Table 2.
Studies of metabolites analyzed in pre-diagnostic blood samples (collected in prospective cohort or general screening settings) for the purpose of risk prediction or early diagnosis of colorectal cancer.
| Reference | Cohort (Design) |
Time from Sampling to Diagnosis (Cohort Setting Only) |
CRC | Adenoma | Contr./ Cohort |
Biomarker/ Platform |
Main Findings | Adapted NOS Scale ** Max: Selection = ★★★★ Comp. = ★★ Exp./Outc. = ★★★ |
|---|---|---|---|---|---|---|---|---|
| Cohort setting | ||||||||
| Kühn et al. BMC Med, 2016 [56] |
EPIC-HEIDELBERG (Case-cohort) |
6.6 years (median) | 163 | - | 774 | 120 metabolites: (acylcarnitines, amino acids, biogenic amines, phosphatidylcholines, sphingolipids, and hexoses) | Top markers: LysoPC a C18:0, PC ae C30:0 LysoPC a C18:0 (4th vs. 1st quartile) OR: 1.84 (95% CI: 1.02–3.34) PC ae C30:0 (4th vs. 1st quartile) OR: 0.50 (95% CI: 0.28–0.90) |
★★★ ★★ ★★★ |
| Shu et al. Int J Cancer, 2018 [58] |
SWHS/SMHS (Nested case control) | Time stratification: <4 years and >4 years |
250 | - | 250 * | Metabolites in plasma: 35 metabolites associated with CRC at FDR-p < 0.05 |
Top 9 panel: AUC: 0.76 Top 2 single metabolites: Picolinic acid: OR: 5.11 (95% CI: 2.33–11.20) PE(20:0/18:2): OR: 0.45 (95% CI: 0.29–0.70) |
★★★★ ★★ ★★★ |
| Cross et al. Cancer, 2014 [55] |
PLCO (Nested case control) | 7.8 years (median) | 254 | - | 254 * | 676 serum metabolites (metabolon) | Leucyl-leucine (90th vs. 10th percentile)OR: 0.50 (95% CI: 0.32–0.80) Glycochenodeoxycholate (90th vs. 10th percentile, sex stratified) OR: 5.34 (95% CI: 2.09–13.68) Significance lost after adjustments |
★★★★ ★★ ★★★ |
| Perttula et al. BMC Cancer, 2018 [57] |
EPIC-TURIN (Nested case control) | 7.5 years (median) | 66 | - | 66 * | Lipophilic metabolites incl. (ULCFAs): 8690 features, 9 selected |
Top markers: IDs: 5080, 3207, 6054 and 839 Classification rate: 72% | ★★ ★★ ★★★ |
| Screening setting | ||||||||
| Farshidfar F et al. Br J Cancer, 2016 [59] |
Screening patients, Calgary, Canada (discovery) |
- | - | A: 31 | 254 | GC-MS untargeted metabolomics | Multi-marker profile: (14 metabolites): AUC 0.81 (0.70–0.92) | ★★★ ★★ ★★★ |
Abbreviations (not including biomarker names, for which the reader is referred to the original study article): CI, confidence interval, CRC, colorectal cancer; EPIC, European Prospective Investigation into Cancer and Nutrition; GC-MS, gas chromatography–mass spectrometry; OR, odds ratio; PLCO, Prostate, Lung, Colorectal, and Ovarian cancer screening trial; SMHS, Shanghai Men’s Health Study; SWHS, Shanghai Women’s Health Study. * Matched controls. ** Newcastle–Ottawa Quality Assessment Scale for case-control studies, adapted for use for assessment of biomarker studies conducted in a screening setting. For case-control studies: selection (max 4 stars), comparability (max 2 stars), exposure (max 3 stars). For cohort studies: selection (max 4 stars), comparability (max 2 stars), outcome (max 3 stars).