How Technology, Big Data, and Systems Approaches Are Transforming Medicine.
Medicine in those days (and even today) was very much like the fable of the elephant and the six blind men--each of the men feels a different part of the elephant and each describes the animal distinctly, as a fan or a stump or a spear. Physicians at that time and today were only able to access a few of the external dimensions that characterized what the elephant/human really was. We did not have a language to think about the complexity of human biology. Later, systems biology allowed us to define and think about these complexities; we could then begin developing systems-driven technologies, systems-driven strategies, and systems-driven computation platforms and algorithms for unraveling this complexity.
In 1973, I read The Structure of Scientific Revolutions by Thomas Kuhn (1970), which describes paradigm changes in physics. Paradigm changes--revolutionary new ways of thinking about or practicing a discipline--are really difficult to conceptualize and even more difficult to achieve. Even if you validate a paradigm shift, it is not always easy to get it accepted, because people are reluctant to give up long-held beliefs. Scientists are extremely conservative with regard to the acceptance of new ideas.
Over my career, I've been involved in seven paradigm changes in biology and medicine--and I'm working on the seeds of an eighth. All of them deal with human complexity. Collectively, they have framed the emergence of 21st-century medicine. The focus of 21st-century medicine will be the eighth paradigm change: bringing systems biology to health care through deep phenotyping of individuals and scientific wellness. I use the term "we" throughout this discussion, as I had many colleagues who helped with every phase of these paradigm changes and my later career.
Seven Paradigm Changes
1. Bringing engineering to biology
The first paradigm change we initiated was bringing engineering to biology in the early 1970s. For the next 25 years or so, we worked on developing instruments that let us analyze DNA and proteins in various ways, including automated DNA and protein sequencers and DNA and peptide synthesizers. These four instruments led us, in the early 1980s, to an integrated view of microchemical facilities. The idea was that, if you could sequence a protein, you could translate the protein sequence into a DNA sequence (via the genetic code dictionary) and synthesize a complementary DNA fragment, which could then be used to clone the corresponding gene. The gene could be sequenced by the DNA sequencer. In that way, you could very rapidly go from proteins to genes. This approach opened up a series of new biological fields in the early and mid-1980s (I remember one issue of the journal Science in which we had four papers on the cloning of new receptor genes). I was a cofounder of the company Applied Biosystems in 1980 or so that manufactured these instruments, bringing them to the scientific community. All scientists should bring their newly gained knowledge, where relevant, to society in the forms of startups, licensed intellectual property, or public disclosures.
The work on protein sequencing led us to the ability to sequence genes available in vanishing small quantities--and clone the corresponding genes. For example, this approach led us to discover that the human platelet growth factor was virtually identical in sequence to an avian oncogene. It was the first time that protein and DNA sequences had been compared, the first bioinformatics analysis. That discovery led to a Science paper on the oncogene hypothesis, the proposal that cancer genes act by inducing the expression of normal genes of growth and development that are present in the cell. A lot of important work has been driven by this hypothesis. We also sequenced the Mad Cow Disease prion protein, and that enabled Stan Prusiner to do the work for which he would win the Nobel Prize in 1997. We sequenced the protein erythropoietin for the first time, and that enabled Amgen to clone its gene, and that led to biotech's first billion-dollar drug. We sequenced the four protein subunits of the torpedo receptor, and that led to cloning of comparable human neuroreceptor genes and opened up the entire field of neurotransmitter receptors. So, too, with alpha and beta interferons and so forth.
The instruments that created the foundational technologies for modern biology--the DNA sequencers and synthesizers--allowed us to automate, integrate, and in some cases, miniaturize chemistries. These instruments led to high-throughput biological measurements (on humans) and in time the accumulation of biological data that led to big data and its analytics.
2. The Human Genome Project
Because we were developing the automated DNA sequencer, I got invited to the first meeting on the Human Genome Project at Santa Cruz in the spring of 1985. I was asked, along with 11 others, to pass judgment on whether the Human Genome Project was a good idea. We decided it was feasible, but we were split six to six on whether it was a good idea. In the mid-1980s, 80 percent of the biologists in the United States were opposed to the Human Genome Project, as was the National Institutes of Health. By the time this project became a reality in 1990, attitudes had started to shift.
I was involved in the genome project in a number of different ways. I developed the first automated DNA sequencer that made this project possible. I was also an early advocate. I helped to create one of the first genome companies, Darwin. My laboratory was one of the 16 US human genome centers, and we sequenced large portions of human chromosomes 14 and 15. We used sequencing to analyze multiple gene families, especially for immune-related genes. We identified disease-related genes in the genome sequences of all the members of a single family in 2010 and opened up the approach of sequencing the genomes of human families to identify more than 15 additional disease-related genes.
The human genome sequence was the second paradigm change; it gave us the ability to correlate genetic variants with both wellness and disease phenotypes--a central pillar of 21st-century medicine.
3. Cross-disciplinary biology
What the automated DNA sequencer led very nicely to--because we had integrated biology, chemistry, engineering, computer science, math, and molecular biology--was the concept of cross-disciplinary biology. To do cross-disciplinary biology, you gather under one roof all the flavors of scientists required for diverse projects, place them in proximity for intimate interactions, and teach them one another's languages.
That interdisciplinary approach makes it possible to take a leading-edge problem in biology, develop the technologies relevant to it, and then develop the computational platforms needed to analyze the results, all in one lab. This approach drove discovery in science, and it drove innovation. We tried to push the idea of creating a cross-disciplinary biology department at Caltech, but the Caltech biologists absolutely vetoed the idea. Bill Gates made it possible for me to move to the University of Washington in 1992, and we established the first cross-disciplinary department--the Department of Molecular Biotechnology--there.
In just eight years, we saw spectacular results. We did major projects in genomics. Phil Green wrote the two key algorithms for quality control and assembly for the Human Genome Program. Ruedi Aebersold and John Yates pioneered two of the first techniques for the newly emerging field of proteomics. We employed inkjet printing to develop a device for the rapid synthesis of DNA fragments and DNA arrays, a technology still used today by Agilent. This cross-disciplinary approach led to the development of a series of technologies and strategies that were critical for the next paradigm change.
4. Systems biology
I wanted to build on the cross-disciplinary approach by adding to it a center for systems biology. That proved challenging in the bureaucracy of a large state university, so I resigned in 2000 to start the independent, nonprofit Institute for Systems Biology, the first it its kind. Systems biology was a holistic, global, integrative, dynamic, and cross-disciplinary approach to biology that differed dramatically from classical reductionistic biology of that time, which studied biological systems one gene or protein at a time. We applied systems approaches to major problems in biology and disease, and that work led to the fundamental concepts that defined personalized medicine and 21st-century medicine.
We called the systems approach to study disease systems medicine. Systems medicine included three things: the use of deep phenotyping to characterize the complexities of disease; the use of network biology to understand the mechanistic underpinnings of various types of diseases and identify biomarkers and candidate drug targets; and the creation and integration of new technologies and computational platforms and diverse data types.
Systems biology was the fourth paradigm change in my career.
5. Conceptualization of 21st-century medicine
The experience with systems biology led to a shift in thinking about the nature of health care itself and the idea that health care needed to be predictive, preventive, personalized, and participatory. We called this P4 health care. The essence of P4 health care is the understanding that there are two domains in health care: wellness and disease. At that time, wellness was virtually ignored in most healthcare systems. The conceptualization of systems medicine and P4 health care lies at the very heart of 21st-century medicine.
6. Scientific (or quantitative) wellness
In 2014, Nathan Price (an ISB colleague) and I put together a pilot project to explore the potential of quantitating wellness. The research was based on the deep phenotyping of 108 individuals longitudinally for a period of 9 months (blood and stool samples taken every 3 months). We had two spectacular results. The first was that we could strikingly change the wellness of individuals with this approach; the second was that the data clouds we generated could be analyzed to identify new approaches and insights into both biology and medicine. Scientific wellness, the quantification of health, which is based on these results, was the sixth paradigm change and is a vital part of 21st-century medicine.
7. Bringing 21st-century medicine to the US healthcare system
The final paradigm change came in 2016, when Rod Hochman, the CEO of Providence St. Joseph Health (PSJH), came to me and said, "I'd like you to become our Chief Science Officer, and have the Institute for Systems Biology affiliate with us to become a research arm of PSJH." We saw this as a way to bring P4 health care to a major healthcare system, so we agreed.
The seven paradigm changes provided fascinating insights about science, data, people, organizations, and leadership.
* Each paradigm change fundamentally changed our views of aspects of biology and medicine and their practice.
* Each paradigm change was met with great resistance and skepticism on the part of scientists and scientific leadership early on.
* Each paradigm change required extended determined optimism on my part--a faith that what I was doing was correct in the face of significant skepticism.
* Each paradigm change required a new organizational structure. Paradigm changes have enormous difficulties emerging from existing organizations, as those bureaucracies are honed by the past and have difficulty dealing with the present, let alone new ideas.
* The key to gaining acceptance of a new paradigm change is to prove that it can provide something new for understanding and executing an aspect of science.
These seven paradigm changes framed the possibilities for 21st-century medicine. We could carry out high-throughput biology on individuals and deal with large data sets. We could correlate genome variation with disease phenotypes and wellness phenotypes. We could use a cross-disciplinary platform to drive new kinds of technologies and strategies to attack specific medical problems. Systems biology transformed how we thought about the complexities of biology and disease. The emergence of scientific wellness, or quantitative wellness, was incredibly important for optimizing the wellness of individuals through deep phenotyping. Scientific wellness, P4 medicine, and systems medicine represent the essence of 21st-century medicine.
The Coming Revolution in Medicine
Twentieth-century medicine was focused almost entirely on disease. In contrast, 21st-century medicine, built on the systems approaches to disease, will be holistic, focused on creating P4 health care--health care that is predictive, preventive, personalized, and participatory. Its essence is using deep phenotyping to understand both wellness and disease and, most importantly, to understand wellness-to-disease transitions and enable early reversal of those transitions.
The convergence of systems biology, scientific wellness, digital health, big data, and social networks has given us a much clearer understanding of what P4 health care is and how different it is from contemporary medicine as currently practiced. P4 health care is
* Proactive rather than reactive. It focuses on preventing disease, or reversing it in its earliest stages, rather than on curing or managing existing disease.
* Focused on bringing health to individuals rather than to populations.
* Focused on wellness and disease, not just disease.
* Data based, using deep phenotyping and personalized data clouds to optimize wellness and avoid or ameliorate disease.
* About both wellness of the body and wellness of the brain. The brain has been almost entirely ignored in health care, and we can do incredible things to optimize and improve cognitive wellness.
Most importantly, it takes a very different view of clinical trials. A typical clinical trial for cancer might require 20,000 patients. In general, researchers give half the patients the drug to be studied and half a placebo and then analyze the averaged data. The fundamental flaw in this approach is that it assumes that all of the objects you're studying (the patients) are identical. Patients are not identical. They differ in genetics, lifestyles, and environment exposures. The fallacy of this assumption is evident in the deplorable efficacy of modern drugs. For the 10 top-selling drugs in the United States today, the best response rate is one in four--the drug works for one person in four. The worst is 1 in 25. Moreover, the average clinical trial offers no insight to distinguish responders from nonresponders.
With P4 health care, on the other hand, you generate an individual data cloud for each of the 20,000 participants. With 20,000 individual data clouds, you can ask relevant questions to stratify the patients and identify relevant biomarkers. For example, one could get biomarkers to distinguish those who respond to the drug as opposed to those who do not. From these stratified subpopulations, you can make inferences about the disease and disease response. Because you can distinguish people who respond to the drug from those who do not, you can identify biomarkers and phenotypes that correlate to this response. This ability leads to the potential for clinical trials to be carried out in two stages. First, you carry out a clinical trial with 50 patients and identify the biomarkers that distinguish responders from nonresponders. Then, you carry out a second clinical trial with 50 responders. Because you will get a 98 percent or better response rate, the FDA is likely to approve the drug. This approach will transform clinical trials, allowing us to identify the actual effectiveness of a given treatment and radically reduce costs.
The 108 Pioneers
In 2014, the time was right to begin a pilot project on wellness. ISB had by this time pioneered a series of technologies that facilitated deep phenotyping (targeted proteomics, single-cell analyses, and computational platforms for analyzing complexity). Moreover, recent studies had been published on the determinants of health, with some surprising results: it turned out that genetics determined 30 percent, lifestyle and environment 60 percent, and the healthcare system just 10 percent of individual health. Deep phenotyping allowed us to assess the contributions to wellness of both genetics and lifestyle and environment.
For the 108 participants, we selected a number of assays to assess human complexity and wellness. First, we determined the complete genome sequence for each participant. Second, every three months, participants had blood drawn to quantify 1,200 different analytes--proteins, clinical chemistries, and metabolites. We also analyzed, every three months, the gut microbiome (the bacteria in your gut that play an important role in your health) by quantifying the bacterial species contained therein. We used digital devices together with self-measurements to collect data on the quantized self (for instance, activity, sleep, pulse, distance traveled). In selected cases, we used computational tools to assess and facilitate the improvement of brain health.
These tests created data clouds that, if analyzed properly, could lead to actionable possibilities to improve wellness and avoid or ameliorate disease. We delivered these actionable possibilities to participants via scientific wellness coaches, trained in psychology, biology, nutrition, and other relevant disciplines, who worked with the participants each month. These coaches enabled a 70 percent compliance rate with the actionable possibilities.
In the first set of data, 91 percent of participants had nutritional limitations, mostly due to genetics; 68 percent had inflammatory difficulties, mostly due to diet; and a large number--51 individuals--were prediabetic. We were able to move the needle on nearly all of these measures in just nine months, making virtually all of the pioneers healthier by these quantitative measures. Eight individuals moved from prediabetic to normal, and most of the rest of the pre-diabetics moved significantly toward normal.
I was one of the participants. The actionable possibilities from genome analyses showed me how I could lose weight more efficiently and optimize my exercise. I lost 20 pounds; I now routinely do 100 push-ups and 100 sit-ups every morning, plus a full hour of other kinds of exercises. I corrected five nutritional deficiencies with supplements. One of my deficiencies was Vitamin D. I tried taking 1,000 units every day; that did nothing. We found I had two gene variants that blocked the uptake of Vitamin D. It took 15,000 units to bring me to a normal level and about 5,000 units a day to maintain that level. The integration of blood chemistries and genetic analysis thus led to a new actionable possibility. This is an example of how increasing the data in the personal data clouds exponentially increased the number of actionable possibilities.
I have one bad allele from the gene that predisposes to cardiovascular disease and Alzheimer's Disease, and I have a family history of Alzheimer's and heart disease. I've initiated specific behaviors to minimize the possibility of getting Alzheimer's and I take statins to keep my cholesterol levels low. I'm also doing Brain HQ--a series of computational game exercises that improve your cognitive functions in striking ways. The outcome of all of these efforts is that I have a biological age (the age your body says you are) 15 years younger than my chronological age. Most important, I learned that, in the end, the individuals have to be responsible for maintaining their own health--the fourth P, participatory. Almost all of the participants--the pioneers, we call them--realized that you have to take an active role in maintaining your own health. We called this deep phenotyping for individual health scientific wellness or quantitative wellness.
Deep Phenotyping a Larger Population
The Pioneers program was successful in changing health for most of the 108 participants. It was able to do that by creating data clouds that led to transformational actionable possibilities in combinations unique to each individual. When a majority of the pioneers explicitly wanted to continue with the scientific wellness program, we decided to create a company. Arivale was founded in 2015 to bring scientific wellness to consumers. Many of the 108 pioneers joined the company; over the next four years, 6,000 clients were recruited. However, the company closed down in mid2019; it failed for two reasons. First, it did not figure out how to recruit sufficient numbers of consumers to make the company profitable. Educating consumers about scientific wellness is an unsolved major challenge. Second, the FDA forbid companies providing genetic information to consumers (of which Arivale was one) to deliver disease-related actionable possibilities to their customers because it was viewed as practicing medicine. To continue our efforts in scientific wellness, we took Arivale's two computational platforms (costing perhaps $20 million) as well as seven computational and clinical biologists into ISB. We will use these scientists to initiate a new program at PSJH to carry out complete genome sequencing and deep phenotyping on a million patients over the next five years. We plan to carry on scientific wellness on a much larger scale in a medical environment.
Arivale did succeed in two regards. First, it demonstrated the power of scientific wellness for the individual. Second, the analyses of the 6,000 personal data clouds we created in this study have been transformational. I liken them to the Hubble telescope, which allowed us to look at the universe with a resolution never before achieved and as a result pioneer new ideas about its origins and its evolution. These data clouds provide entirely new views of human biology and human disease.
* Actionable possibilities. With Arivale as with the P108 project, the number of actionable possibilities increased significantly as the number of data clouds grew and with the integration of different data types. This suggests that if individuals were to take a lifelong journey of scientific wellness and respond to new actionable possibilities as they arose, they could extend their healthy lifespan significantly.
* Scientific wellness. Arivale conducted a three-year clinical trial in scientific wellness, starting with 1,000 patients at PSJH. The trial was terminated a year early with the end of Arivale, but the preliminary data showed significant health improvements for virtually all of the participants in the trial--a likely clinical validation of the power of scientific wellness to transform lives.
* Statistical correlations. In the PI08 wellness pilot, we determined that there were about 3,500 statistical correlations between data bits in one of the six groups of data with data bits in one of the other groups. This hairball of statistical correlations could be broken down into more than 70 discrete communities of blood analytes and other data types. Each community correlated with a physiologically relevant or disease-related phenotype. For example, the cholesterol community has 15 analytes and genetic markers; some of these are wonderful candidates as blood biomarkers or drug targets for assessing or lowering blood LDL. Thus, one can use a statistical approach with deep phenotyping data to identify biomarkers and drug targets.
* Genetic risk. Genome-wide association studies (GWAS) have generated polygenic assessments for more than 100 different diseases and disease phenotypes. These can be used to identify genetic disease risks in individuals whose full genome has been sequenced. This is important in two ways. First, people at high risk for serious diseases may be followed with deep phenotyping to identify the earliest transition point, so action can be taken to try to reverse the disease before it ever manifests itself. Second, it is likely that individuals who have a high genetic risk for, say, LDL cholesterol will be treated differently from those whose genetic risk is low. Arivale demonstrated that individuals with a high genetic risk for LDL cholesterol could not bring those levels down with lifestyle changes alone. They required statins. Conversely, those with low genetic risk could bring high LDL levels down with lifestyle changes. This is likely to be true of other genetic diseases as well.
* Manifestation of analytes in the blood according to genetic risk. It is possible to show a linear correlation between the level of blood LDL and genetic risk, with high-risk individuals having higher blood LDL. This is interesting because statins reduce the levels of LDL and operationally reduce the disease risk. There may be other blood analytes that correlate with genetic risk that can themselves be drug targets (or blood biomarkers) to reduce risk; we have preliminary evidence this will be true generally. This may offer another novel approach to identifying biomarkers and drug target candidates.
* Biological age. The Arivale population age range extended from 21 to 91, which allowed us to analyze participants in 10-year bins. As a result, we were able to demonstrate that the control of the expression patterns in the three classes of blood analytes (clinical chemistries, proteins, and metabolites) decreased with age. From this observation, we could calculate an individual's biological age. If your biological age is younger than your chronological age, you are aging in a healthy manner. Individuals who came into the Arivale program with biological ages five or more years greater than their chronological age lost one year in their biological age for every year they stayed in the wellness program. Hence, biological age may be a metric for healthy aging.
* N = 1 experiments. In a sense, the Arivale cohort constitutes 6,000 N = 1 experiments. This kind of N = 1 experiment may have many applications: to optimize wellness for each individual, to detect wellness-to-disease transitions early enough to allow reversal, to increase understanding of complex human systems such as nutrition, to follow the treatment of a complex disease to identify effective future therapies and make clinical trials more efficient and more informative. Indeed, we are currently carrying out such clinical trials on multiple sclerosis (with Genentech), scientific wellness, and Alzheimer's Disease.
* Wellness-to-disease transitions. Arivale observed more than 100 wellness-to-disease transitions for many different chronic diseases. We can examine the blood analytes that signal the transitions for the most common chronic diseases; that may allow us to identify blood analytes (or their cognate biological networks) that signal the earliest onset of the disease. With enough data and artificial intelligence, we may be able to diagnose common chronic diseases well before they manifest themselves as disease phenotypes and reverse them. This will be the preventive medicine of the 21st century.
* A model for wellness and disease. The ability to identify wellness-to-disease transitions leads to a new way to think about wellness and disease. First, it suggests that medicine must be highly individualized. The only valid control for the wellness state is the data from the individuals themselves, because everyone is different genetically and with regard to lifestyle and environmental exposures, and accordingly, every individual has a unique baseline. Second, it supports a view of all disease as arising from a continuum of wellness to disease. In a mouse model of neurodegeneration, we demonstrated that as the disease progresses, the number of disease-perturbed networks increases exponentially. As the number of disease-perturbed networks increases, disease becomes increasing difficult to stop or reverse. The focus of study for chronic diseases should be at these earliest transition points--and not, as it is today, after the disease phenotype has manifested. The fascinating idea is that our ability to reverse disease at the earliest transition points could lead to the end of many chronic diseases.
Tackling a Specific Disease--Alzheimer's Disease
Alzheimer's is a complex disease, and one that is resistant to traditional treatment approaches. There have been 400 Alzheimer's drug trials in the last 12 years. Zero have worked. We have recruited a team to look at Alzheimer's through the lens of three paradigm changes: 1) dense phenotyping, 2) multimodal therapy, and 3) cognitive brain health.
We plan to use dense phenotyping to analyze high-risk Alzheimer's patients to identify that earliest detectable transition point. Current metabolic PET scanning can detect Alzheimer's pathology 4-10 years before a clinical cognitive diagnosis. We will use PET scanning as the gold standard for identifying blood biomarkers at the earliest Alzheimer's transition point. We will also be able to identify biomarkers to stratify Alzheimer's into its three or more distinct subtypes.
Once we've identified a person with the disease, we'll shift them to the multimodal therapy that was advocated initially by Dale Bredesen (2017), who took a systems approach to optimizing synaptic communication that led to a 36-point regimen for treatment, including drugs, supplements, lifestyle changes, nutrition optimization, exercise, and sleep optimization. Preliminary observational trials on several hundred patients looked promising.
We are also using Brain HQ both to assess cognitive deficiencies and to restore cognitive capacity. This approach is now in its first clinical trial for multimodal Alzheimer's treatment at the Hoag Institute in Santa Monica, California. It is a small clinical trial treating patients with a partial multimodal approach and Brain HQ and comparing those patients against the outcomes for conventional treatment for Alzheimer's. We are about a year and a half into the clinical trial, and the preliminary results look extremely promising.
We're setting up two other clinical trials in the very near future to test the multimodal approach in a combinatorial manner, so that we can isolate the most effective elements in this treatment. We are also taking an artificial intelligence approach to the diagnosis and therapeutic regimens for Alzheimer's, using an expert systems platform to diagnose and treat disease with deep science and deep knowledge. It uses deep learning--having computers do the statistical analysis--to maximize its effectiveness. The basic idea is to set up a deep reasoning platform for a disease like Alzheimer's by delineating from the literature all of the conceptual aspects of the disease. We derived 840 concepts from Dale Bredesen's book, The End of Alzheimer's (2017), in just 60 hours. Each concept is an intelligent building block, and the blocks can communicate with one another and with each patient's data. The idea is to bring the patient's data into these intelligent blocks, let the data interact, and then read out highly accurate diagnosis and therapy for Alzheimer's for each individual patient.
The really important idea is that this is a learning platform; 85 percent of what we need can come from the literature, and the other 15 percent comes from working with clinical experts on N= 1 multimodal therapy experiments. We first use machine learning to convert the data into mechanistic, causal inferences that give us insight into mechanisms, which generates statistical inferences about the disease. We can then deduce causal inferences and refine the deep learning base, creating a continuous learning system. Patient data then can be plugged into the knowledge platform to generate appropriate diagnoses and therapeutic strategies. The Alzheimer's expert will review and optimize the response. By the time we've gone through 100 patients or so, we'll have a system with abilities equivalent to those of an Alzheimer's expert. Furthermore, the system is an clear box; the physician can see the logic used to arrive at a diagnostic or therapeutic recommendation. This deep knowledge and artificial intelligence is the beginning of the eighth paradigm change I've seen in my career.
P4 Health Care at Scale
I am currently working on a new proposition, to fully sequence and do deep phenotyping on one million PSJH patients over five years. PSJH is a large healthcare system; it sees three million patients each year. We're doing a series of deep phenotyping clinical trials, which are going to give us significant insights into relevant biomarkers and drug target candidates and an understanding of a range of disease mechanisms. We can optimize individual wellness. We can use changes in blood analytics to identify early disease transitions and eventually reverse them. We can use the data to identify genetically high-risk individuals to follow for early disease transitions. We can understand the biological complexity behind, for instance, nutrition. All of this activity will build on the previous work.
Deep knowledge computational platforms will change how we deal with the complexities of virtually all chronic disease (starting with Alzheimer's). In the future, diagnosis and treatment will be done with N= 1 studies. We have studied some spectacular examples of wellness-to-disease transitions for several different types of cancer. I hope we will be able to reverse 80 percent of early-stage Alzheimer's disease in perhaps five years. Achieving that goal will result in enormous savings in the half-trillion dollars we currently spend on Alzheimer's each year.
Scientific wellness will become a dominant paradigm in health care, and in time health care will migrate into the home, as the big hospital systems will not be necessary. To enable wellness in the home, in ten years or so, we will have a tricorder-type instrument that can make 5,000 blood measurements and send these data to a computational analysis platform, which will send actionable possibilities to physicians and wellness coaches. This is the essence of 21st-century medicine, which will focus on optimizing wellness and identifying and reversing chronic disease at its earliest stages. It is going to be revolutionary and transformational.
One of the ways to help catalyze this transformation is to carry on my proposed five-year initiative to do complete genome sequence and deep phenotyping of one million patients at PSJH. We have the computational tools to handle the data, to do the analyses that lead to thousands of new actionable possibilities that can transform health care. We will need to recruit a host of strategic partners to enable this ambitious project. It represents a tremendous opportunity to catalyze 21st-century medicine.
Twenty-first-century medicine is going to achieve four transformational objectives. First, it is going to optimize wellness for each individual and make healthy aging possible. Second, it is going to identify the earliest detectable transition for most chronic diseases and, in time, identify the therapies for early reversal. Third, it is going to strikingly decrease the cost of health care. Finally, I believe that it will also migrate wellness health care to where it belongs, in the home. One big challenge for 21st-century medicine is to understand how we can recruit healthcare leaders who understand the vision and are willing to make the commitments necessary to achieve this paradigm change.
Lee Hood received his MD from Johns Hopkins and PhD from Caltech. He was a faculty member at Caltech (1970-1992) and University of Washington (UW) (1992-2000). He founded the first cross-disciplinary department of biology at UW and in 2000 co-founded the independent Institute for Systems Biology; he served as the institute's president until January 1, 2018. He has pioneered important technologies in genomics, proteomics, and single-molecule analyses and co-founded 17 biotechnology companies, including Amgen and Applied Biosystems. In 2016, he became Senior VP and Chief Science Officer of Providence St. Joseph Health, the third largest nonprofit healthcare system in the United States. He is a member of all three US national academies--science, medicine, and engineering. email@example.com
Published by Taylor & Francis. All rights reserved.
Bredesen, Dale E. 2017. The End of Alzheimer's. New York: Avery Publishing.
Kuhn, Thomas S. 1970. The Structure of Scientific Revolutions, 2nd ed. Chicago, IL: University of Chicago Press.
Caption: Lee Hood's work in systems biology uses individual data clouds and artificial intelligence to create truly individualized medicine.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||IRI MEDAL|
|Article Type:||Travel narrative|
|Date:||Nov 1, 2019|
|Previous Article:||The 2019 IRI Medal.|
|Next Article:||Tools for Transformation: Michelin's Digital Journey.|