Table 3 Potential diagnostic, predictive or prognostic biomarkers for CRLM

Mutated genes

BRAF, KRAS, TP53, PIK3CA, and members of the SMAD family

Potential prognostic predictors for overall survival after liver surgery for CRLM

Expression is highly consistent between primary and metastatic tumors, and molecular assays can be performed on resected samples of primary tumors

Biomarkers cannot predict primary tumor nodal status, limiting their use as tools to guide surgical patterns

[110]

Differentiated expressed genes

ERCC1

Potential predictive biomarkers for the response rate to platinum-based chemotherapy in patients with CRLM

A high level of reproducible predictive markers in pathology practice benefits ERCC1 as a predictive biomarker for platinum-based chemotherapy

The interobserver agreement is low; ERCC1 IHC expression in primary tumors is low in association with metastatic liver events, implying the need to use tissue from actual tumor burden to evaluate ERCC1 expression; and treatment alters ERCC1 expression

[112]

EMX2

Down-regulated EMX2 is a strong predictor of CRLM

EMX2 has predictive value as a prognostic factor in CRC and may have a functional role in metastatic spread, so that EMX2 inhibition may be a promising therapeutic strategy

Limited power for statistical inference due to the small sample size; whether it is a driver of metastasis or a coincidence of tumor progression remains to be further explored; Adenoviral vectors serve as delivery systems to restore EMX2, but adenoviral vector therapeutic strategies for cancer patients remain challenging

[113]

Chromosome

Chromosome 4

Chromosome 4 deletion can be a potential predictive biomarker for primary tumor progression, long-term survival, and recurrence after complete metastases resection

It serves as a prognostic biomarker to provide evidence for the decision to use adjuvant chemotherapy in patients with CRLM after metastases resection, early identification of patients with good prognosis who do not require further adjuvant therapy, and reduced side effects

The small sample size for preliminary studies

[114]

miRNAs

Serum exosomal miR-122

As a diagnostic marker of CRC with LM

Serum exosomal miR-122 has high specificity and can distinguish CRC patients with LM from CRC without LM

Small sample size

[116]

Exosomal miR-21, miR-203, and miR-210

Non-invasive prognostic biomarker for predicting liver metastases

Circulating molecules are stable, reproducible, and consistent across individuals. They can be used as predictive biomarkers to distinguish liver metastatic CRC from other metastatic or non-metastatic CRC

The expression of miRNAs varies widely among tumor populations; levels of circulating miRNAs can be significantly altered by hemolysis, and techniques for sample storage, RNA extraction, and miRNA quantification need to be improved

[119]

LncRNA

LncRNA Yiya

Independent prognostic biomarkers for CRLM

Colorectal liver metastases generally rely on portal vein drainage, and Yiya is associated with hematogenous spread, which has considerable value in early prediction and timely clinical intervention

[120]

CircRNA

CircRNA_0001178 and circRNA_0000826

Potential diagnostic biomarkers for CRLM

CircRNAs are tissue-specific and, therefore, suitable as cancer biomarkers

[117]

Others

Apoptotic circulating tumor cells

Potential predictive biomarkers for CRLM

Apoptotic CTCs, CTC fragments but not CTCs, are useful liquid biopsy markers associated with worse overall and progression-free survival, especially in metastatic liver disease

The heterogeneity of patient populations, including different prior treatments and disease genetics, must also be considered

[123]

Leptin and the activity of chitinase

Potential predictive biomarkers for CRC liver metastasis

Serum chitinase activity is an independent risk factor for predicting liver metastasis, and the effect is superior to traditional CEA

Dividing subjects into groups makes the amount of data in each group smaller

[124]