Diagnostic Utility of FAR1 Methylation Levels in Hepatocellular Carcinoma Patients Undergoing Liver Transplantation

Introduction

Hepatocellular carcinoma (HCC) is the sixth most prevalent malignancy worldwide, and the third leading cause of cancer-related mortality [1]. In Korea, it ranks sixth among all cancer types [2]. To facilitate early detection, individuals at elevated risk are enrolled in a national surveillance program incorporating ultrasonography and measurement of serum α-fetoprotein (AFP) levels. However, despite their widespread use, these modalities demonstrate limited sensitivity, up to 63% for early-stage HCC, highlighting the need for more sensitive diagnostic strategies [2,3].

Therefore, DNA methylation has emerged as a promising biomarker for cancer detection. DNA methylation is an epigenetic modification involving the covalent addition of a methyl group (−CH3) to cytosine residues within CpG dinucleotides, thereby affecting the transcriptional activity initiated at gene promoters. Aberrant methylation contributes to tumorigenesis by silencing the expression of mRNA encoding tumor suppressor genes in malignant cells, making it an attractive target for diagnostic applications. Because cancer-specific methylation patterns are often preserved throughout tumor progression, and DNA is structurally stable, these epigenetic signatures can be detected reliably in circulating tumor DNA (ctDNA) released into the bloodstream during cancer cell death, enabling minimally invasive, blood-based cancer diagnostics [4,5].

In this study, we focused on Fatty Acyl-CoA Reductase 1 (FAR1), a gene characterized by a notably high frequency of promoter methylation in HCC, as a potential candidate biomarker. The primary objective of this investigation was to elucidate whether site-specific DNA methylation within FAR1 – an epigenetic alteration strongly implicated in HCC tumorigenesis – could serve as a robust and clinically informative diagnostic marker (published Korean and European patents) [6,7]. Importantly, this work is the first attempt to systematically evaluate FAR1 methylation in ctDNA within the liver transplantation (LT) setting, thereby addressing an unmet need in pre- and post-transplant tumor assessment.

ctDNA was extracted from plasma and subjected to droplet digital polymerase chain reaction (ddPCR). The resulting quantitative methylation data were used to evaluate both the pretransplant diagnostic performance of FAR1 methylation for detecting HCC and its prognostic significance in predicting post-transplant recurrence, thus offering novel insights into the utility of FAR1 methylation as a next-generation biomarker in transplant oncology.

Material and Methods

STUDY DESIGN:

This retrospective clinical study analyzed pretransplant blood samples to evaluate the diagnostic accuracy of FAR1 methylation for HCC, as well as its prognostic value for post-transplant recurrence. The investigation comprised of 2 components. First, a diagnostic arm was used to assess whether pretransplant FAR1 methylation levels could identify HCC. Methylation levels in patients with pathologically-confirmed HCC in explanted liver specimens were measured, a diagnostic cut-off was established, and the accuracy for HCC detection was determined. Second, a prognostic arm was used to examine the association between pretransplant FAR1 methylation and the risk of HCC recurrence following LT.

SELECTION OF CIRRHOTIC-LIVER HCC STUDY SUBJECTS:

Adult recipients (≥18 years) who underwent LT at our institution between January 2018 and December 2020 were considered eligible if they survived for at least 1 month after transplant and had available frozen pretransplant blood samples. Patients were excluded if they had non-HCC primary liver malignancies such as intrahepatic cholangiocarcinoma, or if pathological examination revealed complete necrosis of HCC attributable to pretransplant therapy. The final HCC cohort comprised 23 patients with an even distribution of underlying liver disease etiologies.

SELECTION OF CIRRHOTIC-LIVER CONTROL SUBJECTS:

Adult patients who received LT at our institution between January 2018 and December 2020 were eligible for inclusion as cirrhotic-liver controls if they had no evidence of malignancy, including HCC, survived for at least 1 month after transplant, and had available frozen pretransplant blood samples. From this pool, 12 patients were selected after matching for underlying liver disease etiology, age, Model for End-Stage Liver Disease (MELD) score, and graft-to-recipient weight ratio (GRWR). These patients formed the non-HCC cirrhotic-liver control group.

SELECTION OF NORMAL-LIVER CONTROL SUBJECTS:

Among adults who donated a liver as a living donor at our institution between May and July 2015, 10 individuals who had undergone perioperative tumor marker assessment were randomly selected to serve as the normal-liver control group.

PATIENT FOLLOW-UP:

The HCC and cirrhotic-liver control groups were followed longitudinally to monitor post-transplant survival and HCC recurrence through a systematic review of medical records, as well as data provided by the National Health Insurance Service. Follow-up was continued until patient death or until July 2025.

POST-TRANSPLANT SURVEILLANCE AND TREATMENTS FOR HCC RECURRENCE:

After LT, the patients were monitored closely by conducting follow-up evaluations every 1–2 months during the first year and every 3–4 months thereafter. Detailed protocols for post-transplant surveillance and immunosuppressive therapy have been extensively documented elsewhere [8–10]. The management of recurrent HCC adhered to established clinical guidelines.

PATIENT CONSENT AND ETHICS REVIEW COMMITTEE APPROVAL:

Before LT, all participants provided consent for the collection of blood samples and clinical data. The Institutional Review Board of our institution approved the study protocol (IRB No. 2014-0898 and 2021-1627). All the procedures were performed in accordance with the ethical principles of the Declaration of Helsinki (2013 revision).

STATISTICAL ANALYSIS:

Continuous variables are presented as mean±standard deviation, or as median (range) along with 95% confidence interval (95% CI). Sensitivity, the primary metric for evaluating the diagnostic performance of the FAR1 methylation assay, was defined as the proportion of HCC samples correctly identified as positive. Specificity, the proportion of non-HCC control samples classified correctly as negative, was also assessed to determine diagnostic accuracy. The positive likelihood ratio was calculated as sensitivity divided by (1 – specificity), while the negative likelihood ratio was defined as (1 – sensitivity) divided by specificity. Receiver operating characteristic (ROC) curve analysis was performed to evaluate the trade-off between the sensitivity and specificity across different thresholds. The area under the ROC curve (AUC) was used as a measure of the overall diagnostic accuracy. The optimal cut-off value was determined using Youden’s Index (J=sensitivity+specificity – 1), which identified the threshold maximizing combined sensitivity and specificity. Kaplan-Meier survival analysis was used to estimate HCC recurrence and patient survival, with group comparisons made using the log-rank test. Statistical significance was defined as P<0.05. All analyses were conducted using SPSS version 22 (IBM, New York, NY, USA) and MedCalc version 23.2.1 (MedCalc Software Ltd, Ostend, Belgium).

EPIGENETIC ROLE AND QUANTIFICATION OF FAR1 GENE-SPECIFIC CPG METHYLATION:

DNA methylation refers to the covalent addition of a methyl group (−CH₃) to a DNA base, most commonly to the fifth carbon of cytosine within CpG dinucleotides, generating 5-methylcytosine (5-mC). In the context of this study, methylation of the FAR1 gene specifically denotes the presence or absence of methylation at cytosines of defined CpG sites within the FAR1 locus. When these CpG sites are methylated, the binding of transcription factors is hindered, resulting in suppression of FAR1 gene expression. Conversely, unmethylation or hypomethylation at these sites facilitates transcription factor binding and consequently enhances FAR1 expression.

In mammalian genomic DNA, in addition to the canonical nucleotides A, C, G, and T, 5-mC represents a “fifth base”, generated by methylation of cytosine. CpG dinucleotides (5′-mCG-3′) are the principal substrates for 5-mC formation, and their methylation plays a critical role in repressing the expression of Alu elements, transposons, and other repetitive genomic sequences. Moreover, 5-mC within CpG sites is prone to spontaneous deamination to thymine (T), making CpG sites hotspots for epigenetic alterations in mammalian cells.

The phrase “measurement of methylation level” refers to the quantification of methylation at CpG sites within the FAR1 gene. These levels can be assessed using bisulfite-based or bisulfite-free detection methods, including methylation-specific PCR as a widely used technique. CpG sites of the FAR1 gene are located in the promoter, open reading frame, or terminator regions of the locus, corresponding to chromosome 11 (positions 13,689,588–13,690,724), as reported in Korean and European patent publications [6,7].

TECHNICAL OVERVIEW OF THE DDPCR USED TO DETECT FAR1 METHYLATION:

Pretransplant plasma-derived DNA was subjected to bisulfite sequencing followed by quantitative analysis of site-specific methylation within the FAR1 gene using ddPCR. The methylation status at the targeted FAR1 loci was assessed qualitatively using the QX200™ Droplet Digital PCR System (Bio-Rad Laboratories, Hercules, CA, USA). The assay comprises 3 key steps.

DNA EXTRACTION:

Following the collection of 5 mL of whole blood, 2 mL of plasma was isolated and circulating cell-free DNA (cfDNA) was extracted using a dedicated cfDNA extraction kit.

BISULFITE CONVERSION:

Extracted DNA underwent bisulfite treatment to convert unmethylated cytosines to uracil while leaving methylated cytosines unaltered. This differential conversion enables the discrimination of methylated sequences from unmethylated sequences based on changes in the base sequence.

DDPCR ANALYSIS:

Bisulfite-converted DNA was analyzed by ddPCR to quantify methylation levels at specific FAR1 gene sites.

SAMPLE PREPARATION AND STORAGE:

Briefly, 5 mL of blood per subject was collected into anticoagulant-treated tubes, after which the samples were allowed to stand for 4 h to facilitate plasma separation. A minimum 2 mL of plasma was required for testing. Blood samples were transferred to 15 mL tubes and centrifuged at 2000×g for 15 min at 4°C. Plasma was collected carefully to avoid buffy coat contamination and the buffy coat was subsequently removed by additional centrifugation. The resulting plasma was aliquoted into cryovials, sealed securely, and stored at −70°C or below, with proper recording for sample identification.

EXTRACTION OF CFDNA:

Circulating cfDNA was isolated from 2 mL of plasma using the Cobas^® cfDNA Sample Preparation Kit (Roche Diagnostics, Korea), yielding a final elution volume of 50 μL. For quality control, 10 μL of a positive control solution was mixed with 2 mL of 1X phosphate-buffered saline to prepare the positive control sample, while phosphate-buffered saline alone served as the negative control or no-template control.

BISULFITE CONVERSION:

Bisulfite treatment was performed using the EZ DNA Methylation-Lightning Kit (Zymo Research). Briefly, 50 μL of extracted DNA served as the input, and bisulfite-converted DNA was eluted in a final volume of 11 μL. Both the positive and no-template control samples underwent bisulfite conversion for quality assurance.

DDPCR ANALYSIS:

Methylation analysis was conducted using the QX200™ Droplet Digital PCR System.

Results

CLINICOPATHOLOGICAL CHARACTERISTICS OF THE HCC STUDY AND THE 2 CONTROL GROUPS:

The clinicopathological features of 25 patients in the HCC study cohort and 12 individuals in the non-HCC cirrhotic-liver control group are detailed in Table 1. The non-HCC cirrhotic-liver control group was matched to the HCC study group based on the underlying liver disease, age, and GRWR, resulting in no significant differences between the groups in terms of age, sex, liver disease etiology, MELD score, or ABO blood group incompatibility. All the patients in both cohorts received right liver grafts. Pretransplant serum levels of AFP and protein induced by vitamin K absence or antagonist-II (PIVKA-II) were significantly higher in the HCC group than in the non-HCC cirrhotic-liver control group (Table 2).

The mean and median maximum diameters of HCC were 2.4±1.8 cm and 1.5 cm (1.1–9.1 cm), respectively. The number of tumors was 1 in 13 (56.5%) patients, and multiple in 12 (43.5%) patients. The Milan Criteria were satisfied in 17 patients (73.9%). According to the 8th edition of the American Joint Committee on Cancer (AJCC) stage system for HCC, 11 patients (47.8%) had stage IA disease, 1 (4.3%) had stage IB, 9 (39.1%) had stage II, and 2 (8.7%) had stage IIIA (Table 1).

Among the 10 normal-liver donors, the mean age was 32.2±6.8 years, with males comprising 70% (n=7). All donors provided right liver grafts.

EXPRESSION PATTERNS OF FAR1 METHYLATION, AFP, AND PIVKA-II:

Figure 1A illustrates the distribution of FAR1 methylation levels in the HCC cohort, non-HCC cirrhotic-liver controls, and normal-liver controls. In the HCC group, FAR1 methylation exhibited a median copy number of 4 (range: 0–330) with a mean±SD of 31.6±74.5 copies. By contrast, the non-HCC cirrhotic-liver control group showed a median of zero copies (range: 0–12) and a mean of 1.5±3.5 copies, while the normal-liver control group demonstrated minimal methylation with a median of zero copies (range: 0–1.2) and a mean of 0.1±0.4 copies (p<0.001; Table 2).

Figure 1B and 1C show the serum expression profiles of AFP and PIVKA-II, respectively, across the same groups. A comparative summary of FAR1 methylation, AFP, and PIVKA-II expression levels among the HCC and control cohorts is present in Table 2.

PRETRANSPLANT DIAGNOSIS OF HCC USING FAR1 METHYLATION, AFP, AND PIVKA-II:

The results of the ROC curve analysis are presented in Figure 2 and summarized in Table 3. FAR1 methylation had an AUC of 0.832 (95% CI: 0.691–0.927, P<0.001). The optimal cut-off, determined by Youden’s Index (J=0.599), was 1 copy, yielding a sensitivity of 82.6% and specificity of 77.3%, outperforming the conventional cut-off of 2 copies, which showed a sensitivity and specificity of 56.5% and 86.4%, respectively.

The ROC AUC for AFP was 0.737 (95% CI: 0.585–0.857, P=0.002), with a reference cut-off of 7.5 ng/mL yielding 43.5% sensitivity and 86.4% specificity. PIVKA-II exhibited an AUC of 0.732 (95% CI: 0.578–0.852, P=0.003), with a sensitivity and specificity of 34.8% and 90.9%, respectively, at the reference cut-off of 40 mAU/mL (Table 3).

When combined criteria (FAR1 >1 copy, AFP >7.5 ng/mL, or PIVKA-II >40 mAU/mL) were applied, only 2 of 18 patients (11.1%) who did not meet any criteria were found to have HCC on explant pathology. Conversely, 21 of the 27 patients (77.8%) who met at least 1 criterion were diagnosed with HCC. This combined approach achieved a sensitivity of 91.3% (95% CI: 72.0–98.9%), a specificity of 72.7% (95% CI: 49.8–89.3%), a positive likelihood ratio of 3.35 (95% CI: 1.67–6.70), a negative likelihood ratio of 0.12 (95% CI: 0.03–0.46), a positive predictive value of 77.8% (95% CI: 63.6–87.5%), a negative predictive value of 88.9% (95% CI: 67.5–96.9%), and an overall accuracy of 82.2% (95% CI: 67.9–92.0%) assuming a disease prevalence of 50%. These metrics surpassed those of FAR1 methylation, AFP, or PIVKA-II alone, underscoring the superior predictive power of the combined biomarker panel for HCC diagnosis in explant livers.

HCC RECURRENCE AND SURVIVAL ANALYSIS IN THE HCC STUDY GROUP:

Over a mean follow-up period of 62.5±19.4 months, 2 patients experienced HCC recurrence, corresponding to cumulative tumor recurrence rates of 8.7% at 1, 3, and 5 years. Applying a FAR1 methylation cut-off of one copy, no recurrences were observed among 11 patients with FAR1 levels ≤1 copy, whereas 2 recurrences occurred in 12 patients with FAR1 >1 copy, resulting in cumulative recurrence rates of 10.5% at all-time points (p=0.511; Figure 3A). FAR1 methylation levels in patients with HCC recurrence were 7.8 copies and 136 copies, respectively.

During the follow-up period of up to 89 months, 3 patients died: 2 from HCC recurrence and 1 from pneumonia. All patients with FAR1 methylation ≤1 copy remain alive to date, while 3 patients with >1 copy succumbed, yielding overall survival rates of 94.7% at 1 year and 89.5% at both 3 and 5 years (P=0.426; Figure 3B).

Discussion

DNA methylation is increasingly recognized as a promising biomarker for early cancer detection via liquid biopsy. CpG methylation (ie, the addition of a methyl group to cytosine residues within CpG dinucleotides) is an early and persistent regulatory mechanism for gene expression that remains stable across cancer stages. This inherent structural stability enables reliable detection through liquid biopsy, making methylation profiling of specific genes a valuable method for early cancer diagnosis [4,5].

In this study, we focused on methylation of FAR1, a gene specific for HCC. Initially, we conducted bioinformatics analysis to identify candidate biomarkers from 3 high-throughput microarray datasets [6,7]. Although a comprehensive meta-analysis published in 2016 highlighted several DNA methylation markers associated with HCC, including p16, RASSF1A, GSTP1, CDH1, and others, FAR1 was notably absent [11].

FAR1 encodes fatty acyl-CoA reductase 1, an enzyme critical for lipid metabolism, particularly for the biosynthesis of ether lipids and wax esters through the reduction of fatty acids to fatty alcohols. In humans, mutations in FAR1 have been linked to peroxisomal disorders characterized by intellectual disability, epilepsy, and cataracts. Furthermore, the Human Protein Atlas states that FAR1 can be used as a prognostic marker for ovarian serous cystadenocarcinoma [12].

Preliminary validation studies based on plasma samples demonstrated the diagnostic potential of FAR1 methylation in the context of HCC, confirming its sensitivity and specificity compared to normal controls. Building upon these unpublished pilot data, the present investigation aimed to evaluate the role of methylation in HCC diagnosis by detecting methylation at CpG sites within 1 or more genes selected from a panel comprising FAM110A, FAR1, VIM, LDHB, LIPE, INAFM1, ATL1, CELF6, MTHFD2, PAK1, NXPE3, SLC25A36, and VANGL2 [6,7].

The results from blood samples from adult LT candidates and donors demonstrated that FAR1 methylation had a superior diagnostic performance for detecting HCC in patients awaiting transplantation. FAR1 methylation exhibited an ROC AUC of 0.832, with a Youden’s Index of 0.599 at a cut-off of 1 copy, corresponding to a sensitivity of 82.6% and specificity of 77.3%. Notably, the diagnostic accuracy of FAR1 methylation surpassed that of the traditional serological tumor markers AFP and PIVKA-II. Thus, the application of a 1-copy cut-off for FAR1 methylation appears to offer better diagnostic performance than conventional markers. These findings warrant further clinical validation to establish the utility of FAR1 methylation in routine diagnostic practice.

The observed distribution of FAR1 methylation levels suggests that hypermethylation of FAR1 is a tumor-associated epigenetic alteration that is largely absent in non-malignant liver tissue. FAR1 methylation was markedly elevated in HCC tissues, with substantially higher median and mean copy numbers compared with both non-HCC cirrhotic-liver controls and normal-liver controls. This pronounced difference indicates that FAR1 methylation is not merely a byproduct of chronic liver injury or fibrosis, as cirrhotic tissues without HCC exhibited only minimal methylation. Instead, the data support the interpretation that FAR1 hypermethylation arises specifically during hepatocarcinogenesis. Biologically, the accumulation of FAR1 methylation in HCC may reflect epigenetic silencing of FAR1, potentially disrupting its normal functional role in fatty alcohol and lipid metabolism. Such silencing could contribute to metabolic reprogramming in malignant hepatocytes, a hallmark of cancer development. Moreover, the negligible methylation levels in normal and cirrhotic non-tumor liver tissues underscore the specificity of this epigenetic event, suggesting that FAR1 methylation marks a tumor-specific pathway rather than generalized liver pathology. Taken together, these findings imply that FAR1 hypermethylation is a distinct, HCC-associated epigenetic signature, supporting its biological relevance as a potential biomarker for hepatocarcinogenesis.

In patients with advanced liver cirrhosis awaiting LT, the diagnostic performance of the commonly used biomarkers AFP and PIVKA-II is lower than that in the general patient population with preserved liver function. A previous single-center study involving 2074 adult LT recipients categorized patients into HCC (n=970, 46.8%) and non-HCC (n=1104, 53.2%) groups over a decade. The MELD scores were stratified as follows: <10 (n=464), 10–14 (n=632), 15–19 (n=355), 20–29 (n=340), and ≥30 (n=283). The median pretransplant AFP and PIVKA-II levels were 11.3 ng/mL and 26 mAU/mL, respectively, in the HCC cohort versus 4.2 ng/mL and 22 mAU/mL, respectively, in the non-HCC cohort. ROC curve analysis revealed an AFP AUC of 0.693, with both sensitivity and specificity of 64.5% using a cut-off of 6.8 ng/mL. Notably, the diagnostic accuracy of AFP was inversely correlated with increasing MELD score. PIVKA-II demonstrated a lower ROC AUC of 0.546, with a sensitivity of 53.1% and specificity of 51.8% at a cut-off of 25 mAU/mL. After excluding patients with MELD scores ≥30, the AUC was further reduced to <0.6. These findings indicate that AFP and PIVKA-II have limited sensitivity and specificity in patients with advanced liver disease characterized by high MELD scores [13].

A previous multicenter study involving 3273 adult LT recipients stratified patients according to the presence of HCC and underlying liver disease. The median serum AFP and PIVKA-II levels in the HCC group were 6.3 ng/mL and 29 mAU/mL, respectively, and 3.3 ng/mL and 35 mAU/mL, respectively, in the non-HCC group (P<0.001 for AFP; P=0.037 for PIVKA-II). Subgroup analyses revealed that among patients with hepatitis B virus infection, median AFP and PIVKA-II levels were significantly elevated in those with HCC cohort (6.0 ng/mL and 26 mAU/Ml, respectively) than in those without HCC (3.2 ng/mL and 21 mAU/Ml, respectively; both P<0.001). In hepatitis C virus-infected patients, those with HCC group had median AFP levels of 10.7 ng/mL and PIVKA-II levels of 37 mAU/mL versus 2.6 ng/mL and 21 mAU/mL, respectively, in those without HCC (P<0.001 and P=0.117, respectively). Among individuals with alcoholic liver disease, AFP and PIVKA-II levels were 5.2 ng/mL and 61 mAU/mL, respectively, in the HCC group and 6.4 ng/mL and 75 mAU/mL, respectively, in the non-HCC group (P<0.001 and P=0.419). For patients with other liver disease etiologies, the median AFP and PIVKA-II values were 7.1 ng/mL and 32 mAU/mL, respectively, in the HCC group and 3.3 ng/mL and 28 mAU/mL, respectively, in controls (P<0.001 and P=0.822). These findings underscore the considerable variability in serum AFP and PIVKA-II levels among patients with end-stage liver disease prior to transplantation, thereby limiting their reliability as diagnostic biomarkers for HCC in this population [14].

In the present study, the combined use of FAR1 methylation (>1 copy), AFP (>7.5 ng/mL), or PIVKA-II (>40 mAU/mL) demonstrated superior diagnostic accuracy for predicting HCC in explant livers compared to individual markers. This integrated approach achieved a sensitivity of 91.3%, specificity of 72.7%, positive likelihood ratio of 3.35, negative likelihood ratio of 0.12, positive predictive value of 77.8%, negative predictive value of 88.9%, and overall diagnostic accuracy of 82.2%. These robust diagnostic metrics suggest significant potential for clinical implementation.

Currently, there are no FDA-approved assays for early detection of HCC. Epigenomics, a U.S.-based company, has developed an HCC diagnostic test known as the HCCBloodTest, and reported findings from a prospective clinical study (based on a training and testing design) that evaluated DNA methylation biomarkers using both the HCCBloodTest and a next-generation sequencing (NGS)-based methylation panel. The NGS panel algorithm was trained on a cohort of 41 patients with HCC and 46 cirrhotic non-HCC controls. Plasma samples were obtained from patients with Child-Pugh class A or B cirrhosis, with or without early-stage HCC (Barcelona Clinic Liver Cancer stages 0, A, or B). Subsequently, a blinded validation set was analyzed using a predefined algorithm. The HCCBloodTest demonstrated a sensitivity of 76.7% and specificity of 64.1%, while the NGS panel showed a sensitivity of 57% and specificity of 97%. A post hoc analysis combining the NGS methylation panel with AFP at a 20 ng/mL cut-off improved the sensitivity to 68%, while maintaining 97% specificity, achieving an AUC of 0.9. These results indicate that cell-free plasma DNA methylation biomarkers are a promising novel approach for HCC surveillance [15].

Since 2021, research on DNA methylation biomarkers for cancer diagnosis has increased. A Chinese study evaluated 97 patients with HCC, 80 healthy controls, and 46 individuals with chronic hepatitis B or C virus infection. Circulating cfDNA levels were significantly higher in the HCC group than in healthy controls. Using a methylation ratio cut-off of 15.7%, the assay demonstrated a sensitivity of 78.6% and specificity of 89.4% for HCC. The overall diagnostic accuracy was 85.3%, with positive and negative predictive values of 81.9% and 87.2%, respectively. Additionally, the positive likelihood ratio was 7.40, and the negative likelihood ratio was 0.24, underscoring the potential utility of methylation-based cfDNA analysis as a minimally invasive liquid biopsy approach for HCC diagnosis [16].

A Korean multi-institutional study published in 2023 identified HCC-specific methylation biomarkers, and developed a PCR-based assay to detect these biomarkers in circulating cfDNA. A comprehensive analysis of large-scale methylome datasets identified Ring Finger Protein 135 (RNF135) and Lactate Dehydrogenase B (LDHB) as universal HCC methylation markers, with methylation patterns absent in other tissue types. These markers were incorporated into a Methylation Sensitive High-Resolution Melting (MS-HRM) assay, which was evaluated using cfDNA samples from healthy individuals, at-risk patients, and patients with HCC. The combined MS-HRM analysis of RNF135 and LDHB achieved a sensitivity of 57% for HCC detection, surpassing the 45% sensitivity observed for AFP testing. Integration of methylation profiling with AFP measurements increased the detection sensitivity to 70%. These findings support the use of cfDNA methylation analysis as a promising liquid biopsy approach for HCC diagnosis [17].

Beyond its diagnostic utility, we also evaluated FAR1 expression for its potential ability to predict the post-transplant recurrence of HCC. Typical established prognostic indicators for HCC recurrence include tumor size and number, serum biomarkers such as AFP and PIVKA-II, and fluorodeoxyglucose positron emission tomography (FDG-PET) findings. Composite risk models such as the ADV (AFP, PIVKA-II, and tumor volume) score and SNAPP (tumor size and number, AFP, PIVKA-II, and positron emission tomography) score, which integrate these factors, have been validated [18–23]. However, prognostic assessment based on FAR1 expression in the current study was limited by the small number of recurrence events (n=2) and death of patients (n=3) within the HCC cohort. Nonetheless, these preliminary observations suggest a potential link between FAR1 expression and tumor biology, warranting further investigation. To date, comprehensive oncological studies on FAR1 have been lacking. Interestingly, methylation of PAK1 during hepatocarcinogenesis may operate through analogous mechanisms; in vitro studies have shown that PAK1 facilitates HCC progression by upregulating Snail and inducing epithelial-mesenchymal transition, whereas in vivo studies showed that PAK1 knockdown increased apoptosis and suppressed tumor growth [24,25].

This study has several limitations that warrant consideration. First, the small sample size of the HCC cohort, together with the limited numbers in both control groups, restricted the robustness of our assessments regarding the diagnostic performance for HCC and the prognostic value for post-transplant recurrence. Second, the plasma specimens were stored at ultra-low temperatures for 4–8 years, introducing the potential for gradual DNA degradation despite optimal storage conditions. Future investigations incorporating larger, prospectively enrolled cohorts and rigorously standardized protocols for specimen collection and storage will be essential to validate and extend the present findings.

Conclusions

The data presented herein indicate that FAR1 methylation may have potential as a biomarker for identifying HCC in patients with end-stage liver disease awaiting LT. In addition, lower FAR1 methylation levels may be associated with a reduced likelihood of post-transplant HCC recurrence, although this observation requires further confirmation. Overall, larger and well-designed studies will be necessary to clarify the clinical relevance and utility of FAR1 methylation in this setting.