“If you cannot measure it, you cannot improve it,” a phrase generally attributed to Lord Kelvin (William Thomson, 1824-1907), is a timely admonition for today’s medicine and clinical trials. In today’s world, we can measure countless things. Thus, Kelvin’s advice now applies to the need to measure the right things, under the right circumstances, to improve “it.” “It” presumably, is health, or the subject’s current condition, or more topically, his/her quality of life.
To properly apply Kelvin’s teaching, we must figure out what to measure, when, where, and how often in order to have a chance to improve “it,” whatever “it” may be. Interestingly, as the understanding of human physiology and pathophysiology rises to mind-numbing complexity, the opportunity to measure the wrong thing increases as much as measuring the right thing. Paradoxically, some of the minutiae, while ultra-precise, may never bubble up to the level of a subject’s actual health thanks to the body’s robust homeostatic and self-protective mechanisms. So, we must decide what we are trying to improve and then determine what should we measure, when, where, and how often?
Typically, clinical trials feature scheduled site visits at intervals, often 4, 6, 8, or even more weeks apart. That’s right, someone is taking an unapproved drug, or worse yet, someone [with a disease] is taking nothing at all (placebo) for weeks without anyone knowing what is happening in between their scheduled visits. But that isn’t all that is wrong. The clinic visits, though infrequent, are often intensive, putting subjects through blood tests, imaging studies, performance tests, etc. all within one day. Even the tests intended to simulate what goes on in the home are performed in an artificial environment, in front of several white coats, and on just one day out of many weeks. And, not surprisingly, these visits are among the costliest parts of clinical trials.
Wouldn’t it be more informative, or at least complementary, to understand how the subject is doing in general, in their daily lives, and in their own habitat, between visits? In the last several years new emphasis has been placed on Quality of Life (QoL) as a critical endpoint. Right now, there are more than 10,000 clinical trials registered on clinicaltrials.gov with a QoL endpoint. But it’s more than just the satisfaction of a given subject; it is about living - the things that humans do - such as maintaining their vital signs, moving around, performing their activities of daily living (ADL) and sleeping.
Technology now exists to address many of these problems. First, sensor systems have been designed for use in the home, where the subject lives. Second, the technology exists to measure fundamental physiological metrics and functions: vital signs, body position, activity, and sleep, collectively looking at QoL. While it may be tempting to prefer a blood test for a proteomic biomarker or an MRI scan, do you think a person with cancer, heart failure, or Parkinson’s Disease could have a significant response to medication or a considerable disease deterioration without any measurable effect on their vital signs, sleep, or activity?
Newly available wearable sensor technology gathers data passively, digitally, and continuously without any effort on the part of the subject, in the comfort of his/her own home. Consider as an example, one Huntington’s disease clinical trial contained 20 clinic visits over 2 years that included a subject gait analysis. During the same time, a digital wearable sensor allowed the collection of 14,000 gait assessments. Which do you think might detect a subtle improvement or decline more effectively?
Did you know that sleep efficiency correlates with survival in advanced cancer? It does, and so do heart rate variability and activity counts. Did you know that these same metrics also correlate with outcome in Multiple Sclerosis and Heart Failure? Again, they do. And what if I told you that gait speed is the single best predictor of longevity in all people over age 60? Well, it is, with some even calling it the 6th vital sign. The published significance of all of these metrics is well established. The good news is that they are all easily measurable with the right wearable sensor technology.
I suppose Lord Kelvin should be proud. Today we are starting to measure the correct things, at the right times and in the right places. There is already evidence that by measuring these physiologic and biometric parameters we are improving insights into endpoints of significance and QoL. What’s more, we are discovering the best things to measure – namely, we are starting to establish true “digital biomarkers” and demonstrate that measuring them can improve clinical trial endpoints and health.
Written by Arthur Combs, MD - Chief Medical Officer, MC10,inc.