Table 2. Studies aimed to identify prognostic miRNA for CRC through high-throughput screening and subsequent qPCR validation.
| Study | Country | Sample type | Discovery phase (method and sample population) | Validation phase (method and sample population) | Outcome | Follow-up (months) | Candidate miRNAs (expression) | HR (95% CI), P value | |
|---|---|---|---|---|---|---|---|---|---|
| Univariate analysis | Multivariate analysis | ||||||||
| Cheng et al. 2011 (63) | Texas (USA) | Plasma | Ohio State microRNA microarray (230 miRNAs); 74 CRC, 28 controls | qRT-PCR; 108 CRC, 48 controls | OS | – | miR-141 (↑) | 3.61(1.96–6.65) | 2.40 (1.18–4.86) |
| Imaoka et al. 2016 (51) | Japan | Serum | Microarray analysis; 3 CRC, 3 adenoma, 3 controls | qRT-PCR; 211 CRCs, 56 adenomas, and 57 control subjects | OS | Up to 80 | miR-1290 (↑) | 3.43 (1.83–6.67) | 4.51 (1.23–23.69), P=0.0096 |
| Li et al. 2015 (53) | China | Serum | TaqMan low-density array (applied biosystems) (749 miRNAs); pooled 20 paired pre- and post-operative CRC, 20 controls | qRT-PCR; 175 stage II/III CRC patients, 130 controls | DFS | Median: 32 | miR-17-3p (↑) | 2.72 (1.58–4.69), P<0.0001 | 2.24 (1.28–3.92), P=0.035 |
| miR-106a (↑) | 2.81 (1.64–4.80), P<0.0001 | 3.02 (1.36–6.73), P=0.007 | |||||||
| Sun et al. 2016 (56) | China | Plasma | TaqMan array (life technologies) (754 miRNAs); 40 CRC, 10 controls | qRT-PCR; 187 CRC, 47 controls | OS | Median: 28 | miR-96 (↑) | P=0.002 | 2.27 (1.15–4.51), P=0.019 |
| miR-200b (↑) | P<0.001 | 2.63 (1.29–5.37), P=0.008 | |||||||
| Vychytilova-Faltejskova et al. 2016 (57) | Czech Republic | Serum | small RNA sequencing (Illumina); pooled 144 CRC, 96 controls | qRT-PCR; 261 CRC | OS | Up to 80 | miR-23a-3p, miR-376c-3p (all↑) | – | 2.30 (1.44–3.66), P<0.0004 |
(↑), miRNAs upregulated in CRC patients compared to healthy subjects. OS, overall survival; DFS, disease free survival; CRC, colorectal cancer.