hepatocellular carcinoma (HCC) – accounting for 75-85% of primary liver cancers – is the 3rd leading cause of cancer death worldwide (2, 3).It occurs predominantly in patients with chronic liver disease (viral hepatitis B and C, alcohol, metabolic steatosis), which complicates early detection. Despite recommendations for semi-annual surveillance by ultrasound +/- alpha-fetoprotein (AFP) in cirrhotic patients, more than 60% of HCCs are still diagnosed at an intermediate or advanced stage, limiting access to curative treatments. AFP,a ancient serum biomarker,has moderate sensitivity (60-70%) and imperfect specificity for HCC diagnosis,particularly in cases of active liver inflammation.These limitations have favored the emergence of liquid biopsy strategies, for a non-invasive and dynamic assessment of tumor biology. Liquid biopsies are thus based on the analysis of several circulating analytes, reflecting different aspects of HCC.
Protein biomarkers
to begin with, protein biomarkers remain the most accessible in clinical practice. AFP retains prognostic value, but its isolated diagnostic interest is limited.AFP-L3%,a glycosylated isoform specific to tumor cells,and des-gamma-carboxy-prothrombin (DCP) have shown self-reliant value. Thier combination within the galad score (sex, age, AFP, AFP-L3, DCP) considerably improves HCC detection, with areas under the curve (AUC) reported between 0.78 and 0.94, including for small tumors and AFP-negative patients (1).
Genetic signals
Other elements of interest, circulating free DNA (cfDNA) and its tumor fraction (ctDNA) provide more specific details about the tumor. Beyond somatic mutations, DNA methylation alterations constitute an early signal of liver carcinogenesis. Several models based on differentially methylated regions have achieved high diagnostic performance, with sensitivities close to 90%, specificities greater than 90% and AUCs up to 0.99, including for stage I HCCs (1). these signatures outperform AFP and maintain their performance in AFP-negative forms. In addition, certain methylation signatures are associated with the risk of recurrence after curative treatment, suggesting a potential role in the detection of minimal residual disease.
Okay, I understand the constraints. I will perform an adversarial research check on the provided text, focusing on verifying claims and seeking contradictory information. I will not rewrite, paraphrase, or reuse any part of the provided text. My goal is to assess the reliability of the information presented.
Here’s my plan, broken down by claim, with a focus on recent (2024-2025) information where possible:
I. General Context & Statistics (Implied – GLOBOCAN 2022)
* Claim: The text implicitly relies on the global statistics of Hepatocellular Carcinoma (HCC) incidence and mortality.
* Verification: The text cites Bray et al. (2024) GLOBOCAN estimates. GLOBOCAN 2022 is a standard source for these statistics. I will confirm key figures (incidence, mortality) from GLOBOCAN 2022 via the IARC website (https://www.iarc.who.int/).
* Adversarial Check: I will search for more recent (2024) HCC incidence/mortality updates from other reputable sources (e.g., American cancer Society, National Cancer Institute) to see if there are important discrepancies.
II. Alpha-Fetoprotein (AFP) as a Biomarker (Not explicitly stated, but implied as a baseline)
* Claim: AFP is a commonly used, but imperfect, biomarker for HCC.
* Verification: This is widely accepted in the medical literature. I will confirm this via NCI and ACS resources.
* Adversarial Check: I will look for recent studies questioning the utility of AFP, or highlighting its limitations even further than implied.
III. Circulating Tumor Cells (CTCs)
* Claim: CTCs are a direct marker of metastatic potential, associated with increased risk of recurrence and decreased survival. Sensitivity of 0.95 reported in a meta-analysis. Phenotypic/transcriptomic analysis of CTCs can aid treatment selection. Implementation is early-stage, mostly retrospective. CTC analysis provides complementary information to AFP.
* Verification: I will search PubMed (and Google Scholar) for meta-analyses on CTCs in HCC, specifically looking for studies reporting sensitivity around 0.95. I will also look for studies on the prognostic value of ctcs.
* Adversarial Check: I will search for studies disputing the prognostic value of CTCs in HCC, or highlighting technical challenges in CTC detection that might explain variability in results. I will also look for studies showing no added benefit of CTC analysis beyond AFP.
IV. Exosomes and Circulating RNA (microRNA, long noncoding RNA)
* Claim: Exosomes carry microRNAs involved in angiogenesis, invasion, and resistance. miR-21,miR-221,miR-222 are overexpressed,miR-122 is frequently enough decreased.Panels combining RNA and protein biomarkers have >90% sensitivity.
* Verification: I will search PubMed for studies on the role of these specific microRNAs (miR-21, miR-221, miR-222, miR-122) in HCC, and their association with prognosis. I will also look for studies reporting sensitivity >90% for combined panels.
* Adversarial Check: I will search for studies questioning the specificity of these microRNAs for HCC (i.e.,are they also elevated in other cancers or liver diseases?),or highlighting challenges in standardizing exosome/RNA isolation and analysis. I will look for studies showing lower sensitivity for combined panels.
V.Multi-Analytic Approaches
* Claim: Multi-analytic approaches (proteins, cfDNA methylation, CTCs) are the most performant, with AUC > 0.90, and coudl become standard for personalized surveillance.
* Verification: I will search PubMed for studies comparing the performance of single biomarkers vs.multi-analytic approaches in HCC detection/prognosis, focusing on AUC values.
* Adversarial Check: I will look for studies arguing that the cost and complexity of multi-analytic approaches outweigh their benefits, or that simpler biomarkers remain sufficient for most clinical scenarios.
Tools I will use:
* PubMed (https://pubmed.ncbi.nlm.nih.gov/)
* Google Scholar (https://scholar.google.com/)
* IARC GLOBOCAN ([https://www.iarc.who.int/](https://www.iarc.who.int
