What the FDA Labeling Changes Actually Mean

The Headline Is Not the Label

A man sits down across from me holding his phone: "The FDA dropped the heart warning on testosterone. So it's safe now, and I don't need to jump through the old hoops." The actions he is pointing to are real and clinically important. But his headline has inflated one revision into three separate assertions the agency never put in writing: that the cardiovascular debate is closed, that all testosterone products share the same evidence, and that a single low number now qualifies a man for approved therapy. The wording moved; the work of diagnosis did not get easier.

Each regulatory term has a job. An approved indication defines the group and the condition for which enough evidence exists to market the drug. A boxed warning singles out particular serious risks. "Off label" is neither unlawful nor a synonym for bad medicine; it simply means the specific use sits outside the approved indication, and the right response is a conversation about evidence, not a shutdown.

The 2015 Action

In March 2015 the FDA changed the labels on prescription testosterone. Its message was that these products were meant for men whose low testosterone stemmed from identifiable medical conditions, that neither benefit nor safety had been demonstrated for low values blamed on aging alone, and that a laboratory confirmation was expected before treatment. Labels were also required to carry wording about a possible increase in heart attack and stroke risk.

Because the cardiovascular studies of that era pointed in opposite directions, the agency also obligated manufacturers to run a trial large enough to settle the question. That obligation produced TRAVERSE.

What TRAVERSE Found

The trial randomized 5,246 men, ages forty-five to eighty, who had hypogonadal symptoms, two fasting testosterone values under 300 ng/dL, and either established cardiovascular disease or elevated risk. Its primary outcome, the first of cardiovascular death, nonfatal heart attack, or nonfatal stroke, was reached by 7.0 percent of men on testosterone and 7.3 percent on placebo. The hazard ratio came out at 0.96 (95 percent confidence interval 0.78 to 1.17), satisfying the prespecified noninferiority rule.

"Noninferior" is a technical claim: for that combined endpoint, testosterone was not unacceptably worse than placebo. It is not a claim that testosterone prevented cardiac events. And the population and product were specific, protocol-guided gel in selected men, so the reassurance does not automatically extend to injections, pellets, oral or nasal products, doses above the physiologic range, or men who were never properly diagnosed.

The February 2025 Revision

On February 28, 2025, the FDA announced label changes across the testosterone class, citing TRAVERSE together with ambulatory blood-pressure studies it had required. Two things happened. The boxed-warning language about increased adverse cardiovascular outcomes was removed, and the TRAVERSE results were written into the labels.

Taking a sentence out of the boxed warning is a smaller act than erasing cardiovascular caution altogether. Testosterone can still raise blood pressure, shift hematocrit and fluid balance, and interact with existing heart disease; nothing in the new labels told clinicians to stop monitoring. In fact, the same announcement drew on the blood-pressure data to require or strengthen a class-wide warning that testosterone can raise blood pressure. A class average says nothing about how one man will respond, which is why I take a baseline reading and recheck periodically.

A cardiovascular sentence left the box while a blood-pressure warning arrived, and there is no contradiction: the trial measured a composite of major events over a fixed window; the pressure studies measured a risk factor whose consequences show up later, or in someone already vulnerable. And class-wide describes a change applied across prescription testosterone products; each product's label still carries its own warnings.

2026: Requests and an Invitation

The following year the agency asked manufacturers to remove the age-related limitation-of-use language and to revise prostate and benign-prostatic-hyperplasia safety wording across the class. Asking for a limitation to come out does not turn aging into a disease. A diagnosis still rests on a consistent laboratory result, a clinical picture that fits, and an associated condition or defined indication. Likewise, rewording prostate language does not retire the prostate assessment.

The FDA's own testosterone information continued to say that no product was approved for low testosterone without an associated medical condition. On April 16, 2026, the agency invited manufacturers to discuss potential supplemental applications for a low-libido claim in men with idiopathic hypogonadism. An invitation to bring evidence starts a regulatory process rather than finishing it, and until a review succeeds no clinician or marketer should present that indication as established.

Four Questions I Ask With Every Label Headline

When a patient hands me a headline, I open the actual document, revision date visible to both of us, and we work through four questions. What came out? What went in? Which product does this cover? And what did the agency explicitly decline to do? Every regulatory statement I make gets a date attached, and I never describe something merely requested or invited as if it had been approved.

Apply the same habit to any safety claim: which outcome, which men, which product and dose, and for how long. The best randomized evidence we have is genuinely reassuring for the gel-treated men it studied. It did not establish that testosterone is harmless, that it lowers cardiovascular risk, or that it is approved for every complaint that arrives with age.

Common questions