Skip to main content

The exosome encapsulated microRNAs as circulating diagnostic marker for hepatocellular carcinoma with low alpha‐fetoprotein

Publication: Internatiional Journal of Cancer (2020) • Vol 147 • Page 2934-2947PMID: 32441313View on PubMedSuggested by: SOMA BANERJEE, IPGME&R

Purpose

This paper compared microRNA (miRNA, miR) profiles of tissue and exosomes, as well as levels of nine biochemical markers in liver specimens from patients diagnosed with hepatitis with or without cirrhosis or hepatocellular carcinoma (HCC) and in normal liver biopsy specimens to identify potential biomarkers for HCC development.

Conclusion

Levels of 31 miRNA were higher and 10 were lower in liver specimens from 3 patients with HCC relative to 3 normal liver controls. Further evaluation of the top 14 deregulated miRNAs between liver specimens from 11 control and 12 HCC patients confirmed higher levels of miR-10b-5p, miR-182-5p, miR-221-3p, miR-21-5p, miR-1269b, miR-574-5p, miR-183-5p and miR-216b-5p and lower levels of miR-375, miR-424-5p, miR-223-3p, miR-136-3p, miR-139-5p and miR-451a in HCC.  Ten of the 14 miRNAs deregulated in liver specimens were comparable in exosomes isolated from the plasma of controls and HCC patients, although levels of four miRNAs (miR-10b-5p, miR-221-3p, miR-21-5p, miR-223-3p) were higher in exosomes of patients with HCC than in patients with non-HCC hepatitis (chronic hepatitis or cirrhosis) or controls. Of the four miRNAs that differed, miR-221-3p showed the highest sensitivity (87%) for distinguishing specimens from HCC versus non-HCC individuals, although the  specificity was only 52%., Compared to the use of a single marker, higher sensitivity but lower specificity was observed when the four miRNAs were used combination (95% and 58%, respectively)  or when three miRNAs (without miR-21-5p) were used (85%and 50% , respectively).

Patients with HCC had higher levels of alanine transaminase (ALT), aspartate transaminase (AST), alkaline phosphatase (AP), bilirubin, international normalized ratio (INR), alpha fetoprotein (AFP), and lower albumin than controls or patients with chronic hepatitis; however, higher levels of ALT, AP, and INR and lower levels of AFP and albumin were observed in patients with HCC than those with liver cirrhosis. Low albumin had a sensitivity of 89% and specificity 52% for distinguishing HCC from non-HCC individuals, but when used in combination with miR-221 the sensitivity was 71% while the specificity increased to 85%.  When low albumin was used in combination with both miR-10b and miR-221 the miRNA panel had a sensitivity of 68% and a specificity was 85% in distinguishing between individuals with HCC from those that do not.

Studies (1)

Purpose

This study compared the miRNA profile of tumor and exosomes and levels of nine biochemical markers in patients in patients with hepatitis with or without cirrhosis or HCC and controls to identify potential biomarkers for HCC development. A total of 98 patients were selected for the study based on having chronic hepatitis (B or C), of which 25 had developed liver cirrhosis and 38 HCC. Normal liver biopsies were obtained from 11 patients with gall bladder cancer with normal liver architecture. Tissues were collected in RNAlater, stored for 24 h at 4°C before being frozen at -80°C. RNA was obtained from frozen liver pieces (5 mg) using Trizol. RNA integrity was evaluated by bioanalyzer before sequencing using the Truseq Small RNA Sample Prep Kit and the Truseq Ribo-Zero Gold Kit on an Illumina HiSeq2500. Blood was obtained from 19 individuals without a history of liver disease or alcohol consumption as well as from the hepatitis patients; blood was collected in tubes with and without an anticoagulant and plasma and serum were isolated (details not provided). Biochemical parameters were measured using Bayer clinical chemistry kits and AFP ELISA. Exosomes were isolated from plasma using an ExoEnrich instant exosome isolation kit. Exosome size distributions were evaluated by Nanoparticle tracker and transmission electron microscopy. Exosomes were confirmed by Western blot using antibodies against asialoglycoprotein receptor 2 (Asgr2) and GRP78, and triglyceride content was evaluated using the HDL-cholesterol assay. RNA was isolated from RNase-treated and untreated exosomes with Trizol and miRNA was quantified by real-time PCR.

Summary of Findings

Levels of 31 miRNA were higher and 10 were lower in liver specimens from 3 patients with HCC relative to 3 normal liver controls (log fold change ≥ 1.5 and P<0.05). Levels of miR-10b, miR-182 and miR-21 were also shown to be higher in HCC than non-HCC specimens in data from GEO and TCGA specimen cohorts. Further evaluation of the top 14 deregulated miRNA between liver specimens from 11 healthy and 12 HCC patients confirmed higher levels of miR-10b-5p, miR-182-5p, miR-221-3p, miR-21-5p, miR-1269b, miR-574-5p, miR-183-5p and miR-216b-5p and lower levels of miR-375, miR-424-5p, miR-223-3p, miR-136-3p, miR-139-5p and miR-451a by real-time PCR. Investigation of the same datasets identified lower levels of 142 mRNA and higher levels of 79 mRNA in HCC compared to control patients; pathways analysis suggested that these alterations were indicative of  enrichment for genes in carcinogenesis pathways, and decreases in the complement lectin pathway and glutamate metabolism pathways.

Isolated EVs were between 100 and 150 nm, positive for the exosome marker asialoglycoprotein2 (Asgr2) and negative for HDL and the endoplasmic reticulum marker GRP78. Ten of the 14 miRNAs deregulated in liver  specimens were comparable in exosomes isolated from the plasma of  controls and HCC patients, although levels of four miRNAs (miR-10b-5p, miR-221-3p, miR-21-5p, miR-223-3p) were higher in exosomes of patients with HCC than in patients with non-HCC hepatitis or controls  (P<0.0001, all), and in patients non-HCC than controls (P<0.05, all). Importantly, the same differences were observed in RNAse-treated exosomes. Only miR-221-3p displayed a sensitivity >70% (87%) for distinguishing between patients diagnosed with HCC and those that were not, although the specificity was only 52%. Compared to the use of a single marker, higher sensitivity but lower specificity was observed when four miRNAs (miR-10b-5p, miR-221-3p, miR-21-5p, miR-223-3p) or three miRNAs (miR-10b-5p, miR-221-3p, miR-223-3p) were used in combination, with  a specificity of 95% and 85%, respectively, and a sensitivity of 58% and 50%, respectively. For distinguishing patients with HCC from those with chronic hepatitis, miR-221-3p alone, all four miRNAs used in combination, and three miRNAs used in combination (miR-10b-5p, miR-221-3p, miR-223-3p) had a sensitivity of 87%, 74%, and 50%, respectively and a specificity of 63%, 86%, and 100%, respectively.  Only miR-21-5p could distinguish patients with HCC from those with cirrhosis, with a sensitivity of 74% and a specificity of 68%.

Patients with HCC had higher levels of ALT, AST, alkaline phosphatase (AP), bilirubin, INR, AFP and lower albumin than controls or patients with chronic hepatitis (P<0.05, all); however, higher levels of ALT, AP, and INR and lower levels of AFP and albumin were observed in patients with HCC than those with liver cirrhosis (P<0.05 all). Low albumin had a sensitivity of 89% and specificity 52% for distinguishing HCC from non-HCC individuals, but when used in combination with miR-221 the sensitivity was 71% while the specificity increased to 85%. When low albumin was used in combination with both miR-10b and miR-221 the miRNA panel had a sensitivity of 68% and a specificity was 85% in distinguishing between individuals with HCC from those that do not.

  • Biospecimens

    FluidBlood
    TissueLiver
  • Preservation Types

    RNAlaterFrozen
  • Diagnoses

    NeoplasticCarcinoma
    Hepatitis
    Cirrhosis
    Not specified
  • Platforms

    RNANext generation sequencing
    RNAReal-time qRT-PCR
    ProteinWestern blot
    ProteinClinical chemistry/auto analyzer
    ProteinELISA
    SteroidClinical chemistry/auto analyzer
    LipoproteinClinical chemistry/auto analyzer
  • Pre-analytical Factors

    ClassificationPre-Analytical FactorValue(s)
    PreaquisitionDiagnosis/ patient condition
    Cirrhosis
    Control
    Hepatitis
    Hepatocellular carcinoma
    Biospecimen AcquisitionBiospecimen location
    Exosomes
    Tumor