Skip to main content

Table 4 Risk prediction models for CRC recurrence

From: Risk factors and risk prediction models for colorectal cancer metastasis and recurrence: an umbrella review of systematic reviews and meta-analyses of observational studies

Author, year Country Study design Population Outcome Prediction time horizon Sample size (development/validation) Predictors Model performance (95%CI) Model presentation Internal validation
Hoshino, 2016 [34] Japan D CRC (II) Overall recurrence 5 years 4167 Sex C statistic 0.64; calibration plot Nomogram NA
CEA
Tumor location
Tumor depth
Lymphatic invasion
Venous invasion
No. of positive lymph nodes
Peng, 2010 [37] China D CRC (II–III) Overall recurrence 3 years 95 AJCC stage AUC 0.75 Formula NA
Genetic score
Ying, 2014 [38] China D CRC (I–III, curative resection) Overall recurrence 3 years 205 Tumor size C statistic 0.810/0.890/0.802 Nomogram NA
Tumor differentiation
AJCC stage
NLR
Chemotherapy
Zakaria, 2007 [39] Japan D CRC (liver metastasis + hepatectomy) Overall recurrence 5 years 662 Hepatoduodenal C statistic 0.61 (0.57–0.64)/0.58 (0.550.61) Nomogram NA
Lymph node status
Transfusions
Primary cancer
Regional lymph node
No. of metastasis
Tian, 2017 [36] China D CRC Overall recurrence 3 years 556 Gene signature AUC 0.921 (0.869–0.972); calibration plot Nomogram NA
AJCC stage
Tumor differentiation
*Kim, 2018 [40] Korea D + IV CRC (I) Overall recurrence 5 years 1538 Sex C statistic 0.71; calibration plot Nomogram NA
Tumor location
pT stage
LVI
Tumor size
*Miyoshi, 2016 [41] Japan D + EV CRC (IV with liver and/or lung metastases) Overall recurrence 5 years 113 Preoperative CEA C statistic 0.631 Nomogram NA
Tumor location
Tumor invasion
Lymph node metastasis
Synchronous metastatic lesions
*Saso, 2018 [42] Japan D + EV Colon cancer (II) Overall recurrence 5 years 352/213 CEA level C statistic 0.675; external C statistic 0.552 Nomogram NA
Tumor invasion
Lymphatic invasion
Venous invasion
Renfro, 2014 [35] USA D + EV Colon cancer (III) Overall recurrence 5 years 15,995/1903 Sex C statistic 0.65; calibration plot Nomogram NA
BMI
PS
T stage
Lymph node ratio
Grade
Tumor location
Treatment
Hida, 2017 [43] Japan D Rectal cancer (II–III) Overall recurrence 2 years 792 Tumor differentiation AUC 0.831 Formula NA
Depth
Lymph node
Surgery
Postoperative complication
Tumor height
CEA
Gijn, 2015 [31] Netherlands D + IV Rectal cancer (tis-III) Local recurrence 6 years 1823 Distance from the anal verge C statistic 0.787 (0.761–0.814); Hosmer–Lemeshow test: p = 0.68 Nomogram Cross-validation
pT stage
pN stage
pM stage
Surgery type
Residual tumor status
Radiotherapy
Valentini, 2011 [32] Belgium D + EV Rectal cancer (II–III) Local recurrence 5 years 3458 pT stage External C statistic 0.68 (0.59–0.76); Wald statistic: p = 0.064 Nomogram Random split
cT stage
pN stage
Age
Concomitant chemotherapy
Adjuvant chemotherapy
  1. Reproduced from He et al. [21]
  2. Abbreviations: D model development, D + IV model development with internal validation, D + EV model development with external validation, AJCC American Joint Committee on Cancer, AUC area under the curve, BMI body mass index,  CEA carcinoembryonic antigen, CRC colorectal cancer, LVI lymph vascular invasion, NA non-available, NLR neutrophil to lymphocyte ratio, PS performance status
  3. *Three recently developed prediction models were additionally included, and data was extracted based on the previous criteria