Diagnostic performance of Liver FibraChek Dx©, a blood-based test for the non-invasive detection of liver cirrhosis and cancer

Abstract

BACKGROUND

Metabolic dysfunction-associated steatotic liver disease (MASLD), hepatic fibrosis, and cirrhosis are major risk factors for hepatocellular carcinoma (HCC), yet current blood-based diagnostic assays lack sufficient accuracy for routine clinical use. Identifying a non-invasive molecular signature that accurately detects liver disease could improve early diagnosis and monitoring. We hypothesized that the Liver FibraChek Dx^© serum assay could discriminate MASLD and HCC from healthy controls using a multiplex biomarker-based algorithm.

AIM

To evaluate the diagnostic performance of the Liver FibraChek Dx^© assay for detecting MASLD and HCC.

METHODS

This was a prospective, single-center study conducted in a United States tertiary care setting. Serum samples were collected from 45 participants (14 MASLD, 19 HCC, 12 healthy controls) with liver histology confirmed by biopsy. The Liver FibraChek Dx^© algorithm integrates weighted values of aspartate aminotransferase, alanine aminotransferase, taurocholic acid, L-tyrosine, platelet count, and patient age to generate a risk score. Wilcoxon rank sum tests were used to assess associations with histologic diagnosis, and receiver operating characteristic (ROC) curves quantified diagnostic performance.

RESULTS

Liver FibraChek Dx^© risk scores were significantly elevated in MASLD and HCC compared to controls (median: 6.92 ± 3.86 vs 3.61 ± 1.67, P < 0.001). The area under the ROC curve was 0.890 (95%CI: 0.776–1.000) for distinguishing diseased from healthy individuals. Sensitivity was 93.9%, specificity 75.0%, positive predictive value 91.1%, negative predictive value 81.8%, and overall accuracy 88.9%.

CONCLUSION

The Liver FibraChek Dx^© assay accurately detects liver disease and shows promise as a non-invasive tool for diagnosing and monitoring MASLD and HCC.

Key Words: Biomarkers; Metabolic dysfunction-associated steatotic liver disease; Cirrhosis; Multiplex; Metabolomics; Peripheral blood; Hepatocellular carcinoma; Fibrosis; High pressure liquid chromatography

Core Tip: This retrospective observational study assessed the diagnostic accuracy of Liver FibraChek Dx^©, a non-invasive serum assay for detecting metabolic dysfunction-associated steatotic liver disease (MASLD) and hepatocellular carcinoma (HCC). The test algorithm integrates five serum biomarkers and age to produce a risk score. Among 45 participants with biopsy-confirmed diagnoses (MASLD, HCC, or healthy), Liver FibraChek Dx^© achieved an area under the receiver operating characteristic curve of 0.890, with 93.9% sensitivity and 88.9% accuracy. Risk scores significantly distinguished individuals with liver disease from healthy controls. These findings support the potential clinical utility of Liver FibraChek Dx^© in diagnosing and monitoring liver pathogenesis via blood-based testing.

INTRODUCTION

The incidence of hepatocellular carcinoma (HCC) has almost tripled since the early 1980s in the United States where it is one of the fastest rising causes of cancer-related deaths[1]. HCC is the most common type of primary liver cancer occurring most often in people with chronic liver diseases (CLD), such as advanced liver fibrosis and cirrhosis. CLD is considered a public health problem due to its high prevalence. It is estimated that 1.5 billion people worldwide are affected by CLD[2]. Frequent causes of CLD include hepatitis B and C, followed by alcoholic hepatitis, and metabolic dysfunction-associated steatohepatitis or metabolic dysfunction-associated steatotic liver disease (MASLD). When not diagnosed and treated in a timely manner, MASLD may progress to irreversible liver damage. Such damage to the liver alters the functional and regenerative capacity of hepatocytes, thereby causing hepatic functional loss and irreversible cirrhosis. Recent evidence demonstrates that successfully treating the cause of early-to-moderate hepatic fibrosis can stop or even reverse the disease processes[3,4]. For this reason, detecting hepatic fibrosis at an early stage is crucial.

Liver biopsy remains the gold standard approach for assessing hepatic fibrosis in clinical practice[5,6]; however, liver biopsy is an invasive procedure associated with significant cost and morbidity. Thus, there is an urgent clinical need for the development of blood-based tests that can accurately detect MASLD. Liver FibraChek Dx^© is a non-invasive blood-based test that was developed in China where it is undergoing extensive validation. The assay simultaneously evaluates serum levels of L-tyrosine (Tyr) and taurocholic acid (TCA). A panel of four predictive metabolite markers was selected based on a previous study of serum metabolites of participants with/without liver fibrosis[7]. To select markers, two machine learning methods, least absolute shrinkage and selection operator and random forest, were utilized in univariate analysis against the differential metabolites with P < 0.001. Tyr and TCA, along with alanine aminotransferase (ALT), aspartate aminotransferase (AST), platelet count, and age were the best classifiers (unpublished). We developed an enhanced algorithm that incorporates serum levels of the five biomarkers with patient age to output a unique liver disease risk score. In this study, we tested the performance of the Liver FibraChek Dx^© assay, supplemented with three additional biomarker concentrations and an age-adjusted algorithm, for non-invasively detecting MASLD and HCC in a United States population.

MATERIALS AND METHODS

Subjects and blood samples

This was a retrospective observational diagnostic accuracy study designed to assess the clinical validity of the Liver FibraChek Dx^© assay and associated algorithm in identifying liver disease risk among individuals with histologically confirmed MASLD or HCC, compared to age-matched healthy controls. The study included 14 individuals with documented MASLD, 19 individuals with histologically confirmed HCC and 12 healthy age-matched controls. To meet the diagnostic criteria for MASLD, participants required evidence of hepatic steatosis and ≥ 1 marker of metabolic dysregulation (e.g., elevated triglycerides, low high-density lipoprotein cholesterol). The blood samples from MASLD and HCC patients were drawn before they began treatment. Blood samples from controls were drawn when they visited clinics for annual checkup. All subjects were fasting when the samples were collected. Data were reported according to Strengthening the Reporting of Observational studies in Epidemiology criteria for observational studies. Exclusion criteria were as follows: (1) Age > 89-years; (2) History of liver disease not histologically documented; and (3) Liver disease (fibrosis or cancer) currently being treated. Overnight (12 hours)-fasted peripheral blood samples were collected in a clot activator tube, allowed to stand at room temperature for 30-60 minutes, then centrifuged at 1000 ×g for 10 minutes. Serum supernatants were collected and frozen at -80 °C in aliquots to limit the number of freeze-thaw cycles before metabolomic testing. This study received Institutional Review Board approval and a waiver of consent to use previously banked de-identified serum samples and associated clinical data (e.g., AST, ALT, platelet count, age, and histological report from liver biopsy) from the Cedars-Sinai Medical Center Institutional Review Board, Los Angeles, CA, United States (No. 00002172). Study performance complied with the tenets of the Declaration of Helsinki. All subjects provided written informed consent for blood draws, biopsy, and the use of anonymized data for research purposes. All MASLD patients and HCC patients had documented disease confirmed by histological examination of biopsied liver tissue. Pertinent information on clinical presentation, histology, staging, and grading is provided in Table 1.

Table 1 Demographics and clinicopathological characteristics of study cohort, n (%).

Parameter	Control (n = 12)	Metabolic dysfunction-associated steatotic liver disease (n = 14)	Hepatocellular carcinoma (n = 19)
Age (years) (mean ± SD)	60.9 ± 13.3	59.9 ± 8.0	74.4 ± 7.6
Sex
Male	5 (41.7)	9 (64.3)	12 (63.2)
Female	7 (58.3)	5 (35.7)	7 (36.8)
Race/ethnicity
African-American/Black	9 (75.0)	4 (28.6)	8 (42.1)
Asian	2 (16.7)	7 (50.0)	3 (15.8)
White/Caucasian	1 (8.3)	2 (14.3)	8 (42.1)
Other race	0 (0.0)	1 (7.1)	0 (0.0)
Hispanic	0 (0.0)	0 (0.0)	0 (0.0)
Fibrosis stage
S0-2		9 (64.3)
S3-4		5 (35.7)
Tumor stage
TX T0-1			13 (68.4)
T2-3			5 (26.3)
Unknown			1 (5.3)

Liver FibraChek Dx^© assay

The Liver FibraChek Dx^© assay is manufactured by Human Metabolomics Institute (HMI; Shenzhen, China) and is an ultra-high pressure liquid chromatography (UHPLC)/tandem mass spectroscopy-based assay that measures TCA and Tyr. The kit contains mobile-phase buffer components, standards, internal controls, sample buffers, and 96-well assay plates and covers. Serum samples were defrosted at refrigerator temperature (2-8 °C) and kept on ice until use. The standard curve reagents (n = 6; TCA range 25-618 ng/mL, limit of detection 10 ng/mL; Tyr range 45-830 μmol/L, limit of detection 5 μmol/L) and internal controls (low, medium, and high) are all lyophilized powders and were reconstituted according to manufacturer instructions. Ten μL of each standard was pipetted into appropriate wells of a 96-well V-bottom plate, with deionized water serving as vehicle bank. Next, 10 μL of each serum sample or control were pipetted into the other wells. Internal standard solutions (190 μL/well) were added to appropriate wells, and the microtiter plate was covered with aluminum foil and agitated at 650 rpm for 20 minutes at room temperature (18-22 °C). Next, 60 μL of the supernatant from each well was transferred into a new V-bottom 96-well plate, followed by the addition of 60 μL kit dilution solution into each well. The plate was covered with an adhesive seal, agitated at 650 rpm for 1 minute, then placed into the autosampler of a UHPLC tandem mass spectrometry system (Vanquish Flex UPHLC, Thermo Scientific, UCLA Metabolomics Center, Los Angeles, CA, United States) for testing. UHPLC was performed using an ACQUITY UPLC CSH Fluoro-Phenyl column (50 mm × 2.1 mm, 1.7 μm at 40 °C) and an injection volume of 5 μL, with the following settings for mass spectrometry: (1) Desolvation gas temperature: 550 °C; and (2) Desolvation gas flow: 850 L/hour. Pathologically elevated metabolite levels were Tyr > 118 μmol and/or TCA > 27 ng/mL. The standard curves for both Tyr and TCA were generated by linear regression and demonstrated the assay’s linear range using these settings (Supplementary Figure 1A).

Statistical analysis

The Liver FibraChek Dx^© platform generates a composite liver disease risk score using age and five serum metabolites, analyzed through HMI’s proprietary online tool (Figure 1). Scores were tested for their ability to distinguish between histologically confirmed liver disease (MASLD or HCC) and healthy controls. Wilcoxon rank sum tests were used to determine the association between risk score and confirmed histological evaluation of liver biopsy. Then a nonparametric receiver operating characteristic (ROC) curve was generated that plotted sensitivity against the false-positive rate (1-specificity)[8]. Statistical significance was set at two-tailed P values < 0.05. All analyses were performed using SAS software version 9.3.3 (SAS Institute; Cary, NC, United States).

Figure 1 Risk analysis for liver disease. A comprehensive risk score is calculated using the levels of 5 biomarkers [serum L-tyrosine, taurocholic acid (by Liver FibraChek Dx^©), alanine aminotransferase, aspartate aminotransferase, and platelet count] (by standard laboratory hematological assessments), and patient age. Risk scores are categorized into 4 quartile groups, with 0.0 indicating low risk of hepatic disease and 1.0 indicating maximum risk of serious liver disease.

RESULTS

The cohort of 45 subjects consisted of 14 subjects with MASLD/cirrhosis, 19 subjects with liver cancer and 12 control subjects. Demographic, clinical and pathologic characteristics of the groups are illustrated in Table 1. The mean serum concentration of Tyr was significantly higher in MASLD and HCC patients than in healthy controls (mean 88.2 μmol/L, 63.1 μmol/L, and 41.1 μmol/L, respectively, P < 0.0001) (Supplementary Figure 1 and Supplementary Table 1). On the other hand, although the average TCA concentration was > 2-fold higher in MASLD (103.0 pg/mL) and HCC patients (184.4 pg/mL) compared to controls (45.0 pg/mL), the difference did not reach statistical significance (P = 0.4341) (Supplementary Figure 1 and Supplementary Table 1).

With the 5 biomarkers (AST, ALT, platelet count, TCA, and Tyr) and age, we calculated the risk score using HMI’s web portal, as described in the Materials and Methods. The mean risk score was markedly elevated in MASLD (0.43) and HCC patients (0.45) compared to control (0.31), though this difference did not reach statistical significance (P = 0.0682) (Supplementary Table 1). We analyzed whether the risk score could predict the presence of liver diseases including MASLD and HCC was analyzed using nonparametric ROC analyses. The probability of having MASLD by Liver FibraChek Dx^© was demonstrated by an area under the ROC curve (AUC) of 0.890 (95%CI: 0.776-1.000), as shown in Figure 2. The assay + algorithm combination demonstrated a sensitivity of 93.9%, specificity of 75.0%, positive predictive value of 91.1%, negative predictive value of 81.8%, and an accuracy of 88.9% (Tables 2 and 3). Supplementary Table 2 depicts fibrosis-4 (FIB-4) performance in the same cohort (sensitivity 75.8% and specificity of 58.3).

Figure 2 Liver assay receiver operating curve characteristics. The high likelihood of identifying metabolic dysfunction-associated steatotic liver disease using the Liver FibraChek Dx^© + algorithm combination was demonstrated by an area under the receiver operating curve of 0.890 (95%CI: 0.776-1.000). The Liver FibraChek Dx^© + algorithm combination. ROC: Receiver operating curve.

Table 2 Cases and controls calculated as low vs higher risk of liver disease, n (%).

Parameter	Cases1	Control	Total
> 0.25 score	31 (93.9)	3 (25)	34
≤ 0.25 score	2 (6.1)	9 (75)	11
Total	33	12	n = 45

¹Cases include chronic liver disease and hepatocellular carcinoma.

Table 3 Diagnostic performance of Liver FibraChek Dx^© assay and algorithm.

Sensitivity and specificity
Statistical parameter	Estimate	Standard error	95%CI
Sensitivity	0.9394	0.0415	0.8580-1.0000
Specificity	0.7500	0.1250	0.5050-0.9950
Positive predictive value	0.9118	0.0486	0.8164-1.0000
Negative predictive value	0.8182	0.1163	0.5903-1.0000
Accuracy	0.8889	0.0468	0.7971-0.9807

Additional ROC analysis (Supplementary Figure 2) found that Liver FibraChek Dx^© could distinguish MASLD from controls (AUC = 0.899) and HCC from controls (AUC = 0.884), but not MASLD from HCC (AUC = 0.566), suggesting the assay performs well in detecting liver disease but may be limited in differentiating between disease subtypes.

DISCUSSION

This study demonstrates the clinical utility of a novel multiplex blood assay in identifying risk of chronic liver pathologies. In the United States, approximately 4.5 million adults have MASLD, which is 1.8 percent of the adult population[9]. In 2022, MASLD and cirrhosis resulted in 54802 United States deaths (16.4 deaths per 100000 population). Considering the high prevalence and poor prognosis of MASLD, early and accurate diagnosis is a crucial step for effective disease management in these patients. In addition to MASLD diagnosis, precise assessment of fibrosis severity in the liver is of great interest in clinical settings because the incidence of HCC increases with liver fibrosis stage. Liver biopsy has been the gold standard for evaluating the severity of liver fibrosis and inflammation. However, drawbacks to its clinical application include its invasiveness, sampling errors, and intra-observer and inter-observer variability[10].

Several non-invasive imaging studies have been used to assess MASLD[11]. However, they are not widely incorporated into clinical practice. Several non-invasive blood-based risk-scoring systems have been used to assess MASLD stage, such as enhanced liver fibrosis, Aminotransferase-to-Platelet Ratio Index[12], and FIB-4[13]. These scoring systems use basic clinical information (e.g., age) and laboratory tests (e.g., AST, ALT, and platelet count) to stratify patient risk. Our selection of these variables aligns with established scoring tools such as the NAFLD Fibrosis Score, Hepamet Fibrosis Score, and FIB-4, all of which incorporate routinely collected laboratory parameters and patient age to predict fibrosis progression risk[14,15]. Recently, more specific fibrosis-specific biomarkers have been developed, such as glycosylated mac2bp, M2BP4[16]. None of these new biomarkers have been widely adopted or have surpassed biopsy as the gold standard for MASLD scoring. Specifically, Liver FibraChek Dx^© outperformed FIB-4 with a higher sensitivity and specificity (93.9% and 75% vs 75.8% and 58.3%, respectively).

Accumulating evidence indicates that discrete blood metabolite profiles can reflect organ function and may serve as promising disease biomarkers. The liver is a crucial metabolic organ, and MASLD alters serum metabolite composition, including that of bile acids[17], free fatty acids[18], and amino acids[19]. Previous investigators have conducted metabolomic profiling of a Chinese cohort with MASLD and identified serum metabolites associated with advanced liver fibrosis, including TCA and Tyr[7]. These findings support the growing body of literature demonstrating that targeted metabolomic signatures—especially bile acids and aromatic amino acids (AAA)—serve as reliable non-invasive indicators of liver pathology and fibrosis stage[20]. Based on these findings, researchers developed and validated Liver FibraChek Dx^©, a novel non-invasive scoring assay for assessing MASLD patients. Liver FibraChek Dx^© multiplex blood test assesses TCA and Tyr, and together with laboratory values for AST, ALT, platelet count, and patient age, calculates a MASLD risk score.

In the current study, we validated the potential usefulness of the Liver FibraChek Dx^© in clinical assessments for MASLD in the United States. At a scoring cutoff of 0.25, the Liver FibraChek Dx^© assay provided an AUC of 0.890 for stratifying MASLD severity, demonstrating an overall sensitivity of 93.9% and specificity of 75.0. The results indicate that Liver FibraChek Dx^© test can assist in distinguishing MASLD from healthy individuals with good reproducibility. In the current study, the MASLD cohort included very early stages of fibrosis (F0-2), suggesting that the test may aid in screening for early and efficient detection of MASLD.

Liver FibraChek Dx^© coordinately measures circulating levels of Tyr and TCA. Tyr is an AAA known to be elevated in MASLD patients[21]. Furthermore, previous studies reported that HCC alters amino acid profiles including Tyr[22]. In contrast, branched-chain amino acids (BCAA), such as leucine, isoleucine, and valine, tend to decrease in MASLD patient serum[23]. Based on knowledge about amino acid imbalances in MASLD patients, amino acid indexes such as Fischer ratio (BCAA/AAA ratio) and BCAA/Tyr ratio (BTR) have been proposed for diagnosing liver fibrosis stage[19,24,25]. However, serum BCAA levels may not accurately reflect early-stage MASLD pathophysiology, whereas our current data indicate that circulating Tyr level is a useful marker of early liver disease. Over 98% of Tyr is degraded by the intrahepatic oxidative pathway, and this pathway is disrupted when the hepatic parenchyma is damaged[26]. In contrast, the primary location for BCAA catabolism is in extrahepatic organs such as muscle, and its catabolism is affected by other metabolic pathways such as ammonia detoxification, suggesting BCAA level is not a simple and direct indicator of liver damage[27]. Thus, blood BCAA levels may not accurately reflect liver fibrosis severity in those patients, decreasing the robustness of the Fischer ratio and BTR that uses BCAA as an index. Future studies will elucidate the mechanistic role of Tyr metabolism in MASLD progression.

TCA is another metabolite measured by the Liver FibraChek Dx^© assay. Serum bile acid metabolites including TCA are elevated in patients with chronic liver disorders, including MASLD, cirrhosis, and HCC[28]. TCA can modulate inflammation during liver pathogenesis, including hepatitis[29], and is involved in cirrhosis progression[30]. Furthermore, abnormally elevated bile acids foster an immunosuppressive microenvironment that favors HCC development[31]. Elevated TCA also promotes fibrosis by activating hepatic stellate cells[32]. These insights about TCA and Tyr as detrimental factors of MASLD pathogenesis indicate the potential application of Liver FibraChek Dx^© for predicting future disease progression in MASLD patients.

The primary limitation of this study was the limited sample size, which did not allow for statistical distinction between TCA levels in MASLD/HCC vs control serum samples, even though average levels were more than 2-fold higher in liver disease. Further validation is planned in a large and culturally and ethnically diverse prospective study. Second, we are a tertiary care facility that preferentially admits patients with advanced disease, which is reflected in our cohort. To evaluate the usefulness of Liver FibraChek Dx^© for screening purposes, subsequent studies will assess more blood samples from subjects with low-grade, low-stage disease in community-based practices. Third, the study used banked serum samples that had been stored at -80 °C prior to analysis. It is unknown whether metabolite composition of plasma changes during frozen storage, though the number of freeze-thaw cycles was restricted to 1-2 by storing serum supernatants in multiple small aliquots. It is feasible that freshly collected plasma may provide different results. We are currently investigating the performance of selected biomarkers in plasma processed via several different protocols. Despite these study limitations, it is encouraging that the Liver FibraChek Dx^© assay, in conjunction with other laboratory values and our risk-score algorithm, demonstrated a clear capability for stratifying liver fibrosis risk. Larger prospective studies are planned to verify the clinical utility of Liver FibraChek Dx^© as a non-invasive evaluation of patient risk of harboring liver fibrosis and liver cancer.

CONCLUSION

The Liver FibraChek Dx^© is a standardized multiplex blood test measuring TCA and Tyr alongside routine clinical markers (AST, ALT, platelet count, and age), which accurately stratifies hepatic fibrosis risk. These findings support its potential clinical application as a non-invasive tool for identifying patients at higher risk for advanced liver disease, including MASLD and HCC. By enabling earlier and more precise risk assessment, Liver FibraChek Dx^© may inform clinical decision-making and improve patient management in both primary care and hepatology settings.

ACKNOWLEDGEMENTS

The authors thank Matthew Silverman PhD (Biomedical Publishing Solutions, Kinard, FL, United States) for expert analytical and editorial assistance. Consultant fees were paid by Cedars-Sinai Medical Center (Los Angeles, CA, United States). This work was supported in part by the Shared Resource of Cedars-Sinai Cancer Biostatistics Shared Resource for statistical analysis.