Hepatocellular carcinoma in F0 livers: Clinical and prognostic insights

Abstract

BACKGROUND

Hepatocellular carcinoma (HCC) is typically associated with advanced liver fibrosis or cirrhosis. However, the clinical characteristics, prognosis, and recurrence risk factors of F0 HCC remain uncertain, and data directly comparing F0 with cirrhotic (F4) HCC are limited.

AIM

To clarify the clinical characteristics, prognosis, and risk factors for recurrence of F0 vs F4 HCC.

METHODS

We retrospectively analyzed 315 patients [F0 HCC (n = 42); F4 HCC (n = 105)] who underwent curative hepatic resection for HCC at the Oita University (January 2010-December 2021). Fibrosis stage was determined histologically using Hematoxylin–Eosin and Azan staining according to the New Inuyama Classification. Clinical characteristics, laboratory data, and tumor features were compared. Survival analyses were performed using Kaplan-Meier curves and log-rank tests. Cox proportional hazards regression was applied to identify recurrence risk factors in F0 HCC. The mean observation period was 1463 ± 1110 days.

RESULTS

Patients with F0 HCC were significantly older; more often male; and more frequently had negative for both hepatitis B and C etiology compared with F4 cases (P < 0.05). Platelet counts and prothrombin activity were higher in the F0 group (P < 0.001). Tumor size was significantly larger in F0 HCC, whereas tumor number did not differ between groups. Overall survival (OS) and recurrence-free survival were comparable between F0 and F4 HCC. Cases of F0 and F4 HCC had no significant difference in OS. Among patients with F0 HCC, those without recurrence had significantly better OS than those with recurrence (P = 0.001). However, patients with F4 HCC exhibited similar OS, irrespective of recurrence. Multivariate analysis identified tumor number as the only independent predictor (P < 0.05).

CONCLUSION

F0 HCC presents with distinct clinical features, including older age, male predominance, negative for both hepatitis B and C background, and larger tumor size. Despite preserved liver function, recurrence significantly compromised survival, underscoring recurrence as a key determinant of outcome. Tumor number was an independent predictor of recurrence, highlighting the importance of early detection and aggressive treatment. Given that patients without cirrhosis are not under standard HCC surveillance, interdisciplinary collaboration is essential to improve early diagnosis and outcomes in this population.

Key Words: Hepatocellular carcinoma; Nonfibrotic liver; Prognosis; Recurrence; Collaboration

Core Tip: Hepatocellular carcinoma (HCC) arising in nonfibrotic (F0) livers is uncommon and its clinical behavior remains unclear. This study directly compared F0 and cirrhotic (F4) HCC, revealing that F0 HCC typically occurs in older men with negative for both hepatitis B and C etiology and larger tumors. Despite preserved liver function, recurrence significantly worsened prognosis. Tumor number was the sole independent risk factor for recurrence in F0 HCC. These findings emphasize the need for early detection strategies, even in non-cirrhotic populations.

INTRODUCTION

Hepatocellular carcinoma (HCC) is the most common primary liver malignancy and a leading cause of cancer-related mortality worldwide[1]. It typically occurs in the setting of chronic liver disease, particularly cirrhosis, which is strongly associated with chronic hepatitis B and C infection, alcohol-related liver disease, and nonalcoholic fatty liver disease[2]. The close association of HCC with liver fibrosis and cirrhosis is well-established, with approximately 80%-90% of cases developing in fibrotic livers and one-third of patients with cirrhosis predicted to develop HCC during their lifetime[3,4]. The fibrotic microenvironment plays a critical role in hepatocarcinogenesis, particularly through mechanisms such as extracellular matrix remodeling, immune evasion, and enhanced angiogenesis[5,6]. Chronic inflammation, such as in the setting of viral hepatitis or metabolic disorders, further accelerates the progression of fibrosis and increases the risk of HCC[7].

A distinct subset of HCC arises in nonfibrotic livers, following an alternative oncogenic pathway. This nonfibrotic HCC is often triggered by direct genotoxic or metabolic insults, bypassing the conventional fibrosis-cirrhosis sequence. Histologically, these tumors have minimal changes in the surrounding liver tissue and lack the regenerative nodules or premalignant changes typically seen in cirrhotic livers[8]. Despite being relatively uncommon, nonfibrotic HCC nevertheless accounts for approximately 20% of all cases of HCC[9]. Clinically, these present as large, solitary tumors in younger patients with preserved liver function and normal alpha-fetoprotein levels[10]. Unlike cirrhosis-associated HCC, these tumors do not develop in the setting of chronic inflammation or regenerative nodules. Instead, the risk factors for nonfibrotic HCC include hepatitis B infection, metabolic syndrome and exposure to genotoxic agents[11-13]. Its early detection is difficult due to the absence of underlying fibrosis. Since these patients are not included in routine HCC surveillance programs, the diagnosis is typically made at an advanced stage[9]. Liver fibrosis is classified into five stages: (1) F0 (normal liver, no fibrosis); (2) F1 (portal fibrosis without septa); (3) F2 (portal fibrosis with few septa); (4) F3 (advanced fibrosis without cirrhosis); and (5) F4 (cirrhosis)[14]. While cases of HCC in non-cirrhotic livers have been reported in several studies[8,15], most cohorts include patients with varying degrees of fibrosis (F0–F3)[16-18], resulting in heterogenous analyses. Consequently, the precise clinical and prognostic characteristics of HCC in true nonfibrotic (F0) livers remain unclear.

Surgical resection is the primary treatment for HCC in nonfibrotic livers. Due to their preserved liver function, such cases have more favorable long-term outcomes vs HCC in cirrhotic livers[19]. However, challenges remain for F0 HCC, specifically in terms of optimizing surveillance strategies and understanding its underlying molecular mechanisms[9]. Accordingly, a deeper understanding of their differences can facilitate the development of more targeted therapeutic strategies and preventive measures for this unique patient population.

In this study, we aimed to define the clinical characteristics and prognosis of F0 HCC using a cohort of histologically confirmed cases, with comparisons made to patients with F4 HCC.

MATERIALS AND METHODS

Patient selection

This retrospective analysis included 315 patients who underwent curative hepatic resection for HCC at the Department of Gastroenterological and Pediatric Surgery, Faculty of Medicine, Oita University, between January 2010 and December 2021. The staging of liver fibrosis was determined by experienced liver pathologists via hematoxylin-eosin staining and Azan staining, with the analysis including patients classified as either stage F0 or F4 based on the New Inuyama Classification[20]. Among these patients, only those with background liver histologically diagnosed as either stage F0 (nonfibrotic) or F4 (cirrhotic) were included in the final analysis. Patients with intermediate fibrosis stages (F1-F3) were excluded. The difference in sample size between the F0 and F4 groups reflects the low frequency of HCC arising in histologically normal (F0) livers. The mean observation period was 1463 ± 1110 days.

Data collection and analysis

Data regarding the clinical background, laboratory tests, and tumor characteristics at the time of HCC onset were collected. The preoperative laboratory data included liver function markers [i.e., albumin, total bilirubin, aspartate aminotransferase, alanine aminotransferase (ALT), prothrombin time], tumor markers (i.e., alpha-fetoprotein, des-gamma-carboxy prothrombin), and platelet count. Tumor characteristics (i.e., tumor size, number of tumors, tumor differentiation and vascular invasion) were recorded. Postoperative recurrence was diagnosed based on imaging studies such as contrast-enhanced computed tomography and magnetic resonance imaging. Follow-up imaging was performed every 3-6 months after surgery. The factors contributing to postoperative recurrence were analyzed using the Cox proportional hazards model.

Overall survival (OS) and recurrence-free survival (RFS) were compared between patients with F0 and F4 HCC. OS was defined as the time from surgery until death from any cause, while RFS was defined as the time from surgery until the first documentation of recurrence or last follow-up without recurrence.

As a sensitivity analysis, age-adjusted and sex-adjusted Cox proportional hazards regression analyses were performed to evaluate the impact of recurrence on OS in patients with F0 HCC.

Statistical analysis

All statistical analyses were performed using statistical analysis software (SPSS 27.0, IBM Corp., Armonk, NY, United States). Categorical variables were analyzed using the χ² test, while continuous variables were analyzed using the Mann-Whitney U test or Student’s t-test depending on the normality of distribution. Survival curves were estimated using the Kaplan-Meier method, and differences between groups were assessed using the log-rank test. The multivariate analysis included Cox proportional hazards regression to identify independent factors associated with recurrence in patients with F0 HCC. Statistical significance was set at P < 0.05.

Ethical considerations

This study was approved by the Institutional Review Board of Oita University. Informed consent was obtained using an opt-out method, wherein information about the study was disclosed on the institutional website, and participants were given the opportunity to decline participation. This study was conducted in accordance with the Declaration of Helsinki.

RESULTS

Comparison of background characteristics between F0 and F4 HCC

Among a total of 315 patients who underwent liver resection for HCC, 42 (13%) and 105 (33%) patients were classified as having F0 and F4 liver fibrosis, respectively. The baseline characteristics were compared between patients with F0 and F4 HCC (Table 1). Patients with F0 HCC were significantly older (P < 0.001) and had a higher proportion of male patients (P = 0.002). Regarding etiology, negative for both hepatitis B and C (NBNC) disease was significantly higher in the F0 group (P < 0.001), while hepatitis C infection was more common in the F4 group (P < 0.001). The F0 group exhibited significantly higher platelet counts (P < 0.001) and greater prothrombin activity (P < 0.001), indicating better preserved liver function and the absence of advanced fibrosis or portal hypertension. Consistent with these findings, both the fibrosis-4 index (P < 0.001) and the aspartate aminotransferase to platelet ratio index score (P < 0.001) were significantly higher in the F4 group than in the F0 group, reflecting advanced fibrosis in patients with F4.

Table 1 Comparison of background characteristics between F0 and F4 hepatocellular carcinoma, mean ± SD.

	F0 HCC (n = 42)	F4 HCC (n = 105)	P value
Sex (male/female)	36/6	62/43	0.002
Age (year)	77.9 ± 7.5	70.2 ± 8.1	< 0.001
Body mass index (kg/m2)	23.6 ± 3.2	24.1 ± 3.6	0.481
Platelet count (104/μL)	20.7 ± 5.3	11.9 ± 7.8	< 0.001
Albumin (g/dL)	3.9 ± 0.6	3.8 ± 0.4	0.302
Total bilirubin (g/dL)	0.7 ± 0.3	0.9 ± 0.3	0.011
Aspartate aminotransferase (U/L)	42.3 ± 48.0	42.0 ± 21.6	0.958
Alanine aminotransferase (U/L)	28.1 ± 32.4	34.1 ± 23.2	0.216
Gamma-glutamyl transpeptidase (U/L)	114 ± 171	81.4 ± 82.1	0.249
Prothrombin time (%)	101 ± 17.7	87.7 ± 17.7	< 0.001
Alpha-fetoprotein (ng/mL)	2981 ± 6988	2804 ± 18478	0.956
Des-gamma-carboxy prothrombin (mAU/mL)	34031 ± 138072	4333 ± 28027	0.222
Albumin-bilirubin grade	-2.61 ± 0.46	-2.46 ± 0.39	0.056
Fibrosis-4 index	3.14 ± 2.52	5.47 ± 2.99	< 0.001
Aspartate aminotransferase to platelet ratio index	0.74	1.45	< 0.001
Hypertension (yes/no)	28/14	59/46	0.243
Diabetes mellitus (yes/no)	19/23	35/70	0.176
Cardiovascular disease (yes/no)	10/32	16/89	0.219
Significant alcohol use (yes/no)	16/26	37/68	0.744
Etiology (B/C/negative for both hepatitis B and C)	5/4/33	22/54/30	< 0.001
Tumor number (solitary/multiple)	37/5	79/26	0.371
Largest tumor diameter (mm)	84.7 ± 44.2	57.2 ± 37.6	0.035
Tumor differentiation (well/moderate/poor)	13/26/3	20/65/12	0.337
Vascular invasion (yes/no)	26/16	84/21	0.034

HCC: Hepatocellular carcinoma.

Regarding tumor characteristics, the maximum tumor diameter was significantly larger in cases of F0 HCC (P = 0.035), although both groups had a similar proportion of solitary vs multiple tumors. Additionally, vascular invasion was significantly more frequent in the F0 group (P = 0.034), while tumor differentiation did not differ significantly between the two groups.

Comparison of survival between F0 and F4 HCC

Cases of F0 and F4 HCC had no significant difference in OS (P = 0.789; Figure 1A). A total of 13 patients died in the F0 group during the observation period. Of them, 8 deaths were attributable to HCC progression, while 5 were due to non-liver-related causes. Notably, no liver failure-related deaths were observed in the F0 group. By contrast, 42 deaths occurred in the F4 group, comprising 26 from HCC progression, 7 from liver failure, and 9 from other causes. There were 22 cases and 73 cases of recurrence observed in the F0 and F4 groups, respectively, although there was no significant difference in RFS (P = 0.116; Figure 1B).

Figure 1 Overall survival and recurrence-free survival of patients with F0 and F4 hepatocellular carcinoma. A: Overall survival. Kaplan-Meier analysis showed no significant difference in overall survival between F0 (nonfibrotic liver) and F4 (cirrhotic liver) hepatocellular carcinoma (P = 0.789); B: Recurrence-free survival. Recurrence was seen in 22 patients and 73 patients, respectively, in the F0 and F4 hepatocellular carcinoma groups, although recurrence-free survival was comparable between both groups based on the Kaplan-Meier curves (P = 0.116). HCC: Hepatocellular carcinoma.

Comparison of survival between recurrent and nonrecurrent cases of F0 and F4 HCC

Among cases of F0 HCC, OS was significantly shorter among cases with recurrence vs those without (P = 0.001; Figure 2A), indicating the strong negative impact of recurrence on the survival of these patients. However, among cases of F4 HCC, OS was similar regardless of recurrence (Figure 2B).

Figure 2 Comparison of overall survival based on recurrence status. A: Among patients with F0 hepatocellular carcinoma, those with recurrence had significantly shorter overall survival (P = 0.001); B: Among patients with F4 hepatocellular carcinoma, overall survival was similar regardless of the presence of recurrence (P = 0.830).

As a sensitivity analysis, after adjustment for age and sex, recurrence remained a significant independent predictor of OS in patients with F0 HCC (P = 0.003). Sex was also significantly associated with OS (P = 0.009), whereas age was not a significant predictor.

Factors contributing to recurrence in F0 HCC

A Cox proportional hazards model was used to identify factors contributing to recurrence in patients with F0 HCC. On univariate analysis, the significant risk factors for recurrence were ALT, hemoglobin A1c (HbA1c), tumor number, maximum tumor diameter and vascular invasion (P < 0.05). On multivariate analysis, only tumor number remained an independent risk factor for recurrence (P < 0.05) (Table 2).

Table 2 Factors contributing to recurrence in F0 hepatocellular carcinoma.

	Univariate analysis		Multivariate analysis
	HR (95%CI)	P value	HR (95%CI)	P value
Sex (male)	1.261 (0.606-1.261)	0.535
Age	1.031 (0.974-1.092)	0.292
Body mass index (kg/m2)	0.906 (0.788-1.042)	0.166
Platelet count (104/μL)	0.990 (0.907-1.082)	0.828
Albumin (g/dL)	0.538 (0.231-1.255)	0.151
Total bilirubin (g/dL)	1.809 (0.473-6.925)	0.386
Aspartate aminotransferase (U/L)	1.005 (0.999-1.010)	0.122
Alanine aminotransferase (U/L)	1.021 (1.005-1.037)	0.008	1.010 (0.990-1.031)	0.319
Gamma-glutamyl transpeptidase (U/L)	1.001 (0.999-1.003)	0.195
Prothrombin time (%)	0.997 (0.951-1.004)	0.089
Alpha-fetoprotein (ng/mL)	1.000 (1.000-1.000)	0.810
Des-gamma-carboxy prothrombin (mAU/mL)	1.000 (1.000-1.000)	0.625
Hemoglobin A1c (%)	0.440 (0.205-0.945)	0.035	0.624 (0.310-1.254)	0.185
Albumin-bilirubin grade	2.568 (0.978-6.742)	0.056
Fibrosis-4 index	1.108 (0.987-1.244)	0.083
Hypertension (yes)	1.232 (0.489-3.103)	0.658
Diabetes mellitus (yes)	1.938 (0.771-4.869)	0.159
Cardiovascular disease (yes)	0.641 (0.245-1.673)	0.363
Smoking (yes)	1.265 (0.516-3.101)	0.608
Significant alcohol use (yes)	1.334 (0.531-3.350)	0.540
Negative for both hepatitis B and C (yes)	0.785 (0.262-2.353)	0.666
Hepatitis C virus (yes)	25.586 (0.083-7906)	0.268
Tumor number (multiple)	5.217 (1.801-15.117)	0.002	4.452 (1.288-15.393)	0.018
Largest tumor diameter(mm)	1.012 (1.001-1.022)	0.028	1.003 (0.989-1.018)	0.269
Vascular invasion (yes)	2.530 (1.077-5.940)	0.033	2.133 (0.644-7.062)	0.682

HR: Hazard ratio.

DISCUSSION

This study provides a comprehensive analysis of HCC in nonfibrotic (F0) livers, focusing on clinical characteristics, prognostic outcomes, and recurrence risk factors. Our findings highlight the key differences between F0 and F4 HCC, emphasizing the unique biological behavior of tumors in non-cirrhotic backgrounds.

Regarding background characteristics, F0 HCC was more frequently observed in older male patients, with a higher prevalence of NBNC etiology and a larger tumor size compared to F4 HCC. This clinical profile supports the hypothesis that F0 HCC follows a different oncogenic pathway, possibly influenced by metabolic or environmental factors, rather than fibrosis-associated mechanisms. Consistent with this hypothesis, HCC arising in nonfibrotic livers suggests the involvement of carcinogenic mechanisms distinct from the traditional inflammation-fibrosis-cirrhosis sequence[8,21]. Metabolic dysfunction constitutes one such mechanism, as metabolic syndrome and metabolic dysfunction-associated steatotic liver disease have been associated with an increased risk of HCC even without advanced fibrosis[22]. Insulin resistance, chronic inflammation, and oxidative stress may promote hepatocarcinogenesis independently of fibrosis progression[23].

Another important mechanism relevant to F0-HCC is direct viral carcinogenesis mediated by hepatitis B virus DNA integration. Unlike hepatitis C virus-related hepatocarcinogenesis, hepatitis B virus can induce malignant transformation through genomic integration, resulting in genomic instability and oncogene activation without the requirement for advanced fibrosis[24].

In addition to metabolic and viral factors, genetic and epigenetic alterations may play a critical role in the pathogenesis of F0-HCC. Tumor-intrinsic molecular abnormalities, including somatic mutations and epigenetic dysregulation, are recognized as central events in hepatocarcinogenesis[25]. Although molecular analyses were not conducted in the present study, future investigations incorporating genomic profiling may help elucidate the biological heterogeneity of F0-HCC.

Consistent with our findings, Naganuma and Ishida[10] also reported a predominance of older male patients with solitary large tumors among cases of F0 HCC. However, their review noted a second, smaller peak in young female patients, which was not observed in our cohort. This difference may reflect population differences or referral biases.

Despite these differences in background and tumor characteristics, the OS and RFS did not significantly differ between F0 and F4 HCC. However, within the F0 HCC group, those without recurrence demonstrated a significantly more favorable prognosis (P = 0.001), emphasizing the critical impact of recurrence on survival. Considering that these patients have well-preserved liver function, recurrence is likely the primary determinant of long-term prognosis in F0 HCC. This interpretation is further supported by differences in the causes of death between the two groups. In the F0 group, most deaths were attributable to tumor progression, and no deaths due to liver failure were observed. By contrast, in the F4 group, 17% of the deaths were related to liver failure, indicating the presence of competing risks distinct from tumor progression. Importantly, the prognostic impact of recurrence in F0 HCC remained significant even after adjustment for age and sex, underscoring the robustness of our findings. This underscores the need for improved surveillance and early intervention strategies in these patients. Using a Cox proportional hazards model, ALT, HbA1c, tumor number, maximum tumor diameter and vascular invasion were significant risk factors for recurrence on univariate analysis, but only tumor number remained significant on multivariate analysis. Thus, tumor burden at diagnosis is a crucial determinant of the risk of recurrence, further reinforcing the importance of early detection and radical treatment. Notably, a paradoxical association was observed on univariate analysis, with higher HbA1c levels linked to a lower risk of recurrence (hazard ratio = 0.440). The biological mechanism behind this remains unclear, although this result may reflect unmeasured confounders or heterogeneity in metabolic status. Since this association did not remain significant in multivariate analysis, further investigation is needed, especially considering the rising prevalence of metabolic-associated fatty liver disease as a background for HCC in non-cirrhotic livers.

These findings highlight the distinct characteristics of HCC in nonfibrotic livers and its clinical challenges. Given the high recurrence rates and the significant impact of recurrence on OS, enhanced surveillance strategies are urgently needed in this population. In contrast to patients with cirrhosis, those without cirrhosis do not undergo routine imaging-based surveillance, which often leads to a delayed diagnosis. Although the early detection of HCC is crucial for improving outcomes, identifying at-risk individuals remains a major challenge.

There is currently no established strategy for systematically screening individuals at risk for F0 HCC. Additionally, most patients do not undergo regular follow-up for liver disease, making early detection even more difficult. Therefore, collaboration across medical specialties is essential to ensure that patients who are not under the care of hepatologists can still undergo appropriate imaging for early detection. Increasing awareness among general practitioners, diabetologists, and gastroenterologists regarding the risk of HCC in non-cirrhotic individuals can help in the earlier identification of the disease. In 2023, the Japan Society of Hepatology issued the Nara Declaration, an initiative that encourages the general public to seek medical consultation if their serum ALT levels exceed 30 U/L, as well as advises primary care physicians to perform appropriate evaluations and refer patients to gastroenterologists when necessary. These efforts aimed to promote the early detection of chronic liver diseases, including those associated with metabolic syndrome, alcohol use, or immune-mediated liver injury, helping prevent the development of cirrhosis and HCC. These early detection strategies can facilitate the diagnosis of nonfibrotic HCC cases that would otherwise go unnoticed. This underscores the critical need for interdisciplinary cooperation to improve the early diagnosis and outcomes of this unique patient population.

By strictly defining fibrosis stage through histological evaluation of resected background liver specimens and focusing on recurrence-driven prognosis in F0 HCC, the present study provides a refined clinical framework for understanding outcomes in this distinct patient population. Although related to prior fibrosis studies, this work shifts the emphasis toward prognostic stratification within histologically confirmed F0 HCC[26,27].

Several limitations should be acknowledged. First, the retrospective study design could introduce selection bias. Second, the relatively small sample size of F0 HCC cases compared to the F4 cases potentially limits the statistical power of our findings. Although propensity score matching based on age and sex was explored to address baseline imbalances, the matched cohort was substantially reduced in size and therefore was not adopted as the primary analytical approach. Third, molecular analyses were not performed, and thus, the genetic drivers of HCC in F0 livers remain unclear. To validate these findings, future prospective studies should incorporate larger cohorts and include molecular profiling.

CONCLUSION

This study provides valuable insights into the clinical characteristics, prognosis, and recurrence risk factors of HCC in nonfibrotic livers. Compared to F4 HCC, cases of F0 HCC were more common in older male patients, characterized by an NBNC etiology and larger tumor. Although OS and RFS were comparable between F0 and F4 HCC, recurrence significantly impacted survival in F0 HCC. Tumor number was identified as an independent predictor of recurrence, emphasizing the need for improved early detection and treatment strategies. Given the growing recognition of HCC in non-cirrhotic livers, further research is needed to optimize management strategies and improve outcomes in this distinct population.

ACKNOWLEDGEMENTS

The authors are grateful to the pathology team at Oita University for their expertise in histological assessment of liver fibrosis.