The Patient's Perspective on the Effects of Intranasal Esketamine in Treatment-Resistant Depression

Participants were consecutively recruited from the Psychiatric Unit of Fondazione Policlinico Universitario Agostino Gemelli IRCCS. Inclusion criteria required a primary DSM-5-TR diagnosis of major depressive disorder, an ongoing major depressive episode while taking an SSRI or SNRI, and inadequate response to at least two antidepressants from different classes given at adequate dose and duration. Exclusion criteria included current contraindications to esketamine such as pregnancy, cardiovascular disease, active substance or alcohol use, psychosis, or neurocognitive disorders. Esketamine nasal spray was delivered adjunctively to existing oral antidepressant therapy over three months: twice weekly during the first month, then once weekly for the subsequent eight weeks. Initial dosing was 28 or 56 mg depending on an age cut-off of 65 years; subsequent doses could be increased to 56 or 84 mg according to clinical judgement and tolerability. Concomitant medications were permitted and not altered during follow-up. Cognitive screening with the Mini‑Mental State Examination excluded neurocognitive disorder when MMSE total score exceeded 26. Assessments occurred at baseline and at 4, 8 and 12 weeks in the outpatient setting before ESK-NS administrations. Clinician-rated measures included the Montgomery‑Åsberg Depression Rating Scale (MADRS), Hamilton Anxiety Rating Scale (HARS), Digit Symbol Substitution Test (DSST) for cognition, and specific MADRS items for symptoms such as anhedonia (item 8), sleep (item 4), cognition (item 6) and suicidality (item 10). Self-report instruments comprised the Beck Depression Inventory (BDI) with item-level analyses for sleep (item 16) and suicidality (item 9), the Snaith‑Hamilton Pleasure Scale (SHAPS) for anhedonia, the Perceived Deficits Questionnaire‑5 (PDQ‑D5) for perceived cognitive deficits, and the Zung Self-Rating Anxiety Scale (SAS). Remission was defined as MADRS total <10 and response as ≥50% reduction from baseline at any time-point. For analysis, descriptive statistics summarised baseline characteristics. Relationships between clinician and patient measures used Pearson correlations. Longitudinal changes from baseline to 1, 2 and 3 months were analysed with a mixed model for repeated measurements (MMRM) that treated time as a fixed effect, adjusted for baseline score and the baseline-by-time interaction, and used all available observations (full-analysis set). A two‑tailed significance threshold of p < 0.05 was applied; analyses were performed in SPSS v.28.

Patients presented with moderate to severe depression at baseline and overall improved during the three-month ESK-NS treatment, which was delivered at a mean dose of 78.4 ± 11.43 mg. Strong, statistically significant correlations were observed between corresponding clinician-rated and self-reported measures: MADRS and BDI total scores (r = 0.65, p < 0.001); MADRS item 8 (anhedonia) and SHAPS (r = 0.47, p < 0.001); MADRS item 4 (sleep) and BDI item 16 (r = 0.34, p = 0.01); MADRS item 6 (cognition) and PDQ‑D5 (r = 0.59, p < 0.001); MADRS item 10 (suicide) and BDI item 9 (r = 0.57, p < 0.001); and HARS and SAS total scores (r = 0.86, p < 0.001). Clinician ratings showed significant reductions beginning in the first month for MADRS total score and MADRS subitems addressing anhedonia (item 8), sleep (item 4) and suicidality (item 10), as well as for HARS anxiety scores. Cognitive performance on MADRS item 6 and on the DSST improved from the second month. Patients' self-reports indicated early reductions after one month in BDI total score, BDI item 16 (sleep), and SAS (anxiety). Self-rated measures of anhedonia (SHAPS) and perceived cognition (PDQ‑D5) showed significant improvement at month two, while reduction in BDI item 9 (suicide ideation) began after one month and reached statistical significance at the three‑month endpoint. Six patients discontinued esketamine prematurely for varied reasons: three for medical comorbidities at different time-points (after first administration, after one month, after two months), two during the third month owing to perceived lack of benefit, and one due to relapse into substance use. Among patients with complete available observations (n = 19), 52.6% met response criteria and 31.6% met remission criteria at different assessment points. Specifically, at two months there were four responders and one remitter; at three months six responders and four remitters; and one patient achieved remission after the first month. Adverse effects were generally mild and transient: nausea, short-lived blood pressure increases during the two-hour post‑administration period (most marked in the first 40 minutes), and brief dissociative symptoms detected mainly after initial administrations (CADSS scores >4) that diminished over subsequent sessions. No clinically relevant side effects were reported during the treatment period.

Pepe and colleagues interpret their findings as evidence that three months of intranasal esketamine is associated with overall clinical improvement in a sample of patients with moderately to severely treatment‑resistant depression, and that patient self-reports broadly converged with clinician ratings. The observed response and remission rates at endpoint align with prior real-world studies suggesting sustained benefit at three months, and the authors note that delayed responses may occur in more severely treatment‑resistant cases. The discussion highlights the sample's clinical complexity — multiple lifetime depressive episodes, long current episode duration, frequent suicidality, significant anxiety, and medical and psychiatric comorbidities — all of which support substantial illness burden and may influence the tempo of symptomatic change. Anxiety levels were high at baseline and improved rapidly, consistent with other evidence for anxiolytic effects of ketamine/esketamine. The authors note the potential modifying role of concomitant benzodiazepine use, which has been proposed as a predictor of delayed response, but they also point out that existing reports do not consistently show negative interactions with sedative‑hypnotics. A key observation was discordant timing across symptom domains: clinicians detected early improvement across many domains, while patients reported delayed improvement for anhedonia and suicidality (significant change in SHAPS at month two and in BDI item 9 at month three). The authors discuss neurobiological hypotheses for these delays, referencing fronto‑striatal dopaminergic dysfunction in anhedonia, the role of glutamatergic modulation and prefrontal circuits, and potential contributions of neuroinflammation and neurotrophic pathways such as BDNF and mTORC1 to treatment effects. They suggest that more severe, recurrent, or inflammatory‑linked symptom clusters may need longer to yield noticeable subjective improvement. Practical implications proposed by the investigators include integrating patients' perspectives routinely to better characterise symptomatic phenotypes and to personalise interventions, and combining pharmacological treatment with psychotherapeutic and rehabilitative strategies that might enhance engagement and accelerate perceived benefit — particularly for anhedonia and suicidality. The authors acknowledge important limitations: small sample size, absence of a control group, lack of TRD‑specific standardised tools in the assessment battery, the clinically severe and specific sample, and the single‑centre design limiting generalisability. Strengths cited include emphasis on the patient's viewpoint, a longitudinal three‑month follow-up, and use of validated psychometric instruments including an objective cognitive task. The paper concludes that larger real‑world studies are needed to define symptomatic profiles that predict response to esketamine and to refine personalised treatment approaches.

The authors conclude that careful clinical and psychometric assessment incorporating both clinician ratings and patient self-reports is valuable in TRD, and that targeting prognostic factors such as anhedonia and suicidality may enhance treatment outcomes. They advocate for multimodal approaches — combining pharmacological, psychotherapeutic and rehabilitative strategies — and for inclusion of biological markers alongside psychometric measures in future studies to improve understanding of mechanisms and predictors of response. The study's findings support continued real-world investigation into how symptomatic profiles influence response to intranasal esketamine and argue for larger trials aimed at personalising care for patients with difficult‑to‑treat depression.