TABLE 1.
Comparison of predictive effects of different prediction models with or without gut microbiota.
| Study author | Biologics | Disease | Specimen | Design | Research technique | Population (cohort size) | Predictive markers | Prediction accuracy/AUC | Prediction time | |
|---|---|---|---|---|---|---|---|---|---|---|
| Prediction model with gut microbiota | Arijs et al. (2009) | IFX | UC | Mucosal biopsy | cross-sectional | microarray and qPCR | cohort A: adult (24) | top five genes from cohort A | accuracy:A to A (83%) A to B (59.1%) | week 4–6 |
| cohort B: adult (22) | top five genes from cohort B | accuracy:B to B (90.9%) B to A (70.8%) | Week 8 | |||||||
| Billiet et al. (2015) | IFX | CD | clinical data | retrospective | Matrix model | adult CD (201) | age at first IFX, BMI, and previous surgery | 0.78 < AUC <0.80 | week 14 | |
| Gonczi et al. (2017) | IFX | IBD | serum | prospective | ELISA | adult CD (184) | Infliximab trough level | AUCTLweek2 = 0.72 | week 14 | |
| adult UC (107) | Infliximab trough level | AUC TLweek2 = 0.81 | week 14 | |||||||
| Bar-Yoseph et al. (2018) | IFX | IBD | serum | retrospective | ELISA | adult (140) | Infliximab levels <6.8 μg/ml ATI > 4.3 μg/mL-eq | AUC = 0.68 | week 14 | |
| AUC = 0.78 | ||||||||||
| Dulai et al. (2018) | VDZ | CD | GEMINI 2 and VICTORY Dataset | cross-sectional | Model derivation | discovery cohort: GEMINI 2 (814) Validation cohort: VICTORY (336) | Individual multi-variable logistic regression prediction models | AUC = 0.67 | week 26 | |
| Zhou et al. (2018) | IFX | CD | serum, clinical data | prospective | ELISA | discovery cohort: adult (16) Validation cohort: RISK(668), PRISM(155) | CDAI | accuracy: CD (58.7%) | week 30 | |
| Fecal calprotectin | accuracy: CD (62.5%) | |||||||||
| Engström et al. (2019) | IFX | IBD | feces, serum | cross-sectional | ELISA and near-infrared particle immunoassay | adult (CD: 76 UC: 47) |
Fecal calprotectin >221 μg/g | AUC = 0.71 | week 12 | |
| CRP > 2.1 mg/L | AUC = 0.58 | |||||||||
| Shi et al. (2021) | IFX/ADA | IBD | Mucosal biopsy | cross-sectional | RNA-seq and microarray | GEO and SRA databases | GIMATS module | AUC = 0.720–0.853 | week 4–6 | |
| VDZ | AUC = 0.661–0.728 | |||||||||
| Lee et al., 2021 | IFX/UST/VDZ | IBD | feces, serum | prospective | Random forest classifiers | adult (CD: 108 UC: 77) | clinical features | AUC = 0.624 | week 14 | |
| Prediction model with gut microbiota | Ananthakrishnan et al. (2017) | IFX,VDA | IBD | feces | prospective | 16srRNA | adult (CD: 42 UC: 43) | Gut microbiota | AUC 0.872 | week 14 |
| Zhou et al. (2018) | IFX | IBD | feces | prospective | 16srRNA | Discovery cohart :adult (16) Validati Cohort: RISK (668) PRISM(155) | Gut microbiota | Accuracy.CD (87.5%) UC (79.1%) | week 30 | |
| Zhou et al. (2018) | IFX | IBD | feces, serum,Clinical, data | prospective | 16srRNA ELISA | Discovery cohart :adult (16) Validati Cohort: RISK PRISM(155) | Gut microbiota+FC+CDAI | Accuracy.CD (93.8%) | week 30 | |
| Doherty (2018) | UST | CD | feces, serum,Clinical, data | prospective | 16srRNA | Adult(306) | Gut microbiota | AUC = 0.844 | week 6 | |
| Zhuang et al, (2020) | IFX | CD | feces | prospective | 16srRNA | Adult(49) | Gut microbiota | Clinical response (83.4%) Clinical response (83.4%) endoscopic response(89.1%) | week 30 | |
| Ventin and Holmberg et al, (2021) | IFX | IBD | feces | prospective | 16srRNA | adult (CD: 25 UC: 47) | Gut microbiota | CD AUC = 0.933 UC AUC=0.818 | week 12 | |
| Lee et al. (2021) | IFX/UST/VDZ | IBD | feces, serum | prospective | Metagenomic Sequencing | adult (CD: 108 UC: 77) | Gut microbiota+Clinical Features | AUC = 0.849 | week 14 |