In early February, the media exploded with reports that the FDA was investigating(1) whether low-testosterone therapies like Axiron, Androgel, Testim, and Fortesta might be tied with an increased risk of heart attacks and strokes. Yet, as the FDA said in its safety announcement, “At this time, FDA has not concluded that FDA-approved testosterone treatment increases the risk of stroke, heart attack, or death.” A plaintiff bringing a lawsuit thus can’t run to court and point to the FDA announcement as definitive proof that testosterone therapy caused their (or their husband’s) heart attack or stroke. So evidence can they use to prove or disprove such a causal link? It’s a question I’ve been getting a lot lately, so I thought I’d discuss it here.
The FDA’s investigation is based upon a November 2013 study in the Journal of the American Medical Association(2) and a January 2014 study posted on PLOS ONE(3). The JAMA study found “Among a cohort of men in the VA health care system who underwent coronary angiography and had a low serum testosterone level, the use of testosterone therapy was associated with increased risk of adverse outcomes” like death, myocardial infarction, or stroke, and the PLOS ONE study found “In older men, and in younger men with pre-existing diagnosed heart disease, the risk of [myocardial infarction] following initiation of [testosterone therapy] prescription is substantially increased.”
These results didn’t come entirely out of the blue — for example, a study in 2010 that was supposed to look at the effects of testosterone had to be discontinued when the participants suffered an unusual number of cardiovascular events(4) — but the latest two studies together form the strongest evidence showing a connection between testosterone therapy and cardiovascular events like heart attack and stroke. The situation is not altogether different from Vioxx, the litigation over which also began with a study showing a connection between Vioxx and cardiovascular events. The questions are if, like with Vioxx, future studies will demonstrate that “Low T” therapy causes these conditions and, perhaps, that the drug manufacturer concealed evidence that would have revealed this link sooner.
Let’s back up for a moment. The FDA has approved testosterone therapy to treat a single condition: hypogonadism. The FDA-approved prescribing information describes this as occurring as a result of “testicular failure due to conditions such as cryptorchidism, bilateral torsion, orchitis, vanishing testis syndrome, orchiectomy, Klinefelter’s syndrome, chemotherapy, or toxic damage from alcohol or heavy metals,” or “idiopathic gonadotropin or luteinizing hormone-releasing hormone (LHRH) deficiency or pituitary-hypothalamic injury from tumors, trauma, or radiation.” The FDA did not approve the use of testosterone therapy to treat fatigue, or muscle weakness, or depression, or low libido, or erectile dysfunction, or for “anti-aging” purposes — yet most testosterone therapy prescriptions today are for those conditions, which for most purpose aren’t related to hypogonadism. Some of them are just part of aging.
After about 30 or so, men’s testosterone levels decline as they get older. (A study back in 2012 calculated that athletes’ “peak performance” generally occurred around 26 years old, and that the mean age of world record-setting athletes was 26.1 years.(5)) For some men, this shift can be sublime or even exciting*, but, for most of us, the shift comes too quickly or with severe side effects, and let’s face it: most of us would be more than happy to find the fountain of youth.
That’s where the drug companies saw a chance to make a buck. Although testosterone therapy is only approved to treat hypogonadism (or to be even more specific, hypoandrogenism), doctors are allowed to prescribe virtually any medication for virtually any purpose. There’s nothing wrong with that — doctors and patients need to have room to make their own medical decisions — but the drug companies routinely exploit that situation by engaging “off label” marketing, in which they try to convince doctors to prescribe patients medications for unapproved conditions, or to convince patients (by way of relentless television, Internet, and magazine advertisements) to lobby their doctors for the medication. Off label marketing is illegal, and pharmaceutical companies have paid the federal government billions in penalties for doing it, but they keep doing it anyway, concealing it when they can and then, if they’re caught, writing off the penalties as a cost of business. There’s too much money to be made.
The “Low T” companies have engaged in that sort of off-label marketing in spades, like the website “Is It Low T?”, for the purpose of encouraging men to wrongly believe that a variety of ailments are really the symptoms of low testosterone, and that the ordinary effects of aging can be safely treated with testosterone therapy. E.g., the website asks visitors if they are “sad” or “grumpy,” if they have “a lack of energy,” and if they aren’t as good at sports as they used to be — something that is to of course be expected as a person ages, particularly into their 60s and beyond.
All of that marketing (Abbott Laboratories spends over $80 million annually marketing AndroGel) worked: testosterone therapies are a $4 billion market, well over the size it should be based on the prevalence of the condition the medication was designed to treat. A review of prescribing information on low testosterone therapy found that one third of men on testosterone therapy were diagnosed with “fatigue,” and that one quarter of men on testosterone therapy did not even have their testosterone levels tested before they received a prescription(6).
Now, let’s move to the mixed question of science and law that I used for the title of this post: can we prove in court that testosterone gel causes heart attacks and strokes?
Ideally, we would prove (or disprove) such a causal link with a randomized controlled trial, putting one group of men on testosterone therapy and another on a placebo and then watching for heart attacks and strokes. But we need to be realistic: such a trial would be extraordinarily difficult to construct correctly given the huge number of factors at work, and it might even be dangerous to complete the trial. In many prior examples, like with Vioxx, clinical trials looking for potentially fatal conditions had to be stopped as soon as the initial evidence of a risk became apparent, because it would be unethical to knowingly expose patients to a risky medication.
So let’s start with what we already have. The JAMA study and the PLOS ONE studies are both observational studies, not randomized controlled trials. Medical researchers looked at a bunch of data they already had on patients, performed statistical analysis with the hope of removing any “confounding variables” — e.g., you have to make sure you’re not accidentally comparing heart attacks among obese smokers in their 80s and normal weight non-smokers in their 60s — and then saw the worrying conclusions that have prompted all this attention. These observational studies are a great way to learn about complex medical issues that defy easy analysis; a prominent, and perhaps foreshadowing, example was the Women’s Health Initiative, which identified the potentially fatal risk posed by women’s hormone therapy treatment, particularly as women got older.(7)
As the Executive Editor of Harvard Health notes, “neither was the type of study that can prove cause and effect. They can only show associations, or links. That means there’s no smoking gun here that testosterone therapy is harmful. But the studies do suggest caution.”(8) I disagree. Both as a matter of science and as a matter of law, although observational studies primarily show associations, observational studies can also be enough to show that a drug causes a particular disease — if they are done properly. As the Federal Judicial Center’s Reference Manual on Scientific Evidence says(9):
[O]bservational studies can be very useful. For example, there is strong observational evidence that smoking causes lung cancer. Generally, observational studies provide good evidence in the following circumstances:
• The association is seen in studies with different designs, on different kinds of subjects, and done by different research groups. That reduces the chance that the association is due to a defect in one type of study, a peculiarity in one group of subjects, or the idiosyncrasies of one research group.
• The association holds when effects of confounding variables are taken into account by appropriate methods, for example, comparing smaller groups that are relatively homogeneous with respect to the confounders.
• There is a plausible explanation for the effect of the independent variable; alternative explanations in terms of confounding should be less plausible than the proposed causal link.
In other words, well-done observational studies can be good enough, as long as they’ve approached the subject from different angles, have accounted for other possibilities, and have suggested “a plausible explanation” for the effect.
The JAMA and PLOS ONE studies have already been criticized, but, frankly, the criticisms I’ve seen are just plain wrong, to the point of being irresponsible. Over at Huffington Post, an OB/GYN who sells hormone therapy argues that the studies are “essentially meaningless,” because they didn’t assess testosterone levels “to tell if a patient is a proper candidate for therapy and if they are tolerating the therapy well.”(10) There’s no reason to believe those results would be different, but, the biggest issue is that she missed the point of the study: the point wasn’t to see if a narrow class of “proper” patients were at risk, but whether all patients on testosterone were at risk. Part of the problem with testosterone therapy is that it’s overprescribed, and artificially limiting a study of its risks would be tantamount to willful blindness to the real problem.
That said, the second factor — “effects of confounding variables” — is likely going to be the biggest issue in the testosterone therapy litigation. Heart attacks and strokes aren’t like, say, mesothelioma, which can only be caused by asbestos exposure. As they get older, men are at an increased risk of both, particularly after age 65. Observational studies have shown that there are nine major factors that increase the risk of heart attack in both men and women (11):
- HDL to LDL cholesterol ratio;
- current smoking;
- former smoking;
- abdominal obesity;
- psychosocial (e.g., (depression, locus of control, global stress, financial stress, and life events including marital separation, job loss, family conflict);
- physical inactivity;
- very low alcohol intake;
- high risk diet
If a man had any of those factors, then the drug companies will jump on it and claim that it’s impossible to prove the testosterone therapy increased his risk of suffering a heart attack or stroke.
To which I say: rubbish. We’ve seen those sorts of attacks before by drug companies as a way to avoid responsibility, but this isn’t the place to start explaining how we intend to do that in the litigation, beyond saying those issues are better left for cross-examination at trial. There’s no “smoking gun” here but there’s enough smoke to know there’s a fire, which is why we’re actively pursuing these cases. A company can’t just sit back and reap billions off a false wonder drug while being silent about its potential to kill its users.
(1) “FDA evaluating risk of stroke, heart attack and death with FDA-approved testosterone products,” an FDA Drug Safety Announcement.
(2) “Association of Testosterone Therapy With Mortality, Myocardial Infarction, and Stroke in Men With Low Testosterone Levels” at JAMA Network.
(3) “Increased Risk of Non-Fatal Myocardial Infarction Following Testosterone Therapy Prescription in Men” at PLOS ONE.
(4) “Adverse Events Associated with Testosterone Administration” available at PubMed.
(5) “Exponential growth combined with exponential decline explains lifetime performance evolution in individual and human species” at Springer.
(6) “Trends in Androgen Prescribing in the United States, 2001 to 2011” at JAMA Network.
(7) “The Last Word On Hormone Therapy From the Women’s Health Initiative” at NPR Health News.
(8) “New study adds caution to testosterone therapy for “low T”” at Harvard Health.
(9) Federal Judicial Center’s Reference Manual on Scientific Evidence, at fjc.gov.
(10) “Testosterone Therapy Does Not Cause Heart Attacks” at Huffington Post.
(11) “Risk factors for myocardial infarction in women and men: insights from the INTERHEART study” at European Heart Journal.