Redefining the Nature of Disease

The Oxford Dictionary says disease is a disorder of structure or function, especially one that has a known cause and a distinctive group of symptoms, signs, or anatomical changes. It’s time to rethink that simplistic definition.

By Paul Cerrato, MA, senior research analyst and communications specialist and John Halamka, M.D., Diercks President, Mayo Clinic Platform

At the most basic level, dis-ease refers to a lack of ease or discomfort. Over the centuries, the word has taken on various iterations. It was once thought to occur when a person offended the gods, but took a step into the science realm in the 19^th century with the discovery that disease can be brought on microbes. Now most clinicians rely on an understanding of pathophysiology to define disease, with an emphasis of malfunctioning organ systems. But a recent editorial in Nature Medicine postulates that the future of healthcare will require a transition "from a reactive model to one that is proactive, personalized and preventive. Advances in multi-omics — integrating genomics, transcriptomics, proteomics, and metabolomics — may enable precise prediction of individual disease risks. Continuous monitoring of biomarkers through wearable and implantable biosensors could allow early intervention, long before signs and symptoms manifest.”

This new paradigm can include real-time data from blood glucose meters, blood pressure readings from smart watches, individual genetic work-ups, and virtual simulations of a patient’s physiology, i.e. digital twins. Turning this vision into a daily reality will likely require clinicians to think “outside of the box,” which currently focuses primarily on signs, symptoms, medical histories, and abnormal medical test results. It will also require clinicians and researchers to take a much deeper dive into the root causes of each patient’s condition, and to explore an individual patient’s lifestyle, their exposure to environmental toxins, psychosocial stressors, along with the aforementioned genomics, transcriptomics, proteomics, and metabolomics.

Of course, many experienced physicians and nurses will question this multi-omics approach and want to see proof that it will improve clinical outcomes in the real world. Consider the mounting evidence: British and German investigators measured about 3,000 plasma proteins from more than 41,000 individuals to derive models that might predict the 10-year incidence of 218 common and rare diseases and compared this proteomic data to clinical data with or without clinical assays. They then narrowed down the scope using smaller models that included 5-20 proteins, which outperformed the predictions of basic clinical data with or without assays. The proteomic-based models generated useful insights for 52 diseases, including multiple myeloma, non-Hodgkin lymphoma, motor neuron disease, pulmonary fibrosis, and dilated cardiomyopathy. The researchers acknowledged the limitation of the study, but also pointed out that they "were able to assess generalizability of results for 6 of the 67 diseases for which proteins improved prediction over and above clinical models in the European Prospective Investigation into Cancer (EPIC)-Norfolk study."

In an independent study, Carrasco-Zanini et al. evaluated the potential value of specific blood proteins to improve screening and diagnosis of Type 2 diabetes. Currently, screening and diagnosis rely on hemoglobin A1c (HbA1c) and fasting glucose, but that ignores the many individuals with isolated impaired glucose tolerance, which refers to anyone with an abnormal two-hour glucose tolerance test (GTT) but normal HbA1c and fasting glucose. Unfortunately, routine GTTs are expensive and impractical, so Carrasco-Zanini et al. used machine learning to find a select group of serum proteins that might help pinpoint the presence of isolated impaired glucose tolerance; they identified three markers that improved the detection of the disease, when combined with HbA1c testing. Their conclusion: "Assessment of a limited number of proteins can identify individuals likely to be missed by current diagnostic strategies and at high risk of T2D and its complications.”

There is also evidence to demonstrate that the continuous monitoring of biomarkers through wearable and implantable biosensors can facilitate early intervention, long before signs and symptoms of specific diseases become manifest. At Mayo Clinic, for instance, cardiologists have found that patients who underwent cardiac rehabilitation and used a smartphone-based app to record their weight and blood pressure experienced greater improvements in cardiovascular risk factors and were less likely to be readmitted to the hospital 90 days after discharge, when compared to patients who did not have access to the app.

This is only one of many experiments to demonstrate the value of collecting such data. Some innovative scientists are taking this approach further and measuring a massive collection of biomarkers on themselves and others. Michael Snyder, Ph.D., with Stanford University, is at the forefront of this movement. Dr. Snyder is using himself as a N of 1 experiment on the value of the "quantified self". As director of Stanford’s Center for Genomics and Personalized Medicine, he oversees several projects designed to find the many actionable insights that can be derived from measuring all the omics data currently available, including one’s genome, proteome, microbiome, and transcriptome. With the assistance of several smart watches, a subdermal chip, fingertip pricks to obtain blood samples, and several whole-body MRI scans, he has compiled a large database that can be analyzed for trends. With the help of his physician and this data set, he created an integrative personal omics profile (iPOP), which enabled them to predict the onset of diabetes and Lyme disease.

Unfortunately most clinicians are not in a position to order a long list of biomarkers for their patients, but even a single marker can be revealing. Elevated levels of C-reactive protein, for instance, usually signals the presence of inflammation. And while there can be many causes for this finding, mounting evidence strongly suggests that this risk factor can be modified by non-pharmacological interventions, including intermittent fasting (IF). IF refers to at least three protocols: alternate day fasting, fasting two days per week, or daily time restricted feeding. One systematic review and meta-analysis found that “IF regimens and ERDs [energy restricted diets] significantly reduced C-reactive protein (CRP) concentrations… Additionally, IF regimens were more effective in reducing CRP levels than ERDs…”. That was especially the case in overweight and obese patients. Considering the fact that chronic inflammation contributes to a variety of diseases, this approach may be worth considering in patients willing to try it. Rheumatoid arthritis, ulcerative colitis, psoriatic arthritis, systemic lupus, gout, and sarcoidosis are among the many disorders that have an inflammatory component.

None of these research studies suggest that clinicians should abandon their diagnostic protocols or ignore patients’ symptoms and test results, but they do suggest that we expand our view and think more broadly about disease and its causes.