Table 3.
Studies on use of circulating tumor cells (CTCs) as biomarkers in HCC patients.
| Diagnosis | ||||
|---|---|---|---|---|
| Study | CTCs Definition | Number of Patients | Comparator | Main Findings (Sensitivity/Specificity, AUC) |
| Yao et al., 2005 [227] | CD45 (−) EpCAM (+) then AFP mRNA | 49 HCC 36 CLD or LC 18 healthy subjects |
AFP (cut-off 20 ng/mL) | AFP mRNA (sensitivity/specificity): 72.1%/66.7% Low AFP: sensitivity, 75% High AFP: sensitivity, 71% (p > 0.05) |
| Guo et al., 2007 [228] | CD45 (−) and EpCAM (+), then AFP mRNA | 44 HCC 7 healthy subjects |
AFP (20 ng/mL) | AFP mRNA (sensitivity): 72.7%; 50% in patients with AFP < 20 ng/mL and 86.7% in patients with AFP >1000 ng/mL (p < 0.05) |
| Xu et al., 2011 [229] | ASGPR (+) | 85 HCC 37 CLD or benign liver diseases 20 healthy subjects |
AFP (cut-off 20 or 100 ng/mL) | CTCs (sensitivity/specificity): 81%/100% No significant differences in CTCs level according to AFP values |
| Liu et al., 2013 [210] | CD45 (−) and ICAM-1 (+) | 60 HCC | AFP (cut-off 20 ng/mL) | High levels of CTCs in 83.3% of AFP + and 16.7% of AFP negative patients (p = 0.14) |
| Sun et al., 2013 [204] | CellSearch™ | 123 HCC 5 CLD 10 healthy subjects |
AFP (cut-off 400 ng/mL) | ≥2 CTCs/7.5 mL: Overall (sensitivity/specificity): 41.5%/100% High AFP: sensitivity, 54.7% Low AFP: sensitivity, 31.4% (p = 0.009) |
| Bahnassy et al., 2014 [230] | CD45 (−) and either CK19, CD90 or CD133 (+) | 70 HCC 30 CLD (HCV) 33 healthy subjects |
AFP ratio (undefined) | CTCs have poorer performances compared to AFP. HCC vs. CLD: AFP ratio: 95.7%/90.5%, 0.99 CK19 (+) CTCs: 87.1%/82.5% CD90 (+) CTCs: 81%/89.6% CD133 (+) CTCs: 40%/6.3% |
| Fang et al., 2014 [231] | CellSearch™ | 42 HCC 10 CLD 10 healthy subjects |
AFP (cut-off 40 ng/mL) | CTCs (sensitivity/specificity): 74%/100% Sensitivity 89% among patients with high AFP and 61% among those with low AFP (p = 0.08) |
| Guo et al., 2014 [202] † | CellSearch™ and quantitative PCR for EpCAM in CD45 (−) cells | 122 HCC 25 CLD or LC (HBV) 24 benign tumors 71 healthy subjects |
AFP (cut-off NR) | HCC vs. other groups: EpCAM-mRNA (+) CTCs: 42.6%/96.7%, 0.70 EpCAM-mRNA (+) CTCs + AFP: 73%/93.4%, 0.86 |
| Kelley et al., 2015 [194] | CellSearch™ | 20 HCC 10 CLD |
AFP (400 ng/mL) | CTC detection in 7 of 20 (35%) HCC patients and 0 of 9 CLD (p = 0.04). AFP ≥ 400 ng/mL: sensitivity 70% AFP < 400 ng/mL: sensitivity 10% (p = 0.008) |
| Zhou et al., 2016 [232] | CD45 (−) EpCAM-mRNA (+) | 49 HCC | AFP (cut-off 400 ng/mL) | Any CTCs (sensitivity): Overall: 34.6% Low vs. high AFP: 28.2% vs. 60% (p = 0.06) |
| Kalinich et al., 2017 [233] | PCR assay: AFP, AHSG, ALB, APOH, FABP1, FGB, FGG, GPC3, RBP and TF | 63 HCC 31 CLD 26 healthy subjects |
AFP (cut-off 100 ng/mL) | PCR score +: 9 of 16 (56%) untreated HCC patients, 1 of 31 (3%) CLD and 2 of 26 (7.6%) healthy subjects. 15 patients with both PCR score and AFP available: 4 (27%) PCR score +, 1 (7%) AFP +, 5 (33%) PCR score + and AFP + 6 patients within Milan criteria: 2 (33%) PCR score + and 0 (0%) AFP + |
| Bhan et al., 2018 [234] | CD45 (−) and hydrodynamics, followed by HCC score based on gene expression | 54 HCC 39 CLD 10 healthy subjects |
AFP (cut-off 20 ng/mL) | HCC score outperformed AFP in identifying HCC vs. CLD (sensitivity/specificity): HCC score: 85%/95% AFP: 55%/100% |
| Guo et al., 2018 [200] † | CTC detection panel: PCR for EpCAM, CD133, CD90 and CK19 | Training and validation cohorts: 395 HCC 301 CLD and LC (HBV) 210 healthy subjects |
AFP (cut-off 20 ng/mL) | Validation cohort (sensitivity/specificity, AUC): HCC vs. other groups: CTC detection panel: 72.5%/95%, 0.88 AFP: 57%/90%, 0.77 CTC detection panel + AFP: 76%/95%, 0.89 Early-stage HCC vs. other groups: CTC detection panel: 71.8%/95%, 0.87 AFP: 53.4%/90%, 0.74 CTC detection panel + AFP: 80.9%/87%, 0.88 AUC in different stages: 0.92 (BCLC 0), 0.86 (BCLC A), 0.91 (BCLC B), 0.86 (BCLC C) In AFP negative patients: 77.7%/95%, 0.89 |
| Xue et al., 2018 [235] | CellSearch™ and iFISH (either CD45 (−) CK (+) DAPI (+) and hybridization signal for CEP8 ≥2 or CD45 (−) CK (−) DAPI (+) and hybridization signal for CEP8 > 2) | 30 HCC 10 healthy subjects |
AFP (400 IU/mL) | CTCs measured by CellSearch™ (sensitivity/specificity): 26.7%/100% CTCs measured by iFISH (sensitivity/specificity): 70/100% Low AFP: sensitivity, 90% High sensitivity, 30% (p = 0.002) |
| Yin et al., 2018 [236] | CanPatrol™ | 80 HCC 10 healthy subjects |
AFP (cut-off 20 ng/mL) | Overall cohort (sensitivity/specificity): Any CTCs: 77.5%/100% Twist (+) CTCs: 67.5%/100%Low AFP: sensitivity, 35.3% or 17.7% for any CTCs or Twist (+) CTCs, respectively (p < 0.001) High AFP: sensitivity, 88.9% or 71.8%for any CTCs or Twist (+) CTCs, respectively (p < 0.001) |
| Cheng et al., 2019 [201] | CanPatrol™ | 113 HCC 57 benign liver lesions |
AFP (cut-off 400 μg/L) | CTCs outperformed and provided incremental benefit to AFP.AFP: 44.3%/89.5%, 0.67 Total CTCs (≥3): 62%/89.5%, 0.77 Total CTCs or AFP: AUC = 0.82 |
| Prognosis | ||||
| Study | CTCs Definition | HCC Patients | Stage/Treatment | Main Findings |
| Vona et al., 2004 [203] | Size (diameter > 25 μm) | N = 44 | Stage: 39% multinodular, 39% tumor ≤3 cm, 45% PVT, no EHS Treatment: NR |
Patients with CTCs/circulating tumor microemboli had poorer OS (p = 0.01) No significant association with survival in multivariate analysis. |
| Fan et al., 2011 [208] | CD45 (−) CD90 (+) CD44 (+) | N = 82 | TNM stage I/II/III/IV: 5%/34%/34%/27% Treatment: LR |
CTCs predicted recurrence (sensitivity/specificity): 65.9%/80.5% CTCs (>0.01%) independently associated with poorer: Median recurrence-free survival (6.0 vs. >46.5 months) 2-years recurrence-free survival (22.7% vs. 64.2%) 2-year OS (58.5% vs. 94.1%) (p < 0.001 for all) |
| Liu et al., 2013 [210] | CD45 (−) ICAM-1 (+) | N = 60 | Stage: tumor size >5 cm 72%, multifocal 12% Treatment: LR |
High proportion of ICAM-1 (+) CTCs associated with: Poorer DFS: adjusted HR = 7.15 (2.99–17.05) No independent association with OS: adjusted HR = 2.28 (0.95–7.82) |
| Nel et al., 2013 [237] | CTCs: CD45 (−), DAPI (+), EpCAM (+), ASGPR1 (+) Mesenchymal: either N-cadherin (+) or vimentin (+) Epithelial: pan-CK (+) Mixed: CK (+) and either N-cadherin (+) or vimentin (+) |
N = 11 | Stage: NR Treatment: various (SIRT in 45%) |
Vimentin (+)/CK (+) ratio >0.5 associated with a longer TTP: 1 vs 15 months (p = 0.03) N-cadherin (+)/CK (+) ratio <0.1 associated with a shorter TTP: 1 vs 15 months (p = 0.03) |
| Sun et al., 2013 [204] | CellSearch™ | N = 123 | BCLC stage 0-A/B-C: 82%/18% Treatment: LR |
Presence of CTCs (>2/7.5 mL) before surgery associated with: Increased risk of recurrence: adjusted HR = 5.20 (2.65–10.21) |
| Cheng et al., 2013 [209] | Magnetic cell sorting and PCR for Lin28B | N = 96 | BCLC stage A/B-C: 55%/45% Treatment: LR |
Lin28B positive CTCs associated with: Decreased RFS: adjusted HR = 2.25 (1.01–4.99) Early recurrence (<1 year): adjusted HR = 2.65 (1.02–6.86); also true in earlier stages |
| Schulze et al., 2013 [205] | CellSearch™ | N = 59 | BCLC stage A/B/C: 15%/53%/32% Treatment: NR |
Detection of CTCs was associated with lower OS at the Kaplan-Meier analysis (p = 0.02) |
| Guo et al., 2014 [202] | CellSearch™ and quantitative PCR for EpCAM in CD45 (-) cells | N = 299 | Stage: NR Treatment: LR 53%, TACE 25%, RT 22% |
EpCAM mRNA (+) CTCs associated with worse outcomes Surgery: shorter TTR; adjusted HR = 2.9 (1.6–5.3) TACE: shorter PFS; unadjusted HR = 3.8 (1.4–10) RT: shorter PFS; unadjusted HR = 5.1 (1.4–18.5) |
| Nel et al., 2014 [238] | CD45 (−), EpCAM (+), DAPI (+), pan-CK (+) and IGFBP1 mRNA (+) | N = 25 | TNM stage II/III/IV: 28%/48%/24% Treatment: SIRT |
Low expression of IGFBP1 mRNA in CTCs discriminate progression from disease control (sensitivity 80%, specificity 80%, AUC = 0.8). Low IGFBP1 mRNA in CTCs correlated with shorter TTP (p = 0.04) |
| Li et al., 2016 [222] | Density-based, CD45 (−), pan-CK (+) and either pAkt1/2/3 or pERK1/2 (+) | N = 109 | Stage: advanced Treatment: sorafenib |
High proportion of pERK (+) pAkt (−) CTCs associated with longer PFS: adjusted HR = 9.39 (3.24–27.19) |
| Ogle et al., 2016 [195] | CD45 (−), morphology, size | N = 69 | BCLC stage A/B/C/D: 16%/7%/73%/4% Treatment: LT 6%, LR 4%, ABL 10%, IAT 39%, sorafenib 13%, BSC 28% |
Presence of CTCs (>1/4 mL) at any time (N = 69): Shorter OS: adjusted HR = 2.34 (1.015.43) Shorter TTP (p = 0.006) Presence of CTCs post-treatment (N = 29): Shorter OS: adjusted HR = 6.16 (1.71–22.33) Shorter TTP (p = 0.002) |
| Zhou et al., 2016 [232] | EpCAM mRNA (+) | N = 49 | BCLC stage 0-A/B-C: 90%/10% Treatment: LR |
High EpCAM mRNA (+) CTCs associated with increased risk of recurrence: adjusted HR = 6.69 (1.94–22.88) |
| von Felden et al., 2017 [206] | CellSearch™ | N = 57 | BCLC stage A/B: 92%/8% Treatment: LR |
CTCs status was independently associated with increased risk of recurrence: adjusted HR = 3.1 (1.0–9.4) |
| Guo et al., 2018 [200] | CTC detection panel: PCR for EpCAM, CD133, CD90 and CK19 | N = 395 | Training: BCLC stage 0-A: 66% Treatment: LR 98%, TACE 2% Validation: BCLC stage 0-A: 48% Treatment: LR 67%, TACE 33% |
CTC detection panel was associated with shorter TTR: Training cohort: adjusted HR = 2.69 (1.62–4.48) Validation cohort: adjusted HR = 3.13 (1.36–7.19) Association remained significant in patients with negative AFP and with early-stage (BCLC 0-A) tumor |
| Qi et al., 2018 [211] | Can Patrol™ | N = 112 | BCLC stage 0/A/B/C: 10%/39%/21%/30% Treatment: LR |
CTCs associated with HCC recurrence: CTC count: adjusted HR = 1.02 (1.01–1.04) Mesenchymal CTC percentage: adjusted HR = 1.02 (1.01–1.03) Mesenchymal > epithelial CTC percentage: adjusted HR = 1.00 (0.99–1.02) Mesenchymal = epithelial CTC percentage, mesenchymal < epithelial CTC percentage, epithelial CTC percentage not associated with recurrence at univariate analysis. |
| Sun et al., 2018 [226] | CellSearch™ | N = 73 | BCLC stage 0-A/B-C: 77%/23% Treatment: LR |
Presence of CTCs in different vascular sites. Association with intrahepatic recurrence: Peripheral veins: adjusted HR = 0.77 (0.14–5.19) Peripheral arteries: adjusted HR = 2.54 (0.87–7.42) Peripheral veins CTCs with clusters: adjusted HR = 3.48 (1.40–8.61) Association with lung metastasis: Hepatic vein CTCs: adjusted HR = 0.59 (0.04–9.54) Intrahepatic inferior vena cava CTCs: adjusted HR = 0.67 (0.10–4.40) Hepatic vein CTCs with clusters: adjusted HR = 42.2 (3.73–477.8) |
| Wang et al., 2018 [239] | CanPatrol™ | N = 62 | BCLC stage 0-A/B-C: 37%/63% Treatment: LR |
Association with early recurrence: Total CTCs: unadjusted HR = 2.95 (1.18–7.35); NS after adjustment Mesenchymal CTCs: unadjusted HR = 4.74 (2.04–11.01); adjusted HR = 3.45 (1.39–8.56) Mixed CTCs: unadjusted HR = 2.94 (1.31–6.59); NS after adjustment |
| Yu et al., 2018 [215] | CellSearch™ | N = 139 | BCLC stage 0+A/B+C: 40%/60% Treatment: LR |
4 categories: 1) persistently (+); 2) preoperatively (+) but postoperatively (−); 3) preoperatively (−) but postoperatively (+); 4) persistently (−). For a 1-point increase in category: DFS: adjusted HR = 0.53 (0.41–0.68) OS: adjusted HR = 0.48 (0.36–0.66) |
| Ye et al., 2018 [240] | CanPatrol™ | N = 42 | BCLC stage A-B/C-D: 81%/19% Treatment: LR |
Pre-operative CTC count not associated with OS and PFS Post-operative CTC count (>5): Poorer PFS: adjusted HR = 6.89 (1.64–29.0) No independent association with OS: adjusted HR = 15.65 (0.80–304.64) Increase of post-operative CTC count: Poorer PFS: adjusted HR = 39.58 (4.22–371.64) |
| Wang et al., 2018 [213] | SE-iFISH | N = 14 | Stage: NR Treatment: NR |
Detection of small CTCs with triploid chromosome 8 showed shorter DFS (p = 0.007); HR not reported |
| Court et al., 2018 [241] | NanoVelcro™ | N = 80 | BCLC stage A/B/C/D: 18%/28%/43%/11% Treatment: ABL, TACE, SIRT, systemic therapy, BSC |
Total CTCs were associated with: Shorter TTR in patients with early stage: univariate HR = 9.7 (2.08–45.19); no significant association in multivariate. Shorter PFS in patients with advanced disease: univariate HR = 2.09 (1.11–3.96); multivariate HR =2.09 (1.11–3.96) Vimentin (+) CTCs independently associated with: Poorer OS: adjusted HR = 2.21 (1.38–3.56) Poorer PFS in patients with advanced disease: adjusted HR = 2.16 (1.33–4.42) Trend toward fast TTR in patients with early stage: adjusted HR = 2.45 (0.91–6.57) |
| Shen et al., 2018 [242] | CellSearch™ | N = 97 | BCLC stage A-B/C: 56%/44% Treatment: TACE |
CTC count independently predicted OS: High vs. low level group: adjusted HR = 2.82 (1.22–6.53) Intermediate vs. low group: adjusted HR = 1.30 (0.63–2.69) |
| Ha et al., 2019 [214] | Tapered slit platform (detection based on size and morphology) | N = 105 | BCLC stage 0/A: 19%/81% Treatment: LR |
Presence of pre- and post-operative CTCs not associated with recurrence. Positive ΔCTC (increase of CTC after surgery): Shorter RFS: adjusted HR = 2.28 (1.06–4.90) No associations with OS |
| Hamaoka et al., 2019 [243] | Glypican-3 (+) | N = 85 | Stage: median tumor number 1 and median size 25 mm Treatment: LR |
CTCs associated with: Higher risk of microscopic portal vein invasion: adjusted OR = 14.6 (3.3–106.0) Shorter DFS (p = 0.02) Shorter OS (p = 0.047) |
| Wu et al., 2019 [244] | CD45 (−) and abnormal chromosome 8 amplification by FISH | N = 155 | BCLC stage A/B/C: 38%/14%/48% Treatment: TACE |
Presence of pre-TACE CTCs associated with poorer OS: adjusted HR = 2.84 (1.41–5.73) |
| Chen et al., 2020 [218] | CD45 (-) and imFISH | N = 50 | TNM stage I/II/III/IV: 8%/32%/58%2% Treatment: LT |
CTCs detection was associated with recurrence post-LT: adjusted HR = 5.41 (1.13–25.87) |
| Zhou et al., 2020 [245] | Size and deformability | N = 137 | BCLC stage 0-A/B-C: 57%/43% Treatment: LR |
Presence of CTCs: Independently associated with microvascular invasion: adjusted HR = 1.76 (1.34–2.30) Shorter OS (19.2 months vs. not reached; p = 0.005) |
| Winograd et al., 2020 [223] | CD45 (−), DAPI (+), CK (+), PD-L1 (+) | N = 87 | BCLC stage A/B/C/D: 25%/25%/41%/8% Treatment: various; checkpoint inhibitors 14.3% |
Detection of CTCs expressing PD-L1:Associated with poorer OS (≥4 PD-L1 (+) CTCs): adjusted HR = 3.22 (1.33–7.79) Predicted response to checkpoint inhibitors |
| Wang et al., 2020 [246] | CellSearch™ | N = 344 | BCLC stage 0-A/B-C: 73.8%/26.2% Treatment: LR ± adjuvant TACE |
After propensity score matching, in CTC positive patients’ adjuvant TACE provide benefits in: TTR (45.8 vs. 9.8 months, p < 0.001) OS (not reached vs. 36.4 months; p < 0.001) |
| Wang et al., 2020 [219] | ChimeraX®-i120 platform | N = 193 | Stage: Milan-in 60% Treatment: LT |
Post-operative CTC count ≥1 independently associated with tumor recurrence: adjusted HR = 2.67 (1.50–4.74) |
† Cohort of Guo et al., 2014 [202] and Guo et al., 2018 [200] may overlap. Abbreviations: AFP, alpha-fetoprotein; ABL, ablation; AUC, area under the curve; BCLC, Barcelona Clinic Liver Cancer; BSC, best supportive care; DC, disease control; DFS, disease-free survival; EHS, extrahepatic spread; HBV, hepatitis B virus; HCC, hepatocellular carcinoma; HCV, hepatitis C virus; HR, hazard ratio; IAT, intra-arterial therapies; LR, liver resection; LT, liver transplantation; OS, overall survival; OR, odds ratio; NS, not significant; NR, not reported; PFS, progression-free survival; PVT, portal vein thrombosis; RFS, recurrence-free survival; RT, radiotherapy; SIRT, selective internal radiation therapy; TACE, transarterial chemoembolization; TTP, time to progression; TTR, time to recurrence.