Abstract
Background
Liver resection and local ablation are the only curative treatment for non-cirrhotic hepatocellular carcinoma (HCC). Few data exist concerning the prognosis of patients resected for non-cirrhotic HCC. The objectives of this study were to determine the prognostic factors of recurrence-free survival (RFS) and overall survival (OS) and to develop a prognostication algorithm for non-cirrhotic HCC.
Methods
French multicenter retrospective study including HCC patients with non-cirrhotic liver without underlying viral hepatitis: F0, F1 or F2 fibrosis.
Results
A total of 467 patients were included in 11 centers from 2010 to 2018. Non-cirrhotic liver had a fibrosis score of F0 (n=237, 50.7%), F1 (n=127, 27.2%) or F2 (n=103, 22.1%). OS and RFS at 5 years were 59.2% and 34.5%, respectively. In multivariate analysis, microvascular invasion and HCC differentiation were prognostic factors of OS and RFS and the number and size were prognostic factors of RFS (P<0.005). Stratification based on RFS provided an algorithm based on size (P=0.013) and number (P<0.001): 2 HCC with the largest nodule ≤10 cm (n=271, Group 1); 2 HCC with a nodule >10 cm (n=176, Group 2); >2 HCC regardless of size (n=20, Group 3). The 5-year RFS rates were 52.7% (Group 1), 30.1% (Group 2) and 5% (Group 3).
Conclusions
We developed a prognostication algorithm based on the number (≤ or >2) and size (≤ or >10 cm), which could be used as a treatment decision support concerning the need for perioperative therapy. In case of bifocal HCC, surgery should not be a contraindication.
Keywords: Hepatocellular carcinoma (HCC), non-cirrhotic liver, prognostic factors, recurrence-free survival (RFS), prognostication algorithm
Introduction
Liver cirrhosis is the most common risk factor for hepatocellular carcinoma (HCC). However, non-cirrhotic HCC accounts for 15–20% of HCC, but its incidence widely varies across the world and is higher in Western countries. Only 10% of HCC develop in a non-cirrhotic liver with F0, F1 or F2 in the METAVIR scoring system (1), and the known prognostic factors are hepatopathies such as viral hepatitis, nonalcoholic steatohepatitis (NASH), hemochromatosis and hereditary diseases (2-4). To date, liver resection and local ablation are the only curative treatment for non-cirrhotic HCC. The subgroups of patients treated for non-cirrhotic HCC that are most likely to benefit from liver resection are not clearly defined in the literature.
Moreover, non-cirrhotic HCC studies include small and non-homogeneous cohorts that mix patients with hepatitis, hemochromatosis, or Wilson’s disease and patients without these known risk factors. The inhomogeneity of the studies is subsequent to the absence of consensus on the definition of HCC on a “non-pathological liver”, which is sometimes defined by a non-cirrhotic liver with a fibrosis score of F0-F1 or F0-F1-F2 in the METAVIR scoring system (1). Consequently, there is very little data in the literature on the prognosis of patients who have undergone liver resection for F0, F1 or F2 non-cirrhotic HCC without underlying chronic disease (2,5-7). The aims of this study were to determine the prognostic factors of recurrence-free survival (RFS) and overall survival (OS) and to develop a prognostication algorithm for patients treated for resectable F0, F1 or F2 non-cirrhotic HCC without underlying viral hepatitis. We present the following article in accordance with the STROBE reporting checklist (available at https://hbsn.amegroups.com/article/view/10.21037/hbsn-22-33/rc).
Methods
Ethical statement
The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013) and was approved by the institutional ethics board of the Toulouse University Hospital (No. RnIPH 2022-67). According to French law on ethics, patients were informed that their codified data would be used for the study.
Population
This is a French retrospective multicenter study, including 11 centers. Between January 2010 and December 2018, patients who underwent hepatectomy for HCC in a non-cirrhotic liver ≤ F2 without underlying hepatitis B virus (HBV) or hepatitis C virus (HCV) were included. We excluded all patients with anti-HBc antibodies. Liver resections were performed using laparoscopy or laparotomy, according to the surgeon’s usual practices and with their own parenchymal transection equipment. All types of hepatectomy were included. A non-cirrhotic liver ≤ F2 was defined as the absence or minimal presence of non-tumoral liver fibrosis: F0, F1 or F2 in the “equivalent” METAVIR scoring system (1). The diagnosis of HCC and the extent of non-tumoral liver fibrosis were determined, or confirmed in case of preoperative biopsy, on histological examination of the resected specimen. A non-tumoral liver with fibrosis ≥ F3, iterative hepatectomy, viral hepatitis HBV and/or HCV, metabolic diseases, fibrolamellar HCC and hepatocholangiocarcinoma were excluded. Every patient included in this study had preoperative Magnetic Resonance Imaging (MRI), and when possible, gadoxetic acid enhanced Magnetic Resonance Imaging (EOB-MRI). The eligibility criteria are detailed in Table 1. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). The study was approved by institutional ethics board of Toulouse University Hospital (No. RnIPH 2022-67).
Table 1. Eligibility criteria.
| Inclusion criteria | Exclusion criteria |
|---|---|
| Age ≥18 years | Age <18 years |
| Hepatectomy for HCC | Non-tumoral liver: Fibrosis ≥ F3 in the “equivalent” METAVIR scoring system1 |
| Laparoscopic or open approach | Repeated hepatectomy for HCC recurrence |
| Non-tumoral liver: Fibrosis F0, F1 or F2 in the “equivalent” METAVIR scoring system1 | Viral hepatitis HBV and/or HCV |
| Metabolic diseases (hemochromatosis, Wilson’s disease) | |
| Fibrolamellar HCC | |
| Hepatocholangiocarcinoma |
HCC, hepatocellular carcinoma; HBV, Hepatitis B virus; HCV, Hepatitis C virus.
Data collected
For each patient, variables were retrospectively collected and detailed in a supplementary data table (Table S1). The data collected were (I) demographic data, (II) prognostic factors of HCC, (III) HCC diagnosis, (IV) preoperative biopsy, (V) preoperative treatment, (VI) preoperative biological analysis, (VII) surgery, (VIII) postoperative complications, (IX) anatomopathological analysis of the tumor, (X) anatomopathological analysis of the non-tumoral liver and (XI) follow-up. Severe morbidity was defined for a Dindo-Clavien score ≥3 (8). Postoperative liver failure was defined according to “50-50” criteria (prothrombin time <50% and serum bilirubin >50 µml/L) on postoperative day 5 (9). A strictly normal liver was defined as A0-F0 according to the the “equivalent” METAVIR grading score (1).
Statistical methods
Qualitative variables were described as numbers (percentages) and quantitative variables as means ± the standard deviation. RFS and OS were presented with Kaplan-Meir curves. Univariate followed by multivariate Cox regression were performed using either survival or recurrence free survival as the outcome. Variables significantly (P<0.05) associated with survival in the univariate analysis were then introduced in the multivariate analysis. We used multiple imputation to account for missing data in predictors of multivariate Cox models. Multiple imputation was performed using the R Package MICE. To stratify the patient’s risk prior to surgery, we introduced preoperative variables significantly associated with RFS in the multivariate analysis in a conditional inference tree. For further stratification in the patients with the best survival rate and who were therefore more likely to benefit from surgery, we drew another conditional inference tree of this population subset and introduced anatomopathological variables significantly associated with recurrence free survival. We used a conditional inference tree using recursive partitioning with a Bonferroni correction for multiple tests. This algorithm finds the variable the most strongly associated with survival and makes the best split in this variable (10,11). We performed this analysis using the partykit package in R. For all analyses, a P value of less than 0.05 was considered statistically significant. All of the statistical analyses were performed using R software.
Results
Population study
Between January 2010 and December 2018, 467 hepatectomies for non-cirrhotic HCC without underlying viral hepatitis, according to the inclusion and exclusion criteria, were performed in the 11 centers, 6 of which included more than 40 patients.
Preoperative characteristics
The preoperative characteristics are listed in Table 2. Median age was 69.2 (IQR: 61.8, 75.8 years), with a male/female gender ratio of 3/1. Most (53.3%) the patients in this study had a BMI >25 kg/m2 and more than a quarter of them had diabetes, hypercholesterolemia, high blood pressure and/or metabolic syndrome. For almost half of them, HCC was discovered incidentally. A total of 10% of the patients had preoperative treatment for HCC [transarterial chemoembolization (TACE) or sorafenib] and 14.3% had preoperative portal vein embolization. Half of the patients (54.6%) had a preoperative tumoral biopsy and one third of the patients (35.8%) had a non-tumoral liver biopsy. A total of 42.8% of the patients had a normal alpha-fetoprotein (AFP) level (AFP <5 ng/mL). For the 57.2% of the patients with abnormal AFP (≥5 ng/mL), the median AFP was 50.3 ng/mL with an IQR of 489.6.
Table 2. Preoperative characteristics of the total population.
| Characteristics | Study group (n=467) |
|---|---|
| Demographic data | |
| Age (years), median [IQR] | 69.2 [61.8, 75.8] |
| Gender, n (%) | |
| Female | 117 (25.1) |
| Male | 350 (74.9) |
| BMI, median [IQR] | 26 [5.8] |
| <25 | 164 (35.1) |
| 25–30 | 158 (33.8) |
| ≥30 | 91 (19.5) |
| Missing data | 54 (11.6) |
| Diabetes, n (%) | 132 (28.3) |
| Dyslipidemia, n (%) | 119 (25.5) |
| High blood pressure, n (%) | 258 (55.2) |
| ASA score, n (%) | |
| 1 | 80 (17.1) |
| 2 | 205 (43.9) |
| 3 | 141 (30.2) |
| 4 | 2 (0.4) |
| Missing data | 39 (8.4) |
| Risk factors of HCC, n (%) | |
| Alcohol consumption | 161 (34.5) |
| Metabolic syndrome | 126 (27.0) |
| Tobacco consumption | 158 (33.8) |
| HCC diagnosis, n (%) | |
| Symptoms at diagnosis | 203 (43.5) |
| Incidental discovery | 227 (48.6) |
| Screening | 23 (4.9) |
| Preoperative jaundice | 6 (1.3) |
| Alteration in general health status | 65 (13.9) |
| Preoperative biopsy, n (%) | |
| Tumoral biopsy | 255 (54.6) |
| Non-tumoral biopsy | 167 (35.8) |
| Preoperative treatment, n (%) | |
| Transarterial chemoembolization | 42 (9.0) |
| Sorafenib | 7 (1.5) |
| Radiotherapy | 0 |
| Biliary drainage | 2 (0.4) |
| Portal vein embolization | 67 (14.3) |
| Preoperative biological analysis | |
| AFP (ng/mL), median [IQR] | 9.5 [106.5] |
| AFP ≥5 ng/mL, n (%) | 171 (57.2) |
| AFP among patients with AFP ≥5 ng/mL (ng/mL), median [IQR] | 50.3 [489.6] |
| Missing data, n (%) | 168 (36.0) |
| MELD, median [IQR] | 7 [3] |
| PR (%), median [IQR] | 97 [14] |
| Platelets (G/L), median [IQR] | 250.5 [115.2] |
| Serum bilirubin level (μmol/L), median [IQR] | 8 [5] |
IQR, interquartile range; BMI, body mass index; ASA, American Society of Anesthesiologists; HCC, hepatocellular carcinoma; AFP, alpha-fetoprotein; MELD, model for end-stage liver disease; PR, prothrombin ratio.
Peroperative and postoperative characteristics
Peroperative and postoperative characteristics are detailed in Table 3. The majority of the HCC resections were performed by open approach, with anatomical resection of the HCC. Pedicular clamping was used in more than half of the hepatectomies with a median clamping duration of 30 min (IQR: 26.8). Most resections (52.5%) were major hepatectomies with a median number of resected hepatic segments of 3 (IQR: 2). During surgery, 3.0% of patients had capsular effraction. During 90 postoperative days, the overall morbidity was 73.4% (n=343): 64.0% (n=299) of the patients had at least one medical complication and 28.7% (n=134) at least one surgical complication. A total of 9.4% (n=44) had a postoperative biliary fistula. The reoperation rate was 6.9% (n=32). The severe morbidity rate was 18.7% (n=87). Seventeen patients died within 90 post-operative days (3.6%); 4 patients from a biliary fistula, 2 from a cardiac cause, 3 from a respiratory cause, 4 from liver failure and 4 from other miscellaneous causes. The median length of hospitalization was 10 days (IQR: 7).
Table 3. Peroperative and postoperative characteristics of the total population.
| Characteristics | Study group (n=467) |
|---|---|
| Surgery | |
| Surgery duration (min), median [IQR] | 228 [120] |
| Surgical technique, n (%) | |
| Laparoscopic approach | 119 (25.5) |
| Conversion to laparotomy | 29 (6.2) |
| Open approach | 348 (74.5) |
| Anatomic resection, n (%) | 315 (67.5) |
| Capsular effraction, n (%) | 14 (3.0) |
| Number of resected segments, median [IQR] | 3 [2] |
| Major hepatectomy (≥3 resected segments), n (%) | 245 (52.5) |
| Pedicular clamping, n (%) | 261 (55.9) |
| Clamping duration (min), median [IQR] | 30 [26.8] |
| Vascular exclusion of the liver, n (%) | 13 (2.8) |
| Blood loss (mL), median [IQR] | 500 [700] |
| Blood transfusion, n (%) | 80 (17.1) |
| Number of packed red blood cells, median [IQR] | 2 [2] |
| Procedures associated with surgery, n (%) | 52 (11.1) |
| Thermoablation | 6 (1.3) |
| Bilio-enteric anastomosis | 5 (1.1) |
| Vascular resection | 16 (3.4) |
| Digestive resection | 4 (0.9) |
| Diaphragm resection | 15 (3.2) |
| Veinous thrombectomy | 6 (1.3) |
| Postoperative complications | |
| Surgical complication, n (%) | 134 (28.7) |
| Biliary fistula, n (%) | 44 (9.4) |
| Hemorrhage, n (%) | 11 (2.4) |
| Medical complication, n (%) | 299 (64.0) |
| Liver failure, n (%) | 16 (3.4) |
| Reoperation, n (%) | 32 (6.9) |
| Dindo-Clavien score ≤ POD 90, n (%) | |
| Dindo-Clavien 1 | 133 (28.5) |
| Dindo-Clavien 2 | 123 (26.3) |
| Dindo-Clavien 3 | 46 (9.9) |
| Dindo-Clavien 4 | 24 (5.1) |
| Dindo-Clavien 5 | 17 (3.6) |
| Length of hospitalization (days), median [IQR] | 10 [7] |
IQR, interquartile range; POD, postoperative day.
Pathological characteristics
Pathological characteristics are detailed in Table 4. A total of 407 patients (87.2%) had a single lesion and median tumor size was 90 mm (IQR: 90) for the largest nodule. The non-tumoral liver analysis identified mainly F0 (n=237, 50.7%), F1 (n=127, 27.2%) and F2 (n=103, 22.1%) fibrosis according to the “equivalent” METAVIR grading score. A total of 52.2% of the patients had steatosis ≥5%. Among them, the median steatosis rate was 20% (IQR: 20%). More than 80% of the patients had R0 resection (n=391, 83.7%). One quarter of the patients had satellite nodules (n=113, 24.2%) and almost a third had microvascular invasion (n=146, 31.3%). Tumor differentiation was good, moderate and poor in 27.4% (n=128), 45.8% (n=214) and 9.2% (n=43) respectively.
Table 4. Anatomopathological characteristics.
| Characteristics | Study group (n=467) |
|---|---|
| Anatomopathological analysis of the tumor | |
| Number of lesions, n (%) | |
| 1 | 407 (87.2) |
| 2 | 40 (8.6) |
| >2 | 20 (4.3) |
| Maximal tumor size (mm), median [IQR] | 90 [90] |
| Tumor differentiation, n (%) | |
| Poor | 43 (9.2) |
| Moderate | 214 (45.8) |
| Well | 128 (27.4) |
| Missing data | 82 (17.6) |
| R status, n (%) | |
| R0 | 391 (83.7) |
| R1 | 61 (13.1) |
| R2 | 3 (0.6) |
| Missing data | 12 (2.6) |
| Satellite nodules, n (%) | 113 (24.2) |
| Peritumoral capsule, n (%) | 187 (40.0) |
| Microvascular invasion, n (%) | 146 (31.3) |
| Macrovascular invasion, n (%) | 27 (5.8) |
| Perineural invasion, n (%) | 6 (1.3) |
| Anatomopathological analysis of the non-tumoral liver | |
| Normal (A0, F0), n (%) | 146 (31.3) |
| Fibrosis, n (%) | |
| F0 | 237 (50.7) |
| F1 | 127 (27.2) |
| F2 | 103 (22.1) |
| Steatosis, n (%) | 244 (52.2) |
| Steatosis ≥5% (%), median [IQR] | 20 [20] |
| Steatohepatitis | 29 (6.2) |
IQR, interquartile range.
Follow-up
The median patient follow-up was 30.84 months (Q1: 16.08, Q3: 56.16). A total of 209 patients had a recurrence, more than half of whom had a recurrence confined to the liver (51.2%, n=107). A total of 86.6% (n=181/209) of the patients with a recurrence received treatment for their recurrence of which 32.6% (n=59/181) received curative treatment for their recurrence.
Independent prognostic factors of OS (multivariate analysis)
The OS of this study population was 91.8% at one year, 73.9 % at 3 years and of 59.2% at 5 years, with a median OS of 88 months (Figure 1). The prognostic factors of OS included in the multivariate model based on the results of the univariate analysis (Table S2) were two statistically significant pathological prognostic factors: the presence of microvascular invasion (P=0.003) and tumor differentiation (P=0.011) (Table 5).
Figure 1.
Overall survival.
Table 5. Prognostic factors of overall survival (multivariate analysis).
| Hazard ratios | 95% CI | P | |
|---|---|---|---|
| Number of HCC >2 | 1.23 | 0.75–2.02 | 0.416 |
| Maximal HCC size >10 mm | 1.47 | 0.96–2.26 | 0.079 |
| Metabolic syndrome | 0.91 | 0.58–1.41 | 0.662 |
| Preoperative AFP | 1 | 1–1 | 0.101 |
| Steatosis of the non-tumoral liver | 0.83 | 0.56–1.24 | 0.365 |
| Perineural invasion | 1.36 | 0.33–5.61 | 0.675 |
| Satellite nodules | 1.05 | 0.92–1.2 | 0.442 |
| Microvascular invasion | 1.9 | 1.26–2.85 | 0.003* |
| HCC differentiation | 1.89 | 1.17–3.05 | 0.011* |
*, P<0.05. CI, confidence interval; HCC, hepatocellular carcinoma; AFP, alpha-fetoprotein.
Independent prognostic factors of RFS (multivariate analysis)
The RFS of this study population was: 68.7% at one year, 41.5 % at 3 years and of 34.5% at 5 years, with a median RFS of 28 months (Figure 2). The prognostic factors of disease recurrence included in the multivariate model based on the results of the univariate analysis (Table S3) were 4 statistically significant prognostic factors: two preoperative criteria: >2 lesions (P=0.003) and maximal tumor size >10 cm (P<0.001); and two pathological criteria: the presence of microvascular invasion (P=0.004) and tumor differentiation (P=0.008) (Table 6).
Figure 2.
Recurrence-free survival.
Table 6. Prognostic factors of disease recurrence (multivariate analysis).
| Hazard ratios | 95% CI | P | |
|---|---|---|---|
| Number of HCC >2 | 1.71 | 1.21–2.42 | 0.003* |
| Maximal size of HCC >10 cm | 1.99 | 1.46–2.72 | <0.001* |
| Metabolic syndrome | 0.84 | 0.61–1.15 | 0.271 |
| Preoperative AFP | 1 | 1–1 | 0.083 |
| Steatosis of the non-tumoral liver | 0.88 | 0.66–1.17 | 0.377 |
| Perineural invasion | 1.14 | 0.33–4.01 | 0.836 |
| Satellite nodules | 1.07 | 0.97–1.17 | 0.169 |
| Microvascular invasion | 1.64 | 1.18–2.28 | 0.004* |
| HCC differentiation | 1.58 | 1.13–2.2 | 0.008* |
*, P<0.05. CI, confidence interval; HCC, hepatocellular carcinoma; AFP, alpha-fetoprotein.
Prognostication algorithm based on RFS
Based on the prognostic criteria for RFS noted in the multivariate analysis (Table 6), the population was stratified, which provided a prognostication algorithm to help in the therapeutic decision. The population was stratified according to the two statistically significant preoperative prognostic factors of RFS identified in the multivariate analysis: the number of HCC and the size of the largest nodule. Three groups were defined by this algorithm based on preoperative criteria: (I) Group 1 (n=271) included patients with 1 or 2 nodules of less than 10 cm; (II) Group 2 (n=176) included patients with 1 or 2 nodules, the largest of which was greater than 10 cm; (III) Group 3 (n=20) included patients with 3 or more nodules. Group 3 could not be stratified according to the size of the largest nodule because of the small size of this group.
Group 1 had the best prognosis by far, with a 5-year RFS of 52.7% and a median RFS of 46 months. Conversely, in Group 2, when the largest nodule was >10 cm, the RFS was 30.1% at 5 years and the median RFS was 14 months. In Group 3 (≥3 nodules) the RFS at 5 years was 5% and the median RFS was 10 months (Figure 3).
Figure 3.
Recurrence-free survival according to two preoperative criteria: the number of nodules and the size of the largest hepatocellular carcinoma nodule. Group 1 (n=271) included patients with 1 or 2 nodules of less than 10 cm, Group 2 (n=176) included patients with 1 or 2 nodules, the largest of which was greater than 10 cm and Group 3 (n=20) included patients with 3 or more nodules.
In Group 1 with the best survival (n=271, 1 or 2 nodules ≤10 cm), we performed a new stratification using the two statistically significant pathological prognostic factors of RFS identified in the multivariate analysis: the presence of microvascular invasion and the HCC differentiation. Through this stratification, 3 subgroups of patients (Figure 4) were defined. In the subgroup of patients with 1 or 2 well-differentiated nodules of less than 10 cm and without vascular emboli, in group 1a, the RFS was 62.6% at 5 years with a median RFS of 88.8 months. In the absence of invasion with one or two poorly or moderately differentiated nodules, in group 1b, the RFS was 40.0% at 5 years with a median RFS of 33.6 months. Finally, in the presence of microvascular invasion on the specimen, in group 1c, the RFS was 37.8% at 5 years with a median RFS of 25.5 months (Figure 4). The prognostication algorithm is detailed in Figure 5.
Figure 4.
Recurrence-free survival in the subgroups of patients with 1 or 2 nodules of less than 10 cm: Group 1a (n=72) included patients without microvascular invasion with one or two well differentiated nodules, Group 1b (n=115) included patients without microvascular invasion with one or two poorly or moderately differentiated nodules and Group 1c (n=77) included patients with microvascular invasion on the specimen.
Figure 5.
Prognostication algorithm. HCC, hepatocellular carcinoma; RFS, recurrence-free survival.
Discussion
The aims of our study were to develop a prognostication algorithm and to highlight the prognostic factors of RFS and OS for patients treated for strictly non-cirrhotic resectable HCC without underlying viral hepatitis. We excluded patients with advanced fibrosis beyond F2 to exclude the influence of liver disease. Although the prognosis for patients after HCC surgery is known to be partly correlated with the underlying liver disease, only few authors have studied liver resection for HCC in non-cirrhotic liver. These studies were unicentric and included few patients (less than 40 patients) (2,5,6). Therefore, most of the series in the literature on non-cirrhotic HCC were heterogeneous, including patients with underlying liver diseases, such as viral hepatitis (5,7,12-16). The largest cohort included thousands of patients but the study was only conducted on epidemiological criteria, based on the Surveillance, Epidemiology, and End Results (SEER) database, and included patients with underlying liver disease (17). To our knowledge, our study is the largest multicenter cohort of resected non-cirrhotic HCC patients without underlying viral hepatitis.
All the centers in this study were expert liver surgery centers that practice all types of hepatectomy by laparoscopy or laparotomy, including complex liver surgery. The rates of postoperative morbidity and mortality in this study were similar to those in the literature (3). The 5-year RFS and OS of the population were respectively 34.5% and 59.2%, which is similar to the surgical series in the literature (2,7,16-21).
The multivariate analysis identified two anatomopathological prognostic factors of OS and RFS previously described in HCC: tumor differentiation and the presence of microvascular invasion (3,5,21,22). Steatosis in the non-tumoral liver, although present in more than half of patients and metabolic syndrome, present in more than a quarter of patients, did not emerge as a prognostic factor of OS or RFS in the multivariate analysis. In addition, AFP did not emerge as a prognostic factor in this study. AFP is a widely used biomarker in the surveillance of patients at risk for developing HCC and in the surveillance of patients treated for HCC (17,23). AFP rates are correlated to vascular invasion and tumor size. However, 30% to 40% of HCC do not secrete AFP and several authors highlight the lack of consistency as a prognostic factor, reporting limited correlation of AFP rates with disease progression or tumor stage (12,17,24-26). In our study, we could not draw a conclusion on the role of AFP as a prognostic factor for patients with non-cirrhotic HCC without underlying viral hepatitis due to the large quantity of missing AFP data, which is a bias in a retrospective study.
Two preoperative criteria were found to be regular prognostic factors of RFS in the multivariate analysis: the number of HCC and the size of the largest nodule. The number of nodules strictly superior to 2 and the size of the largest nodule strictly superior to 10 cm were prognostic factors of RFS. In the literature, large HCCs have a worse prognosis as they are often correlated to the presence of satellite nodules and microvascular invasion, positive resection margins and differentiation grade (17,27). In patients with non-cirrhotic HCC, median tumor diameter is usually greater than in patients with chronic liver disease and is often a factor of poor prognosis (3,12). The large diameter of non-cirrhotic HCC could be explained by the absence of screening in these patients without a known underlying liver disease (7).
The prognostic factors identified in this study are well known in the literature, but the originality of this study is based on the prognostication algorithm developed from the results. In fact, through the stratification of the population based on prognostic factors of RFS, 3 groups of patients with different prognoses were identified. The two preoperative prognostic factors used for this stratification (number and size of nodules) are known at diagnosis, making it possible to classify non-cirrhotic HCC patients in one of these three different prognostic groups before surgery. Moreover, the prognosis of these patients can be estimated in the preoperative period and therefore be considered in the treatment decision-making.
Through the algorithm built in this study a group of patients with good prognosis were identified, for which the benefit of surgery seems to be the very significant. Indeed, in the presence of one or two non-cirrhotic HCC nodules with the largest nodule diameter ≤10 cm (Group 1), the prognosis is favorable, with a median RFS multiplied by almost 4 compared to the other groups of patients (Groups 2 and 3). Thus, surgery should not be a contraindication in case of case of bifocal HCC on non-cirrhotic liver. Furthermore, in this good prognosis group, the RFS is further improved (up to 7 years) if the HCC is well differentiated and without vascular emboli (Group 1a). These anatomopathological criteria of poor prognosis are currently difficult to obtain preoperatively because only half of the patients in this study had a routine preoperative biopsy. Moreover, when a preoperative biopsy is performed, it is often inefficacious for microvascular invasion and sometimes even for differentiation, which is why some authors do not recommend preoperative biopsy, especially for large resectable HCC (28).
The current trend is increasingly towards neoadjuvant treatment for patients with other solid-organ cancer in order to downstage the malignancy and obtain improved rates of survival. However, the role of (neo)adjuvant therapies in the management of resected HCC is not currently defined (29). Moreover, in case of HCC in patients with NASH, immunotherapy is reported to be ineffective (30). Due to the lack of literature on (neo)adjuvant therapies in HCC patients, current guidelines do not have a clear stance on the use of (neo)adjuvant strategies in resected HCC patients (31,32). However, in the subgroups of patients with moderate prognosis (1 or 2 HCC with or without tumor vascular emboli, poor or moderate differentiation, with the largest nodule diameter ≤10, Groups 1b and 1c), the use of adjuvant therapy could help to decrease the incidence of HCC recurrence after resection. In fact, at a time when more and more systemic treatments are being developed for HCC, particularly immunotherapy treatments, the value of combining surgery with systemic treatment is increasingly interesting, especially in patients with a moderate or poor prognosis, and should be studied in prospective cohorts (33).
Furthermore, based on the results of the algorithm, in case of non-cirrhotic bifocal HCC, contrary to the practice in some centers surgery should not be a contraindication, even if extensive resection is required and especially in the presence of nodules ≤10 cm.
In HCC patients with the poorest prognosis (3 or more nodules or one or two HCC nodule >10 cm, Groups 2 and 3), a perioperative treatment should be explored, combining neoadjuvant and adjuvant liver resection treatments. The two criteria used to define these two groups or patients are two simple preoperative criteria (the number and size of the nodules), which enables multidisciplinary discussion of the use of neoadjuvant therapy prior surgery. In addition, in these groups of patients, the combination of resection and systemic treatment, radioembolization or TACE as perioperative treatment could improve survival outcomes, especially in these patients with non-cirrhotic liver who can tolerate both regional and systemic treatments associated with surgery. This prognostic model requires validation in a prospective cohort to confirm the reproducibility of the results.
Conclusions
In this multicenter retrospective study, RFS analysis enabled the development of a simple algorithm based on two preoperative criteria: the number (≤2 or >2) and size of nodules (≤10 or >10 cm, if less than 2 nodules). This algorithm highlighted distinctive groups of non-cirrhotic HCC patients with different prognoses. It could be used as a treatment decision support for multidisciplinary discussion before surgery concerning the need for perioperative therapy, based on each patient’s estimated prognosis. In case of bifocal HCC, surgery should not be a contraindication. This prognostication algorithm could be the basis of a prospective clinical trial.
Supplementary
The article’s supplementary files as
Acknowledgments
The authors gratefully acknowledge all the centers that participated in this study and the French Association of Hepato-Biliary Surgery and liver Transplantation (ACHBT) group.
Funding: None.
Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). The study was approved by the institutional ethics board of the Toulouse University Hospital (No. RnIPH 2022-67). According to French law on ethics, patients were informed that their codified data would be used for the study.
Footnotes
Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://hbsn.amegroups.com/article/view/10.21037/hbsn-22-33/rc
Data Sharing Statement: Available at https://hbsn.amegroups.com/article/view/10.21037/hbsn-22-33/dss
Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://hbsn.amegroups.com/article/view/10.21037/hbsn-22-33/coif). LB receives grant to the institution from the national transplant agency (agence de la biomédecine) for a separate project, and payment for legal expertise in liability cases. LB participates on national RCT on antibiotics in appendicitis (ABAP study) without payment. LB is a past board member of the French association for the study of liver diseases (AFEF), past board member of the French Surgical Association (association francaise de chirurgie), and a past board member of the French Association of HPB Surgery and Transplantation (ACHBT) with no payment. The other authors have no conflicts of interest to declare.
References
- 1.Bedossa P, Poynard T. An algorithm for the grading of activity in chronic hepatitis C. The METAVIR Cooperative Study Group. Hepatology 1996;24:289-93. 10.1002/hep.510240201 [DOI] [PubMed] [Google Scholar]
- 2.Lubrano J, Huet E, Tsilividis B, et al. Long-term outcome of liver resection for hepatocellular carcinoma in noncirrhotic nonfibrotic liver with no viral hepatitis or alcohol abuse. World J Surg 2008;32:104-9. 10.1007/s00268-007-9291-0 [DOI] [PubMed] [Google Scholar]
- 3.Desai A, Sandhu S, Lai JP, et al. Hepatocellular carcinoma in non-cirrhotic liver: A comprehensive review. World J Hepatol 2019;11:1-18. 10.4254/wjh.v11.i1.1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Liu Y, Li H, Ye N, et al. Non-Cirrhotic Liver is Associated with Poor Prognosis of Hepatocellular Carcinoma: A Literature Review. Med Sci Monit 2019;25:6615-23. 10.12659/MSM.915722 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Lang H, Sotiropoulos GC, Dömland M, et al. Liver resection for hepatocellular carcinoma in non-cirrhotic liver without underlying viral hepatitis. Br J Surg 2005;92:198-202. 10.1002/bjs.4763 [DOI] [PubMed] [Google Scholar]
- 6.Yokoi Y, Suzuki S, Baba S, et al. Clinicopathological features of hepatocellular carcinomas (HCCs) arising in patients without chronic viral infection or alcohol abuse: a retrospective study of patients undergoing hepatic resection. J Gastroenterol 2005;40:274-82. 10.1007/s00535-004-1536-1 [DOI] [PubMed] [Google Scholar]
- 7.Laurent C, Blanc JF, Nobili S, et al. Prognostic factors and longterm survival after hepatic resection for hepatocellular carcinoma originating from noncirrhotic liver. J Am Coll Surg 2005;201:656-62. 10.1016/j.jamcollsurg.2005.05.027 [DOI] [PubMed] [Google Scholar]
- 8.Dindo D, Demartines N, Clavien PA. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg 2004;240:205-13. 10.1097/01.sla.0000133083.54934.ae [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Balzan S, Belghiti J, Farges O, et al. The "50-50 criteria" on postoperative day 5: an accurate predictor of liver failure and death after hepatectomy. Ann Surg 2005;242:824-8, discussion 828-9. 10.1097/01.sla.0000189131.90876.9e [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Hothorn T, Hornik K, Zeileis A. Unbiased Recursive Partitioning: A Conditional Inference Framework. J Comput Graph Stat 2006;15:651-74. 10.1198/106186006X133933 [DOI] [Google Scholar]
- 11.Hothorn T, Zeileis A. Partykit: a modular toolkit for recursive partytioning in R. J Mach Learn Res 2015;16:3905-9. [Google Scholar]
- 12.Giannini EG, Marenco S, Bruzzone L, et al. Hepatocellular carcinoma in patients without cirrhosis in Italy. Dig Liver Dis 2013;45:164-9. 10.1016/j.dld.2012.08.018 [DOI] [PubMed] [Google Scholar]
- 13.Pandey D, Lee KH, Wai CT, et al. Long term outcome and prognostic factors for large hepatocellular carcinoma (10 cm or more) after surgical resection. Ann Surg Oncol 2007;14:2817-23. 10.1245/s10434-007-9518-1 [DOI] [PubMed] [Google Scholar]
- 14.Schütte K, Schulz C, Poranzke J, et al. Characterization and prognosis of patients with hepatocellular carcinoma (HCC) in the non-cirrhotic liver. BMC Gastroenterol 2014;14:117. 10.1186/1471-230X-14-117 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Wörns MA, Bosslet T, Victor A, et al. Prognostic factors and outcomes of patients with hepatocellular carcinoma in non-cirrhotic liver. Scand J Gastroenterol 2012;47:718-28. 10.3109/00365521.2012.677952 [DOI] [PubMed] [Google Scholar]
- 16.Nagasue N, Ono T, Yamanoi A, et al. Prognostic factors and survival after hepatic resection for hepatocellular carcinoma without cirrhosis. Br J Surg 2001;88:515-22. 10.1046/j.1365-2168.2001.01732.x [DOI] [PubMed] [Google Scholar]
- 17.Dasari BV, Kamarajah SK, Hodson J, et al. Development and validation of a risk score to predict the overall survival following surgical resection of hepatocellular carcinoma in non-cirrhotic liver. HPB (Oxford) 2020;22:383-90. 10.1016/j.hpb.2019.07.007 [DOI] [PubMed] [Google Scholar]
- 18.Bismuth H, Chiche L, Castaing D. Surgical treatment of hepatocellular carcinomas in noncirrhotic liver: experience with 68 liver resections. World J Surg 1995;19:35-41. 10.1007/BF00316977 [DOI] [PubMed] [Google Scholar]
- 19.Chang CH, Chau GY, Lui WY, et al. Long-term results of hepatic resection for hepatocellular carcinoma originating from the noncirrhotic liver. Arch Surg 2004;139:320-5; discussion 326. 10.1001/archsurg.139.3.320 [DOI] [PubMed] [Google Scholar]
- 20.Xu L, Huang L, Li BK, et al. Clinicopathologic features and long-term outcomes of Chinese patients with hepatocellular carcinoma in non-cirrhotic liver. Dig Surg 2008;25:376-82. 10.1159/000170881 [DOI] [PubMed] [Google Scholar]
- 21.Arnaoutakis DJ, Mavros MN, Shen F, et al. Recurrence patterns and prognostic factors in patients with hepatocellular carcinoma in noncirrhotic liver: a multi-institutional analysis. Ann Surg Oncol 2014;21:147-54. 10.1245/s10434-013-3211-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Bège T, Le Treut YP, Hardwigsen J, et al. Prognostic factors after resection for hepatocellular carcinoma in nonfibrotic or moderately fibrotic liver. A 116-case European series. J Gastrointest Surg 2007;11:619-25. 10.1007/s11605-006-0023-9 [DOI] [PubMed] [Google Scholar]
- 23.Witjes CD, Polak WG, Verhoef C, et al. Increased alpha-fetoprotein serum level is predictive for survival and recurrence of hepatocellular carcinoma in non-cirrhotic livers. Dig Surg 2012;29:522-8. 10.1159/000348669 [DOI] [PubMed] [Google Scholar]
- 24.Burnett NP, Dunki-Jacobs EM, Callender GG, et al. Evaluation of alpha-fetoprotein staging system for hepatocellular carcinoma in noncirrhotic patients. Am Surg 2013;79:716-22. 10.1177/000313481307900717 [DOI] [PubMed] [Google Scholar]
- 25.Wang Y, Chen Y, Ge N, et al. Prognostic significance of alpha-fetoprotein status in the outcome of hepatocellular carcinoma after treatment of transarterial chemoembolization. Ann Surg Oncol 2012;19:3540-6. 10.1245/s10434-012-2368-5 [DOI] [PubMed] [Google Scholar]
- 26.Di Martino M, Saba L, Bosco S, et al. Hepatocellular carcinoma (HCC) in non-cirrhotic liver: clinical, radiological and pathological findings. Eur Radiol 2014;24:1446-54. 10.1007/s00330-014-3173-2 [DOI] [PubMed] [Google Scholar]
- 27.Arii S, Yamaoka Y, Futagawa S, et al. Results of surgical and nonsurgical treatment for small-sized hepatocellular carcinomas: a retrospective and nationwide survey in Japan. The Liver Cancer Study Group of Japan. Hepatology 2000;32:1224-9. 10.1053/jhep.2000.20456 [DOI] [PubMed] [Google Scholar]
- 28.Young AL, Malik HZ, Abu-Hilal M, et al. Large hepatocellular carcinoma: time to stop preoperative biopsy. J Am Coll Surg 2007;205:453-62. 10.1016/j.jamcollsurg.2007.04.033 [DOI] [PubMed] [Google Scholar]
- 29.Akateh C, Black SM, Conteh L, et al. Neoadjuvant and adjuvant treatment strategies for hepatocellular carcinoma. World J Gastroenterol 2019;25:3704-21. 10.3748/wjg.v25.i28.3704 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Pfister D, Núñez NG, Pinyol R, et al. NASH limits anti-tumour surveillance in immunotherapy-treated HCC. Nature 2021;592:450-6. 10.1038/s41586-021-03362-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.European Association for the Study of the Liver . Electronic address: easloffice@easloffice.eu; European Association for the Study of the Liver. EASL Clinical Practice Guidelines: Management of hepatocellular carcinoma. J Hepatol 2018;69:182-236. 10.1016/j.jhep.2018.03.019 [DOI] [PubMed] [Google Scholar]
- 32.Benson AB, D'Angelica MI, Abbott DE, et al. Hepatobiliary Cancers, Version 2.2021, NCCN Clinical Practice Guidelines in Oncology. J Natl Compr Canc Netw 2021;19:541-65. 10.6004/jnccn.2021.0022 [DOI] [PubMed] [Google Scholar]
- 33.Finn RS, Qin S, Ikeda M, et al. Atezolizumab plus Bevacizumab in Unresectable Hepatocellular Carcinoma. N Engl J Med 2020;382:1894-905. 10.1056/NEJMoa1915745 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
The article’s supplementary files as