Current landscape of preoperative neoadjuvant therapies for initial resectable colorectal cancer liver metastasis

Table 1 Overview of biological behavior assessment systems for colorectal cancer liver metastasis

Evaluation System	Components included	Advantages	Limitations	Ref.
CRS	Five clinical indicators	Widely used; identifies high-risk patients	Limited evidence on improving outcomes for high-risk CRS patients	[21-23]
CMS	Molecular classification into 4 subtypes	Offers nuanced view of disease	Application in identifying neoadjuvant treatment beneficiaries remains unexplored	[24]
m-CS	RAS gene status, size of liver metastases, lymph node status of the primary tumor	Enhanced system compared to previous scores	Lacks granularity in weighing diverse high-risk factors and does not account for chemotherapy sensitivity	[29,30]
TBS	Categorizes patients into low, intermediate, and high-risk groups based on tumor size and number	Superior discriminatory ability	Limited by excluding chemotherapy as an evaluation parameter	[31]
GAME	Genetic and morphological factors	Outperformed CRS in external validation	Potential limitation in excluding chemotherapy as an evaluation parameter	[33,34]
CERR	Integrates mTBS with additional parameters	CERR	Mathematical complexity and abstract metrics pose challenges for widespread clinical application	[35]
AI model	Machine learning-based model predicting recurrence risks	Remarkable accuracy in predicting recurrence risk	Inherent limitations include model overfitting and the black-box nature, hindering seamless integration into clinical practice	[36]

CRS: Clinical risk score; CMS: Consensus molecular subtypes; m-CS: Modified clinical score; mTBS: Refined tumor burden score; GAME: Genetic and morphological evaluation; CERR: Comprehensive evaluation of relapse risk; AI: Artificial intelligence.