News outlets this weekend reported the results of a trial published in the New England Journal of Medicine, that evaluated the safety of testosterone replacement therapy (TRT) in aging men. Specifically, the trial was designed to evaluate cardiovascular safety. The reporters generally implied that the trial proved TRT is safe on the heart.
The trial was prompted by prior evidence, albeit weak evidence, that TRT may be bad for the heart. For example, some observational trials have correlated TRT use with heart disease and all cause mortality. Other trials have demonstrated that TRT increases certain cardiovascular risk factors, such as: high blood pressure, sleep apnea, plaque scores, and red blood cell counts leading to thicker blood. Furthermore, women live longer than men, and castrated animals live longer than non-castrated animals. These suspicions, along with the rapidly increasing usage of TRT despite no FDA approval, led the FDA to issue a "black box warning" in 2015.
Abbvie, the makers of the multibillion dollar drug Androgel, funded this current trial to address this issue. They recruited over 5,000 patients and randomized them to TRT vs placebo specifically to measure the risk of cardiovascular events.
As most of you know, I am a big proponent of randomized controlled trials. They are the strongest kind of evidence in medicine. However, a trial is only as good as its methodology. Unfortunately, this trial had serious methodological flaws. I will review them below.
Flaws of the trial
The major flaw of the trial was under-dosing and sub-therapeutic testosterone levels. Typically, we treat men with TRT when their testosterone level falls below 300 ng/dl on a fasting morning blood sample. The goal of treatment is to get levels between 600 ng/dl and 800 ng/dl - the levels of a young healthy male. If we can't achieve these levels, we increase the dose, or switch to injectables.
The major flaw of the trial was under-dosing and sub-therapeutic testosterone levels.
In this study, however, the average testosterone level prior to treatment was 250 ng/dl, and was only brought to 350 ng/dl (see graph below from the trial). Therefore, it is quite obvious the treatment group was under-dosed and did not reach therapeutic or real-world levels. Thus, the hypothesis, TRT is safe for the heart, was never really tested in this trial. As far as I am concerned, the cardiovascular safety of TRT is still yet to be determined.
Other limitations of the trial include a short follow up time of only two years (plaque build-up and cardiovascular disease is a long term process), and a dropout rate of 60%.
This is unsurprising. The best strategy for Abbvie to prevent it's drug from looking dangerous was to adopt this methodology. Expose the treatment group to as little drug as possible, for as short a time as possible.
Abbvie strategy: Expose the treatment group to as little drug as possible, for as short a time as possible.
Although atherosclerotic heart disease was not found in the treatment group, there was a statistically significant increase in two other cardiac-related issues: atrial fibrillation and pulmonary embolism (blood clotting). Whether these are real or not will require further investigation. It seems this trial has generated more questions than answers.
The reality of TRT
I have discussed the problems with TRT in the past (here and here). In summary TRT is not likely to make you live longer - maybe a little shorter - but may improve quality of life.
A helpful analogy I believe, is to think of your aging body like a 25-year-old Ford Taurus. A system made up of many parts. It has a 25-year-old engine block, a 25-year-old fuel injector, 25-year-old break pads, etc. When you give testosterone, it is like adding a nitro-booster to the engine. It will result in a temporary boost in performance allowing you to accelerate and drive faster.
However, as a trade off, the other parts in the system are taking a beating. The 25 year-old engine block, 25 year-old fuel-injector, and 25 year-old brake pads, couldn't handle a nitro-booster in their prime, let alone in their current state. The nitro-booster will inevitably accelerate their breakdown. The car will cease to function at a sooner point in time. In other words, the nitro-booster accelerated the aging of the other parts, which ultimately accelerated the aging of the entire system. It was a trade-off. Performance for longevity.
The other parts in the system take a beating as a trade-off.
With TRT, we give a 65-year-old the muscles of a 25-year-old; yet, the joints, the tendons, the heart, and the vessels, are all still 65 years old. It's reasonable to hypothesize that we may be wearing out these other parts prematurely. For example, men on TRT are observed to have a 50% increase risk of musculoskeletal injuries.
This brings up some interesting arguments regarding the physiology of healthy aging. Especially when the body is viewed as a complex system:
What happens when we put the muscles of a 25 year-old on a 65 year-old frame?
Does the augmentation of one part of our aging body lead to the premature aging of the rest of the system?
Instead, should the goal of healthy aging be to age all the parts at an equal pace?
Despite these nuances, hormone and anti-aging clinics neglect to properly inform their patients of the potential harms of TRT. They manipulate lab results to make every patient look "deficient", and end up bringing testosterone levels way past the levels of a healthy young person (often far greater than 1000 ng/dl). And to make things more confusing, they simultaneously start multiple interventions their clinic pushes (other hormones, supplements, exercise, diets, etc), so the patient never really knows which intervention or interventions made them feel better.
The bottom line
As with many interventions in medicine, it is helpful to think of TRT as a quality-of-life versus quantity-of-life dilemma. A trade-off with benefits and harms.
Unfortunately, both the benefits of TRT, and the harms of TRT, are hard to assess. The benefits are subjective (how you feel), vague (fatigue, low sex drive, depression), and hard to measure. Additionally, many patients do not feel any different on TRT, so a personal trial is essential to see if it's right for you. However, these personal trials are prone to confounding and placebo effects. Any concurrent changes in your psychology, exercise, sleep, nutrition, weight, medications, and supplements, can mimic the vague effects of TRT. This is why if you decide to try TRT, carry it out in a controlled fashion. Ensure everything else in your life is stable, and meticulously document how you feel before and after treatment.
The harms of TRT are also hard to assess. The totality of the weak evidence speculate some harms involved. However, there is no strong evidence. No successful randomized controlled trials. All those that have attempted have failed. Either they did not enroll enough subjects, they did not last long enough, or they did not reach therapeutic levels (like the trial we are reviewing today). Unfortunately, now that most testosterone products are generic, no pharmaceutical company will fund such a trial. We are left to speculate.