Psychiatry · Medical Literature
Brain iron-sensitive markers (magnetic susceptibility and R2*) predict antidepressant response to ketamine in treatment-resistant depression.
Abstract
Ketamine may alleviate treatment-resistant depression (TRD) primarily through glutamatergic modulation, with downstream dopaminergic activation. Iron plays an important role in monoaminergic metabolism, that is also implicated in the pathophysiology of TRD. Both Quantitative Susceptibility Mapping (QSM) and Effective Transverse Relaxation Rate (R2*) mapping can determine the extent of iron deposition in the brain. Given that abnormal iron accumulation may reflect dopamine dysfunction, we hypothesized that baseline magnetic substances could predict ketamine's antidepressant effects in patients with TRD. We used data from a double-blind, randomized placebo-controlled trial followed by an extended single-arm open-label study to assess the efficacy of repeated intravenous ketamine in Japanese patients with TRD (jRCTs031210124). This study analyzed the data from the participants who underwent QSM and R2* mapping before receiving ketamine in either phase. Multivariable regression analyses were conducted to explore the association between baseline magnetic susceptibility and R2* with change in MADRS total and subdomain scores. This study included 17 patients with TRD (7 women; mean ± standard deviation age, 42.9 ± 10.6 years). Baseline magnetic susceptibility in the right nucleus accumbens negatively correlated with the change in MADRS retardation symptom scores (β = -0.73, p = 0.003). Moreover, baseline R2* in the left amygdala was negatively associated with the change in MADRS vegetative symptom scores (β = -0.71, p = 0.004). Baseline magnetic substances in the right nucleus accumbens and the left amygdala may be biomarkers to predict the effect of repeated ketamine infusions in patients with TRD.
Clinical relevance
Do baseline brain iron-sensitive MRI markers — quantitative susceptibility mapping (QSM) and effective transverse relaxation rate (R2*) — predict the antidepressant response to repeated intravenous ketamine in patients with treatment-resistant depression (TRD)? __SNAPSHOT__ {"studyDesign":"Secondary analysis of a double-blind, randomized, placebo-controlled trial followed by an extended single-arm open-label study (jRCTs031210124); specifically a multivariable regression-based neuroimaging biomarker study","setting":"Japan; academic psychiatric center (Keio University School of Medicine, Tokyo, with MRI conducted at University of Tokyo)","sampleSize":"n=17 TRD patients who underwent baseline QSM and R2* mapping prior to ketamine administration in either the RCT or open-label phase; this represents a neuroimaging subsample of the parent trial","population":"Japanese adults with treatment-resistant major depressive disorder (mean age 42.9 ± 10.6 years; 7 women, 10 men); required MADRS total score ≥20 at screening and baseline; had failed ≥2 adequate antidepressant trials during the current episode","intervention":"Repeated intravenous ketamine (0.5 mg/kg per infusion; specific number of infusions not reported in abstract); baseline MRI biomarkers assessed via QSM (measuring magnetic susceptibility, a proxy for iron deposition) and R2* mapping (effective transverse relaxation rate, another iron-sensitive parameter) prior to ketamine","comparator":"Not applicable for this biomarker analysis; the parent RCT included an IV placebo arm, but the imaging analyses are reported without placebo-arm comparisons in the available abstract","primaryOutcome":"Association between baseline magnetic susceptibility (QSM) and R2* values in specific brain regions with change in Montgomery–Åsberg Depression Rating Scale (MADRS) total and subdomain scores following ketamine","followUp":"Not reported in the available abstract (duration from baseline imaging to MADRS reassessment not specified)","fundingSource":"Not reported in the available abstract; the parent trial (jRCTs031210124) was conducted at Keio University; multiple investigators in the broader study group have disclosed pharmaceutical industry relationships (see Limitations)"}
Key findings
In 17 TRD patients who underwent baseline neuroimaging, two region-specific iron-sensitive MRI markers independently predicted ketamine's antidepressant effects on MADRS subdomains in multivariable regression analyses. First, higher baseline magnetic susceptibility (QSM signal, reflecting greater iron deposition) in the right nucleus accumbens was negatively associated with improvement in MADRS retardation symptom scores (β = −0.73, p = 0.003), indicating that patients with higher right nucleus accumbens iron at baseline showed less improvement in psychomotor retardation symptoms. Second, higher baseline R2* in the left amygdala was negatively associated with improvement in MADRS vegetative symptom scores (β = −0.71, p = 0.004), indicating that patients with greater left amygdala iron burden showed less improvement in vegetative (neurovegetative) symptoms. No results for MADRS total score prediction were reported in the abstract. Effect sizes for absolute change in MADRS scores, confidence intervals, NNT, and NNH were not reported. Whether these associations held independently for the RCT versus open-label phases was not reported.
Practice implications
This study provides preliminary hypothesis-generating evidence that brain region-specific iron deposition, measured non-invasively by QSM and R2* MRI, may modulate the antidepressant response to ketamine in TRD — particularly for psychomotor retardation (nucleus accumbens) and neurovegetative (amygdala) symptom clusters. The mechanistic rationale is biologically coherent: iron is a required cofactor for tyrosine hydroxylase and monoamine oxidase, and abnormal iron accumulation in limbic-striatal circuits may reflect dopaminergic dysfunction relevant to ketamine's therapeutic mechanism. However, the findings are substantially limited: only 17 participants were analyzed (a neuroimaging subsample from a larger RCT), the study was not prospectively powered to test this hypothesis, the abstract does not report whether QSM or R2* predicted the primary MADRS total score, and no confidence intervals are provided. The two significant associations involve subdomain scores rather than the primary clinical endpoint. The findings do not establish that QSM/R2* should be used clinically to guide ketamine prescribing decisions. The population was exclusively Japanese, limiting generalizability to other ethnic groups. Only the abstract was available for this review, restricting full appraisal of statistical methodology, covariate selection, correction for multiple comparisons, and the precise ketamine protocol used. __LIMITATIONS__ Very small sample (n=17), a neuroimaging subsample of the parent RCT; severely underpowered for biomarker discovery and prone to spurious associations,Post-hoc/secondary analysis not prospectively powered to test iron-marker hypotheses; inflated risk of Type I error,No correction for multiple comparisons reported in the available abstract (multiple brain regions and multiple MADRS subdomains tested),Associations reported only for MADRS subdomain scores (retardation and vegetative); prediction of MADRS total score not reported in the abstract,Combined analysis of RCT ketamine recipients and open-label participants introduces heterogeneity in exposure, timing, and selection,No confidence intervals reported for regression coefficients; statistical uncertainty around effect estimates cannot be assessed,Follow-up duration for MADRS reassessment not reported; timing of outcome relative to ketamine course is unclear,Exclusively Japanese sample; generalizability to non-Asian populations is unknown,Funding source not reported in the abstract; several investigators in the broader study group have disclosed pharmaceutical industry affiliations,Only the abstract was available for full review; covariate selection, iron quantification methods, and placebo-arm comparisons could not be independently verified __BOTTOM_LINE_CATEGORY__ promising_but_preliminary __BOTTOM_LINE__ In a small neuroimaging subsample (n=17) of a Japanese TRD ketamine trial, higher baseline iron deposition in the right nucleus accumbens and left amygdala — measured by QSM and R2*, respectively — was associated with less improvement in specific MADRS symptom subdomains following ketamine. These findings are mechanistically plausible but entirely preliminary, given the very small sample, post-hoc design, absence of multiple comparison correction, and lack of total-score prediction data. QSM/R2* biomarkers should not currently inform clinical ketamine prescribing decisions; replication in adequately powered, prospectively designed studies is required.