Women improved more than men on esketamine — and more on placebo too

Precision psychiatry's premise is that the right treatment for a given patient can be predicted from something measurable about that patient. Sex assigned at birth is the cheapest such variable available — it is already in every trial dataset — and it is among the least analysed.

A pooled analysis published in Molecular Psychiatry on February 18 went back through five randomised, double-blind, placebo-controlled trials of intranasal esketamine in treatment-resistant depression to ask what sex-assigned-at-birth explains [s1].

The analysis

Across the five trials, adults with treatment-resistant depression received intranasal esketamine or placebo twice weekly for four weeks, alongside a newly initiated oral antidepressant [s1]. That design detail matters for interpreting everything that follows: every participant, in both arms, was also starting a conventional antidepressant.

The authors evaluated the effect of sex-assigned-at-birth on overall depression severity, measured by total Montgomery-Åsberg Depression Rating Scale (MADRS) score, and then across four symptom factors — sadness, negative thoughts, detachment, and neurovegetative symptoms — plus rates of clinical response and remission [s1].

What they found

Esketamine improved total MADRS scores in both sexes [s1].

But the sex differences the analysis surfaced mostly did not track the drug. Toward the end of the trials, females showed greater improvement in total MADRS scores and higher odds of treatment response than males — in both the placebo and the esketamine arms [s1].

The same pattern held at the symptom level. Females showed more pronounced reductions in the sadness and detachment factors at the end of the trials, and in the neurovegetative factor on day 15, regardless of treatment group [s1].

One finding did appear specific to the drug and to males: males showed a significant reduction in sadness symptoms after esketamine on day 2 [s1].

Why "in both arms" is the whole story

A sex difference that appears in the placebo arm as well as the treatment arm is not a sex difference in how the drug works. It is a sex difference in how people improve during a depression trial — which could reflect the natural course of the illness, the effect of the concurrently started oral antidepressant, the placebo response itself, expectancy, symptom reporting, or the psychometric behaviour of the MADRS across sexes.

This is a distinction that gets flattened in summary. "Women respond better to esketamine" would be a claim about the drug. What the data show is closer to "women's scores improved more over the course of these trials, whether or not they received esketamine."

The one result that is arm-specific — the day-2 sadness reduction in males after esketamine — is also the most fragile, being a single symptom factor at a single early timepoint.

Limits

This is a pooled secondary analysis, not a prospective test. The sex comparison was not what any of the five trials were designed or powered to answer, and the analysis inherits every eligibility criterion and measurement choice those trials made.

Sex-assigned-at-birth is a binary variable standing in for a large bundle of biological and social factors — hormonal status, body composition and therefore drug exposure, comorbidity patterns, help-seeking behaviour, and how symptoms are described to a rater. The analysis identifies that the bundle matters. It does not identify which part of it does.

Multiple comparisons are a live concern: four symptom factors across multiple timepoints, in two arms, in two sex groups.

And these are trials of esketamine specifically, with an oral antidepressant started at the same time. Nothing here transfers automatically to ketamine given intravenously, to esketamine given without a concurrent antidepressant, or to any other rapid-acting agent.

What the authors take from it

Their conclusion is a methodological one: sex-assigned-at-birth influences overall antidepressant response and shapes the trajectory and symptom profile of improvement, and should be incorporated as a key variable in optimising treatment-resistant depression strategies [s1].

That is a defensible reading. The stronger claim — that sex should inform which drug a given patient receives — would require the effect to be drug-specific, and in these data it largely is not.

What to watch

Whether any prospective trial pre-specifies sex as a stratification variable and reports arm-by-sex interaction rather than within-arm improvement, and whether the day-2 male sadness signal appears anywhere else.

This article describes a pooled secondary analysis of clinical trial data. Nothing here is medical advice or a recommendation about any treatment.

Sources

• [s1] Huc M, Siddiqi S, Myers M, Colman I, Salmaso N, Jaworska N, Aguilar-Valles A. Sex differences in placebo and antidepressant response to intranasal esketamine for treatment-resistant depression. Molecular Psychiatry, published online 2026-02-18.