Evolving landscape of biomarkers in colorectal cancer: From interleukin-8 to integrated multi-marker strategies

Table 1 Key clinical trials and studies evaluating circulating biomarkers in colorectal cancer

Biomarker	Study/trial	Population	Key findings	Level of evidence	Ref.
ctDNA (MRD detection)	DYNAMIC-III	Stage III colon cancer (n = 1002)	ctDNA-guided de-escalation reduced oxaliplatin exposure (88.6% → 34.8%) without compromising 3-year RFS in low-risk patients; non-inferiority not formally met	Phase II/III randomized	[45,46]
	TRACC Part B	Stage I-III CRC (n = 214 evaluable)	Tissue-free MRD detection; 2-year RFS 91.1% (ctDNA-) vs 50.4% (ctDNA+); HR 6.5; lead time 73 months	Prospective cohort	[44]
	Faulkner meta-analysis	37 studies (n = 3002)	Postoperative ctDNA positivity strongly associated with poorer PFS (HR 6.92; 95%CI 4.49-10.64)	Meta-analysis	[42]
	Barault et al[47]	mCRC (n = 182)	Five-gene methylation panel (EYA4, GRIA4, ITGA4, MAP3K14-AS1, MSC) detected in 85.7%; dynamics correlated with response and PFS	Prospective cohort	[47]
Methylation markers	Wang et al[48] MethylDT	Multicenter (n = 1194)	NTMT1/MAP3K14-AS1 methylation: Sensitivity 91.2% for CRC, 87.4% for stage I-II; specificity 92.4%	Multicenter validation	[48]
Protein markers	Duan et al[49]	Discovery (n = 72) + validation (n = 191)	SETD7 identified by proteomics; sensitivity 92.2%, specificity 81.1%, AUC 0.947; correlated with stage and MSI	Case-control + functional	[49]
IL-8 (prognostic)	Meta-analysis	Multiple studies	High IL-8 levels significantly correlated with shorter OS and PFS; effect stronger for circulating than tissue IL-8	Meta-analysis	[15]
	Meta-analysis	Multiple studies	Pooled diagnostic accuracy: AUC 0.92, sensitivity 70%, specificity 91%	Meta-analysis	[20]
	Björkman et al[31]	328 CRC patients	IL-8 was the only significant univariate prognostic factor after FDR correction (HR = 4.88); machine learning identified IL-8 and IL-2Ra for multivariate model	Prospective cohort	[31]
	Park et al[55]	400 CRC patients	IL-8 and LMR were independent prognostic factors for PFS and OS; composite classification enhanced prognostic value	Prospective cohort	[55]
IL-8 (Predictive)	Suenaga et al[58]	mCRC on bevacizumab/regorafenib (n = 125)	High IL-8 associated with liver metastasis and shorter OS (HR = 4.31)	Pooled analysis	[58]
	Sanmamed et al[59]	Melanoma/NSCLC on anti-PD-1	Changes in serum IL-8 reflected and predicted response to immunotherapy	Prospective cohort	[59]
IL-8 (Genetic)	Lurje et al[57]	Stage III colon cancer (n = 125)	IL-8 T-251A polymorphism predicted tumor recurrence; combined with VEGF C+936T identified highest risk group	Retrospective cohort	[57]
CEA	Beom et al[27]	Discovery (n = 2021) + validation (n = 171)	High-normal CEA (2.1-5.0 ng/mL) independently prognostic; optimal cutoff 2.1 ng/mL; influenced by smoking and tumor location	Retrospective + prospective validation	[27]
MMR	ATOMIC	Stage III dMMR colon cancer (n = 700)	mFOLFOX + atezolizumab improved DFS vs mFOLFOX alone	Phase III RCT	[26]
	NICHE-2	Locally advanced dMMR colon cancer (n = 112)	Neoadjuvant nivolumab + ipilimumab → pCR > 65%, excellent 3-year DFS	Phase II	[25]
KRAS + laterality	Kamphues et al[28]	International retrospective, n = 1093 resected non-metastatic CRC	KRAS mutation prognostic only in left-sided tumors; not in right-sided or rectal	Retrospective cohort	[28]
	Tran et al[32]	NCDB analysis, n = 5292 stage I-III colon cancer	Interaction between KRAS and laterality: In mutant KRAS, left-sided worse OS; in wild-type, left-sided better OS	Retrospective cohort	[32]
HER2	HERACLES	KRAS wild-type, HER2-positive mCRC (n = 27)	Trastuzumab + lapatinib active in HER2-positive mCRC	Phase II	[33]
NTRK	Drilon et al[37]	NTRK fusion-positive cancers (n = 55, including CRC)	Larotrectinib demonstrated high response rates across tumor types	Phase I/II	[37]

ctDNA: Circulating tumor DNA; MRD: Minimal residual disease; CRC: Colorectal cancer; mCRC: Metastatic colorectal cancer; RFS: Recurrence-free survival; HR: Hazard ratio; CI: Confidence interval; PFS: Progression-free survival; OS: Overall survival; AUC: Area under the curve; MSI: Microsatellite instability; FDR: False discovery rate; IL-2Ra: Interleukin-2 receptor alpha; LMR: Lymphocyte-to-monocyte ratio; CEA: Carcinoembryonic antigen; MMR: Mismatch repair; dMMR: Deficient mismatch repair; pCR: Pathological complete response; DFS: Disease-free survival; RCT: Randomized controlled trial; NCDB: National Cancer Database; HER2: Human epidermal growth factor receptor 2; NTRK: Neurotrophic tyrosine receptor kinase; IL-8: Interleukin-8; VEGF: Vascular endothelial growth factor; NSCLC: Non-small cell lung cancer; PD-1: Programmed cell death protein 1; SETD7: SET domain containing 7; KO: Knockout; mFOLFOX: Folinic acid, fluorouracil, and oxaliplatin; NTMT1: Pro-Lys N-methyltransferase 1; MAP3K14-AS1: MAP3K14 antisense RNA 1.

Table 2 Current status and evidence gaps for interleukin-8 as a colorectal cancer biomarker

Application	Current evidence level	Key findings	Evidence gaps	Future directions
Diagnosis	Moderate (case-control)	AUC 0.68; sensitivity/specificity 63.6%; independent predictor (OR 1.050)[16]; meta-analysis AUC 0.92[20]	Lack of validation in independent cohorts; optimal cut-off undefined; heterogeneity in assays	Multi-center validation; integration into multi-marker panels; harmonization of assays
Early detection	Limited	Performance varies widely; correlation with stage and metastasis[19,20,56]	Limited data on stage I-II; no screening cohort studies	Evaluation in asymptomatic populations; comparison to FIT and colonoscopy
Prognosis	Moderate (retrospective cohorts)	Correlates with OS and PFS in multiple studies[19,20,31,54-56]; independent of other inflammatory markers[55]; associated with liver metastasis[53,58]	Predominantly retrospective; heterogeneous cut-offs; limited validation in prospective cohorts	Large prospective cohorts with long-term follow-up; integration with ctDNA and CEA; define standardized cut-offs
Predictive (anti-angiogenic)	Preliminary	Baseline IL-8 associated with bevacizumab response in mCRC[58]	Retrospective analyses; heterogeneous populations	Prospective validation in biomarker-directed trials
Predictive (immunotherapy)	Exploratory	IL-8 dynamics correlate with immunotherapy response in melanoma/NSCLC[40]; relevance to CRC uncertain	Limited to melanoma/NSCLC; scarce CRC data, particularly in pMMR/MSS	Evaluation in CRC ICI trials, particularly pMMR/MSS; prospective biomarker-stratified studies
Monitoring	Not established	No longitudinal data; IL-8 lacks tumor specificity[21]; cfDNA shows earlier recurrence detection[65]	Cross-sectional design only; no comparison to ctDNA kinetics	Serial measurement studies; comparison to ctDNA dynamics; assessment of confounding factors
Therapeutic targeting	Preclinical	IL-8/CXCR2 inhibitors in development; KO models show reduced tumor growth[63,64]	No clinical trials in CRC	Phase I/II trials of IL-8/CXCR2-targeted agents; combination with chemotherapy and immunotherapy

IL-8: Interleukin-8; CRC: Colorectal cancer; AUC: Area under the curve; OR: Odds ratio; FIT: Fecal immunochemical test; CEA: Carcinoembryonic antigen; ctDNA: Circulating tumor DNA; cfDNA: Cell-free DNA; OS: Overall survival; PFS: Progression-free survival; mCRC: Metastatic colorectal cancer; NSCLC: Non-small cell lung cancer; pMMR: Proficient mismatch repair; MSS: Microsatellite stable; ICI: Immune checkpoint inhibitor; KO: Knockout; CXCR2: C-X-C motif chemokine receptor 2.

Table 3 Summary of established and emerging biomarkers in colorectal cancer

Biomarker	Prevalence	Clinical utility	Evidence level	Guideline recommendation	Ref.
MMR/MSI	15% stage II/III	Predictive for immunotherapy; prognostic in stage II	Level I	Universal testing recommended	[23-26]
KRAS	Approximately 40%	Prognostic (interacts with laterality); predictive in metastatic setting	Level II	Recommended for all patients	[28,29,32]
BRAF V600E	5%-10%	Prognostic in MSS; predictive in metastatic setting	Level II	Recommended for all patients	[30,31]
HER2	3%-5%	Predictive for anti-HER2 therapy in metastatic disease	Level II	Recommended in metastatic setting	[33-35]
NTRK	< 1%	Predictive for TRK inhibitors	Level II	Recommended (tumor-agnostic)	[36,37]
PIK3CA	10%-15%	Predictive for aspirin benefit (secondary prevention)	Level II	NCCN recommends testing	[38-41]
CEA	N/A	Prognostic; surveillance	Level II	Recommended perioperatively	[22,27]
ctDNA	N/A	MRD detection; treatment monitoring	Level II (prognostic)	Not routine outside trials; NCCN recommends in clinical trials	[42-46]
			Level I (emerging interventional)
IL-8	Variable	Investigational; prognostic potential	Level III	Not recommended	[10,19-21,31,35-37,59]
Methylation (SEPT9, etc.)	Variable	Screening; diagnostic	Level II	FDA-approved for screening	[48]
SETD7	Variable	Investigational; diagnostic potential	Level III	Not recommended	[49]
Metabolomic signatures	Variable	Investigational; diagnostic potential	Level III	Not recommended	[50]

MMR: Mismatch repair; MSI: Microsatellite instability; MSS: Microsatellite stable; KRAS: Kirsten rat sarcoma viral oncogene homolog; BRAF: V-Raf murine sarcoma viral oncogene homolog B; HER2: Human epidermal growth factor receptor 2; TRK: Tyrosine receptor kinase; NTRK: Neurotrophic tyrosine receptor kinase; PIK3CA: Phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha; CEA: Carcinoembryonic antigen; ctDNA: Circulating tumor DNA; MRD: Minimal residual disease; IL-8: Interleukin-8; SEPT9: Septin 9; SETD7: SET domain containing 7, histone lysine methyltransferase; FDA: Food and Drug Administration; NCCN: National Comprehensive Cancer Network; N/A: Not applicable.