Fezolinetant for Menopausal Hot Flashes

A hot flash starts in your hypothalamus. It does not start in your skin. A small cluster of cells called KNDy neurons, named for the kisspeptin, neurokinin B and dynorphin they carry, sits in the arcuate nucleus and feeds the brain circuits that dump heat. Fezolinetant is a neurokinin 3 receptor antagonist that blocks neurokinin B from binding the KNDy neuron, which modulates activity in the thermoregulatory center. Its affinity for the NK3 receptor is more than 450 times higher than for NK1 or NK2, so it works on that one signal rather than across the whole tachykinin family.

Estrogen normally restrains those neurons, and they enlarge once it withdraws. The KNDy hypothesis was built on exactly that observation, together with evidence that these neurons project to preoptic areas carrying the NK3 receptor and that ablating them reduces skin vasodilatation and blunts estrogen's effect on temperature control. Neurokinin 3 receptor neurons in the median preoptic nucleus modulate heat-defense pathways in female mice, and this drug is the pharmacological mirror of that experiment. The plain-language section of the label states it outright: this medicine is not a hormone.

These trials counted flashes rather than scoring them. In Trial 1, 174 women on the 45 mg dose started at a mean of 10.4 moderate to severe flashes per 24 hours and dropped 6.4 by week 12, against 3.9 on placebo, a difference of 2.6 flashes with a 95% confidence interval of -3.4 to -1.7. In Trial 2, 167 women started at 11.8 and dropped 7.5, against 5.0 on placebo, a difference of 2.5 with an interval of -3.6 to -1.5. The label treats a reduction of at least 2 flashes over 24 hours as the clinically meaningful threshold.

Severity shifted far less than frequency did. Both arms began at 2.4 on the severity measure; by week 12 the drug arm fell 0.6 against 0.4 on placebo in Trial 1, and 0.8 against 0.5 in Trial 2. Look at how much the placebo arms moved on their own: more than half of the total improvement in Trial 2 appeared without any active drug. SKYLIGHT 1 ran at 97 sites across seven countries in women aged 40 to 65, and SKYLIGHT 2 saw frequency and severity improve by week 1 and hold through week 52.

The realistic promise is a partial reduction. These trials measured how far the daily count falls, never whether flashes stop altogether.

The indication is narrow: moderate to severe vasomotor symptoms due to menopause, and nothing on the label reaches sleep, mood, bone density or genitourinary symptoms. The registration population averaged 54 years old, self-identified as 81% Caucasian and 17% African American, and included prior hysterectomy in 32.1% and prior oophorectomy in 21.6%. Anyone on hormone therapy went through a washout first, so the combination was never tested. The label also records that too few women over 65 enrolled to say whether they respond differently.

The candidate this fits is someone with frequent moderate to severe flashes who cannot take estrogen or has decided against it. The 2023 nonhormone therapy position statement of The North American Menopause Society places fezolinetant among its Level I recommendations, alongside cognitive behavioral therapy, clinical hypnosis, SSRIs and SNRIs, and gabapentin. That same statement concludes hormone therapy remains the most effective treatment for vasomotor symptoms and should be considered within 10 years of the final menstrual period. Being a recommended option is not the same as being the first thing a clinician reaches for, and a licensed clinician decides which applies to you.

A Cochrane review of 24 placebo-controlled trials covering 3,329 women found oral hormone therapy cut weekly hot flush frequency by a weighted mean of 17.92, equivalent to a 75% reduction with a 95% confidence interval of 64.3 to 82.3. Those trials ran from three months to three years and reported weekly counts; fezolinetant's reported daily counts over 12 weeks. Stacking the two percentages against each other would be arithmetic theatre. What genuinely separates them is who can take which. Estrogen carries its own boxed warning and its own exclusions, and fezolinetant carries none of the estrogen-related ones because it is not a hormone.

Paroxetine 7.5 mg is the only SSRI with an FDA indication for moderate to severe vasomotor symptoms, and its label states it is not indicated for any psychiatric condition. In its 606-woman 12-week trial, median flash frequency fell 5.9 from a baseline median of 10.4, against 5.0 on placebo. That is a median from one trial set beside least-squares means from another, which is context rather than a scoreboard. Where the two diverge in daily life is the paperwork: one asks for repeat liver panels, the other does not, and neither one wins that trade for everyone.

Fezolinetant was first, and for about two and a half years it stood alone. Elinzanetant, a neurokinin 1 and neurokinin 3 receptor antagonist for the same indication, was approved as a new molecular entity on 24 October 2025. It is taken as 120 mg, two 60 mg capsules, once daily at bedtime. Any article still describing the neurokinin approach as a single drug is working from stale information, and so is any comparison that treats hormone therapy as the only alternative on the shelf.

The monitoring gap between the two is the part worth understanding before you choose. The newer label reports transaminase elevations at or above three times the upper limit of normal in 0.6% on drug against 0.4% on placebo across 12 weeks, asks for baseline bloodwork, and then a single follow-up liver test three months after starting. Fezolinetant asks for five follow-up draws inside nine months and carries a boxed warning on top. The newer drug is also contraindicated in pregnancy, which the fezolinetant label does not state.

Same receptor family, materially different obligations.

The regimen is a single 45 mg tablet once daily, with or without food, at about the same time each day. Swallow it whole with liquid; the label instructs that tablets are not to be cut, crushed or chewed. Miss one and you take it as soon as you can, unless fewer than 12 hours remain before the next scheduled dose, then resume the normal schedule the next day. Only one strength exists, supplied in bottles of 30 or 90 tablets.

There is no starting dose, no titration ladder and no written dose reduction.

Peak concentration lands at a median 1.5 hours, steady state arrives after two once-daily doses, and the effective half-life is 9.6 hours in women with vasomotor symptoms. A high-calorie, high-fat meal did not meaningfully change absorption, so timing around meals is not something you need to manage. Both registration trials read a co-primary endpoint at week 4, and SKYLIGHT 2 recorded improvement in frequency and severity as early as week 1. If your own count has not moved by week 4, that is early and useful information to bring back to whoever prescribed it.

FDA approved fezolinetant on 12 May 2023 as the first NK3 receptor antagonist cleared to treat moderate to severe hot flashes from menopause, with no boxed warning. On 12 September 2024 the agency issued a communication about rare serious liver injury, and on 16 December 2024 it added a Boxed Warning, which it describes as its most prominent warning. That communication describes a postmarketing report in which liver test values rose and signs of liver injury appeared after roughly 40 days on the medicine. Drugs@FDA records the matching labeling supplement approved 16 December 2024.

The trials had not surfaced this. A pooled analysis of the three phase 3 studies, with 1,100 women on 45 mg against 952 on placebo, reported transaminase elevations in 1.5% to 2.3%, usually without symptoms, resolving on treatment or after stopping, and no case meeting Hy's law criteria. The label's postmarketing section records peak elevations reaching 50 times the upper limit of normal, with bilirubin up to 5 times normal, all resolving gradually after the drug was stopped. LiverTox summarizes it as a low rate of aminotransferase elevations plus rare instances of clinically apparent liver injury with jaundice. A separate 52-week randomized study covered long-term safety.

Fezolinetant is a CYP1A2 substrate, and concurrent use with any CYP1A2 inhibitor, weak, moderate or strong, is contraindicated outright. The measured magnitudes explain why the label draws the line so hard rather than asking for a dose cut. Fluvoxamine raised area under the curve by 840% and peak concentration by 80%, mexiletine raised area under the curve by 360%, and even cimetidine, classed as a weak inhibitor, doubled it. That makes a complete medication list, including anything you take without a prescription, load-bearing here rather than a formality.

Known cirrhosis is a contraindication, and the drug has not been studied in Child-Pugh Class C impairment at all. Milder liver impairment still raises exposure: area under the curve rose 56% in Child-Pugh Class A and 96% in Class B. Severe renal impairment, meaning an eGFR of 15 to under 30, and end-stage renal disease below 15, are contraindications, while mild and moderate impairment need no dose change. The kidney reasoning is indirect and worth knowing: fezolinetant exposure itself did not change across renal impairment, but the major metabolite's exposure rose about 380% in severe impairment.

There is one public, verifiable dollar figure here and it is not a retail price. The Medicaid National Average Drug Acquisition Cost file published on 2 September 2026 lists the 45 mg tablet at $18.58505 per tablet, with a brand classification for rate setting. That is what surveyed pharmacies pay to acquire it, roughly $558 for a 30-tablet bottle, before any pharmacy markup, dispensing fee, insurance adjudication or manufacturer copay program. We publish no single monthly price for what you would be charged, because nothing we can verify supports one. Nothing generic sits behind that number yet. Drugs@FDA shows the original 2023 New Drug Application as the reference listed drug with no abbreviated applications listed against it, which is the structural reason the acquisition cost has barely moved.

Budget the bloodwork separately as well: a baseline liver panel plus five follow-up panels inside the first nine months is six separate lab events, and whether they land against a deductible depends entirely on your coverage. Ask about the lab charges before the first tablet.

This is the criterion that should decide where you get the prescription, remote or otherwise. The current label requires baseline ALT, AST, alkaline phosphatase and total and direct bilirubin before the first dose, then follow-up hepatic testing monthly for the first 3 months, at 6 months and at 9 months. FDA's own May 2023 approval announcement described routine bloodwork every three months for the first nine months, a schedule the December 2024 boxed warning replaced. If a service quotes you that older cadence, it is reading superseded labeling, and you now know more than the intake form does.

So judge on capability rather than on shipping speed. Can the service order a baseline panel and read the result before anything is dispensed? Will it decline to start you if ALT or AST is at or above twice the upper limit of normal, or total bilirubin is at or above twice normal? Does it schedule and act on all five follow-up draws? Whether the medication suits you is a decision for the prescriber, and this page is information rather than medical advice. And ask what makes them stop: the label sets explicit stopping rules, discontinue if transaminases exceed 5 times normal, or exceed 3 times normal with bilirubin above twice normal.