ResearchPod Summary
Schizophrenia is characterized by both positive symptoms (e.g., hallucinations) and negative symptoms (e.g., social withdrawal, blunted affect, and avolition). While antipsychotic medications effectively manage positive symptoms, they often fail to address negative symptoms, which are major drivers of functional impairment. This systematic review and network meta-analysis evaluated the efficacy of adding various antidepressants to existing antipsychotic regimens to treat these persistent negative symptoms.
The authors conducted a comprehensive search of PubMed and Web of Science for randomized, double-blind, placebo-controlled trials published up to April 2025. They included 15 studies involving 655 patients. The primary outcome was the change in negative symptom scores, measured using standardized scales like the Scale for the Assessment of Negative Symptoms (SANS) or the Positive and Negative Syndrome Scale (PANSS-N).
The meta-analysis found that adjunctive antidepressant treatment is superior to placebo in reducing negative symptoms. Specifically, mirtazapine and duloxetine emerged as the most effective options. Mirtazapine showed statistically significant superiority over placebo and outperformed several other antidepressants, including reboxetine, escitalopram, and bupropion. Duloxetine also demonstrated significant efficacy compared to placebo. The authors suggest that the mechanism of action for these two drugs—specifically their effects on serotonin and norepinephrine receptors—may be particularly relevant for addressing the underlying pathophysiology of negative symptoms.
Negative symptoms represent a significant, often unaddressed therapeutic gap in schizophrenia care that profoundly impacts a patient's ability to work and maintain social relationships. These findings provide clinicians with evidence-based options for augmenting standard antipsychotic treatment. By identifying mirtazapine and duloxetine as potentially effective adjunctive therapies, this research offers a promising strategy for improving the quality of life for patients who do not respond to antipsychotics alone.
[[RP_SECTION:antidepressants-for-negative-symptoms|Antidepressants for negative symptoms]]
Alex: Mirtazapine and duloxetine come out as the most effective add-on antidepressants for easing negative symptoms in stable schizophrenia — the social withdrawal, the flattened affect, the loss of drive that antipsychotics rarely touch. That's the headline from a network meta-analysis by Yuting Li and colleagues.
Sam: That's a striking claim, given how resistant negative symptoms usually are to treatment. Did they account for the obvious confound — that some of these patients might just be depressed, rather than presenting core schizophrenia symptoms?
Alex: They did. The inclusion criteria explicitly filtered out studies where patients had a primary depressive disorder. What's left is fifteen randomized controlled trials testing these agents as add-ons to existing antipsychotic treatment, specifically in patients whose negative symptoms aren't attributable to a separate depressive episode. [[RP_SECTION:network-meta-analysis-methodology|Network meta-analysis methodology]]
Sam: So they're isolating the drug effect from the confound rather than adjusting for it statistically after the fact. How does a network meta-analysis actually let you rank drugs like this, when mirtazapine and duloxetine were probably never tested head to head?
Alex: That's the whole point of the method. None of these trials pit mirtazapine directly against duloxetine — each is compared against placebo in separate trials. A network meta-analysis stitches those separate comparisons together through the shared placebo arm, generating an indirect ranking across drugs that were never in the same study. It's a way of borrowing statistical power across a fragmented evidence base.
Sam: Which also means the ranking is only as trustworthy as the assumption that those placebo arms are comparable across trials. Setting that aside — what's the actual effect size? [[RP_SECTION:effect-size-and-mechanism|Effect size and mechanism]]
Alex: Both drugs beat placebo, and mirtazapine came out on top, with a standardized mean difference of 1.73. In this literature, that's a large effect — well above what you'd typically see even for antipsychotics targeting positive symptoms.
Sam: A strong number for a symptom domain where most trials report null or marginal results. What's the proposed mechanism? Why would an antidepressant outperform a drug engineered to sit on the dopamine system?
AI-generated third-party summary by ResearchPod. Not official content or an endorsement by the paper authors or affiliated organizations.
Alex: Think of standard antipsychotics as brakes on a dopamine system running too hot. They're good at damping down hallucinations and delusions, but braking a system doesn't restore the drive or emotional range that negative symptoms take away. Mirtazapine and duloxetine work through an entirely different set of receptors — serotonin and norepinephrine, including 5-HT2 and alpha-2 adrenergic sites.
Sam: So rather than suppressing an overactive circuit, they're acting directly on the circuits tied to motivation and affect.
Alex: That's the proposed logic, yes. It's plausible, and it fits with why dopamine-focused antipsychotics have been such a poor match for this symptom domain for decades. [[RP_SECTION:evidence-base-limitations|Evidence base limitations]]
Sam: Plausible is doing some work in that sentence, though. How many trials are actually behind the mirtazapine ranking specifically? If it's one or two studies driving that 1.73, the confidence interval around it is going to be wide.
Alex: You're right to flag that. The authors are upfront that some of these rankings rest on a small number of trials per drug, which limits precision and makes the point estimate more fragile than the headline number suggests. This isn't a failure of the method — a network meta-analysis is doing legitimate work synthesizing what exists — but the underlying evidence base is thin. There's no preregistered mega-trial here, no dose-response testing, and no look at whether the benefit holds over longer follow-up. [[RP_SECTION:clinical-implications-and-future|Clinical implications and future]]
Sam: So it's less "we've found the answer" and more "we've built the best possible synthesis of a small, scattered literature."
Alex: That's a fair way to put it. It gives clinicians and trialists a ranked, evidence-based starting point rather than a guess, but it's not the kind of result that should shift prescribing on its own — it tells the field where to point the next, larger trial.
Sam: A meaningful step forward, then, not yet a verdict.
Alex: If you want the figures and the method choices we skipped, you can generate a deep dive of this paper. The paper has the rest either way.
Sam: Thanks for listening.