30-Second Takeaway
- Dose-specific lithium and select antipsychotics remain core for bipolar maintenance, with LAIs as viable long-term options.
- Connectome-guided targeting and focal neuromodulation are improving outcomes in depression, social anxiety, PTSD, and autism.
- Conversational AI now has RCT-level evidence as a scalable intervention for anxiety and depression in students.
- Integrated approaches modestly improve pain and trauma, while topiramate may enhance trauma-focused therapy in PTSD with AUD.
- Emerging biomarkers in postpartum psychosis, major depression, and bipolar disorder may sharpen risk stratification over time.
Week ending April 18, 2026
Targeted treatments, neuromodulation, and biomarkers refining psychiatric care
Dose-specific evidence for bipolar maintenance: lithium and select antipsychotics stand out
This network meta-analysis of 44 RCTs (10,867 participants) compared pharmacologic maintenance options for bipolar disorder across age groups. Asenapine 20 mg/day, aripiprazole 20–30 mg/day, several quetiapine doses, lithium 800 mg/day, and lurasidone 80 mg/day reduced relapse versus placebo. Carbamazepine 400–650 mg/day, valproate 71–125 μg/mL, olanzapine 20 mg/day, and LAI aripiprazole or risperidone were also superior to placebo. Tolerability differences, reflected in discontinuations from adverse effects, necessitate balancing relapse prevention against side-effect burden at these doses. Findings align with guidelines but refine dose-specific and LAI choices for long-term bipolar maintenance planning.
DLPFC–subgenual pathway length helps explain TMS antidepressant response
This study used connectome modeling to map polysynaptic fiber pathways linking dorsolateral prefrontal TMS targets to the subgenual cingulate in depression. Route length between the stimulated DLPFC site and subgenual cingulate explained treatment response in two independent patient cohorts. The same anatomical measure also accounted for the efficacy of fMRI-guided TMS targeting approaches. Results support incorporating individual white-matter connectivity metrics into TMS target selection, beyond standard anatomical coordinates.
Conversational AI outperforms waitlist and often group therapy for student anxiety and depression
This 3-arm RCT randomized 995 psychologically distressed university students to 12 weeks of conversational AI, face-to-face group therapy, or waitlist. The AI platform produced greater anxiety reduction than both group therapy and waitlist, and greater depression reduction than waitlist. Well-being improved more with AI than both comparators, while PTSD symptoms did not differ across groups. Perceived therapeutic alliance with the AI was linked to greater engagement and symptom improvement, suggesting alliance-like processes in digital care.
Topiramate adds mid-treatment gains to prolonged exposure for PTSD with alcohol use disorder
This session-level analysis examined veterans receiving prolonged exposure plus topiramate versus prolonged exposure plus placebo for comorbid PTSD and alcohol use disorder. Topiramate produced greater PTSD symptom reduction beginning at session 9 compared with placebo, consistent with reaching maximum dose. Alcohol cravings improved more with topiramate starting at session 6, preceding the PTSD symptom separation between groups. Depressive symptoms did not differ by condition, and cross-lagged models suggested that craving reductions may drive subsequent PTSD improvement.
References
Numbered in order of appearance. Click any reference to view details.
Additional Reads
Optional additional studies from this edition.