Gut microbiota and metabolomic features associated with hemoglobin improvement in patients with aplastic anemia: an exploratory clinical multi-omics study
Background Hematologic recovery after treatment varies substantially among patients with aplastic anemia (AA). This exploratory clinical multi-omics study aimed to characterize gut microbial and metabolomic features associated with hemoglobin improvement in patients with AA. Methods Nineteen patients with AA who received kidney-nourishing and marrow-benefiting traditional Chinese medicine combined with conventional Western medical treatment were included. According to hematologic response at approximately 6 months, patients were classified as responders ( n = 14) or non-responders ( n = 5). Paired fecal and serum samples collected before and after treatment were analyzed using 16S rRNA sequencing and untargeted metabolomics. Hb_change, defined as post-treatment hemoglobin minus baseline hemoglobin, was used as the primary clinical variable. Associations between omics features and Hb_change were assessed using Spearman correlation with Benjamini–Hochberg correction. Leave-one-out analysis, bootstrap confidence intervals, baseline Hb-adjusted models, and paired longitudinal analyses were used to further examine selected associations. Results Hb_change was greater in responders than in non-responders (31.5 vs. 10.0 g/L, p = 0.006). In the pretreatment microbiome, Methyloversatilis was negatively correlated with Hb_change (rho = −0.712, p = 0.000627, q = 0.0881), whereas Shuttleworthia was positively correlated with Hb_change (rho = 0.703, p = 0.000779, q = 0.0881). For Methyloversatilis, leave-one-out correlations remained negative (rho range, −0.813 to −0.673), the bootstrap 95% confidence interval was −0.883 to −0.340, and the baseline Hb-adjusted coefficient was −8.34 g/L per 1-SD increase (95% CI, −13.89 to −2.80; p = 0.00567). In the metabolomic analyses, the post-treatment fecal feature putatively annotated as N-Acetylgalactosamine_6-sulfate and the serum features Quinic acid and Proline, together with features putatively annotated as PC(37:4) and 4-(Thiophen-2-yl)-6-(trifluoromethyl)pyrimidin-2-ol, showed nominal positive associations with Hb_change, but none met q < 0.1 in the corresponding full-feature analyses. Longitudinal microbiome analyses identified FDR-supported within-patient changes, whereas no fecal or serum metabolomic feature met q < 0.1 in the longitudinal analyses. Conclusions Pretreatment gut microbial features showed FDR-supported associations with hemoglobin improvement, while the highlighted fecal and serum metabolic features showed nominal associations that did not remain significant after full-feature multiple-testing correction. These findings are exploratory and require validation in larger independent cohorts.
Authors
- Zhenzhen Wang (ORCID: https://orcid.org/0000-0001-7936-3792)
- Yuping Si
- Yaru Wang (ORCID: https://orcid.org/0009-0009-6088-709X)
- Lulu Li (ORCID: https://orcid.org/0000-0001-6483-3528)
- Jie Xu (ORCID: https://orcid.org/0000-0002-1897-1644)
- Siyuan Cui (ORCID: https://orcid.org/0000-0003-0560-2388)
- Changnian Li (ORCID: https://orcid.org/0009-0003-6738-7182)
- Zezhou Zang
- Yueli Liu
- Yan Wang
- Ruirong Xu
Institutions
- Shandong University of Traditional Chinese Medicine (CN)
- Beijing University of Chinese Medicine (CN)
- Dongzhimen Hospital Affiliated to Beijing University of Chinese Medicine (CN)
- Affiliated Hospital of Shandong University of Traditional Chinese Medicine (CN)
Publication Details
- Journal
- Frontiers in Medicine
- Published
- 2026-11-13
- DOI
- https://doi.org/10.3389/fmed.2026.1935254
- Primary Topic
- Gut microbiota and health
- Type
- article
- Field-Weighted Citation Impact
- 0.00