A group of researchers has identified and described rhythmic, time-regulated human physiological characteristics in a non-controlled setting — a finding with wide-ranging implications for precision medicine.
Clinicians and researchers widely acknowledge time-of-day variability in disease expression and drug efficacy. Heart attacks and strokes, for example, are more likely to occur in the morning, notes a 2016 editorial in Circulation. Cardiologists often instruct patients to take cholesterol-lowering statins at night when enzymes that produce cholesterol are most active. Individuals with asthma are more likely to experience coughing at night or in the early morning, reports the National Heart, Lung, and Blood Institute. Researchers at the Salk Institute for Biological Studies in California have found that disruptions in the circadian rhythm can set the stage for the development of cancerous tumors.
“Every tissue in the body has its own, independent circadian clock,” says Michael Hughes, PhD, Assistant Professor of Medicine and Genetics in the Division of Pulmonary Critical Care Medicine at Washington University School of Medicine in St. Louis. “These clocks are synchronized by the central oscillator in the hypothalamus. Consequently, most every aspect of normal physiology is under the influence of circadian rhythms.”
The quest to understand human biorhythms is relatively new and has traditionally taken place in controlled settings using forced desynchrony protocols, according to Carsten Skarke, MD, Research Assistant Professor of Medicine at the Institute for Translational Medicine and Therapeutics at Perelman School of Medicine at the University of Pennsylvania. Those procedures are appropriate only for small groups of patients and yield an incomplete picture of the biological clock because they do not take environmental cues into account, Dr. Skarke says. He and his team of researchers at Penn needed to solve the problems of scale and setting if they were to characterize the human chronobiome.
“In order to scale up easily, we would need to be able to study a large number of patients more easily,” Dr. Skarke says. “The easiest way to study is if we don’t really have to ask patients to come for frequent visits to the hospital, but just observe them while they go about their daily lives.”
So, for four months, researchers used wearable devices and smartphone apps to collect physiological data from six healthy young men who went about their normal routines. Those technologies yielded information about activity, sleep patterns, nutrition, light exposure and communication. Additionally, the participants visited the Center for Human Phenomic Science at Penn for two 48-hour periods during the study so the researchers could collect plasma, serum and saliva, as well as oral and rectal samples. The researchers collected approximately 2.2 million data points.
“We observed, collected a lot of data and asked, ‘Are we able, in these healthy volunteers, despite the fact that they each follow individual daily schedules of activity, food intake, etc., to discern rhythmic signals?’” Dr. Skarke says. “[O]nly if we are able to cut through the noise of daily life and ... see these rhythmic signals, only then would we be able to detect disruption of these rhythmic signals in patients who have crossed over from this healthy phenotype and readout to a signal that associates with some sort of pathological change, underlying change, that leads to disease.”
The Chronobiome in Focus
Sixty-two percent of the study’s sensor readouts showed time-specific variability, including, crucially, results that validated the researchers’ expectations, according to findings that appeared in Scientific Reports. Cortisol levels, for example, were higher in the morning than in the evening, and blood pressure was higher during the day than at night. Those data proved the validity of the team’s methodology — critical for their ability to conduct larger studies of the human chronobiome (see “Scaling Up”). Other findings were surprising, including the extent to which certain genera of bacteria in the microbiome were more abundant in the evening and less so in the morning, according to Dr. Skarke.
Hughes, who was not involved with the study, finds its results exciting and thinks the first implications will be felt in personalized pharmacology.
“[C]an we predict which drugs and humans will benefit most from timed administration?” he asks. “Related to this, clinical studies will hopefully be better designed from here onward, with more care given to the time of administration, and time itself being treated as a variable in all statistical modeling.”
Dr. Skarke sees great potential for the human chronobiome in the development of precision medicine, and he thinks his team’s pilot study could provide a model for the field of how to integrate multidimensional data sets. Wearable devices, he says, could fill informational gaps for clinicians between patient visits, provided there are ways to identify and act on anomalies.
“If there’s an actionable finding, who’s taking care of that?” Dr. Skarke asks. “Is the physician reviewing all of that data? Probably not, so we need to have smart algorithms in place to flag actionable clinical signals and transmit them to a physician to take action.”