir William Osler defined “the good physician [as one who] treats the disease, [and] the great physician [as one who] treats the patient who has the disease.” What both Hippocrates and Osler understood was that medicine should consider the context of a patient’s illness in order to find the best treatment.
This concept has existed for centuries, though it has reemerged with new technological advances leading the scientific and medical communities to question whether health management approaches should be treating the patient, rather than the disease itself. The current model of medicine, which is based on the treatment of diseases via a “one-size-fits-all” approach, is outdated, inefficient, costly, and in some cases harmful. New technological and scientific discoveries have demonstrated that diseases are extremely heterogeneous and vary from one patient to another.
The form of health management that relies on making patient-specific (intrapatient) diagnostic and treatment decisions based on the patient’s background, clinical history, molecular data, and environmental data is defined as precision or personalized medicine. Over time, the scientific and medical communities have embraced the diagnostic and therapeutic potential of personalized medicine. As such, in 2018, more than 40% of FDA-approved drugs were cataloged as personalized medicines, defined as “therapeutic products for which the label includes reference to specific biological markers, identified by diagnostic tools, that help guide decisions and/or procedures for their use in individual patients.” Personalized medicine is beginning to improve patient outcomes in particular diseases, though it still has many hurdles to overcome before it can be fully incorporated into the health system.
Multi-omics Techniques and Market Value
To make personalized medicine more effective, a set of patient-specific criteria must be defined. These criteria can be defined by multi-omics approaches—high-throughput techniques used to characterize global changes in the status and composition of the genome, transcriptome, epigenome, proteome, metabolome, and microbiome (Table 1). These “omes” paint a unique molecular picture of an individual from which medical decisions can be made.
Table 1. High-throughput -omics techniques employed or under development for clinical use.
|Genome: Entire DNA and gene set from a cell, tissue, or organism.||Genomics: sequences DNA and analyzes the structure of genomes to identify specific mutations, translocations, or copy number variants in a host’s genome.||Novartis|
|Transcriptome: Collection of all gene read-outs/transcripts within a cell, tissue, or organism.||Transcriptomics: sequences complementary DNA generated from RNA to identify changes in gene expression.||10x Genomics|
|Epigenome: All chemical compounds and modifications added to a genome that regulate gene expression in a cell, tissue, or organism.||Epigenomics: identifies the localization of transcription factors (TF) or proteins attached to DNA that modify genomic accessibility (DNA methylation, TF binding, histone modification, among others).||Evosep|
|Proteome: Collection and levels of proteins produced by a cell, tissue, or organism.||Proteomics: purifies and separates proteins using unique characteristics (mass and retention time) to pinpoint changes in protein levels.||Oryzon|
|Metabolome: Collection of metabolites (low molecular weight molecules) produced by a cell, tissue, or organism.||Metabolomics: Like proteomics, metabolites are purified and separated to measure differences in composition between samples for key readouts of metabolic function.||Metabolon|
|Microbiome: Collection of genetic material within the microbiota living in a tissue, organism, or environment.||Metagenomics: microbial DNA is sequenced to identify bacterial species and microbial genes to define microbial populations in different samples.||Novome|
A single -omics technique can point to a biological characteristic that is associated with a particular disease (e.g. 2D picture), while multi-omics allows for visualization of biological characteristics at multiple levels, granting a more complete picture of the systems driving a disease phenotype (e.g. 3D picture). Generally, these techniques are applied to a bulk sample—a piece of tissue/bodily fluid in which every cell within this tissue is processed together. However, with the emergence of single cell analysis technologies, scientists are now able to apply -omics techniques to single cells within a sample, increasing data resolution and allowing them to pinpoint cell specific differences and study disease heterogeneity.
The impact of single cell multi-omics for personalized medicine is immense and has the potential to reshape the medical market. In 2018, the single cell multi-omics market was valued at $1.83 billion and is projected to reach $5.32 billion by 2025. On a wider scale, the global personalized medicine market, estimated at $92.4 billion in 2017, is projected to reach $194.4 billion by 2024. The diagnostic and therapeutic potential of multi-omics personalized medicine is undeniable, but is it feasible to implement this new modality into the health system?
Scientific and Technological Advancements in Personalized Medicine
Recent scientific and technological advances towards personalized medicine have been enormous. The biopharmaceutical and biotechnology industry, along with academic institutions, are developing new biological approaches and therapies for precision medicine. Examples include CRISPR-based gene editing, using nanoparticles and viral vehicles to deliver genetic material to correct genetic aberrations, approaches to reshape the microbiome and the metabolome, generating patient-derived xenografts and animal models that resemble human diseases, and developing autologous and allogenic immunotherapies, among others.
Successful personalized medicine therapeutics based on genomics data include long-term treatments such as Herceptin to treat HER-2+ breast cancers and Zelboraf to treat melanomas with BRAF mutations. Recently, Spark therapeutics received approval for Luxturna to correct RPE65 gene mutations associated with retinitis pigmentosa and Leber’s congenital amaurosis. Likewise, Novartis received approval for Zolgensma to correct a defective/missing SMN1 gene to treat spinal muscular atrophy for which there were minimal treatment options. Both Luxturna and Zolgensma are next-generation one-time targeted therapies with the potential to have lifelong effects that demonstrate the emerging power of personalized medicine approaches.
These scientific efforts in personalized medicine are accompanied by a remarkable technological advancement characterized by improved data accessibility, analysis, visualization, storage, and integration. These technological advances will assist health providers in the interpretation of multi-omics data and in the use of Electronic Health Records (EHR) to make accurate medical decisions. For example, Persistent is developing Persistent Systems Multi-omics to systematically analyze heterogenous multi-omics data, facilitating its integration and visualization to better understand complex biological systems, Qiagen’s OmicSoft sequencing analysis suite analyzes and compares multi-omics data across thousands of curated datasets, and M2Gen develops software to integrate a patient’s clinical and molecular data. Another example is a program led by clinical artificial intelligence (AI) expert Dr. Beau Norgeot to develop a user-friendly and interpretable learning healthcare system using EHRs from millions of patients. It is designed to be used by caregivers for the diagnosis and treatment of molecularly similar diseases in new patients. These and other clinical and -omics datasets are pivotal in the development of machine-learning and other AI approaches to create digital programs to integrate, analyze, and interpret multi-omics data.
Fundamental to implementation of these tools is training physicians in their use. Dr. Michael Snyder, an expert in personalized medicine at Stanford University, explains “we need to change the curriculum of medical schools to include courses on genomics, integration of large datasets, and the use of -omics and other big data in the clinic”. The current medical school model relies heavily on shaping the thinking of doctors-to-be in terms of diseases and not the patient’s environmental and molecular imprints that will define disease phenotypes and treatments. Genetics and bioinformatics should become core competency in medical schools’ curricula, and career tracks for physicians to specialize in medical genetics and multi-omics data analysis should be established.
The Government’s Role in Enabling Personalized Medicine
Government investments in healthcare infrastructure to meet personalized medicine needs and incentives for all stakeholders involved are necessary to move the field forward and fast-track its incorporation into the clinic. In 2015, the Administration launched the Precision Medicine Initiative and All of Us Research Program with the goal of “[enabling] a new era of medicine through research, technology, and policies that empower patients, researchers, and providers to work together toward development of individualized care.” This led to many of the aforementioned scientific and technologic advances and to the development of initiatives by the National Human Genome Research Institute including the Clinical Sequencing Evidence-Generating Research (CSER) consortium, Implementing Genomics in Practice (IGNITE), and the Electronic Medical Records and Genomics (eMERGE) Network, which allows patients to share and update their environmental and background data.
In 2018, the new Administration aimed at cutting the National Institutes of Health budget by $5.8 billion. This was the largest cut proposed from the $15 billion budget reduction to the Department of Health and Human Services. In this front, Dr. Henri Michael von Blanquet, founder of the Precision Medicine Alliance, stated that: “[The] healthcare system in the US is very fragmented with fragmented interests. The main handicap today is that national economic systems are not incentivizing better patient outcomes.” Government budget cuts set a barrier against the Precision Medicine Initiative and crushed the momentum of all precision medicine stakeholders. In addition to Government cuts, other financial, social, and insurance obstacles must be overcome to integrate precision medicine into the health system.
Dealing with High Costs and Insurance for Personalized Medicine
High treatment costs may be the biggest hurdle for personalized medicine to overcome. Zolgensma is estimated at $2.1 million per treatment, Luxturna at $425,000 per eye, and Bluebird’s LentiGlobin at $2.1 million. These high costs originate from drug development, production expenses, and the high manufacturing prices involved in making drugs for small patient populations. Undoubtably, for the average insured US resident who makes $61,937 annually, paying any fraction of the total cost for personalized therapeutics is challenging. Currently, insurance companies are moving towards value-based contracting for FDA-approved personalized medicines, where they only pay if treatment works and if the cost of the cure (personalized medicine) is less than the cost of care (life-long treatments).
Covering and reimbursing such expensive treatments poses financial challenges to insurance companies that may not fully realize their investment return, given that this occurs over several years and patients are unlikely to stay with the initial payer for >3-5 years. Thus, there must be legislative incentives, relief plans, and restructuring of reimbursement plans to support insurance companies. Everett Crosland, a Digital Therapeutics Executive at AppliedVR, suggests the need to i) develop a payment system where the primary insurance company pays for a therapeutic with subsequent plans, similarly paying for the incremental value realized on an annual basis from the treatment, and ii) allow more dynamic premium rate setting to the beneficiary pool, which would enable cost sharing when used within the pool that year, but force premiums back down if a treatment is not used in a given year.
The problem then becomes whether a personalized medicine risk pool should be used. Currently, there is limited data on the long-term clinical and cost value of personalized medicine and on how that will apply to the general public. Thus, business and financial risk analyses must be performed to accurately show the value of personalized medicine.
Educating the General Public on the Value of Personalized Medicine
Another obstacle to incorporating personalized medicine in health systems is acceptance. As most individuals are not willing to pay extra insurance premiums, it is imperative to convince the general public that adding this premium to their insurance will provide a long-term benefit. First, there is a need to educate the general public on how personalized medicine will positively reshape their lifestyle and reduce health costs. Dr. Snyder explains that “there is no financial system to incentivize staying healthy.” Dr. Snyder proposes the use of a health-tracker for baseline physiology information, genome sequencing, and a whole-body MRI as the minimum to define a patient’s baseline for personalized medicine. Second, there is a need for factual evidence that using personalized medicine will decrease health costs in the long-run since these patients will not receive life-long treatments, ultimately relieving the long-term financial load on the health system creating value-based healthcare. Third, the public should be assured that their personal information is secured and will only be used for medical and scientific purposes. This should coincide with the government implementing guidelines and regulations for the use of patient data.
Personalized medicine is no longer a utopic dream. It is an attainable reality. All stakeholders involved—patients, physicians, scientists, payers, lawmakers, educators, AI developers, etc.—must work together to develop an organized and workable roadmap to integrate this new model of medicine into a reformed health system. Finding this middle-ground where personalized medicine can be optimally integrated and beneficial for all stakeholders seems more utopic than personalized medicine itself. This integration will unquestionably come with complications and will require sustained discussions, energy, and effort. Most importantly, it will take courage from all stakeholders to forgo the comfort of the familiar but inefficient health system for one that will positively revolutionize medicine. Stakeholders should work together to make intrapatient multi-omics and personalized medicine the new normal, shifting the health system to care for the patient who has a disease.
Wadie D. Mahauad-Fernandez, PhD. is a Postdoctoral Research Scholar in the Department of Medicine at the Stanford University School of Medicine, Stanford, California.