Large hepatocellular carcinoma: Does fibrosis really impact prognosis after resection?

+Model

ARTICLE IN PRESS

JVS-757; No. of Pages 9

Journal of Visceral Surgery (2018) xxx, xxx—xxx

Available online at

ScienceDirect www.sciencedirect.com

ORIGINAL ARTICLE

Large hepatocellular carcinoma: Does fibrosis really impact prognosis after resection? N. Golse a,∗, A. El Bouyousfi b, F. Marques a, B. Bancel c, K. Mohkam b, C. Ducerf b, P. Merle d, M. Sebagh e, D. Castaing a, A. Sa Cunha a, R. Adam a, D. Cherqui a, E. Vibert a, J.-Y. Mabrut b a

Liver Transplantation and Hepato-Biliary Surgery, centre hépato-biliaire, hôpital Paul-Brousse, université Paris-Sud, Assistance publique—Hôpitaux Paris, 12, avenue Paul-Vaillant-Couturier, 94800 Villejuif, France b Hospices civils de Lyon, Digestive Surgery and Liver Transplant Department, Croix-Rousse Hospital, 69004 Lyon, France c Hospices Civils de Lyon, Department of Pathology, Croix-Rousse Hospital, 69004 Lyon, France d Hospices Civils de Lyon, Hepatology and Gastroenterology Department, Croix-Rousse Hospital, 69004 Lyon, France e Department of Pathology, centre hépato-biliaire, hôpital Paul-Brousse, université Paris-Sud, Assistance publique—Hôpitaux Paris, 94800 Villejuif, France

KEYWORDS Hepatocellular carcinoma; Fibrosis; Cirrhosis; Non-cirrhotic; Prognostic factors; Hepatectomy

Summary Background: Hepatectomy remains the standard treatment for large hepatocellular carcinoma (LHCC) ≥ 5 cm. Fibrosis may constitute a contraindication for resection because of high risk of post-hepatectomy liver failure, but its impact on patient outcome and cancer recurrence remains ill defined. Our aim was to compare predictors of survival in patients with and without cirrhosis following hepatectomy for LHCC. Methods: The data on consecutive patients undergoing hepatectomy for LHCC in two tertiary centres between 2012 and 2016 were reviewed. The outcomes of cirrhotic (F4) and non-cirrhotic (F0—F3) patients were compared. Patients with perioperative medical (sorafenib) or radiological (transarterial chemoembolization, radiofrequency) treatments were excluded. Results: Sixty patients were included. Preoperative and intraoperative features were identical between both groups. Cirrhotics (n = 15) presented more satellite nodules on specimens (73% vs. 44%; P = 0.073) but better differentiated lesions than non-cirrhotics (P = 0.041). The median overall survival of cirrhotics was 34 vs. 29 months for non-cirrhotics (P = 0.8), and their diseasefree survival was 14 versus 18 months (P = 0.9). Fibrosis stage did not impact overall (P = 0.2) nor disease-free survivals (P = 0.6).

Abbreviations: AFP, alpha-fetoprotein; BMI, body mass index; CL, cirrhotic liver; DFS, disease-free survival; HCC, hepatocellular carcinoma; Hx, hepatectomy; ICG, indocyanine green retention; LHCC, large hepatocellular carcinoma; NCL, non-cirrhotic liver; OS, overall survival; SD, standard deviation; TACE, transarterial chemoembolization. ∗ Corresponding author. E-mail address: [email protected] (N. Golse). https://doi.org/10.1016/j.jviscsurg.2017.10.015 1878-7886/© 2017 Elsevier Masson SAS. All rights reserved.

Please cite this article in press as: Golse N, et al. Large hepatocellular carcinoma: Does fibrosis really impact prognosis after resection? Journal of Visceral Surgery (2018), https://doi.org/10.1016/j.jviscsurg.2017.10.015

+Model JVS-757; No. of Pages 9

ARTICLE IN PRESS

2

N. Golse et al. Conclusion: Hepatectomy for LHCC in cirrhotics can achieve acceptable oncological results when compared to non-cirrhotic patients. Curative resection of LHCC should be attempted if liver function is acceptable, whatever the fibrosis stage. © 2017 Elsevier Masson SAS. All rights reserved.

Introduction Hepatocellular carcinoma (HCC) is the most prevalent primary liver cancer, the sixth most common cancer worldwide and the third leading cause of cancer-related mortality [1,2]. About 80—90% of all HCC occur on underlying liver pathology, mainly cirrhosis. Others may develop on healthy (or slightly fibrotic) livers. In this situation, metabolic syndrome and non-alcoholic fatty liver disease or steatohepatitis are growing risk factors for HCC in Western countries, even without cirrhosis [3—7], so that the concept of HCC on healthy liver is being replaced by that of HCC in non-cirrhotic livers (NCL). Theses tumours require specific management because they are often large, symptomatic and affect young patients. Liver transplantation (LT) has little place in this setting since major hepatectomy (Hx) can be proposed with a low risk of postoperative liver failure when compared to cirrhotics [8—10]. Many prognostic factors have been reported to impact survival after the resection of HCC on NCL, mainly age, alpha-fetoprotein [AFP] level, histological grade, multiple tumours, satellite nodules, resection margins, vascular invasion and adjuvant therapy [11—13]. Tumour size is a well-recognised prognostic factor [14—20] and the 5 cm cut-off point is nowadays widely accepted and validated in the AJCC staging [21—27]. Tumours larger than 5 cm have an increased risk of satellite nodules and vascular invasion, and are usually less differentiated [13,23,28,29]. The surgical management of such high-risk tumours therefore requires a strict selection of patients without extra-hepatic spreading. It is largely accepted that cirrhosis is an independent negative risk factor for survival after the resection of large HCC [LHCC] [29—35], but publications often contain bias because of pathological and clinical differences between groups, mainly concerning tumour size, vascular invasion, encapsulation, gender and age. Moreover, some authors have not demonstrated any negative impact of severe fibrosis on overall (OS) and/or disease-free survival [DFS] [15,16,27,36—39]. Finally, specific analysis of the influence of parenchymal fibrosis on the survival and the risk of recurrence has not been widely studied with respect to exclusive large HCC resected either from NCL or strictly cirrhotic liver (CL), because the principal cohorts reported on a mixed population including both cirrhotic and non-cirrhotic patients carrying small and large lesions [40—44]. The aim of this study was therefore to assess, in recent Western populations, the perioperative and long-term outcomes of patients with and without cirrhosis following hepatectomy for a LHCC > 5 cm. The impact of moderate or intermediate fibrosis stage was specifically evaluated.

Patients and methods Selection of patients The experience of two tertiary centres (Lyon — HepatoBiliary Unit, Croix-Rousse hospital and Villejuif — HepatoBiliary Centre, Paul-Brousse hospital) was reviewed from

a prospective secure website database from January 2012 (opening date of this National Observatory website) to January 2016. The inclusion criteria was the presence of at least one large HCC (≥ 5 cm) on pathological specimen. Repeat hepatectomies and macrovascular invasion were excluded, as well as patients with neoadjuvant therapy (transarterial chemoembolisation [TACE] or sorafenib) and those with intraoperative ablation (radiofrequency) in order to specifically assess oncological impact of surgery. Data of 30 consecutive patients who underwent a resection for LHCC in Lyon were collected and pooled with the 56 consecutive patients resected in Villejuif. According to these pre-cited criteria, sixty patients were finally included and analysed. We compared patients with non-cirrhotic liver (F0—F3 fibrosis, HCC-NCL group, n = 45) to those with underlying cirrhosis (F4 fibrosis, HCC-CL group, n = 15), and evaluated prognostic factors for survival in each group.

Preoperative evaluation Both the diagnosis of HCC and treatment strategy were implemented according to the EASL and BCLC algorithms, except hepatic venous pressure gradient not measured [2]. Liver function was assessed by a combination of MELD score, liver biological tests, and more recently with plasma indocyanine green retention clearance (ICG). Child-Pugh B status was a contraindication to extensive surgery unless the ICG test was normal (< 15% at 15 min). A limited resection was performed if the ICG test was > 20%. Exclusion criteria for resection were extra-hepatic spread or decompensated cirrhosis.

Surgical procedure Laparoscopic resections were performed according to surgeon’s preferences. An intraoperative ultrasound investigation was performed systematically for parenchyma and vascular assessment. Parenchymal transection was performed using an ultrasonic device and/or dipolar electric coagulation. Curative resection was planned for each patient (anatomical resection or margins ≥ 2 cm, if possible [45]). An intermittent Pringle manœuvre was performed when necessary (15 min/5 min). Lymphadenectomy was not performed systematically and was not evaluated in this study. No patient with ablative therapy (exclusive or associated treatment) was included in this series.

Postoperative management All patients received regular clinical follow-up that included the determination of serial alpha-fetoprotein levels and ultrasonography or a CT scan, every 3 months. In the event of a recurrence (defined as the appearance of a new lesion with radiological features compatible with HCC), treatment modalities were proposed in accordance with the decision of a multidisciplinary team, and included salvage LT, surgical iterative resection, ablative

Please cite this article in press as: Golse N, et al. Large hepatocellular carcinoma: Does fibrosis really impact prognosis after resection? Journal of Visceral Surgery (2018), https://doi.org/10.1016/j.jviscsurg.2017.10.015

+Model JVS-757; No. of Pages 9

ARTICLE IN PRESS

Fibrosis does not impact prognosis therapy (radiofrequency or cryotherapy), TACE, radiotherapy, systemic chemotherapy (sorafenib) or supportive care.

Definitions Major hepatectomy was defined as the resection of ≥ 3 segments according to the Couinaud classification. Tumour size was defined as the largest diameter of the tumour specimen. Tumour grade was assessed using the nuclear grading scheme proposed by Edmonson and Steiner [46]. The degree of fibrosis was determined using the METAVIR score [47]. The HCC-NCL group included patients with fibrosis F0—F3, while the HCC-CL group included F4 patients. Histopathology and METAVIR score were reviewed independently and blindly to outcomes.

Statistical analysis Continuous data were presented as mean ± standard deviation (SD). Comparisons between groups were made with Student’s t-test for continuous variables (or Mann—Whitney Wilcoxon for non-parametric data) and Fisher’s exact test for categorical variables. Overall survival (OS) was calculated from the date of resection to the last visit or loss to follow-up (or death). Disease-free survival was calculated from the date of resection to the first multidisciplinary confirmation of recurrence (evocative imaging without biopsy required). Survival was analysed according to the Kaplan—Meier method and compared using the log rank test. A Cox proportional hazards regression model was used to identify factors independently associated with OS or DFS. Variables achieving a P-value < 0.1 were included in a multivariate Cox regression analysis. A P-value of ≤ 0.05 was considered to be significant. All statistical analyses were performed using SPSS version 13.0 software (SPSS Inc., Chicago, IL).

3

Entire cohort, postoperative data The median follow-up of this cohort was of 20.8 months [0—52] (Figs. 1—3). The median OS reached 33.6 ± 3 months and median DFS was of 16.2 ± 4 months. One and 3-year overall survival rates were of 87.4% and 41.9% respectively, while 1- and 3-year DFS rates were of 58% and 34.2%, respectively. A comparison of OS and DFS between HCC-NCL and HCC-CL patients revealed no statistical difference (29.3 vs. 33.6 months; P = 0.825, and 17.8 vs. 13.7 months; P = 0.918, respectively). By comparing the survivals of patients with low (F0—F2) versus severe (F3—F4) fibrosis, we could not demonstrate any significant difference with respect to OS (P = 0.145) or DFS (P = 0.522). Median OS was better in highly fibrotic patients (52.3 months if fibrosis ≥ F3 vs. 27.9 months if fibrosis < F3). Patients in both groups experienced the same rate of recurrence (53% vs 47%; P = 0.77). A recurrence occurred in the liver only in 67% vs. 86% of cases (P = 0.64) and the same percentage of patients benefited from iterative surgery. We observed no statistical difference for postoperative morbidity or death.

Entire cohort, prognostic factors for survival A univariate analysis of prognostic factors for OS showed that positive resection margins (R1 or R2 status) exerted a negative impact on survival (P = 0.013) (Table 2). Univariate analysis of the predictors for DFS demonstrated that microvascular invasion, satellite nodules, poor differentiation grade, and AFP level had a negative impact on DFS. Microvascular invasion, satellite nodules and AFP levels were associated with poorer DFS under multivariate analysis (P = 0.007, P = 0.06 and P = 0.06 respectively). The degree of fibrosis (F0/F1/F2/F3/F4) was not significant with respect to either overall survival (P = 0.20) or DFS (P = 0.62).

Results

HCC-NCL group, prognostic factors for survival

Entire cohort, preoperative data

Univariate analysis showed that age (P = 0.1), high BMI (P = 0.085), MELD (P = 0.01) and non-R0 resection (P = 0.01) negatively impacted OS (Table 3). These factors remained statistically significant under multivariate analysis. The fibrosis stage (P = 0.145) had no influence on survival. Univariate analysis of the prognostic factors for DFS demonstrated that microvascular invasion had a negative impact on DFS (P = 0.002).

Clinical and epidemiological comparisons between the two groups did not reveal any differences with respect to age, gender, body mass index (BMI), AFP level, MELD score, incidence of diabetes, or hepatitis B infections (Table 1). In the HCC-CL group, patients presented a trend toward higher incidence of hepatitis C virus-related hepatopathy (27% vs. 9%; P = 0.098).

Entire cohort, intraoperative data Laparoscopic approach was more frequent in the HCC-CL group (27% vs. 9%, P = 0.098). The number of resected segments was slightly higher in the HCC-NCL group (2.8 vs. 2.2; P = 0.13) but the rate of major Hx was identical between both groups (P = 1). There was no difference regarding the maximal tumor size (P = 0.36), extent of resection margins (P = 0.44), the R status (P = 0.48), encapsulation (P = 0.35) nor microvascular invasion (P = 0.77). In the HCC-CL group, lesions presented with more satellite nodules (or multiple lesions) than in the HCC-NCL group (73% vs. 44%; P = 0.07) but lesions were more frequently well differentiated (P = 0.04).

Secondary analysis including patients with perioperative non-surgical treatment A secondary analysis assessing all patients resected for LHCC from 2007 in the Croix-Rousse Center (Lyon) and 2012 in Paul Brousse, including those with neoadjuvant therapy (TACE) or sorafenib and those with intraoperative ablation (radiofrequency) was also performed in order to evaluate the ‘‘real life’’ management of such patients (onco-surgical management) and to present a larger experience. One hundred and twenty two patients were analysed. Cirrhotic patients (n = 39) were more frequently HCV-infected (28% vs. 10%, P = 0.014), presented smaller tumor size (7.7 cm vs. 9.9 cm, P = 0.01) and more satellite nodules on specimens (68% vs. 44%; P = 0.018) than non-cirrhotics (n = 83).

Please cite this article in press as: Golse N, et al. Large hepatocellular carcinoma: Does fibrosis really impact prognosis after resection? Journal of Visceral Surgery (2018), https://doi.org/10.1016/j.jviscsurg.2017.10.015

+Model JVS-757; No. of Pages 9

ARTICLE IN PRESS

4

N. Golse et al. Table 1

Comparison of perioperative features in the two groups.

Preoperative features Age Gender M/F (%) BMI (kg/m2 ) AFP MELD Platelets (G/L) Diabetes (%) HBV (%) HCV (%) Alcohol consumption (%) Preoperative portal vein embolization (%) Intraoperative and pathological features Laparoscopic resection (%) Number of resected segments Major hepatectomy (%) Transfusion Y/N (%) Operative time (min) Pedicle clamping (%) Tumour size (cm) Tumour size ≥ 10 cm (n, %) Resection margins (mm) R0/R1/R2 (%) F0/F1/F2/F3 (%) Encapsulation (intact capsule) (%) Differentiation grade (well/moderate/poor) (%) Microvascular invasion (%) Satellite nodules (%) Postoperative features Length of stay (days) Clavien ≥ III (%) Clavien V (%) HCC recurrence (%) Exclusively intra-hepatic recurrence (%) Surgical management of recurrence (%) LT after recurrence (n) Mean follow-up (months) Median patient survival (months) Median disease-free survival (months)

F0—F3 n = 45

F4 n = 15

P-value

66.6 ± 12 82/18 26.3 ± 4.4 1107 ± 3346 7.8 ± 2.2 253 ± 97 20 13.3 8.9 14.7 4 (9)

64.1 ± 10 67/33 26.7 ± 5 1477 ± 3212 7.7 ± 2.1 188 ± 90 26.7 6.7 26.7 26.7 2 (13)

0.397 0.279 0.740 0.150 0.943 0.072 0.719 0.668 0.098 0.431 0.634

8.9 2.8 ± 1.2 40 15/85 254 ± 100 62.5 9.5 ± 4.5 16 (35.6) 4.9 ± 5.9 70/25/5 9/9/40/42 34.1 16/39/45 46.5 44.2

26.7 2.2 ± 1.1 44.4 14/86 276 ± 84 72.7 7.5 ± 1.6 2 (13.3) 5.1 ± 10.4 53/40/7 — 25 46/39/15 53.3 73.3

0.098 0.133 1 1 0.352 0.726 0.360 0.191 0.441 0.480 — 0.354 0.041 0.767 0.073

15.5 ± 12.7 20 8.9 53.3 66.7 12.5 0 22 29.3 ± 3.4 17.8 ± 5.9

14.6 ± 8.7 6.7 6.7 46.7 85.7 14.3 1 18.1 33.6 ± 10.3 13.7 ± 1.7

0.949 0.426 1 0.768 0.639 1 — 0.186 0.825 0.918

Values are expressed as mean ± standard deviation. AFP: alphafeto protein; BMI: body mass index; F: fibrosis stage; HBV: hepatitis B virus; HCV: hepatitis C virus; LT: liver transplantation; M/F: male/female; R: resection margins.

The median overall survival of non-cirrhotic patients was 63 vs. 52 months for cirrhotic patients (P = 0.253), and their disease-free survival was 18 versus 16 months (P = 0.592). After a recurrence, repeat hepatectomy or salvage transplantation rates were statistically identical (P = 1). Fibrosis stage did not impact survivals. Under multivariate analysis, non-radical (≥ R1) resection was associated with poor survival, while vascular invasion and satellite nodules were predictors of poor disease-free survival (data not shown).

resection. This study also demonstrates that the degree of fibrosis has no impact on OS and DFS after resection of LHCC ≥ 5 cm: cirrhotic and non-cirrhotic patients achieved the same median survival, and even within the HCC-NCL group, fibrosis stage (F0 to F3 METAVIR Score) did not impact survival. Another important finding of this work was the identical pattern of recurrence after curative resection in both groups (interval, localization) and the same profile of treatment with curative intent, namely liver transplantation or repeat hepatectomy.

Discussion Statement of principal findings

Strengths and weaknesses of the study

Many known risk factors of OS and DFS after resection of LHCC were confirmed in our series, mainly related to histopathological features and the completeness of

The principal strength of this study was that we collected data from two Western transplant teams that covered a recent study period involving homogeneous practices. In

Please cite this article in press as: Golse N, et al. Large hepatocellular carcinoma: Does fibrosis really impact prognosis after resection? Journal of Visceral Surgery (2018), https://doi.org/10.1016/j.jviscsurg.2017.10.015

+Model JVS-757; No. of Pages 9

ARTICLE IN PRESS

Fibrosis does not impact prognosis

5

Figure 1. Overall survival in the whole cohort for each fibrosis stage. The degree of fibrosis has not impact on overall survival.

Table 2

spite of the limited cohort size, this bicentric study is one of the largest European cohort reporting management and prognostic factors of non-cirrhotic LHCC. In many reports, cirrhotic patients often display epidemiological and pathological differences from non-cirrhotic groups [17,32,36,48]. Our series presented comparable groups in witch almost all preoperative, intraoperative and pathological data are identical, except the rate of satellite nodules, higher in the cirrhotic group (73% vs. 44%, P = 0.07), and the differentiation grade (tumours better differentiated in the HCC-CL group, P = 0.04). A propensity score matching could have prevented these biases, but this was not realistic due to the sample size. Our aim was to answer to an unsolved question (the influence of fibrosis on survival following the resection of LHCC) and to evaluate prognostic factors at different fibrotic stages. Preoperative, pathological and surgical data relative to the HCC-NCL and HCC-CL groups were largely identical, allowing us to compare these groups. The principal limitation was the sample size. This reflects the scarcity of resected LHCC in our daily practice. Our present series could have been larger if we had also included patients with neoadjuvant therapy. In fact, we excluded 23 patients preoperatively treated (19 TACE, 3 sorafenib and 1 radiofrequency) because most of them presented highly

Prognostic factors for overall and disease-free survivals in the whole cohort. Univariate

Overall survival Age Female Diabetes BMI AFP MELD Major Hx Tumour size Resection margins Resection R0/R1/R2 Encapsulation Microvascular invasion Satellite nodules Differentiation grade Fibrosis stage (F0/F1/F2/F3/F4) Cirrhosis Disease-free survival Age Female Diabetes BMI AFP MELD Major Hx Tumour size Resection margins Resection R0/R1/R2 Encapsulation Microvascular invasion Satellite nodules Differentiation grade Fibrosis stage (F0/F1/F2/F3/F4) Cirrhosis

Multivariate

P-value

HR (95% CI)

P-value

0.157 0.212 0.605 0.175 0.564 0.121 0.776 0.358 0.080 0.013 0.430 0.337 0.959 0.906 0.203 0.825

1.031 1.806 1.273 0.936 0.99 1.167 1.130 1.041 0.875 2.280 0.682 1.542 1.023 0.964 0.795 0.885

(0.998—1.076) (0.704—4.632) (0.511—3.172) (0.851—1.030) (0.972—1.002) (0.960—1.418) (0.487—2.622) (0.995—1.136) (0.755—1.016) (1.188—4.374) (0.264—1.764) (0.637—3.733) (0.433—2.417) (0.530—1.755) (0.558—1.132) (0.298—2.627)

Not tested

0.730 0.319 0.381 0.830 0.100 0.552 0.487 0.195 0.138 0.317 0.906 < 0.001 0.052 0.090 0.624 0.918

1.005 1.309 0.684 1.008 0.954 0.941 1.292 1.054 0.940 1.383 0.955 5.685 2.065 1.567 0.923 1.048

(0.976—1.036) (0.771—2.223) (0.293—1.600) (0.937—1.084) (0.540—1.760) (0.769—1.150) (0.627—2.662) (0.973—1.142) (0.866—1.020) (0.733—2.607) (0.443—2.056) (2.363—13.680) (0.979—4.353) (0.919—2.673) (0.669—1.272) (0.427—2.575)

HR (95% CI)

0.060

0.953 (0.863—1.020)

0.007 0.061 0.665

3.942 (1.456—10.673) 2.548 (0.960—6.433) 1.141 (0.629—2.070)

Please cite this article in press as: Golse N, et al. Large hepatocellular carcinoma: Does fibrosis really impact prognosis after resection? Journal of Visceral Surgery (2018), https://doi.org/10.1016/j.jviscsurg.2017.10.015

+Model JVS-757; No. of Pages 9

ARTICLE IN PRESS

6

N. Golse et al.

Figure 2. Overall survival in the whole cohort according to the presence of cirrhosis. The presence of cirrhosis has not impact on overall survival.

Table 3

Figure 3. Disease-free survival in the whole cohort according to the presence of cirrhosis. The presence of cirrhosis has not impact on overall survival.

Prognostic factors for overall and disease-free survivals in the HCC-NCL group. Univariate P-value

Overall survival Age Female Diabetes BMI AFP MELD Major Hx Tumour size Resection margins Resection R0/R1/R2 Encapsulation Microvascular invasion Satellite nodules Differentiation grade Fibrosis stage (F0/F1/F2/F3) Disease-free survival Gender BMI Age Diabetes AFP MELD Major Hx Tumour size Resection margins Resection R0/R1/R2 Encapsulation Microvascular invasion Satellite nodules Differentiation grade Fibrosis stage F0/F1/F2/F3

Multivariate HR (95% CI)

P-value

HR (95% CI)

0.100 0.722 0.685 0.085 0.884 0.011 0.920 0.303 0.110 0.014 0.464 0.330 0.853 0.897 0.145

1.040 1.253 1.228 0.901 1.360 1.308 1.049 1.048 0.878 2.416 0.677 1.617 0.910 1.046 0.722

(0.992—1.090) (0.361—4.353) (0.456—3.307) (0.800—1.015) (0.287—3.877) (1.063—1.610) (0.413—2.668) (0.959—1.145) (0.748—1.030) (1.197—4.874) (0.238—1.925) (0.615—4.255) (0.336—2.466) (0.532—2.056) (0.466—1.119)

0.023

1.074 (1.012—1.139)

0.058

0.869 (0.791—1.004)

0.034

1.295 (1.041—1.611)

0.004

3.543 (1.500—8.370)

0.750 0.245 0.788 0.448 0.115 0.802 0.651 0.188 0.126 0.264 0.661 0.002 0.150 0.197 0.505

1.111 1.050 1.005 0.698 0.754 1.031 1.205 1.057 0.929 1.478 0.825 4.328 1.833 1.483 0.870

(0.583—2.116) (0.967—1.139) (0.972—1.038) (0.275—1.769) (0.870—1.345) (0.814—1.305) (0.537—2.703) (0.973—1.149) (0.845—1.021) (0.745—2.932) (0.349—1.951) (1.739—10.773) (0.803—4.187) (0.815—2.699) (0.579—1.309)

Not tested

Please cite this article in press as: Golse N, et al. Large hepatocellular carcinoma: Does fibrosis really impact prognosis after resection? Journal of Visceral Surgery (2018), https://doi.org/10.1016/j.jviscsurg.2017.10.015

+Model JVS-757; No. of Pages 9

ARTICLE IN PRESS

Fibrosis does not impact prognosis fibrotic liver and their inclusion may have biased the results. Conversely, our aimed was to assess the impact of surgical management (alone) in LHCC-bearing patients. For the same reason, we also excluded intraoperative radiofrequency procedures (n = 3) associated to resection. Despite of this, we were able to identify some of the most classical prognostic factors, therefore validating the robustness of the entire statistical analysis. The secondary analysis briefly presented at the end of the ‘‘results section’’ confirms all the detailed results of the main cohort and reinforces our message. Another limitation was that we did not include nonresected patients (surgical cohort only) so that we could not definitively assess the impact of fibrosis on survival in the event of non-surgical management. Particularly, F4 patients bearing LHCC with poor liver function were not resected and not assessed.

Interpretation with reference to other studies Evaluating the impact of the degree of fibrosis on postoperative outcomes following resection of LHCC is a crucial question to determine if the same strategies must be applied whatever the underlying fibrosis level (Table 4). However, this evaluation remains challenging, and a review of the literature does not provide a definitive answer to the question. In fact, most authors stratified their cohorts according to tumour size rather than the degree of fibrosis. It seems necessary to evaluate two situations separately: the impact of cirrhosis on OS and DFS (vs. non-cirrhotic patients) and the impact of progressive fibrosis (F0 to F3) within the non-cirrhotic group. Published series have often displayed bias between the studied groups, and it seems difficult to draw any definitive conclusions. Some authors compared the influence of fibrosis on non-cirrhotic liver, but included heterogeneous tumour sizes. Dupont et al. [12] and Wang et al. [49] did not find any differences between F0—F1/F2—F3 groups, but they included both small and large HCC. Shrager et al. reported the same survival rates in F0—F2 vs. F3—F4 patients, but this group presented with significantly smaller tumour (5.7 vs. 8.2, P < 0.01) and they also included small ones. Surprisingly, F2 patients achieved a longer survival than F0—F1 patients (90 vs. 39 months, P = 0.02) [16]. Truant et al. (HCC > 8 cm) reported that F0 patients achieved better survival than F1—F3 patients, but the DFS rates were identical (P = 0.11) and the influence of the fibrosis stage was not clearly evaluated [13]. Fong et al. [18] did not find any difference in survival between cirrhotic and non-cirrhotic groups (P = 0.09), but this farmer population presented with larger tumours (median: 10 cm vs. 6 cm) and also included very small (2 cm) HCC. In the sub-group of HCC > 5 cm, they observed a poorer survival of cirrhotic patients but they did not analyse the fibrosis stage impact in the non-cirrhotic group and no detailed data were available in this selected population. Four other authors reported a negative impact of cirrhosis [29,30,32,35], but two of these series only included HCC > 10 cm [32,35], Lim et al. included small tumours, while Hanazaki et al. did not evaluate the impact of progressive fibrosis on non-cirrhotic liver. Two recent series evaluated the impact of cirrhosis on LHCC resected: 1/Zhao et al. [27] did not find any negative impact of cirrhosis on OS (P = 0.985) whereas 2/for Chang et al. [20], cirrhosis negatively impacted OS (HR = 1,37; P = 0.013) but this result can not be compared to our because 30% of their patients were operated before 2006 and 75% of patients presented HBV and/or HCV infection(s).

7 All these reports in the literature have added some confusion to the real impact of fibrosis on LHCC (≥ 5 cm). Our study attempted to clarify this situation by showing that the degree of fibrosis within the HCC-NCL-group, or by comparing the HCC-CL and HCC-NCL groups, did not impact OS nor DFS in the case of strictly ≥ 5 cm HCC. In spite of the limitations of this work, our findings have led us to propose radical resection for LHCC whatever the fibrosis stage. A rigorous evaluation of the hepatic functional reserve in F4 patients is strongly advocated (and preoperative portal vein embolization highly recommended) but allows to perform major hepatectomy with reasonable morbi-mortality rate [50]. As most centers, we are used to perform TACE associated with preoperative portal vein embolization before major hepatectomy on cirrhosis, but these patients were excluded from analysis to get a better inter-groups comparability [50,51]. At last, our fibrosis scoring was based on final pathology analysis. No preoperative evaluation was systematically obtained (potential sampling bias of biopsies) and noninvasive stiffness assessment was not routinely performed in both centres since we know that large right-side tumours could prevent a reliable evaluation (unless sonographicguided left-side measure). However, elastography is proven to be a valuable tool to predict recurrence after resection [52] and it would be very interesting to assess liver stiffness as DFS or OS predictor in LHCC setting, as we know that cirrhotic livers present a very wide range of stiffness measures (< 10 kPa to > 50 kPa). In the case of large tumours, we may hypothesize that METAVIR score might not be a predictor as reliable as elastometry because of the huge overlap values between each category (no consensual and predictable cut-off for each fibrotic stage). The practical message of this study is to consider 1/LHCC on NCL as severe as LHCC occurring on CL, requiring extensive resection when possible, and 2/to consider that resection of LHCC on CL may offer the same survival benefit as that occurring on NCL. Surgical resection should therefore be considered as a first-intent curative treatment, even on cirrhosis. Liver transplantation may probably be indicated in case of poor liver function, even for LHCC potentially transplantable outside Milan score while AFP score ≤ 2 [53].

Conclusion We demonstrated that the degree of fibrosis affecting the underlying parenchyma had no impact on postoperative outcomes following the resection of LHCC (≥ 5 cm). In such tumours, prognosis appeared to be correlated to pathological features (microvascular invasion, satellite nodules, grade of differentiation and resection margins) rather than to the underlying parenchyma. In the event of fibrosis ≤ F3, hepatic resection remains the gold standard for curative treatment. In cases of strictly selected patients with F4 fibrosis, the same survival rates can be achieved without over morbi-mortality.

Ethical statement All procedures performed in this study were in accordance with the ethical standards of the institutional committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. For this retrospective study, formal consent was not required.

Please cite this article in press as: Golse N, et al. Large hepatocellular carcinoma: Does fibrosis really impact prognosis after resection? Journal of Visceral Surgery (2018), https://doi.org/10.1016/j.jviscsurg.2017.10.015

+Model JVS-757; No. of Pages 9

ARTICLE IN PRESS

8

N. Golse et al.

Contribution of authors Nicolas Golse: acquisition of data and manuscript writing. Aminah El Bouyousfi, Frédéric Marques, Brigitte Bancel, Kayvan Mohkam, Mylène Sebagh: acquisition of data and analysis. Christian Ducerf, Philippe Merle, Denis Castaing, Antonio Sa Cunha, René Adam, Daniel Cherqui, Eric Vibert: data analysis and study design. Jean-Yves Mabrut: coordinator of the work.

[14]

[15]

[16]

[17]

Appendix A. Supplementary data Supplementary data (Table 4) associated to this article can be found at http://www.sciencedirect.com et https://doi.org/10.1016/j.jviscsurg.2017.10.015.

[18]

[19]

Disclosure of interest

[20]

The authors declare that they have no competing interest.

[21]

References

[22]

[1] Ferlay J, Parkin DM, Steliarova-Foucher E. Estimates of cancer incidence and mortality in Europe in 2008. Eur J Cancer 2010;46:765—81 [Elsevier]. [2] European Association For The Study Of The Liver, European Organisation For Research And Treatment Of Cancer. EASL-EORTC clinical practice guidelines: management of hepatocellular carcinoma. J Hepatol 2012;56:908—43. [3] Paradis V, Zalinski S, Chelbi E, et al. Hepatocellular carcinomas in patients with metabolic syndrome often develop without significant liver fibrosis: a pathological analysis. Hepatology 2009;49:851—9. [4] Perumpail RB, Wong RJ, Ahmed A, et al. Hepatocellular carcinoma in the setting of non-cirrhotic non-alcoholic fatty liver disease and the metabolic syndrome: US experience. Dig Dis Sci 2015;7:1—7 [Springer US]. [5] Baffy G. Hepatocellular carcinoma in non-alcoholic fatty liver disease: epidemiology, pathogenesis, and prevention. J Clin Transl Hepatol 2013;1:131—7. [6] Leung C, Yeoh SW, Patrick D, et al. Characteristics of hepatocellular carcinoma in cirrhotic and non-cirrhotic nonalcoholic fatty liver disease. World J Gastroenterol 2015;21: 1189—96. [7] Shrager B, Jibara GA, Tabrizian P, et al. Resection of nonalcoholic steatohepatitis-associated hepatocellular carcinoma: a Western experience. Int J Surg Oncol 2012;2012:915128. [8] Lerut J, Mergental H, Kahn D, et al. Place of liver transplantation in the treatment of hepatocellular carcinoma in the normal liver. Liver Transpl 2011;17:S90—7 [Wiley Subscription Services, Inc., A Wiley Company]. [9] Mergental H, Adam R, Ericzon B-G, et al. Liver transplantation for unresectable hepatocellular carcinoma in normal livers. J Hepatol 2012;57:297—305. [10] Golse N, Bucur PO, Adam R, et al. New paradigms in post-hepatectomy liver failure. J Gastrointest Surg 2013;17:593—605. [11] Chiche L, Menahem B, Bazille C, et al. Recurrence of hepatocellular carcinoma in noncirrhotic liver after hepatectomy. World J Surg 2013;37:2410—8. [12] Dupont-Bierre E, Compagnon P, Raoul J-L, et al. Resection of hepatocellular carcinoma in noncirrhotic liver: analysis of risk factors for survival. J Am Coll Surg 2005;201:663—70. [13] Truant S, Boleslawski E, Duhamel A, et al. Tumor size of hepatocellular carcinoma in noncirrhotic liver: a controversial

[23]

[24]

[25]

[26] [27]

[28]

[29]

[30]

[31]

[32]

[33]

[34]

predictive factor for outcome after resection. Eur J Surg Oncol 2012;38:1189—96. 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. Faber W, Sharafi S, Stockmann M, et al. Long-term results of liver resection for hepatocellular carcinoma in noncirrhotic liver. Surgery 2013;153:510—7. Shrager B, Jibara G, Schwartz M, et al. Resection of hepatocellular carcinoma without cirrhosis. Ann Surg 2012;255: 1135—43. Young AL, Adair R, Prasad KR, et al. Hepatocellular carcinoma within a noncirrhotic, nonfibrotic, seronegative liver: surgical approaches and outcomes. J Am Coll Surg 2012;214:174—83. Fong Y, Sun RL, Jarnagin W, et al. An analysis of 412 cases of hepatocellular carcinoma at a Western center. Ann Surg 1999;229:790—9 [discussion799-800]. Capussotti L, Muratore A, Amisano M, et al. Liver resection for large-size hepatocellular carcinomas in 47 non-cirrhotic patients — no mortality and long-term survival. Hepatogastroenterology 2006;53:768—72. Chang YJ, Chung KP, Chen LJ. Long-term survival of patients undergoing liver resection for very large hepatocellular carcinomas. Br J Surg 2016;103:1513—20. Smoot RL, Nagorney DM, Chandan VS, et al. Resection of hepatocellular carcinoma in patients without cirrhosis. Br J Surg 2011;98:697—703. 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. Wu T-H, Yu M-C, Chen T-C, et al. Encapsulation is a significant prognostic factor for better outcome in large hepatocellular carcinoma. J Surg Oncol 2012;105:85—90. Yamashita Y-I, Takeishi K, Tsuijita E, et al. Beneficial effects of preoperative lipiodolization for resectable large hepatocellular carcinoma (≥ 5 cm in diameter). J Surg Oncol 2012;106:498—503 [Wiley Subscription Services, Inc., A Wiley Company]. Cho YB, Lee KU, Lee HW, et al. Outcomes of hepatic resection for a single large hepatocellular carcinoma. World J Surg 2007;31:795—801. Vauthey J-N, Lauwers GY, Esnaola NF, et al. Simplified staging for hepatocellular carcinoma. J Clin Oncol 2002;20:1527—36. Zhao H-C, Wu R-L, Liu F-B, et al. A retrospective analysis of long term outcomes in patients undergoing hepatic resection for large (> 5 cm) hepatocellular carcinoma. HPB (Oxford) 2016;18:943—9 [Elsevier]. Liau K-H, Ruo L, Shia J, et al. Outcome of partial hepatectomy for large (> 10 cm) hepatocellular carcinoma. Cancer 2005;104:1948—55 [Wiley Subscription Services, Inc., A Wiley Company]. Hanazaki K, Kajikawa S, Shimozawa N, et al. Hepatic resection for large hepatocellular carcinoma. Am J Surg 2001;181:347—53. Lim C, Mise Y, Sakamoto Y, et al. Above 5 cm, size does not matter anymore in patients with hepatocellular carcinoma. World J Surg 2014;38:2910—8. Pandey D, Lee KH, Wai C-T, 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. Ng KM, Yan TD, Black D, et al. Prognostic determinants for survival after resection/ablation of a large hepatocellular carcinoma. HPB (Oxford) 2009;11:311—20. Huang J, Hernandez-Alejandro R, Croome KP, et al. Hepatic resection for huge (> 15 cm) multinodular HCC with macrovascular invasion. J Surg Res 2012;178:743—50. van Meer S, van Erpecum KJ, Sprengers D, et al. Hepatocellular carcinoma in cirrhotic versus noncirrhotic livers: results from a large cohort in the Netherlands. Eur J Gastroenterol Hepatol 2015.

Please cite this article in press as: Golse N, et al. Large hepatocellular carcinoma: Does fibrosis really impact prognosis after resection? Journal of Visceral Surgery (2018), https://doi.org/10.1016/j.jviscsurg.2017.10.015

+Model JVS-757; No. of Pages 9

ARTICLE IN PRESS

Fibrosis does not impact prognosis [35] Pawlik TM, Poon RT, Abdalla EK, et al. Critical appraisal of the clinical and pathologic predictors of survival after resection of large hepatocellular carcinoma. Arch Surg 2005;140:450—7 [discussion457-8]. [36] Kadri HS, Blank S, Wang Q, et al. Outcomes following liver resection and clinical pathologic characteristics of hepatocellular carcinoma occurring in patients with chronic hepatitis B and minimally fibrotic liver. Eur J Surg Oncol 2013;39:1371—6. [37] Lang H, Sotiropoulos GC, Brokalaki EI, et al. Survival and recurrence rates after resection for hepatocellular carcinoma in noncirrhotic livers. J Am Coll Surg 2007;205:27—36. [38] Laurent C, Blanc J-F, 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. [39] Wörns MA, Bosslet T, Victor A, et al. Prognostic factors and outcomes of patients with hepatocellular carcinoma in noncirrhotic liver. Scand J Gastroenterol 2012;47:718—28. [40] Yang L-Y, Fang F, Ou D-P, et al. Solitary large hepatocellular carcinoma: a specific subtype of hepatocellular carcinoma with good outcome after hepatic resection. Ann Surg 2009;249:118—23. [41] Zhou L, Rui J-A, Wang S-B, et al. Early recurrence in large hepatocellular carcinoma after curative hepatic resection: prognostic significance and risk factors. Hepatogastroenterology 2014;61:2035—41. [42] Lee NH, Chau GY, Lui WY, et al. Surgical treatment and outcome in patients with a hepatocellular carcinoma greater than 10 cm in diameter. Br J Surg 1998;85:1654—7 [Blackwell Science Ltd]. [43] Beard RE, Hanto DW, Gautam S, et al. A comparison of surgical outcomes for noncirrhotic and cirrhotic hepatocellular carcinoma patients in a Western institution. Surgery 2013;154:545—55. [44] Grazi GL, Cescon M, Ravaioli M, et al. Liver resection for hepatocellular carcinoma in cirrhotics and noncirrhotics. Evaluation

9

[45]

[46]

[47]

[48]

[49]

[50]

[51]

[52]

[53]

of clinicopathologic features and comparison of risk factors for long-term survival and tumour recurrence in a single centre. Aliment Pharmacol Ther 2003;17:119—29. Shi M, Guo R-P, Lin X-J, et al. Partial hepatectomy with wide versus narrow resection margin for solitary hepatocellular carcinoma: a prospective randomized trial. Ann Surg 2007;245:36—43. Edmonson HA, Steiner PE. Primary carcinoma of the liver: a study of 100 cases among 48,900 necropsies. Cancer 1954;7:462—503. The French METAVIR Cooperative Study Group. Intraobserver and interobserver variations in liver biopsy interpretation in patients with chronic hepatitis C. Hepatology 1994;20: 15—20. Trevisani F, Frigerio M, Santi V, et al. Hepatocellular carcinoma in non-cirrhotic liver: a reappraisal. Dig Liver Dis 2010;42:341—7. Wang Q, Fiel MI, Blank S, et al. Impact of liver fibrosis on prognosis following liver resection for hepatitis B-associated hepatocellular carcinoma. Br J Cancer 2013;109:573—81. Farges O, Belghiti J, Kianmanesh R, et al. Portal vein embolization before right hepatectomy: prospective clinical trial. Ann Surg 2003;237:208—17. Aoki T, Imamura H, Hasegawa K, et al. Sequential preoperative arterial and portal venous embolizations in patients with hepatocellular carcinoma. Arch Surg 2004;139:766—74 [American Medical Association]. Jung KS, Kim SU, Choi GH, et al. Prediction of recurrence after curative resection of hepatocellular carcinoma using ® liver stiffness measurement (FibroScan ). Ann Surg Oncol 2012;19:4278—86. Duvoux C, Roudot-Thoraval F, Decaens T, et al. Liver transplantation for hepatocellular carcinoma: a model including ␣-fetoprotein improves the performance of Milan criteria. Gastroenterology 2012;143:986—94 [e3—quize14-5].

Please cite this article in press as: Golse N, et al. Large hepatocellular carcinoma: Does fibrosis really impact prognosis after resection? Journal of Visceral Surgery (2018), https://doi.org/10.1016/j.jviscsurg.2017.10.015