This post hoc analysis of the TRANSFORM-2 trial assessed the effects of esketamine plus an oral antidepressant (AD) using the Patient Health Questionnaire (PHQ-9) and Montgomery-Asberg Depression Rating Scale (MADRS). The odds of improving in those treated with esketamine plus AD were at least two times greater than with placebo plus AD.
Objective
The objective of this study was to determine which symptoms measured by the Patient Health Questionnaire (PHQ-9) and Montgomery-Asberg Depression Rating Scale (MADRS) improve in those treated with esketamine nasal spray in combination with oral antidepressant (AD) compared with those treated with placebo plus AD for adult patients with treatment-resistant depression (TRD). These results complement the interpretation of PHQ-9 and MADRS total scores.
Methods
The TRANSFORM 2 study evaluated the efficacy and safety of esketamine nasal spray in combination with AD. This posthoc analysis used PHQ-9 and MADRS data to evaluate symptom changes. The total scores change and proportions of individual item change scores on the PHQ-9 and MADRS were evaluated at days 15 and 28; analysis of variance was used to test differences in total scores. Generalized estimation equations of logistic regression models were used to estimate the likelihood of improvement on instrument items.
Results
The mean total score reduction of the PHQ-9, indicating improvement, was greater in the esketamine plus AD arm compared with placebo plus AD at day 15 (- 1.8; p = 0.045) and day 28 (- 2.8; p = 0.006). Proportions of those who improved (≥ 1 point on a 4-point scale and ≥ 2 points on a 7-point scale for the PHQ-9 and MADRS, respectively) was greater in the esketamine plus AD group compared with the placebo plus AD group across all items. The odds of improving for those in the esketamine plus AD group compared with the placebo plus AD group were over two times greater on the PHQ-9 items: Little interest/pleasure in things (OR 2.252, 95% CI 1.165-4.355); Feeling down, depressed, or hopeless (OR 2.767, 95% CI 1.400-5.470); and Feeling tired or having little energy (OR 2.171, 95% CI 1.153-4.087). The mean reduction in total scores on the MADRS, indicating improvement, was numerically greater at day 15 (- 2.0; p = 0.189) and statistically significantly greater at day 28 (- 4.4; p = 0.017) in the esketamine plus AD arm compared with placebo plus AD. The odds of improving for those in the esketamine plus AD group compared with the placebo plus AD group were over two times greater on the MADRS items measuring Apparent sadness (OR 2.007, 95% CI 1.096-3.674); and Inability to feel (OR 2.099, 95% CI 1.180-3.735).
Conclusion
Improvement in mean total scores in those treated with esketamine plus AD compared with placebo plus AD are important results to confirm efficacy. The odds of improving in those treated with esketamine plus AD was at least two times greater than with placebo plus AD on three patient- and two clinician-reported individual symptoms of TRD. These findings provide patient-relevant quantification of the esketamine plus AD treatment benefit, adding understanding as to which symptoms are most improved with treatment.
Depression is a leading cause of long-term disability worldwide, and around 30% of patients fail to respond to at least two pharmacological treatments, meeting criteria for treatment-resistant depression (TRD). Individuals with TRD experience greater morbidity, poorer quality of life, higher relapse rates and greater healthcare costs than patients with more treatment-responsive major depressive disorder. Prior to esketamine's approval in 2019, few pharmacological options were specifically licensed for TRD, and many augmentation strategies had limited efficacy or tolerability. Clinical outcome assessment instruments such as the Patient Health Questionnaire-9 (PHQ-9; a patient-reported outcome) and the Montgomery–Åsberg Depression Rating Scale (MADRS; a clinician-reported outcome) are commonly used to quantify overall depression severity, but total scores collapse across distinct symptoms and may obscure which specific symptoms drive observed change. Floden and colleagues set out to examine symptom-level change on the PHQ-9 and MADRS using data from the Phase III TRANSFORM-2 randomised trial. The objective was to determine which individual items (symptoms) show greater improvement with esketamine nasal spray plus a newly initiated oral antidepressant (esketamine plus AD) versus intranasal placebo plus a newly initiated oral antidepressant (placebo plus AD) over the 4-week double-blind induction period. By analysing item-level shifts and the likelihood of improvement for each symptom, the study aimed to provide more patient-relevant interpretation of the trial's total-score findings.
The analyses used data from TRANSFORM-2, a Phase III, multicentre, double-blind, active-controlled trial with flexible dosing of esketamine nasal spray (56 mg or 84 mg) administered twice weekly for 4 weeks alongside a newly initiated open-label oral antidepressant. Adults meeting DSM-5 criteria for major depressive disorder without psychotic features and meeting criteria for TRD (non-response to at least two antidepressants in the current episode, and an Inventory of Depressive Symptomatology–Clinician-rated 30-item total score ≥ 34) were randomised 1:1 to receive esketamine plus oral AD or intranasal placebo plus oral AD. The trial excluded patients at serious risk for suicide as determined by other assessments. Patient-reported symptoms were measured with the PHQ-9 (nine items, each scored 0–3, recall 2 weeks, total range 0–27). Clinician-assessed symptoms were measured with the MADRS (ten items, each scored 0–6, total range 0–60); MADRS ratings were collected by independent remote raters who received standardised training. Electronic data capture was used for PROs, and translated versions were deployed where appropriate. The analytic population comprised patients in the intent-to-treat population with at least one clinical outcome assessment. Outcomes were assessed at baseline, day 15 and day 28. Total-score changes were compared by treatment arm using analysis of variance (ANOVA). Item-level improvement was defined using within-patient thresholds: a decrease of ≥ 1 point on a PHQ-9 item (a 25% shift on the 4-point item scale) or ≥ 2 points on a MADRS item (a 29% shift on the 7-point item scale). The analysis also referenced meaningful change thresholds (MCTs) for total scores (PHQ-9 MCT = 6 points; MADRS MCT = 10 points) and noted that the minimal important difference (MID) for MADRS mean differences is commonly considered about 2 points. Proportions of patients achieving item-level improvement at days 15 and 28 were calculated, and generalized estimating equations (GEE) logistic regression models with an autoregressive correlation structure for repeated measures were used to estimate odds ratios (ORs) and 95% confidence intervals (CIs) for the likelihood of improvement, with fixed effects for treatment, time point and their interaction.
The analytic population was defined as adult patients with COA assessments at any time point within the intent-to-treat population. These exploratory post-hoc analyses were performed using the statistical analysis system (SAS) Version 9.4. Distributions and descriptive statistics of demographic and clinical characteristics were described at baseline for each treatment group and overall. Total scores and change from baseline on the PHQ-9 and MADRS were displayed by treatment arm from baseline to days 15 and 28; treatment differences were calculated using analysis of variance (ANOVA). Categorical distributions of change from baseline in each item of the PHQ-9 and MADRS were displayed graphically for days 15 and 28. The definition of within-patient item-level improvement corresponded to the categorical shift of at least the magnitude of the total score meaningful change thresholds (MCTs). The within-patient MCTs for the PHQ-9 and MADRS are 6 points, or 22% of the total score range and 10 points, or 17% of the total score range, respectively. For example, the minimal important difference (MID) of the MADRS is generally agreed to be 2 points. Individual patients were classified as improved if they had a decrease of at least 1 point on the PHQ-9 items (representing a 25% shift, measured on a 4-point scale) or 2 points on the MADRS items (representing a 29% shift, measured on a 7-point scale). It is important to note that the MCT differs from the threshold for clinical relevance both conceptually and often numerically. While the threshold for within-person meaningful change refers to the amount of change each individual needs to achieve to be classified as improving or deteriorating, the MID is a measure of clinical relevance used to judge clinical significance of mean difference in change between groups. The proportions of patients with categorical improvement from baseline of PHQ-9 and MADRS items, by treatment arm, were calculated at Days 15 and 28. Generalized estimation equations (GEE) of logistic regression models were used to estimate the likelihood of improvement. The GEE model of an 'improved' binary event was regressed on fixed effects of treatment arm, time point, and their interaction terms, as well as an R-side random effect with autoregressive correlation structure to account for repeated measures. Models generated odds ratios (OR) with 95% confidence intervals (CI) to compare the likelihood of 'improvement' between treatment groups.
Create a free account to open full-text PDFs.
Baseline characteristics were similar between arms. The sample was, on average, 46 years old (SD 11.89), predominantly non-Hispanic (93%), White (93%), and female (62%), with a mean duration of TRD of about 12 years (SD 10.2). Baseline PHQ-9 and MADRS scores were comparable across treatment groups. PHQ-9 total scores improved from baseline in both groups at days 15 and 28, but mean reductions were larger with esketamine plus AD. The extracted text reports a greater mean reduction favouring esketamine at day 15 (mean difference −1.8; p = 0.045) and at day 28 (mean difference −2.8; p = 0.006). At the item level, most patients showed improvement on nearly all PHQ-9 items except item 9 (suicidal ideation), where many patients showed no change — likely reflecting low baseline endorsement and trial exclusion criteria for serious suicide risk. The proportion of patients with a ≥ 1-point improvement on each PHQ-9 item was greater in the esketamine plus AD arm at both assessment points. GEE models estimated that the odds of item-level improvement were more than twofold for several PHQ-9 items: item 1 (“Little interest/pleasure in things”) OR 2.252 (95% CI 1.165–4.355); item 2 (“Feeling down, depressed, or hopeless”) OR 2.767 (95% CI 1.400–5.470); and item 4 (“Feeling tired or having little energy”) OR 2.171 (95% CI 1.153–4.087). Item 6 (“Feeling bad about yourself…”) also showed a nominally significant increase in odds: OR 1.878 (95% CI 1.000–3.527). MADRS total scores likewise improved in both arms, with a numerically larger change at day 15 (mean difference −2.0; p = 0.189) and a statistically significant greater reduction at day 28 favouring esketamine plus AD (mean difference −4.4; SE 1.85; p = 0.017; 95% CI −8.10 to −0.80). Item-level MADRS data showed higher proportions of ≥ 2-point improvement across all items in the esketamine arm, especially by day 28. Five MADRS items had statistically significant greater odds of improvement with esketamine plus AD: item 1 (“Reported sadness”) OR 1.844 (95% CI 1.014–3.354); item 2 (“Apparent sadness”) OR 2.007 (95% CI 1.096–3.674); item 3 (“Inner tension”) OR 1.891 (95% CI 1.080–3.313); item 6 (“Concentration difficulties”) OR 1.880 (95% CI 1.054–3.354); and item 8 (“Inability to feel”) OR 2.099 (95% CI 1.180–3.735). Reduced appetite (item 5) was the symptom least likely to improve on both instruments. The extracted text does not provide a full table of absolute responder counts or adverse event rates in these sections.
Floden and colleagues interpret the item-level findings as congruent with the total-score benefits observed for esketamine plus oral antidepressant versus placebo plus oral antidepressant during the 4-week induction. Multiple individual symptoms showed nominally significant greater likelihoods of improvement with esketamine: four PHQ-9 items (including the two cardinal symptoms, low interest/pleasure and feeling down) and five MADRS items (including apparent sadness and inability to feel/anhedonia). The authors emphasise that analysing individual items offers greater specificity about which symptoms are likely to respond, which can aid clinician–patient discussions and shared decision-making about expected benefits versus risks. The discussion notes that the greatest item-level odds ratios (OR > 2.0) were observed for PHQ-9 items assessing anhedonia, depressed mood and low energy, and for MADRS items assessing apparent sadness and anhedonia. The authors also highlight that items reflecting reduced appetite were least responsive. Key limitations acknowledged include the short follow-up (only baseline and two post-baseline time points within a 4-week double-blind period), and the absence of analyses comparing outcomes by the specific oral antidepressant used, despite clinicians prescribing one of four oral ADs. The investigators further note that these are results from a single randomised trial and that replication in other datasets would strengthen confidence in the symptom-level findings.
Analysing single items on the PHQ-9 and MADRS provides a more granular picture of how esketamine plus a newly initiated oral antidepressant affects specific depressive symptoms in adults with TRD. Treatment with esketamine plus AD led to greater improvement in total scores and to higher odds of improvement on several individual patient- and clinician-reported symptoms compared with placebo plus AD, with the largest gains observed for cardinal depressive symptoms and for anhedonia as rated on the MADRS. These item-level results are intended to enhance interpretability of overall score change for patients and clinicians.
This study demonstrates a pattern of item-level results congruent with the total PHQ-9 and MADRS score change, favoring treatment with esketamine nasal spray plus oral AD versus oral AD plus intranasal placebo. Four out of the nine items on the PHQ-9 and five of the ten items on the MADRS achieved statistical significance, providing a detailed account of the magnitude of which specific depressive symptoms, as measured by single items, are likely to improve from treatment with esketamine nasal spray. In this study, there was a favorable response to all items on both the PHQ-9 and MADRS, indicating overall improvement in depression for those treated with esketamine nasal spray plus oral AD. While the change in total score is an important indicator of change in depression symptom severity, understanding the specific items contributing more, or less, to the total score change is helpful in interpreting treatment efficacy. Each item represents symptoms of depression which are important diagnostic criteria. By isolating the results of each item, it is possible to determine which symptoms are more likely to improve, enhancing patient understanding of potential treatment effects. For example, based on these results, a clinician can explain that the odds of not feeling down, depressed, or hopeless may be twice that for those treated with esketamine plus AD compared with AD plus placebo after 1 month of treatment. As part of the shared decision-making process, patients and clinicians can weigh the likelihood of individual symptom improvement against treatment risks. The particularly favorable results in PHQ-9 items 1, 2, 4, and 6 and MADRS items 1, 2, 3, 6 and 8 suggest improved efficacy of esketamine/oral AD over placebo plus AD in these specific symptoms among TRD adult patients. PHQ-9 items 1 and 2 ("Little interest/pleasure in doing things" and "Feeling down, depressed, or hopeless") are both considered cardinal symptoms of depression; a patient will be diagnosed as depressed only if endorsing at least one of these two items. As all items are important for diagnostic criteria, patients and clinicians can see what other items are driving overall score response for those on esketamine plus AD compared with those on placebo plus AD by highlighting items with statistically significant Fig.Montgomery-Asberg Depression Rating Scale (MADRS) items from baseline to days 15/28 in esketamine plus AD versus placebo plus AD. *The odds ratio represents the likelihood of improv-ing over the course of the study and includes both day 15 and day 28 data. An odds ratio > 1 favors esketamine over placebo. AD antidepressant, CI confidence interval improvement. On the PHQ-9, these were items 4 and 6 ("Feeling tired or having little energy" and "Feeling bad about yourself -or that you are a failure or have let yourself or your family down"). On the MADRS, reported sadness, apparent sadness, inner tension, concentration difficulties and the inability to feel (anhedonia), statistically significantly improved. Together, these results inform the ways in which patients experience clinical improvement with higher specificity than is conveyed by the total score. Of note, the likelihood of improving was the greatest (OR > 2.0), relative to the other items, for "Little interest/pleasure in things", "Feeling down, depressed, or hopeless", and "Feeling tired or little energy" on the PHQ-9. Similarly, on the MADRS, the likelihood of improving was greatest (OR > 2.0) on the items measuring apparent sadness (item 2) and anhedonia (item 8). On both instruments, the symptom that was least likely to improve (other than suicidal thoughts) was having a reduced appetite, as measured by item 5 for both the PHQ-9 and MADRS. One limitation of the study was the short follow-up time; measurements were taken at baseline and two other time points during the double-blind treatment period. The 4-week duration of the induction phase was chosen to provide sufficient time for the onset of efficacy in the oral AD + placebo group. Meta-analysis results suggest that treatment difference is consistent for trials of 4-8 weeks' duration, suggesting a duration of 4 weeks is sufficient. Another limitation is the absence of evaluation of the oral AD used. Clinicians prescribed one of four oral ADs for the concomitant treatment; however, any differences of each AD were not analyzed. Moreover, since these are results from one randomized trial, further studies are needed to replicate these findings.