What’s a biomarker?
What is a biomarker? Drs. Kyle Strimbu and Jorge Tavel attempted to provide a working definition as follows: biomarkers are expected to be objective, quantifiable characteristics of biological processes. 1
The recent two Eating Disorder Institute papers on diabetes mellitus type 2 and hypercortisolism addressed the screening of several biomarkers: serum glucose levels, A1c and serum cortisol levels. Even the dexamethasone suppression test is measuring the response of a biological process by chemically and temporarily suppressing its usual function.
Conversely, a clinical endpoint is a patient’s subjective sense of illness and wellbeing. When clinical endpoints frame the practice of medicine (or medical research for that matter too), then that subjective sense of illness or wellbeing in a patient directs investigation and treatment.
“The goal of clinical practice is to improve morbidity and mortality, not to change quantifiable features of patients’ innate biochemistry, for instance, with no outward clinical effect. Similarly, patients seek treatment for their diseases, not for the numerical measures that frequently but not perfectly correlate with their illnesses.” 2
In most cases these days, we tend to use biomarkers as a harbinger of a clinical endpoint. We determine that these numerical measures foretell of the eventual illness and death of a patient even as the existing clinical endpoint (the patient’s subjective experience) involves no sense of illness or disease. Doing so means that biomarkers act as surrogate clinical endpoints: “because we think you might get sick, based on current biomarker readings, we will treat you now to prevent future illness.”
“Biomarkers could only serve as true replacements for clinical relevant endpoints if we completely understood the normal physiology of a biological process, the pathophysiology of that process in the disease state, and effects of an intervention – pharmacological, device, or otherwise – on these processes.” 3
The practice of medicine today skews much more towards biomarker measurement than to clinical endpoint. For those with an anxiety disorder (an eating disorder) biomarker screening effectively reinforces fear and threat responses: you feel fine now, but your biomarker screen tells us you will soon be sick and dying.
And sometimes that fear is not so much about any harbinger of illness and death, but rather about how such biomarkers might suggest being ostracized and stigmatized in society. It’s common for those with eating disorders to fear that a biomarker screening result represents something that stands between them and a societally-acceptable body weight.
Let’s take a look at thermoregulation for a moment.
Every human being is hypothermic at points in the day and hyperthermic at other points in the day, relative to an average that we define as normal body temperature: 37°C or 98.6°F. My body temperature range on any given day sits below the norm even as it fluctuates between higher and lower levels of hypothermia.
However, my persistent hypothermia is not a condition requiring intervention. It may or may not be correlated with a risk of a future disease state, but these risks would be tiny (just as they are for many metabolic variations in human beings). An average range for me will not be an average range for the next person and a certain amount of variability merely reflects the greater normal ranges within our population.
If we decide to take me to an extremely hot and tropical country for a couple of weeks, then lo and behold, my body temperature range over any 24-hour period appears to go from normal to mildly hyperthermic. If I then head up to hang out in Iqualuit in January for two weeks, then I’m unsurprisingly hypothermic and perhaps more so than my usual average range. I will die and there will be a proximate cause listed on my death certificate. However, short of my body temperature sitting beyond the range at which any human being can survive at the time of my death (i.e. proximate cause of death is actually hyper- or hypothermia), any contribution my persistent lower-than-average body temperature over my lifespan made to my proximate cause of death will be just a guess.
Thermoregulation in our bodies is fairly narrow in its adaptive range. There is a point at which extreme hyperthermia or hypothermia is not survivable for any of us. But what is really critical to understand is that the body has mechanisms in place to attempt to maintain its temperature within a comfortable range to allow for periods of wakefulness and sleep throughout the day and night. In fact, were we to attempt to maintain our body temperature at a static point throughout the entire day and night, we would actually develop illness as a direct result of not allowing the body to vary its internal temperatures.
The ways in which the body maintains its ranges of various metabolic functions are exceedingly complex and (critical point coming up) not static. And unlike both hypo- and hyperthermia, normal metabolic ranges are exceedingly elastic and have to be at absolute extremes before the individual is actually at risk of dying directly due to the presence or absence of the metabolic hormone under observation.
The body dynamically fluctuates glucocorticoids, blood sugar levels, reproductive hormones, neurotransmitters etc., constantly all day and all night long. Medicine hits the broad side of a barn with averages for all sorts of metabolic biomarkers just as we’ve calculated an average body temperature for all human beings as well. As Drs. Stimbu and Tavel pointed out above, we do not understand the variability, physiology and biology of “normal” enough that we can use these averages as surrogates for any clinical endpoint.
Perseverating on these biomarkers, and treating them as surrogate clinical endpoints, will create an internal homeodynamic space that is constantly optimized to both escape or battle a predator all while also generating physical changes in the body that optimize your ability to survive physical trauma and subsequent chance of infection as well.
Your living system doesn’t differentiate between the immediate mortal threat of a charging elephant about to gore you; the vague existential sociocultural threat that you won’t taper in weight due to current biomarker readings; and the more concrete existential threat that you feel well and yet are a walking time-bomb of illness and early death. I use the term “existential threat” here in its literal definition: threat to existence or survival.
Anxiety is a threat response system on high alert and if there’s one thing that metabolic function is designed to do, it is to ensure survival when under threat. It is therefore a circular irony that the act of framing biomarkers as surrogate clinical endpoints will create a threat response that requires adaptive metabolic responses that will be reflected in subsequent metabolic biomarker readings. And repeat.
Surrogate clinical endpoint surveillance and treatment is more akin to deciding to set up an unprotected tent in a field of agitated elephants looking to protect their young, than it is protecting you from any worst-nightmare future existential threat.
It’s all getting to the point where we actually convince ourselves that it’s better to live in that tent in the middle of the field of elephants (so we might see the charging elephant ahead of time) rather than to simply find another place to set up house. Worse still, we’ve been told that if we “eat well and exercise” while hanging out in our tent in the middle of the field, the elephants will be soothed and less likely to gore us.
Consider the possibility that surrogate clinical endpoints are no substitute for actual clinical endpoints. If you live in a field of agitated elephants then you have put your anxiety, and not your health or longevity, at the center of your existence. You exist in the now and only in the now. That’s where your actual clinical endpoint lives too – in the now.
Yes, it’s important to investigate actual clinical endpoints—if you have symptoms that are impacting your quality of life, then you and your doctor should discuss them. But before you walk out of the of appointment with a requisition for biomarker screening tests, ask the doctor what is being investigated and why.
Medical doctors have merely a vague sense of the connection between hypervigilance and metabolic biomarker results. However, there’s a reasonable body of evidence that the correlation of metabolic biomarker states treated as surrogate clinical endpoints (i.e. diabetes mellitus type 2) with psychological distress is strong.
“Diabetic patients displayed high rates of affective and anxiety disorders over time, relative to community adults: 60% higher for MDD [major depressive disorder], 123% for GAD [generalized anxiety disorder], 85% for PANIC [panic disorder], 7% for DYS [dysthymia].” 4 [emphasis mine]
As with all correlation, we have no real sense of causation— do metabolic anomalies instigate affective disorders, or the other way around, or possibly does a third or multitude of other causative agents play a role? However, we do know that affective (emotional) distress worsens over time when a patient is diagnosed with type 2:
“Time × treatment interactions were found for anxiety, diabetes-related distress, and self-efficacy; notably, intensively treated patients showed more distress and less self-efficacy in the 1st year, and usual-care patients reported more distress and less self-efficacy 2–3 years after diagnosis.” 5
We also have some intriguing initial evidence that burnout at work is predictive of the onset of diabetes mellitus type 2 in previously healthy individuals. 6 Such evidence might point to emotional distress having a causative role in the onset of biomarker results as surrogate clinical endpoints.
Where does all of that clinical data leave you as someone with a clear existing anxiety disorder (eating disorder) and a desire to prevent illness and early death? It might be worthwhile to discuss with your therapist the relative merits of focusing on a very human, but ultimately futile, desire for perpetual good health and immortality against the immediate positive impacts of adopting behaviors that lower hypervigilance and threat responses in the short term.
Ultimately, the elephants will be less agitated if you take down the tent and move along.
1. Strimbu, Kyle, and Jorge A. Tavel. "What are biomarkers?." Current Opinion in HIV and AIDS 5, no. 6 (2010): 463.
4. Fisher, L., M. M. Skaff, J. T. Mullan, P. Arean, R. Glasgow, and U. Masharani. "A longitudinal study of affective and anxiety disorders, depressive affect and diabetes distress in adults with type 2 diabetes." Diabetic Medicine 25, no. 9 (2008): 1096-1101.
5. Thoolen, Bart J., Denise T. De Ridder, Jozien M. Bensing, Kees J. Gorter, and Guy E. Rutten. "Psychological outcomes of patients with screen-detected type 2 diabetes." Diabetes Care 29, no. 10 (2006): 2257-2262.
6. Melamed, Samuel, Arie Shirom, Sharon Toker, and Itzhak Shapira. "Burnout and risk of type 2 diabetes: a prospective study of apparently healthy employed persons." Psychosomatic medicine 68, no. 6 (2006): 863-869.