Non-Hormonal Hot Flash Drugs: What the Trials Actually Showed

The honest bottom line

Fezolinetant, sold as Veozah, is the first FDA-approved NK3-receptor antagonist for moderate-to-severe hot flashes of menopause, and the trial record shows a benefit that is real, statistically clear, and modest in size. In the phase 3 SKYLIGHT program, women taking the approved 45 mg dose had roughly two to three fewer moderate-to-severe hot flashes per day than women on placebo, measured against a starting point of around ten or more episodes a day. That margin earned regulatory approval in May 2023, but it arrived alongside a liver-monitoring requirement that the FDA later escalated to a boxed warning. Whether the drug is worth it depends on how a given person weighs a modest, non-hormonal benefit against a rare but real safety burden.

A genuinely new mechanism

Hot flashes are not simply a hormone-deficiency symptom waiting to be reversed. During menopause, a small cluster of neurons in the hypothalamus that helps regulate body temperature becomes overactive as estrogen falls, driven in part by a signaling molecule called neurokinin B. Fezolinetant blocks the neurokinin 3 (NK3) receptor those neurons rely on, dampening the misfiring that triggers a flush. That is a different lever from estrogen, which is why the approval mattered: many women cannot or prefer not to take hormone therapy, and the non-hormonal alternatives before this were largely repurposed antidepressants and older agents with their own limits.

The FDA approved fezolinetant on May 12, 2023, describing it as the first drug in its class for this use. The efficacy case centered on two 12-week randomized, double-blind, placebo-controlled trials, SKYLIGHT 1 and SKYLIGHT 2, which together enrolled roughly a thousand women and tested 30 mg and 45 mg once daily against placebo, each carried into a longer extension, with a separate year-long study assessing safety.

What "statistically significant" actually bought

The SKYLIGHT 1 results, published in The Lancet in 2023, and the SKYLIGHT 2 results in the Journal of Clinical Endocrinology and Metabolism, used the same co-primary endpoints: change in the frequency and severity of moderate-to-severe vasomotor symptoms at week 4 and week 12. On each of those endpoints, the 45 mg dose beat placebo with high statistical confidence, the separation appeared within the first week, and it held through 52 weeks.

Here is where evidence literacy matters. Statistical significance answers "is this difference likely real?" not "is this difference large?" The placebo-adjusted reduction in SKYLIGHT 2 was about 2.5 fewer moderate-to-severe episodes per day at both week 4 and week 12, with a severity difference near 0.3 on the symptom scale. Those are the numbers that cleared the pre-specified bar. For a woman having a dozen disruptive flushes a day, a couple fewer can genuinely matter. But it is a couple fewer beyond placebo, not a couple fewer overall.

The placebo problem

That distinction carries unusual weight here because hot flashes respond enormously to placebo. In this field, women assigned to a dummy pill routinely report their flushes dropping by roughly a third to a half over a few weeks. That is not fakery; expectation, attention, seasonal change, and natural fluctuation all push the numbers down. So the raw before-and-after improvement on fezolinetant looks dramatic, with frequency falling by the large majority from baseline, yet a big share of that drop would have happened on placebo anyway. The drug's true contribution is the gap between the two curves, and that gap is the modest figure, not the headline percentage. Anyone reading a menopause-drug study, or a supplement ad borrowing the same before-and-after visuals, should train their eye on the placebo arm first.

The liver signal and the boxed warning

The second half of the ledger is safety. Across the trials, a small number of participants showed liver enzyme elevations above three times the upper limit of normal, generally transient, and no case met Hy's law, the classic laboratory signature of dangerous drug-induced liver injury. That is a reassuring but not entirely clean signal, and the FDA reflected it in the original label by requiring liver bloodwork before starting and periodically during the first nine months of use.

Then the picture sharpened after launch. In December 2024, the FDA added a boxed warning, its most prominent, for rare but serious liver injury, prompted by a post-marketing report of a patient who developed symptoms and markedly abnormal liver tests within weeks of starting the drug. The agency also expanded the monitoring schedule to include the first and second months. A boxed warning is not a recall and does not declare the drug unsafe for everyone; it says the risk is serious enough that prescribers and patients must actively watch for it. It is also a reminder that a 12-week trial of a thousand people cannot detect a one-in-many-thousands event. Rare harms often surface only after a drug reaches a far larger population, which is exactly why post-marketing surveillance exists.

How to read a modest effect

None of this makes fezolinetant a good or bad choice in the abstract, and this article is educational, not medical advice. It makes it a well-characterized one. The trials show a real mechanism producing a modest, durable, statistically solid benefit, paired with a monitoring burden and a rare serious risk that regulators took seriously enough to headline. The useful question is not whether it works, since it does, nor whether it is dangerous, since serious harm is rare with vigilance. The question is how a specific person values two or three fewer flushes a day against periodic blood draws and a small tail risk, compared with the other options open to her. That trade-off belongs to a patient and her own clinician, informed by the actual numbers rather than the marketing gloss. The broader lesson travels well past menopause: when a new drug arrives, read the effect size and the placebo arm together, and give the safety label the same attention you give the efficacy claim.