This article is more than 5 years old. Information may no longer be current.
FDA committees near-unanimous on need to investigate testosterone therapies
COLLEGE PARK, MD — Two FDA advisory committees voted 20-1 in favor of revising the current indication for testosterone replacement therapies following several hours of charged presentations by clinicians, industry representatives and the public at a joint meeting Wednesday.
Members of the Bone, Reproductive and Urologic Drugs and Drug Safety and Risk Management panels were divided 16-4 on whether the FDA should require further investigation into potential cardiovascular risks for only certain indications or regardless of the indication, respectively, with a single vote for no investigation.
“It’s a weak cardiovascular safety signal. We need to inject some reality into what has really been an uncontrolled use of the drug,” committee member A. Michael Lincoff, MD, of Cleveland Clinic, said.
Definition and Populations Treated
Early on, guest presenter Peter J. Snyder, MD, of the University of Pennsylvania highlighted the Endocrine Society Guidelines from 2010 that clinically define hypogonadism as having unequivocally low morning serum testosterone.
Snyder said, “300 ng/dL is the lower limb of normal for testosterone assays; unequivocal depends not just on an absolute number, but the time of day and the number of times.”
But Snyder urged consideration for symptoms in deciding treatment, a sentiment echoed by presenters throughout the day.
“We always have to pick a number for practical purposes, but it’s likely that there is no one number; the lower, the more certain,” Snyder said. “What if it’s associated with a disease known to cause low testosterone, or if someone has 100 ng/dL and a large pituitary adenoma? This has to be part of our education.”
An explanation of primary and secondary hypergonadism, with a breakdown of “classic” vs. age-related, or andropause, by presenter Mark Sigman, MD, of the Warren Alpert Medical School of Brown University, gave the room a clear picture of the range of potential treatment populations.
Sigman underscored low-testosterone related comorbidities including obesity, type 2 diabetes and reduced bone mineral density as particularly controversial areas for treatment since causation and association remain uncertain.
“We cannot really separate those who benefit from those who do not,” Sigman said. “That’s likely because signs and symptoms are multi-factorial.”
Current State and Clinical Benefits
Industry representatives mounted their evidence for the efficacy and safety of prescribing therapies to distinct demographics, including the extent of testosterone use by age and the reasons for use.
In an introduction, Kraig Kinchen, MD, senior medical director, global urology, at Eli Lilly, said the FDA has not yet required randomized placebo-controlled trials or demonstrated improvement in symptoms and has preferred class labeling except for safety labels to reflect formulations and modes of delivery.
“Registration study populations have been enrolled without regard to etiology and have included men with age-related hypergonadism,” Kinchen said. “The primary efficacy endpoint of most phase 3 registration studies is to demonstrate that at least 75% of patients achieve and average serum testosterone concentration level within the normal range, between 300 and 1,000 ng/dL.”
The FDA’s consideration to revise the indicated patient population, based on lack of clear etiology differentiation and varied interpretation, was encouraged.
“The current indications for the population is not who is being treated right now,” committee member Toby Chai, MD, of Yale School of Medicine, said. “The age range is not the range of most people getting prescriptions.”
Data presented from the SHA Integrated Database showed usage in 2013 was highest among men aged 55 to 64 years, followed by 45 to 54 years. The total number of testosterone therapy prescriptions jumped ninefold from 2000 to 2013, with primary care physicians writing the most.
Adrian S. Dobs, MD, of Johns Hopkins University and a member of the advisory boards for three sponsors present at the hearing, offered a look at the benefits of testosterone replacement therapy, from body composition to sexual function to mood and fatigue, and limitations to data, from heterogeneous populations to variable baseline levels, endpoints and study design.
The sponsors vocalized their willingness to find clarity.
“I thank the sponsors for coming today with an open mind,” committee member John R. Teerlink, MD, of the University of California, San Francisco School of Medicine, said.
Risks and Recommendations
A key presentation by Shalender Bhasin, MD, of Harvard Medical School, with advisory board seats for two of the sponsors, broached the biggest topic of concern in the room during a presentation on cardiovascular events and testosterone therapies.
“Testosterone’s effects in preclinical and clinical models are diverse,” Bhasin said. “Some effects are beneficial, and some may be potentially deleterious.”
Bhasin conceded biologic plausibility for the association of testosterone with cardiovascular events. He presented data from a meta-analysis demonstrating lower testosterone levels were associated with higher all-cause mortality, epidemiologic studies showing mixed results and a small randomized placebo-controlled trial showing increased cardiovascular events with therapy.
“It’s important to distinguish between classical hypogonadism and age-related decline,” Bhasin said. “In young men with classical hypogonadism, testosterone replacement therapy improves symptoms with low adverse event frequency. In older men, with age related decline or frailty, neither benefits nor risks have been clearly demonstrated.”
Two speakers in the public portion raised safety concerns about testosterone transdermal gel —one sharing a personal story about a thromboembolism in the spine that led to paralysis and the other telling about a heart attack still taking its financial toll.
Rita Jain, MD, vice president of global pharmaceutical research and development at AbbVie, noted several on-going randomized placebo-controlled trials by sponsors — T-Trial, NCT01816295, TEAAM —and NIH-funded epidemiological studies — University of North Carolina, Kaiser Foundation, Seattle Institute of Biomedical and Clinical Research.
“New data from the T-trial is expected to report in the first half of 2015,” Jain said. “These results may alter the approach on the benefits and cardiovascular risks of testosterone replacement therapy.”
Fair labeling, education
Educating consumers about the benefits and risks of testosterone therapies was a topic visited throughout the day, with a presentation by Trung-Hieu Brian Tran, PharmD, a FDA regulatory review officer in the Office of Prescription Drug Promotion (OPDP), shedding light on the specifics of various methods.
“Prescription drugs are misbranded if the labeling or advertising is false or misleading, or if it fails to reveal material facts, including about consequences which may result from the use of the drug as suggested in the labeling or advertising,” Tran said.
Tran advised the room that a category called “disease awareness communications” are not subject to the requirements of the Federal Food, Drug, and Cosmetic Act or FDA regulations; OPDP has already received complaints regarding disease awareness communications for hypogonadism.
Product claim promotional materials must comply with the regulations, Tran noted; the submission of such materials to OPDP for testosterone therapies has increased in the past several years.
The sponsors are proposing label changes, based on the limitations of data in less-defined populations, including age-related, and to include elements of professional society guidelines for testing and monitoring testosterone.
“Right now there’s no proof that in this older age group there’s any efficacy at all and I think the label should reflect that,” committee member Michael Domanski, MD, of Mount Sinai Hospital, said.
Hylton V. Joffe, MD, of the FDA, assured the room the committees would carefully review what was discussed and come back with a decision Thursday. — by Allegra Tiver
For More Information: See the Endocrine Today Twitter feed for a live update of the committee meeting. www.Twitter/com/EndocrineToday
Perspective
Back to Top
Clearly, things needed to be refined and definitions to be perfected. The committee decided, on that basis, that we need to set out new criteria for the testing and outline the various clinical states. To that extent, I agree.
We need to perfect the diagnosis based on free testosterone concentrations rather than the old total — something everyone agreed to today — because sex hormone binding globulin concentrations vary, which leads to deceptive rises or falls in testosterone levels.
The use of the terms classical and the not classical is something that I don’t agree with fully. What is classical? By classical, they imply we know the abnormality and we find the low testosterone and treat it. What they did not recognize is that today, as per our work that defines type 2 diabetes as causing phenomenal amount of hypogonadism, is one in three men with type 2 diabetes is hypergonadal, and one in four men with obesity without diabetes is hypergonadal. The result: free testosterone.
If your diagnosis of the so-called classical conditions is based on free testosterone, how can you deny these novel conditions such as obesity and type 2 diabetes as not being hypergonadal? We are not saying they are not hypergonadal, but there is shadow of doubt, and that is something that I totally disagree with.
I’m not denying that patients should have symptoms to go with the low testosterone and then you should go on to treat. However, a lot of endocrine disease starts without the patients even being symptomatic. For example, diabetes is measured now on the basis of blood glucose levels and the patient may be totally asymptomatic. Similarly hypothyroidism is often asymptomatic.
We are moving into an era when blood tests tell us results, and that’s why we do screenings. In the two major conditions that we’ve been talking about with hypergonadism — type 2 diabetes and obesity — we need to diagnose it on the basis of blood tests and then, in association with that, we determine whether the patient has clinical symptomatology and measure it. Sometimes it’s only after the normal tests, that the questioning reveals something.
The next step is considering a patient has no symptoms and low testosterone and begins treatment; this is the answer that will come out of our NIH study. Are there any silent benefits that you don’t perceive? Yes, there are, because lack of testosterone causes increased insulin resistance and improved inflammatory state. Treatment, then, is likely to reduce your cardiovascular risk. But patient will not feel any of this because it is happening silently inside his body. Why would you deny him that benefit?
Lastly, we need outcome studies for cardiovascular disease, properly conducted trials and not two papers that were badly done and badly conducted. One of them demonstrated a 20% reduction in cardiovascular events, but when they subjected them to all kinds of statistical analysis and confounding factors, showed a 20% increase. This kind of statistical gimmickry is not right. The other included 100 women being treated with testosterone in the analysis. How can I trust a paper that is so careless about the population that is being included?
I just don’t believe the data of cardiovascular risks. But I do believe that a cardiovascular risk study ought to be conducted, just like we do for all other anti-diabetic drugs. These drugs are like testosterone and likely be an internal part of diabetes. I’m actually anticipating the pleasure of seeing a positive effect.
Perspective
Back to TopIn only a relatively small population of those patients who have low serum testosterone levels and symptoms do we actually know the causal mechanism. When they talk about classical hypergonadism, it makes a lot of sense and seems attractive; classical always sounds great, it sounds real, as if it’s been around for centuries. But we don’t really know that a majority of people who are being treated for low testosterone are benefitting or not benefitting or what’s causing the low testosterone.
Some people have pituitary tumors, which I see often in my practice; some have had a football injury or they had a surgical removal of the testes because they had cancer. But that represents a relatively small percentage of patients that fall into the category of hypergonadism.
Saying that we’re going to limit treatment to a small group of individuals who have “classical” hypergonadism is attractive but simplistic. We really need to have a better understanding of who’s going to benefit from treatment, and those people who benefit from treatment are obviously people where the risks becomes important. It’s also an important issue to know what level of testosterone will give a patient with some type of symptoms benefit.
If you could reverse symptoms of disease, if you could fix things that make people less well, you would do it in the simplest fashion with the least amount of risk. To say that you’re going to do bariatric surgery, as was suggested is a simpler approach to a low testosterone level, would certainly be subject to some discussion. If we could treat diabetes and make the diabetes well, we would be much more successful above and beyond what that does for serum testosterone levels. If it was easy to treat diabetes and give everyone perfect blood glucose control, we would have a lot less blindness, a lot less kidney failure, a lot less amputations. If testosterone was an adjunctive beneficial treatment for insulin resistance, then that might be a positive thing, but you need better data in order to demonstrate that.
The purpose of this meeting is very valuable. It defines the fact that we want to understand what the risk level of treatment is, and that’s true for all medications. But we have to balance the benefits we might see, how you define who is best going to benefit, and then understand what risks might be involved.
Almost everybody in the room agreed that right now we don’t have a very good handle on whether or not testosterone replacement has any impact on cardiovascular risks — whether it makes the risk less, whether it makes the risk greater, or whether it does nothing to the risk.
We don’t have the data to answer the question, but everybody who spoke on it argued that a long-term safety study for cardiovascular and prostate disease would be very desirable. The question is who’s going to pay for it?
If the FDA would agree to split the cost between government and industry, since they both have a vested interest, then that might be the solution to the problem. The difficulty is that it’s a very large-scale study that needs to be done because the background risk of cardiovascular disease in an older population is high.
For age-associated hypergonadism, the question is what is the cost and benefit of intervention and treatment? Cancer goes up with age, it doesn’t mean that cancer is OK. If rheumatoid arthritis goes up with age, it doesn’t mean rheumatoid arthritis is OK. Just because something goes up with age doesn’t mean it’s not a clinically important problem. We have to look at each type of medical problem and decide whether there are treatments that will offer more benefits than risks and what the cost might be.