Multiband ALFF in patients with major depressive disorder treated with esketamine and sertraline: A preliminary study

Introduction: This study investigated changes in multiband amplitude of low-frequency fluctuations in patients with major depressive disorder following a 2-week combination treatment regimen of esketamine and sertraline, and examined the associations between ALFF changes and clinical symptom improvement.

Methods: Forty patients with major depressive disorder (MDD) received esketamine (0.25 mg/kg, intravenous infusion, three times per week) combined with sertraline (50-150 mg/day, oral) for 2 weeks. The Hamilton Anxiety Scale, the Hamilton Depression Scale, the Beck Scale for Suicide Ideation, and the Montreal Cognitive Assessment were administered at baseline and week 2, and resting-state functional MRI data were collected. ALFF values were calculated for the typical band (0.01-0.08 Hz), the slow-4 band (0.027-0.073 Hz), and the slow-5 band (0.01-0.027 Hz) using DPABI 5.1. Clinical scores were compared between baseline and week 2 with paired t-tests or Wilcoxon signed-rank tests according to the Shapiro-Wilk test. Whole-brain ALFF maps were compared with voxel-wise two-tailed paired t-tests in all participants (Gaussian random field correction; voxel-level p < 0.001, cluster-level p < 0.05), and the extracted mean ALFF value of each surviving cluster was then compared between time points at the region level with a paired t-test or Wilcoxon signed-rank test according to normality; normality was therefore assessed only for the extracted regional values, not at the voxel level. Partial correlation analyses, with sex, age, education, baseline clinical scores, and mean framewise displacement as covariates, were conducted to examine associations between ALFF measures and clinical outcomes. Three families of ROI × frequency-band × clinical-outcome tests were defined according to the question addressed (baseline ALFF versus week-2 outcome; post-treatment ALFF versus concurrent outcome; change in ALFF versus change in clinical score; 32 tests per family, 96 tests in total), and Benjamini-Hochberg FDR correction within each family (q < 0.05) was the primary multiplicity control, with Bonferroni correction within family (adjusted α =0.0016) and FDR across all 96 tests treated as a single family as sensitivity analyses.

Results: After treatment, HAMA, HAMD, and BSI scores decreased and MoCA scores increased (all p < 0.001). ALFF values in the left inferior frontal gyrus increased in the typical and slow-4 bands, whereas ALFF in sensorimotor and visual cortical regions decreased across multiple bands (GRF-corrected, voxel p < 0.001, cluster p < 0.05). Treatment-related increases in L-IFG ALFF in the typical band were negatively correlated with reductions in BSI scores (partial r = -0.500, uncorrected p = 0.004); however, this association did not remain significant after FDR correction across the 32 ΔALFF-Δscore tests (q = 0.114) or after Bonferroni correction (adjusted α =0.0016), and no ROI-clinical correlation survived correction in any family. It is therefore reported as a nominal, exploratory finding. Post-treatment ALFF in the L-IFG, right precentral gyrus, and right middle occipital gyrus showed nominally significant partial correlations with concurrent symptom scores; none of these remained significant after FDR correction within the corresponding family (smallest q = 0.153). Baseline L-IFG ALFF was not independently associated with treatment outcome after covariate adjustment.

Conclusion: A 2-week combination treatment regimen of esketamine and sertraline was associated with frequency-specific ALFF changes in prefrontal, sensorimotor, and visual cortical regions in patients with major depressive disorder. No ROI-clinical correlation survived correction for multiple comparisons across the three families of tests; the nominal association between L-IFG ALFF change and suicidal-ideation reduction is exploratory and hypothesis-generating, and requires independent replication before prefrontal ALFF modulation can be considered a candidate marker for monitoring treatment-related neurofunctional changes.