Predicting Risk of Post-Treatment Relapse in Patients with Inflammatory Bowel Disease Based on Intestinal Microbiota
Bibliographic record
Abstract
Tao Zhang,1,* Binbo He,1,* Chao Lan,1,* Jie Liu,1 Qingyu Zeng,1 Wenfeng Pu,1 Lifeng Zhou,1 Qian Zhou,1 Dan Hu,1 Yanan Chen,1 Yiming Peng,1 Guobing Li,1 Qing Wang,1 Long Chen,1 Zonghan Du,1 Shiqing Li,1 Xiaobo Tang,1 Jian Chen,1 Chuanxing Xiao2 1Department of Gastroenterology, Nanchong Central Hospital, The Second Clinical Medical College, North Sichuan Medical College, Nanchong, Sichuan, 637000, People’s Republic of China; 2Xiamen Institutes of Respiratory Health, Xiamen, Fujian, 361000, People’s Republic of China*These authors contributed equally to this workCorrespondence: Tao Zhang, Department of Gastroenterology, Nanchong Central Hospital, The Second Clinical Medical College, North Sichuan Medical College, Nanchong, Sichuan, 637000, People’s Republic of China, Email 305514271@qq.com Chuanxing Xiao, Xiamen Institutes of Respiratory Health, Xiamen, Fujian, 361000, People’s Republic of China, Email xiaoxx@xmu.edu.cnBackground: Inflammatory bowel disease (IBD) is a chronic inflammatory disorder of the gastrointestinal tract. Post-treatment relapse is a major clinical challenge, and gut microbiota dysbiosis is hypothesized to be involved.Methods: We enrolled 88 patients with IBD (46 UC, 42 CD) to investigate gut microbiota features associated with post-treatment relapse. Fecal samples collected before and after therapy were analyzed by 16S rRNA sequencing. A random forest (RF) model was developed to evaluate the predictive value of microbiota signatures for recurrence.Results: At baseline (pre-treatment), no differences were observed in gut microbiota diversity between patients with UC and CD. However, significant compositional differences were observed, with Fusobacterium and Parabacteroides enriched in CD patients, and Anaerostipes and Enterococcus enriched in UC patients. Post-treatment, there was no significant difference in the α-diversity across IBD patients; however, β-diversity exhibited significant alterations, marked by enrichment of Akkermansia and Lachnoclostridium. Patients maintaining remission exhibited significant post-treatment beta-diversity shifts and enrichment of Erysipelatoclostridium, Delftia, Tyzzerella, Sphingomonas, Subdoligranulum, Proteus, and Enterococcus. Conversely, patients experiencing recurrence showed a significant reduction in Shannon alpha-diversity post-treatment and enrichment of UCG-002, Odoribacter, Delftia, Flavonifractor, and Erysipelotrichaceae_UCG-003. Post-treatment microbiota composition differed significantly between recurrent and non-recurrent patients, with higher alpha-diversity in the non-recurrent group. Non-recurrent patients exhibited enrichment of Eubacterium_hallii_group, Clostridioides, UCG-002, Paraprevotella, Bilophila, Desulfovibrio, Butyricimonas, Clostridium_sensu_stricto_1, Megamonas, Romboutsia, Parabacteroides, and Enterococcus, while Delftia was predominantly enriched in recurrent patients. The RF model, built using differentially abundant genera to distinguish recurrence status, achieved an area under the curve (AUC) of 0.721 in the validation set and 0.861 in the test cohort, indicating good predictive performance.Conclusion: Our findings suggest that gut microbiota composition may hold clues for predicting IBD relapse. The RF model is a proof-of-concept that warrants external validation in prospective, multi-center studies before clinical application.Keywords: inflammatory bowel disease, relapse, gut microbiota, random forest model, 16S rRNA sequencing
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.000 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".