ImportanceThere is growing interest in the role of gut microbiome composition in schizophrenia. However, lifestyle factors are often neglected, and few studies have investigated microbiome composition in treatment-resistant schizophrenia.ObjectiveTo explore associations between the gut microbiome and schizophrenia diagnosis, treatment resistance, clozapine response, and treatment-related adverse effects while adjusting for demographic and lifestyle factors.Design, Setting, and ParticipantsIn this case-control study of adults aged 20 to 63 years, stool samples and data on demographic characteristics, lifestyle, and medication use were collected and gut microbiome measures obtained using shotgun metagenomics. Participants with a schizophrenia diagnosis were referred through psychiatric inpatient units and outpatient clinics. Data were collected for 4 distinct groups: control individuals without a psychiatric diagnosis (past or present), individuals with treatment-responsive schizophrenia taking nonclozapine antipsychotic medications, clozapine-responsive individuals with treatment-resistant schizophrenia, and clozapine-nonresponsive individuals with treatment-resistant schizophrenia. Participants were recruited between November 2020 and November 2021. Control individuals were recruited in parallel through posters and online advertisements and matched for age, sex, and body mass index (BMI) to the individuals with schizophrenia. Participants were excluded if taking antibiotics in the past 2 months, if unable to communicate in English or otherwise follow study instructions, were pregnant or planning to become pregnant, or had any concomitant disease or condition making them unsuited to the study per investigator assessment. Data were analyzed from January 2022 to March 2023.Main Outcomes and MeasuresOmics relationship matrices, α and β diversity, and relative abundance of microbiome features.ResultsData were collected for 97 individuals (71 [74%] male; mean [SD] age, 40.4 [10.3] years; mean [SD] BMI, 32.8 [7.4], calculated as weight in kilograms divided by height in meters squared). Significant microbiome associations with schizophrenia were observed at multiple taxonomic and functional levels (eg, common species: b2, 30%; SE, 13%; adjusted P = .002) and treatment resistance (eg, common species: b2, 27%; SE, 16%; adjusted P = .03). In contrast, limited evidence was found for microbiome associations with clozapine response, constipation, or metabolic syndrome. Significantly decreased microbial richness was found in individuals with schizophrenia compared to control individuals (t95 = 4.25; P < .001; mean [SD] for control individuals, 151.8 [32.31]; mean [SD] for individuals with schizophrenia, 117.00 [36.2]; 95% CI, 18.6-51.0), which remained significant after a covariate and multiple comparison correction. However, limited evidence was found for differences in β diversity (weighted UniFrac) for schizophrenia diagnosis (permutational multivariate analysis of variance [PERMANOVA]: R2, 0.03; P = .02), treatment resistance (R2, 0.02; P = .18), or clozapine response (R2, 0.04; P = .08). Multiple differentially abundant bacterial species (19) and metabolic pathways (162) were found in individuals with schizophrenia, which were primarily associated with treatment resistance and clozapine exposure.Conclusions and RelevanceThe findings in this study are consistent with the idea that clozapine induces alterations to gut microbiome composition, although the possibility that preexisting microbiome differences contribute to treatment resistance cannot be ruled out. These findings suggest that prior reports of microbiome alterations in individuals with chronic schizophrenia may be due to medication or lifestyle factors and that future studies should incorporate these variables in their design and interpretation.

中文翻译：

---

肠道微生物组与精神分裂症治疗耐药性的关系

重要性人们对肠道微生物组组成在精神分裂症中的作用越来越感兴趣。然而，生活方式因素常常被忽视，很少有研究调查难治性精神分裂症中的微生物组组成。目的探索肠道微生物组与精神分裂症诊断、治疗抵抗、氯氮平反应和治疗相关不良反应之间的关联，同时根据人口和生活方式进行调整设计、环境和参与者在这项针对 20 至 63 岁成年人的病例对照研究中，收集了粪便样本和有关人口特征、生活方式和药物使用的数据，并使用鸟枪法宏基因组学获得了肠道微生物组测量值。诊断为精神分裂症的参与者通过精神病住院部和门诊诊所转诊。收集了 4 个不同组的数据：没有精神病学诊断（过去或现在）的对照个体、服用非氯氮平抗精神病药物的治疗有反应的精神分裂症个体、有治疗抵抗性精神分裂症的氯氮平有反应个体、以及有治疗抵抗性的氯氮平无反应个体精神分裂症。参与者在 2020 年 11 月至 2021 年 11 月期间招募。对照个体是通过海报和在线广告并行招募的，并将年龄、性别和体重指数 (BMI) 与精神分裂症患者进行匹配。根据研究者的评估，如果参与者在过去 2 个月内服用抗生素、无法用英语交流或无法遵循研究说明、怀孕或计划怀孕、或有任何伴随疾病或状况使他们不适合本研究，则被排除在外。分析了 2022 年 1 月至 2023 年 3 月的数据。主要结果和测量组学关系矩阵、α 和 β 多样性以及微生物组特征的相对丰度。结果收集了 97 名个体的数据（71 [74%] 男性；平均 [SD] 年龄，40.4 [10.3] 岁；平均 [SD] BMI，32.8 [7.4]，计算方法为体重（公斤）除以身高（米）的平方）。在多个分类学和功能水平上观察到微生物组与精神分裂症的显着关联（例如，常见物种：乙2, 30%; 东南欧，13%；调整过的磷= .002）和治疗抗性（例如，常见物种：乙2, 27%; 东南欧，16%；调整过的磷= .03)。相比之下，微生物组与氯氮平反应、便秘或代谢综合征相关的证据有限。与对照个体相比，精神分裂症患者的微生物丰富度显着降低（t95= 4.25；磷< .001; 对照个体的平均值 [SD]，151.8 [32.31]；精神分裂症患者的平均 [SD] 为 117.00 [36.2]；95% CI，18.6-51.0），在协变量和多重比较校正后仍然显着。然而，关于精神分裂症诊断的 β 多样性（加权 UniFrac）差异的证据有限（排列多元方差分析 [PERMANOVA]：右2, 0.03;磷= .02), 治疗阻力 (右2, 0.02;磷= .18），或氯氮平反应（右2, 0.04;磷= .08)。在精神分裂症患者中发现了多种不同丰度的细菌种类 (19) 和代谢途径 (162)，这主要与治疗抵抗和氯氮平暴露有关。结论和相关性本研究的结果与氯氮平诱导肠道改变的观点一致微生物组组成，尽管不能排除预先存在的微生物组差异导致治疗耐药性的可能性。这些发现表明，先前关于慢性精神分裂症患者微生物组改变的报告可能是由于药物或生活方式因素造成的，未来的研究应将这些变量纳入其设计和解释中。