SUMMARY

• SPRAVATO should be administered in conjunction with an oral antidepressant (AD).^1-3
• In a propensityscore-matched, realworld cohort evaluating 55,480 patients with TRD over 5 years, SPRAVATO combined with serotonin-norepinephrine reuptake inhibitors (SNRIs) was associated with lower risks of allcause mortality, hospitalization, and depression relapse than SPRAVATO combined with selective serotonin reuptake inhibitors (SSRIs). SPRAVATO+SSRIs had a lower risk of suicide attempts. The 5year survival in the SNRI group was higher than that of SSRIs group (91.4% vs 86.9%; P<0.001).^4,5
  • Some limitations included missing information on treatment dose, duration, and adherence. No matching occurred based on clinical factors such as comorbidities and other treatment-related factors.^4,6
• In phase 3 clinical trials for treatment-resistant depression (TRD), a newly initiated, open-label, oral AD (one of a choice of 4 oral ADs; SNRIs: duloxetine, venlafaxine extended-release [XR]; SSRIs: escitalopram, sertraline) was used concomitantly with SPRAVATO nasal spray.^1-3
• A post hoc pooled analysis of adults with TRD from the TRANSFORM-1² and TRANSFORM-2¹ studies showed no major differences in Montgomery-Åsberg Depression Rating Scale (MADRS) total score change from baseline to day 28 by the class of AD or individual AD. Response and remission rates were generally similar by AD class and individual AD. For individual AD, the response and remission rate of SPRAVATO + sertraline was lower than with other SPRAVATO+AD groups; however, the difference was not statistically significant.³
• Based on the post hoc analysis, the rate of treatment-emergent adverse events (TEAEs) reported in ≥5% of patients in the SSRI and SNRI subgroups of the SPRAVATO+AD groups were 53.7% and 56.5%, respectively. The rate of TEAEs reported in ≥5% of patients in the SSRI and SNRI subgroups of the AD + placebo (PBO) groups were 22.9% and 23.0%, respectively.
  • The incidence of TEAEs was generally similar among the individual ADs; however, higher rates of somnolence, increased blood pressure, and sedation were observed in the SPRAVATO + duloxetine vs other SPRAVATO+AD groups. While anxiety and nausea were reported at higher rates in the SPRAVATO + sertraline group compared with other SPRAVATO+AD groups, these data should be interpreted with caution due to the small sample receiving SPRAVATO + sertraline.³

BACKGROUND

Dosing of Oral Antidepressants

At the start of the double-blind induction phase, a newly initiated, open-label, oral AD was used concomitantly with SPRAVATO or PBO nasal spray. The oral AD was assigned by the investigator from four choices (duloxetine, escitalopram, sertraline, or venlafaxine XR).³

Doses of the oral AD were not to exceed the maximum doses defined in the titration schedule. If higher doses were not tolerated, a down-titration was permitted based on clinical judgment.

At the end of the induction phase, the patients were receiving an oral AD to which they had been adherent at or above the minimum therapeutic dosage.¹

In a post hoc analysis of two short-term studies in TRD^1,2, the mean modal daily doses of each oral AD were: duloxetine, 59.6 mg; escitalopram, 17.1 mg; sertraline, 105.8 mg; and venlafaxine, 187.5 mg.³

Clinical data

Del Casale et al (2025)⁴ reported a retrospective, propensityscore-matched, comparative-effectiveness cohort study using the TriNetX global EMR network to compare the risk of all-cause mortality, hospitalization, depressive relapse, and suicide attempt between adult patients with TRD receiving SPRAVATO in combination with SSRIs and SNRIs. Outcomes were assessed for 5 years starting 1 day after the first esketamine prescription. Note that the study did not provide information on duration of treatment over the 5-year period.

Study Design/Methods

Two exposure cohorts were defined based on the concomitant oral antidepressant class at SPRAVATO initiation. The SSRI cohort included citalopram, escitalopram, fluoxetine, fluvoxamine, paroxetine, sertraline, and vilazodone. The SNRI cohort included venlafaxine, milnacipran, duloxetine, desvenlafaxine, and levomilnacipran. Propensity score matching (1:1) balanced age and sex. Outcomes were analyzed using risk differences, risk ratios, and Kaplan-Meier survival with log‑rank testing. Prior occurrences of hospitalization, depression relapse, and suicide attempts were excluded from those respective analyses.^4,5

Results

After matching, each cohort included 27,740 patients. The SSRI cohort had a mean age of 46.0 years and female proportion of 57.7%. The SNRI cohort had a mean age of 45.9 years and female proportion of 58.6%. Outcomes are presented in Table: Outcomes Over 5 Years.

There was a statistically significantly lower risk of deaths, hospitalizations, and depression relapses in the SPRAVATO+SNRI cohort compared with the SPRAVATO+SSRI cohort (all, P<0.001). The SPRAVATO+SSRI cohort had a lower risk of nonfatal suicide attempts (P=0.04).⁴

In Kaplan-Meier analyses, the 5year survival probability was 86.9% with SSRIs and 91.4% with SNRIs; the logrank test revealed a statistically significant difference in survival between the cohorts, (P<0.001); the hazard ratio was 1.68 (95% confidence interval, 1.58-1.80), favoring the SNRI cohort.⁴

Limitations

Matching was limited to age and sex without adjustment for comorbidities or other clinical/treatment factors (e.g., dose, duration, adherence); outcome ascertainment relied on EMR/international statistical classification of disease and related health problem, 10th revision (ICD10) codes (eg, T14.91 for suicide attempts, F32/F33 codes for depression relapse) that had variable sensitivity and may have under or misclassified events; mortality data from EMR sources alone may have been incomplete without linking to other data sources.4,6

Apparent inconsistencies between reported mortality and survival percentages, very low hospitalization rates over 5 years, unexpectedly low prewindow hospitalizations, and questions about feasibility of nearuniversal 5year followup given SPRAVATO 2019 availability were noted. The authors’ reply explained that survival rate estimators were not meant to be mathematically complementary and that conservative definitions of hospitalization and suicide attempts were meant to mitigate overestimation and bias. They further mention that the low hospitalization and mortality rates were consistent with postdischarge risk in psychiatric units.⁶

Adigun et al (2019)³ conducted a post hoc, pooled analysis of two 4-week, double-blind, placebo-controlled studies^1,2 in TRD to examine the impact of individual AD (duloxetine, escitalopram, sertraline, and venlafaxine XR) and AD class (SSRI or SNRI) on the efficacy and safety of SPRAVATO+AD.

Efficacy - Pooled Short-Term Studies (TRD3001 and TRD3002)

MADRS

The SPRAVATO+AD group showed greater improvement from baseline to day 28 in mean MADRS total score compared with the AD+PBO group (-19.9 vs -16.4, respectively; P=0.002). Subgroup analyses in a pooled adult population with TRD from the TRANSFORM-2 (3002) and TRANSFORM-1 (3001) studies showed no significant differences in the change in MADRS total score from baseline to day 28 by the class of AD or individual AD (Figure: Montgomery-Åsberg Depression Rating Scale (MADRS) Total Score: Least Squares Mean Treatment Difference of Change from Baseline to Day 28 by [A] Treatment Group, [B] Antidepressant Class, and [C] Individual Antidepressant [Pooled Studies TRANSFORM-1 AND TRANSFORM-2]).³

Response and Remission Rates

The proportion of patients who met the criteria for response (≥50% improvement in MADRS total score) at day 28 was greater in the SPRAVATO+AD group compared with the AD+PBO group (58.7% vs 45.2%, respectively; P<0.001). The response rates on day 28 were similar by AD class and individual AD. For individual AD, a lower response rate was observed with SPRAVATO + sertraline; however, the difference was not statistically significant (Figure: Response Rates at Day 28 by [A] Treatment Group, [B] Antidepressant Class, and [C] Individual Antidepressant).

Remission (MADRS total score ≤12) rates at day 28 was also greater in the SPRAVATO+AD group compared with the AD+PBO group (42.3% vs 30.8%, respectively; P=0.003). The remission rates at day 28 were similar by AD class and individual AD. For individual ADs, a lower remission rate was observed with SPRAVATO + sertraline; however, the difference was not statistically significant (Figure: Remission Rates at Day 28 by [A] Treatment Group, [B] Antidepressant Class, and [C] Individual Antidepressant).

Safety - Pooled Short-Term Studies (TRD3001 and TRD3002)

For the total SPRAVATO group, more patients received an SNRI (61%) compared with an SSRI (39%). The rate of TEAEs reported in ≥5% of patients from the SSRI and SNRI subgroups were 53.7% and 56.5%, respectively, in the SPRAVATO+AD group and 22.9% and 23.0%, respectively, in the AD+PBO group (Table: Most Frequently Reported Treatment-Emergent Adverse Events Reported in ≥5 Patients by Antidepressant Class).³

The only frequently reported TEAE (5%), which occurred at a higher rate in patients from the SNRI subgroup compared with patients from the SSRI subgroup (difference in incidence of 5%) was somnolence (20.1% vs 13.4%, respectively). Frequently reported TEAEs (5%) which occurred at a higher rate in patients from the SSRI subgroup compared with patients from the SNRI subgroup (difference in incidence of 5%) were nausea (32.1% vs 25.8%, respectively) and anxiety (14.9% vs 4.8%, respectively).

For individual AD, the rate of somnolence, increased blood pressure, and sedation were higher in the SPRAVATO + duloxetine vs other SPRAVATO+AD groups.  The incidence of other TEAEs was generally similar among the individual ADs (Table: Most Frequently Reported Treatment-Emergent Adverse Events in 5% Patients by Individual Antidepressant). Anxiety and nausea were reported at higher rates in the SPRAVATO + sertraline group compared with other SPRAVATO+AD groups; however, these data should be interpreted with caution due to the small sample receiving SPRAVATO + sertraline (n=64).

Literature Search

A literature search of MEDLINE^®, Embase^®, BIOSIS Previews^®, and Derwent Drug File (and/or other resources, including internal/external databases) pertaining to this topic was conducted on 12 January 2026.