This editorial refers to “Baseline and longitudinal biomarkers predict hepatocellular carcinoma in chronic hepatitis C: A cohort study” by Li et al, 2026; https://dx.doi.org/10.4251/wjgo.v18.i5.117512.
INTRODUCTION
Direct-acting antivirals (DAAs) have revolutionized the management of hepatitis C virus (HCV) infection, achieving cure rates that exceed 95% across diverse populations[1]. However, global treatment uptake remains suboptimal, reflecting persistent gaps in the cascade of care and leading to the establishment of a substantial treatment-naïve chronic HCV population worldwide[2]. Between 2014 and 2023, only approximately 21% (approximately 12.7 million individuals) of those living with chronic HCV infection globally received DAA therapy[3]. In these untreated chronic HCV cohorts, the risk of hepatocellular carcinoma (HCC) is cumulative and time-dependent as the population ages[4]. This creates a clinical imperative for low-cost, repeatable, and scalable risk stratification tools to guide surveillance and prioritize treatment, especially in resource-limited settings.
Current HCC risk assessment relies heavily on baseline “snapshots,” such as cirrhosis status, baseline noninvasive fibrosis surrogates such as the fibrosis-4 (FIB-4) index, and alpha-fetoprotein (AFP) thresholds[5]. However, carcinogenesis evolves over time; therefore, risk assessment strategies should account for temporal changes. In a large cohort study, longitudinal changes in FIB-4 (ΔFIB-4) were strongly associated with subsequent cirrhosis and HCC, underscoring the prognostic value of repeated biomarker assessment beyond a single baseline measurement[6].
A recent retrospective cohort study published in the World Journal of Gastrointestinal Oncology by Li et al[7] supported the clinical value of dynamic monitoring in a long-term, treatment-naïve chronic HCV population. In a cohort of 271 patients followed for a median of 9.4 years, 29.9% developed HCC. Crucially, longitudinal analysis demonstrated that a 3-year absolute increase in the FIB-4 score (ΔFIB-4 ≥ 1.0) served as a powerful, independent risk factor for incident HCC (adjusted hazard ratio = 2.67). When stratified by trajectory, the “increase” group (ΔFIB-4 ≥ +1.0) exhibited a high 15-year cumulative HCC incidence of 72.7%. Conversely, patients who maintained persistently low FIB-4 values (< 1.45) experienced no observed HCC events during the same period, suggesting a subgroup with very low observed risk in this cohort. These findings underscore that increasing FIB-4 scores are associated with HCC risk and suggest that longitudinal FIB-4 monitoring may refine risk stratification beyond baseline assessment, although the timing and thresholds for clinical action require further validation. Importantly, cumulative HCC risk increases over time across trajectory groups in untreated cohorts. In Li et al’s study[7], separation between trajectories appeared more pronounced later during follow-up, which suggests that longitudinal patterns may be most informative for long-term risk stratification rather than short-term decision-making.
Positioning this longitudinal approach is highly timely. The broader literature indicates that time-dependent ΔFIB-4, especially ΔFIB-4/year, is consistently associated with HCC risk in chronic HCV cohorts across various clinical contexts, including post-sustained virologic response populations[8]. In parallel, contemporary guidance in metabolic dysfunction-associated steatotic liver disease (MASLD) reinforces the broader concept that repeated assessment of noninvasive fibrosis indices can refine longitudinal risk stratification, underscoring that dynamic fibrosis assessment is not unique to chronic HCV[9]. Importantly, the clinical utility of fibrosis assessment extends beyond predicting hepatic malignancy and is linked to broader systemic outcomes, including cardiovascular risk, while current management frameworks increasingly emphasize comprehensive risk reduction alongside liver-focused care[10,11].
This underlying biology and the accumulating evidence support the clinical use of longitudinal monitoring in patients with untreated chronic HCV. However, longitudinal cutoffs have not yet been universally standardized for all treatment-naïve chronic HCV patients. In this editorial, we summarize the rationale for longitudinal FIB-4 monitoring in patients with untreated chronic HCV, discuss its mechanistic basis, and propose a pragmatic clinical framework to guide clinical decision-making and integrated surveillance. We then outline key standardization needs to support future validation and implementation.
MECHANISTIC DRIVERS OF LONGITUDINAL CHANGES IN FIB-4 AND HCC PROPENSITY
To understand why longitudinal ΔFIB-4 may provide additional risk information beyond static measurements alone, it is necessary to deconstruct the score into its biological determinants. FIB-4 is a composite biomarker that reflects three pathophysiological domains: Cumulative duration of disease exposure (age), ongoing hepatic injury [aspartate aminotransferase (AST)/alanine aminotransferase], and architectural distortion leading to portal hypertension (platelet count)[6,12,13]. While a baseline FIB-4 score provides a snapshot of the existing fibrosis burden, the rate of ΔFIB-4 approximates the velocity of fibrogenesis. Recent clinical evidence indicates that patients with rapidly increasing FIB-4 scores have a significantly higher hazard ratio for HCC than those with stable elevated scores do, suggesting that the biomarker’s kinetic trajectory captures the active acceleration of disease[6,13]. Conversely, a decrease in FIB-4, often observed after viral suppression, is correlated with a reduction in oncogenic risk, reflecting the reversibility of the inflammatory driver[8].
Three specific biological mechanisms underpin the correlation between increasing FIB-4 and HCC propensity: (1) Progressive extracellular matrix deposition and tissue mechanics: An increasing FIB-4 trajectory primarily driven by declining platelets reflects worsening portal hypertension caused by stiffening of the hepatic extracellular matrix. This mechanical stiffness is not inert; rather, it directly activates mechanotransduction pathways that promote cellular proliferation and survival, creating a biophysical niche permissive for malignant transformation[14]; (2) Chronic inflammatory activation: Sustained alterations in transaminase profiles, specifically the progressive increase in AST within the FIB-4 numerator, indicate persistent viral necroinflammation. This state creates a “field effect”, where chronic cycles of hepatocyte death and compensatory regeneration increase the accumulation of double-strand DNA breaks and oncogenic mutations[15]; and (3) Vascular remodeling: The platelet component of FIB-4 serves as a surrogate for portal pressure and splenic sequestration. A progressive decrease in platelets may reflect worsening portal hypertension and fibrosis-related vascular remodeling, including enhanced angiogenic signaling, thereby shaping a procarcinogenic hepatic microenvironment that favors HCC development[13].
Interpretation of longitudinal FIB-4 requires caution because age is embedded in the formula. As a result, FIB-4 can increase over time because of aging alone, even without true disease progression[16]. Kimura et al[17] reported that the specificity of FIB-4 for advanced fibrosis decreases in cohorts aged > 65 years. This age-related “background drift” can inflate the score and may mimic pathological changes. Therefore, clinical interpretation should distinguish physiological drift from a genuinely steep slope suggestive of progression. Consistent with this, age-adjusted cutoffs have been proposed, including a higher threshold (≥ 2.0) for individuals ≥ 65 years to improve specificity and reduce false-positive classification of advanced fibrosis[18].
Longitudinal increases in FIB-4 may reflect accelerated fibrosis progression and portal remodeling. Accordingly, they could plausibly signal HCC risk beyond baseline staging. However, direct validation of specific ΔFIB-4 thresholds in patients with untreated chronic HCV remains important. Interpretation of longitudinal FIB-4 trajectories requires caution. Decreased platelet counts may reflect nonhepatic factors, such as hypersplenism unrelated to fibrosis progression, immune thrombocytopenia, medications, or hematologic disease. Transient elevations in transaminases may arise from alcohol exposure, steatohepatitis flares, myositis, or drug effects rather than sustained necroinflammation[19]. In addition, biological fluctuation and assay variability may contribute to spurious temporal “slopes”[20]. Accordingly, ΔFIB-4 signals should be interpreted in the clinical context and confirmed with repeated testing before risk is escalated or management decisions are made.
PREDICTIVE ACCURACY AND CLINICAL ACTIONABILITY: FROM BIOMARKER VALIDATION TO STANDARDIZED HCC CARE PATHWAYS
Baseline FIB-4 stratification provides prognostic value for HCC risk in patients with untreated chronic HCV[5]. In this setting, older age, AFP ≥ 20 ng/mL, and cirrhosis are additional independent predictors of HCC[5]. The 3.25 cutoff is also widely used as a rule-in threshold for advanced fibrosis or severe fibrosis in chronic hepatitis C[21]. While emerging evidence suggests that longitudinal FIB-4 kinetics correlate with HCC risk, these findings await prospective validation in treatment-naïve cohorts[22].
Until prospective validation is available, a pragmatic interpretation approach can be used to mitigate age drift. Clinicians may apply either a threshold-crossing strategy (escalating attention when FIB-4 crosses ≥ 1.45 or FIB-4 crosses ≥ 3.25) or a slope-based strategy (identifying ‘rapid progressors’ using upper ΔFIB-4/year quantiles)[23]. Reassessment intervals can be aligned with routine follow-up and local capacity, and 1-year to 3-year intervals may serve as a pragmatic starting point rather than a prescriptive mandate[24,25]. Shorter intervals may be reasonable in older patients or those approaching key thresholds, but the optimal cadence for longitudinal risk reclassification in strictly untreated cohorts remains to be defined. Taken together, these considerations support a targeted triage approach: Scores crossing into or presenting at ≥ 3.25 may warrant prioritization for DAA access and intensified surveillance[5]. Intermediate scores with rapid upward trajectories may indicate an upward shift in risk and justify closer monitoring[13]. While persistently low scores may help triage in constrained settings, de-escalation should be avoided pending prospective safety data[24].
INTEGRATED LONGITUDINAL SURVEILLANCE INCORPORATING FIB-4, AFP, AND DEMOGRAPHIC MODIFIERS
Guidelines alignment and modality comparisons
While vibration-controlled transient elastography and magnetic resonance elastography demonstrate superior cross-sectional diagnostic accuracy for staging advanced fibrosis, their reliance on specialized equipment, limited availability, and high costs limit their utility for high-frequency longitudinal monitoring. In head-to-head comparisons with other blood-based noninvasive tests, such as the AST-to-platelet ratio index or HCV-specific algorithms, FIB-4 has shown comparable prognostic performance for predicting incident HCC, likely because it incorporates age as a key risk modifier[26]. Current clinical practice guidelines from the American Association for the Study of Liver Diseases and the European Association for the Study of the Liver recommend ultrasound surveillance in patients with established cirrhosis and acknowledge FIB-4 for initial risk stratification; however, these guidelines do not explicitly operationalize a longitudinal biomarker pathway for strictly treatment-naïve chronic HCV cohorts[9,27]. To bridge this clinical gap, a sequential “step-up” algorithm is proposed: Treatment-naïve patients are initially stratified using widely available FIB-4 and AFP measurements. Those demonstrating a high-risk baseline (≥ 3.25) or a rapid longitudinal increase are subsequently prioritized for second-line elastography to confirm architectural distortion, aligning the scalability of serial FIB-4 testing with the imaging standards recommended by major guidelines.
Integrating multimodal biomarkers for comprehensive risk stratification
Longitudinal FIB-4 addresses the need for high-frequency fibrosis monitoring, but a comprehensive surveillance framework should also account for tumor biology. Abdominal ultrasound and isolated AFP measurements remain the traditional cornerstone of screening; however, both have limitations in early-stage sensitivity and operator dependence. An integrated surveillance approach is biologically plausible because longitudinal FIB-4 and AFP capture distinct yet complementary pathophysiological domains. Time-dependent changes in FIB-4 primarily reflect active fibrogenesis, architectural distortion, and portal remodeling, whereas AFP more directly reflects tumor-associated biological activity[28]. Combining these indices enables a dual-domain assessment that integrates fibrosis burden with tumor-associated biomarker activity.
When constructing this multimodal framework, it is crucial to distinguish validated evidence from proposed clinical heuristics. In patients with untreated chronic HCV, a baseline AFP threshold (e.g., ≥ 20 ng/mL) has been established to have independent prognostic value for HCC development, particularly when integrated with baseline FIB-4 stratification[5]. Conversely, incorporating AFP kinetics (e.g., longitudinal trends, parametric trajectories, or AFP velocity) is biologically plausible and has been reported to improve surveillance sensitivity[29,30]. However, in strictly untreated chronic HCV cohorts, this approach should currently be presented as a proposed model pending direct quantitative validation.
Recent studies suggest the feasibility and potential predictive value of integrated surveillance models that combine these biomarkers into unified clinical algorithms[5]. A conceptual, pragmatic integrated tiering system can operationalize these variables into actionable pathways, emphasizing routine laboratory tests that are low-cost and highly scalable. The cutoffs below are presented as pragmatic, literature-informed examples rather than validated thresholds for strictly untreated cohorts.
Low risk: Characterized by a baseline FIB-4 < 1.45, an absence of upward threshold crossing during longitudinal reassessment, and a stable, nonelevated AFP. Notably, for patients aged ≥ 65 years, an age-adjusted lower threshold of < 2.0 may be considered to reduce age-related false-positive classification[18]. Incorporating age-adjusted FIB-4 cutoffs may improve specificity in older adults and reduce false-positive classification of advanced fibrosis[17,18]. This may help avoid unnecessary escalation to specialized testing in lower-risk patients.
Intermediate risk: Defined by a baseline FIB-4 score between 1.45 and 3.25[5,7], with the lower boundary explicitly adjusted to 2.0 for patients aged ≥ 65 years[18]. This tier also includes patients who exhibit modest temporal variability without a sustained pathological upward trend, alongside a stable, non-rising AFP profile. Patients in this category warrant routine reassessment aligned with local clinical capacity.
High risk: Identified by a baseline FIB-4 ≥ 3.25, an upward threshold crossing into the ≥ 3.25 tier, or a sustained rapid increase in FIB-4 over time. When slope-based criteria are applied, thresholds reported in prior cohorts may inform longitudinal interpretation, but they should not be regarded as validated triggers for strictly untreated populations[22,23]. This core risk is significantly amplified by the presence of demographic modifiers (older age, male sex) and/or an elevated AFP level[5].
To visualize this workflow in clinical practice, a sequential heuristic is proposed in which clinicians identify the baseline band, perform a longitudinal check, assign a corresponding risk tier, and then determine the appropriate surveillance intensity and treatment prioritization. This stepwise framework transforms static laboratory values into a dynamic, integrated surveillance continuum, allowing more tailored monitoring of high-risk, treatment-naïve patients.
Imaging-based risk stratification approaches, including radiomic “virtual biopsy” strategies and longitudinal magnetic resonance imaging-based HCC prediction models, offer complementary precision in specialized settings[31,32]. However, these modalities are resource-intensive and not universally accessible. A pragmatic tiered strategy is therefore warranted, using longitudinal, laboratory-based FIB-4 for population-level triage, with elastography or advanced imaging reserved for patients approaching clinically relevant decision thresholds. Defining when laboratory-only strategies are sufficient and when imaging integration is indicated remains an important translational question.
STANDARDIZATION AND GENERALIZABILITY: REFINING CUTOFFS ACROSS POPULATIONS AND ETIOLOGIES
Establishing universal longitudinal cutoffs for FIB-4 is clinically challenging because of biological and methodological variation. Because age acts as a continuous multiplier in the formula, FIB-4 tends to increase over time even without true fibrosis progression[18]. Furthermore, population-level differences in baseline platelet counts and transaminase levels, together with inconsistencies in longitudinal measurement cadence across different healthcare systems, can significantly distort the interpretation of rigid thresholds and kinetic change metrics[6].
Despite these complexities, robust evidence has demonstrated that the conceptual framework of longitudinal risk stratification is highly generalizable across different liver disease states. For instance, recent guidance and large-scale cohort studies in MASLD support the prognostic value of repeated FIB-4 assessments in predicting HCC risk, specifically for differentiating outcomes between persistently low, persistently high, or rapidly rising trajectories[6,32,33]. However, because the underlying fibrogenic drivers differ, the specific mathematical cutoffs, threshold crossing values, and hazard ratios derived from MASLD or virally suppressed cohorts cannot be directly applied to treatment-naïve chronic HCV populations without rigorous, context-specific recalibration.
Consequently, validating longitudinal FIB-4 for clinical use in patients with untreated chronic HCV requires a structured research agenda. Future validation should use multicenter, multiethnic prospective cohorts of strictly treatment-naïve chronic HCV patients and adopt a harmonized follow-up cadence (e.g., annual or biennial evaluations) to minimize measurement bias. Studies should directly compare the clinical performance and usability of threshold-crossing vs continuous slope-based (ΔFIB-4/year) strategies. Where feasible, AFP should be incorporated, with explicit evaluation of static thresholds vs dynamic kinetics, to optimize integrated surveillance. Given shifting demographics, validation should include patients with mixed etiologies, particularly those with concurrent untreated HCV and metabolic risk factors[34]. At the population level, temporal point process and sequence-modeling frameworks may further refine risk migration capture by integrating multievent electronic medical record streams rather than relying on single-metric trajectories. Such approaches could enable scalable deployment of dynamic HCC risk stratification across large healthcare systems, particularly in regions with a substantial untreated chronic HCV burden.
A major limitation of this longitudinal approach is the current lack of specific quantitative criteria for defining a clinically dangerous trajectory. Although assessing risk via the velocity of FIB-4 (e.g., an increase of > 0.5 points/year) is proposed within this framework, the literature on strictly treatment-naïve HCV populations is insufficient to support a definitive quantitative threshold. Consequently, a vital future direction is to rigorously validate and standardize this dynamic threshold in large, multicenter prospective cohorts of untreated HCV patients to establish concrete clinical guidelines.
CONCLUSION
The clinical paradigm of HCC risk stratification in treatment-naïve patients with chronic HCV should evolve from static staging to dynamic monitoring. While a baseline FIB-4 score identifies the existing fibrosis burden, ΔFIB-4 may better capture ongoing architectural remodeling and progressive risk. By integrating ΔFIB-4 with AFP and demographic modifiers, clinicians may construct a scalable, multimodal surveillance continuum. Although the optimal cut-off values require prospective validation to address age-related drift and mixed metabolic etiologies, the biological rationale for longitudinal monitoring remains compelling. A trajectory-based triage strategy may therefore help identify high-risk progressors and inform the prioritization of antiviral therapy and surveillance intensity in treatment-naïve populations.
Peer review: Externally peer reviewed.
Peer-review model: Single blind
Specialty type: Oncology
Country of origin: Viet Nam
Peer-review report’s classification
Scientific quality: Grade C
Novelty: Grade C
Creativity or innovation: Grade C
Scientific significance: Grade B
P-Reviewer: You L, PhD, China S-Editor: Bai SR L-Editor: A P-Editor: Xu J
