Pooled analyses of three short-term randomised controlled trials found that esketamine nasal spray plus a newly initiated oral antidepressant produced greater reductions in MADRS total score than antidepressant plus placebo in both women and men with treatment‑resistant depression, with no significant sex effect or treatment-by-sex interaction. Overall safety was similar between sexes, although nausea, dissociation, dizziness and vertigo were reported more frequently in women.
The objective of this analysis was to determine if there are sex differences with esketamine for treatment-resistant depression (TRD). Post hoc analyses of three randomized, controlled studies of esketamine in patients with TRD (TRANSFORM-1, TRANSFORM-2 [18–64 years], TRANSFORM-3 [≥ 65 years]) were performed. In each 4-week study, adults with TRD were randomized to esketamine or placebo nasal spray, each with a newly initiated oral antidepressant. Change from baseline to day 28 in Montgomery-Åsberg Depression Rating Scale (MADRS) total score was assessed by sex in pooled data from TRANSFORM-1/TRANSFORM-2 and separately in data from TRANSFORM-3 using a mixed-effects model for repeated measures. Use of hormonal therapy was assessed in all women, and menopausal status was assessed in women in TRANSFORM-1/TRANSFORM-2. Altogether, 702 adults (464 women) received ≥ 1 dose of intranasal study drug and antidepressant. Mean MADRS total score (SD) decreased from baseline to day 28, more so among patients treated with esketamine/antidepressant vs. antidepressant/placebo in both women and men: TRANSFORM-1/TRANSFORM-2 women—esketamine/antidepressant -20.3 (13.19) vs. antidepressant/placebo -15.8 (14.67), men—esketamine/antidepressant -18.3 (14.08) vs. antidepressant/placebo -16.0 (14.30); TRANSFORM-3 women—esketamine/antidepressant -9.9 (13.34) vs. antidepressant/placebo -6.9 (9.65), men—esketamine/antidepressant -10.3 (11.96) vs. antidepressant/placebo -5.5 (7.64). There was no significant sex effect or treatment-by-sex interaction (p > 0.35). The most common adverse events in esketamine-treated patients were nausea, dissociation, dizziness, and vertigo, each reported at a rate higher in women than men. The analyses support antidepressant efficacy and overall safety of esketamine nasal spray are similar between women and men with TRD. The TRANSFORM studies are registered at clinicaltrials.gov (identifiers: NCT02417064 (first posted 15 April 2015; last updated 4 May 2020), NCT02418585 (first posted 16 April 2015; last updated 2 June 2020), and NCT02422186 (first posted 21 April 2015; last updated 29 September 2021)).
Women have about a twofold greater lifetime risk of major depressive disorder than men and frequently differ from men in clinical presentation, comorbidities and treatment course. Previous large observational studies have reported earlier age at onset, greater episode severity and longer episode duration in women, and sex differences have been reported for response and adverse effects with conventional antidepressants. Findings have been mixed for newer treatments: some studies report no sex difference in rapid-acting ketamine response, while others show variable effects related to menopausal status or hormone therapy with biogenic-amine antidepressants. Jones and colleagues conducted a post hoc analysis of three short-term, phase 3 randomized controlled TRANSFORM trials of esketamine nasal spray plus a newly initiated oral antidepressant in patients with treatment-resistant depression (TRD). The analysis aimed primarily to determine whether efficacy (change in depressive symptoms, response and remission rates, and comorbid anxiety) and overall safety differ between women and men. Secondarily, among women the investigators examined whether outcomes varied by menopausal status or by use of systemic hormonal therapy, and whether clinical factors differentiated responders from non-responders to inform clinical decision-making for women with TRD.
The analysis used data from three phase 3, double-blind, active-controlled, multicentre TRANSFORM trials. Each trial had three phases: a 4-week screening/prospective observational phase to identify non-responders to current antidepressant therapy, a 4-week double-blind treatment phase during which patients received twice-weekly intranasal esketamine or matching placebo together with a newly initiated daily oral antidepressant (SSRI or SNRI), and a post-treatment safety follow-up. TRANSFORM-1 and TRANSFORM-2 enrolled adults aged 18–64 and were pooled for analyses; TRANSFORM-3 enrolled adults aged ≥65 and was analysed separately. Key eligibility required non-response to at least two prior oral antidepressants during the current episode; major exclusions included recent suicidality, psychotic or bipolar disorder, recent moderate-to-severe substance use disorder, and positive drug screens. Dosing varied by study: TRANSFORM-1 used fixed 56 mg and 84 mg doses; TRANSFORM-2 used flexible dosing beginning at 56 mg with possible escalation to 84 mg; TRANSFORM-3 used flexible dosing with options 28 mg, 56 mg and 84 mg. Independent, blinded raters assessed depressive symptoms with the Montgomery–Åsberg Depression Rating Scale (MADRS) at baseline and during treatment. Secondary and patient-reported measures included the Sheehan Disability Scale (SDS), Patient Health Questionnaire-9 (PHQ-9), and Generalized Anxiety Disorder-7 (GAD-7; assessed in TRANSFORM-1/2). Female reproductive lifecycle and hormone use were systematically documented with the Massachusetts General Hospital questionnaire. Efficacy analyses included all randomized patients who received at least one dose of intranasal study drug and one dose of oral antidepressant; safety analyses included all patients who received at least one dose of either medication. TRANSFORM-1 and TRANSFORM-2 data were pooled. The primary endpoint was change from baseline to day 28 in MADRS total score and was analysed by mixed-effects model for repeated measures (MMRM) with baseline MADRS as covariate and fixed effects for treatment, study (pooled TRANSFORM-1/2), region, oral antidepressant class (SSRI or SNRI), day, sex and relevant interactions; patient was a random effect. SDS and PHQ-9 used the same MMRM framework with their respective baselines as covariates. Change in GAD-7 was analysed by analysis of covariance. Response (≥50% MADRS reduction) and remission (MADRS ≤ 12) at day 28 were compared by treatment and sex using the generalized Cochran–Mantel–Haenszel test. Adverse events were summarised by treatment group and sex. The trials had ethics approval and all participants provided written informed consent.
Of 711 randomized patients across the TRANSFORM trials, efficacy was evaluated in 702 patients (464 women, 238 men) and safety in 705 patients. Completion rates of the double-blind phase were high for both sexes (women 90.7% [427/471]; men 86.3% [207/240]). Baseline demographics and psychiatric history were generally similar between women and men within each study; by design TRANSFORM-3 enrolled older patients (≥65). In pooled TRANSFORM-1/2 about 48.0% (182/379) of women were pre-menopausal. Common medical comorbidities included hypertension, cardiovascular disease, diabetes and thyroid disease, with thyroid disease and levothyroxine use more common in women; cardiovascular disease and diabetes were more common in older men. Mean MADRS scores decreased from baseline to day 28 in all groups, with greater improvement in patients treated with esketamine plus an oral antidepressant versus antidepressant plus placebo for both women and men. Reported mean (SD) MADRS change at day 28 was, in pooled TRANSFORM-1/2: women esketamine/antidepressant −20.3 (13.19) versus antidepressant/placebo −15.8 (14.67); men esketamine/antidepressant −18.3 (14.08) versus antidepressant/placebo −16.0 (14.30). In TRANSFORM-3 the corresponding values were: women −9.9 (13.34) versus −6.9 (9.65); men −10.3 (11.96) versus −5.5 (7.64). Least-squares mean differences versus antidepressant/placebo reported in the discussion were in the clinically meaningful range (about 2–3 points): in pooled TRANSFORM-1/2 the treatment difference was numerically larger in women (LS mean difference −4.5, SE 1.41; 95% CI −7.26, −1.70) than men (LS mean difference −1.6, SE 2.04; 95% CI −5.60, 2.41); in TRANSFORM-3 the point estimates favored men numerically but confidence intervals overlapped. The MMRM analyses found no statistically significant main effect of sex or treatment-by-sex interaction (p > 0.35). Responder and remission proportions at day 28 were numerically higher with esketamine/antidepressant than with antidepressant/placebo for both sexes. In TRANSFORM-3 the remission (but not response) rate was numerically higher among older women versus older men, but numbers were small. By menopausal status (TRANSFORM-1/2), pre-menopausal and post-menopausal women receiving esketamine achieved similar response rates; a small peri-menopausal cohort (n ≈ 24–25) had numerically lower response. Hormonal therapy use showed a differential pattern: in the antidepressant/placebo arm hormone users had higher response at day 28 (73.7% [14/19]) than non-users (40.3% [48/119]), whereas in the esketamine/antidepressant arm response was 54.2% (28/48) in hormone users versus 62.3% (104/167) in non-users. Improvements in functioning (SDS) and self-reported depressive symptoms (PHQ-9) mirrored MADRS findings, with no significant sex effect or treatment-by-sex interaction (p > 0.20). For GAD-7 there was no treatment-by-sex interaction (p > 0.52); a trend for a sex effect (p = 0.07) suggested greater anxiety score change in women irrespective of treatment. The most common treatment-emergent adverse events (incidence > 20% in esketamine groups) were nausea, dissociation, dizziness and vertigo. Nausea and dissociation occurred at higher rates in women than men across age groups. Patterns for dizziness and vertigo varied by age subgroup. Increased blood pressure was reported overall in 9.3% of esketamine-treated patients but was numerically higher in younger men (12.8%) and older women (17.8%). Most adverse events were mild to moderate, occurred on dosing days, resolved the same day, and were generally not treatment-limiting. Median duration of dissociation events ranged from about 0.7 to 1 hour; 3.1% were classified as severe, none were considered serious. Serious adverse events during esketamine treatment were reported for two women (0.7%) and four men (2.9%); one male participant died on day 55 following a road traffic accident that was considered doubtfully related to study medication. Discontinuations due to adverse events occurred in 4.8% (20/415) of esketamine-treated patients versus 1.7% (5/287) in the placebo arm; among esketamine discontinuations 12 were women and 8 were men, with a range of events cited (e.g., increased systolic blood pressure, dizziness, nausea, depression, headache).
Jones and colleagues interpret the results as showing similar and clinically meaningful antidepressant benefit of esketamine nasal spray plus an oral antidepressant in both women and men with TRD over 4 weeks. Although point estimates of treatment difference versus antidepressant/placebo varied by sex and by study (pooled TRANSFORM-1/2 versus TRANSFORM-3), the treatment-by-sex interaction was not statistically significant and confidence intervals overlapped, supporting no robust sex-based efficacy difference. The magnitude of the between-group effect for both sexes was stated to be within the range considered clinically meaningful (about 2–3 points on MADRS) and comparable with effects seen in trials of recently approved biogenic-amine antidepressants versus placebo. The authors note that the absence of sex differences may relate in part to similar pharmacokinetics of esketamine between males and females, and they situate their findings amid inconsistent prior literature on sex differences for other antidepressant classes. Regarding menopausal status and hormone use, the analysis suggested that post-menopausal women achieved benefit comparable to pre-menopausal women; peri-menopausal women constituted a small group with numerically lower response. Use of hormonal therapy was associated with higher response in the antidepressant/placebo arm but did not increase response in the esketamine arm. The investigators highlight that adverse events such as nausea and dissociation were reported more commonly in women and that increased blood pressure was relatively more frequent in older women. Key limitations acknowledged by the authors include the post hoc nature of the sex-stratified analyses, lack of stratification by sex at randomization, small sample sizes for subgroup analyses (particularly menopausal subgroups and TRANSFORM-3), differences in dosing in TRANSFORM-3 (including a lower 28 mg option), exclusion of patients with some common comorbidities, and limited racial diversity, all of which constrain generalisability. The authors recommend cautious interpretation of subgroup findings and note that the overall findings support data-informed decision-making for women with TRD treated with esketamine.
These post hoc analyses support that esketamine nasal spray, combined with an oral antidepressant, provides antidepressant efficacy and an overall safety profile in women with TRD that is not notably different from that observed in men. The findings are presented as additional evidence to inform clinical decision-making for women with treatment-resistant depression.
The methods of the TRANSFORM studies are published elsewhere. Study methods salient to the work reported here are summarized below.
Efficacy was analyzed in a data set that included all randomized patients who received at least 1 dose of intranasal study drug (esketamine or placebo) and 1 dose of oral antidepressant, and adverse events were analyzed in a data set that included all patients that received at least 1 dose of either medication. Data from the TRANSFORM-1 and TRANSFORM-2 studies were pooled. Baseline characteristics and psychiatric history were summarized by sex using descriptive statistics. The prevalence of comorbid anxiety at baseline was determined using one of the following: generalized anxiety disorder current, panic disorder current, social anxiety disorder current, posttraumatic stress disorder current, or obsessive-compulsive disorder current based on the MINI, or having screening and baseline GAD-7 total score ≥ 10 (for TRANSFORM-1/ TRANSFORM-2 only). The primary efficacy endpoint in the TRANSFORM studies-change from baseline to endpoint (day 28) in MADRS total score-was analyzed by sex using a mixedeffects model for repeated measures (MMRM). The model included baseline MADRS total score as a covariate, and treatment, study (TRANSFORM-1/TRANSFORM-2 only), region, oral antidepressant class (SNRI or SSRI), day, sex, day-by-treatment, treatment-by-sex, and day-by-treatmentby-sex interaction as fixed effects, and a random patient effect. Changes in SDS and PHQ-9 were analyzed using the MMRM model described for the primary efficacy endpoint, but using the respective baseline score (SDS, PHQ-9) as covariate. Change in GAD-7 was analyzed using analysis of covariance with baseline GAD-7 as a covariate and treatment, region, oral antidepressant class, sex, and treatmentby-sex as factors. Response rate (defined as ≥ 50% decrease from baseline MADRS total score) and remission rate (defined as MADRS ≤ 12) at day 28 were analyzed by treatment group and sex using the generalized Cochran-Mantel-Haenszel (CMH) test. Response rate was also evaluated by menopausal status and by use of sex hormones, yes or no) among women. In other analyses, frequency distributions were provided by treatment group and sex for adverse events as a measure of safety.
Similar and robust improvement of depressive symptoms from baseline was observed with esketamine nasal spray compared to placebo nasal spray, in conjunction with an oral antidepressant, among both women and men with TRD. While the between-group difference observed with esketamine/antidepressant vs. antidepressant/placebo was numerically higher among women than men in TRANS-FORM-1/TRANSFORM-2 (LS means (SE) -4.5 (1.41) 95% CI -7.26, -1.70 for women vs. -1.6 (2.04) 95% CI -5.60, 2.41 for men) and vice versa in TRANSFORM-3 (LS means (SE) -3.4 (2.41) 95% CI -8.14, 1.41 for women vs. -5.0 (3.05) 95% CI -11.05, 1.03 for men), the treatment-by-sex interaction was not statistically significant (p > 0.35) and the 95% CIs for the differences overlap. Furthermore, the betweengroup difference observed vs. antidepressant/placebo for both sex subgroups in TRANSFORM-1/TRANSFORM-2 and TRANSFORM-3 was in the range considered clinically meaningful (2-point to 3-point difference)) and is consistent with that observed in clinical trials of the most recently approved biogenic amine antidepressants compared with only a placebo rather than an active comparator). The proportions of patients who were responders at day 28 and the proportion of patients in remission at day 28 were numerically higher among both women and men treated with esketamine/antidepressant as compared to antidepressant/placebo. While it is noted that the remission rate, but not response rate, was numerically higher among the older women, compared to the older men in TRANSFORM-3, the small cohort sizes limit a conclusion being made from the comparison. Additionally, as noted in the "Materials and methods" section, the TRANSFORM-3 study, unlike TRANSFORM-1 and 2, included a lower 28 mg dose and post hoc analyses of TRANSFORM-3 dataMean (SD) change from baseline to day 28 for SDS, PHQ-9, and GAD-7 total score in pooled TRANSFORM-1/ TRANSFORM-2 trials SDS total score ranges from 0 to 30; a higher score indicates greater impairment. PHQ-9 total score ranges from 0 to 27; a higher score indicates greater depression. GAD-7 total score ranges from 0 to 21; a higher score indicates more anxiety. Negative change in SDS total score, PHQ-9 total score, and GAD-7 total score indicates improvement for each GAD-7 Generalized Anxiety Disorder 7-item; LS least squares; PHQ-9 Patient Health Questionnaire 9-item; SD standard deviation; SDS Sheehan Disability Scale revealed several factors that potentially contributed to its failure to achieve statistical significance on the primary endpoint. The treatment benefit of esketamine, regardless of sex, was also observed based on the patient-reported outcomes of functioning (SDS total score), severity of anxiety (GAD-7 total score), and depressive symptoms (PHQ-9 total). The absence of evidence of sex-based differences may be explained, in part, by the absence of differences in the pharmacokinetics of esketamine between male and female subjects (Spravato TM Prescribing Information 2020). Sex differences in efficacy outcomes have been reported for other antidepressants, although findings have been inconsistent. In some studies, older men responded better to tricyclic antidepressants than women, and in other studies women responded better to SSRIs, and to a lesser extent SNRIs, than men. Alternatively, other studies have reported no sex-based differences in antidepressant efficacy. Differences in study design (prospective, retrospective, meta-analysis), patient selection criteria (e.g., age of patients, clinical presentation, severity/duration of depression), study drug (mechanism of action, dosage, duration of treatment), and response criteria may explain the inconsistency in these findings across studies. As noted in the "Introduction" section, findings regarding the effect of menopausal status on response to antidepressants have also been mixed, with some research groups reporting greater response among women treated with an SSRI vs. a tricyclic antidepressant, driven by between-group differences in premenopausal women, and not in those who were post-menopausal. In a pooled analysis of data from 8 randomized, controlled trials, older women exhibited lower remission rates on SSRI than younger women, a trend that was reversed for those taking hormone replacement therapy; remission rates were higher for women treated with SNRI, irrespective of age and hormone replacement therapy. Although small sample size precluded analysis of efficacy by concomitant oral antidepressant class (SSRI, SNRI), by sex or menopausal status, oral antidepressant class was included in the MMRM as a fixed effect. With esketamine, while the response rate with esketamine/antidepressant was numerically lower in the small sample of peri-menopausal women (n = 24), it appears that post-menopausal women with TRD achieve the same benefit that pre-menopausal women do. These findings are consistent with those of, who found no difference between women and men treated with intravenous ketamine for rapid reduction of depressive symptoms, with no difference observed based on menopause status, suggesting antidepressants with a glutamatergic mechanism of action, unlike biogenic amines, may not be impacted by the reproductive life cycle of women. In the TRANSFORM-1/2 studies, use of hormonal therapy increased response rate at day 28 in the antidepressant/ placebo arm, but not in the esketamine/antidepressant arm. Others have also observed that hormone therapy impacts response among women receiving SSRI or SNRI. The most common adverse events experienced by esketamine-treated patients were dissociation, headache, nausea (each reported at a rate higher in women than men), dizziness (reported at a rate higher in older women than older men [in TRANSFORM-3]), and vertigo (reported at a rate higher in younger women than younger men [in TRANS-FORM-1/TRANSFORM-2]). Increased blood pressure was reported most often among older women. This trend is in line with the higher risk for women having adverse drug reactions, in general, and for specific types of events during treatment with antidepressants (e.g., weight gain with SSRIs). Fig.Difference in least square means for SDS, PHQ-9, and GAD-7 total score by sex in pooled TRANSFORM-1/TRANSFORM-2 trials. CI = confidence interval; GAD-7 = Generalized Anxiety Disorder 7-item; LS = least squares; PHQ-9 = Patient Health Questionnaire 9-item; SDS = Sheehan Disability Scale. Notes: SDS total score ranges from 0 to 30; a higher score indicates greater impairment. PHQ-9 total score ranges from 0 to 27; a higher score indicates greater depression. GAD-7 total score ranges from 0 to 21; a higher score indicates more anxiety. Negative change in SDS total score, PHQ-9 total score, and GAD-7 total score indicates improvement for each, and a negative difference favors esketamine
The purpose of this study is to compare the efficacy and safety of switching treatment-resistant depression (TRD) subjects from a prior antidepressant treatment (to which they have not responded) to either intranasal esketamine plus a new oral antidepressant or switching to a new oral antidepressant plus intranasal placebo.
Papers cited by this study that are also in Blossom
Chen, X., Hou, X., Bai, D. et al. · American Journal of Psychiatry (2023)