Laparoscopic ablation therapies or hepatic resection in cirrhotic patients with small hepatocellular carcinoma

Digestive and Liver Disease 48 (2016) 189–196

Contents lists available at ScienceDirect

Digestive and Liver Disease journal homepage: www.elsevier.com/locate/dld

Liver, Pancreas and Biliary Tract

Laparoscopic ablation therapies or hepatic resection in cirrhotic patients with small hepatocellular carcinoma Roberto Santambrogio a,∗ , Savino Bruno e , Michael D. Kluger b,c , Mara Costa a,b , Juan Salceda b , Andrea Belli b,f , Alexis Laurent b , Matteo Barabino a , Enrico Opocher a , Daniel Azoulay b , Daniel Cherqui b,d a

Surgical Unit 2, Hepatobiliary-Pancreatic and Digestive, San Paolo Hospital, University of Milan, Italy Department of Digestive and Hepatobiliary Surgery and Liver Transplantation, Henri Mondor Hospital, Créteil, France c Division of Gastrointestinal and Endocrine Surgery, Columbia College of Physicians and Surgeons, New York-Presbyterian Hospital, New York, USA d Hepato-Biliary Centre, AP-HP Hospital Paul Brousse, Villejuif, France e Humanitas University Medicine and Humanitas Research Hospital, Rozzano, Milano, Italy f Department of Oncological Abdominal Surgery, National Tumour Institute “G. Pascale”, Naples, Italy b

a r t i c l e

i n f o

Article history: Received 21 May 2015 Accepted 16 November 2015 Available online 23 November 2015 Keywords: Hepatic resection Hepatocellular carcinoma Laparoscopic ablation therapies Liver cirrhosis

a b s t r a c t Background: The Barcelona Clinic Liver Cancer staging system recommends radiofrequency ablation as treatment of choice for patients with “small” (up to 2 cm in size) hepatocellular carcinoma. Aims: Aim of the study was to assess whether laparoscopic ablation therapies or hepatic resection could be proposed as alternative option if percutaneous approach is not feasible. Methods: Overall survival and tumour recurrence rate were compared in a retrospective cohort of 176 consecutive patients with small hepatocellular carcinoma on cirrhosis treated by laparoscopic ablation therapies or surgery. To balance the covariates between the two groups, a propensity case-matched analysis was developed to generate a matched sample, which included 76 patients in each arm. Results: Local tumour progression (p = 0.005), intra-segmental recurrence (p = 0.0001), and 5-year recurrence rates (80% vs. 60%; p = 0.0014) were significantly higher in the ablation therapies group. The 5-year survival rate were 48% after ablation therapies and 69% after hepatic resection (p = 0.0006). Multivariate analysis showed that MELD score, alpha-fetoprotein value, procedure category and intraoperative restaging were associated with survival, while the surgery was the only independent predictor of intra-hepatic recurrence. Conclusions: The present study suggests that, if percutaneous ablation is not feasible, hepatic resection may be considered as a sound option in the treatment of small hepatocellular carcinoma. © 2015 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Ltd. All rights reserved.

1. Introduction Several comparative studies [1–5] provided evidence that percutaneous radiofrequency ablation (RFA) may be considered the treatment of choice for patients with single, small hepatocellular carcinoma (HCC) (up to 2 cm in size), even if there are no contraindications to hepatic resection (HR) [6–8]. As a matter of fact, previous reports showed that percutaneous RFA gave good results: complete tumour ablation rates of 90–95% and low local tumour progression rates of 5–10% were documented in the majority of series [6,9,10]. However, in daily practice, it is common to find patients with

∗ Corresponding author at: Ospedale San Paolo, via A. di Rudinì 8, 20142 Milano, Italy. Tel.: +39 02 81843981; fax: +39 02 50323075. E-mail address: [email protected] (R. Santambrogio).

HCC, which were not eligible for this procedure, mainly because the lesion is not identifiable at ultrasound examination [11]. Three studies [11–13] have shown that percutaneous RFA was not feasible in 25–55% of candidates, above all because the impossibility to visualize the nodule with percutaneous ultrasound (due to the subphrenic location of the nodule and the presence of macronodular cirrhosis) prevented this approach. On the other hand, multicentre Italian studies have demonstrated that surgical resection was considered for small HCC only in few patients [14,15]. Laparoscopic ablation therapies (LATs), which were introduced in the past few years [16–18], have been proposed as an alternative option for patients ineligible for other procedures. These treatments allow an improvement in tumour staging by intraoperative ultrasound (IOUS) and gross examination for tumour spread [19]. Additionally, recent studies have proved that a laparoscopic approach provides safe treatment for sub-capsular lesions,

http://dx.doi.org/10.1016/j.dld.2015.11.010 1590-8658/© 2015 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Ltd. All rights reserved.

190

R. Santambrogio et al. / Digestive and Liver Disease 48 (2016) 189–196

lesions located in close proximity to the gallbladder or other visceral structures, or lesions with locations unfavourable for percutaneous ablation [16–18,20,21]. Finally, despite no consistent study exists, the laparoscopic approach has been recommended as a reasonable alternative for patients with concurrent severe co-morbidities or poor liver function [17,18]. Aim of the present study was to evaluate the efficacy and safety of laparoscopic ablation in comparison to surgical resection as initial treatment for small HCC in a large cohort of cirrhotic patients ineligible for percutaneous ablation.

2. Patients and methods 2.1. Patients Prospective databases of treated patients (total, 855 patients: 529 surgical and 326 LATs) from two institutions (San Paolo Hospital, Milan, Italy and Hospital Henri Mondor, Creteil, France) between February 1997 and January 2012 were retrospectively analyzed. Both Centres assessed disease staging according to the Barcelona Clinic Liver Cancer (BCLC) criteria, since they became available [22]. BCLC staging was retrospectively assessed in all patients enrolled before its accessibility. A multidisciplinary team, including surgeons, radiologists, and hepatologists determined patient eligibility for an invasive treatment. Criteria for staging and treatment evolved over time. However, both were similar amongst the two centres, with only slight differences during the study period. In the Creteil Centre, the possibility of liver transplantation was considered first. Alternatively, liver resection was considered the treatment of choice. For patients referred to the Italian Centre, a transplant centre evaluation was performed even if liver transplant was not available on-site, despite the availability of this procedure was rather limited before 2000. Until 2012 (when the modified BCLC therapeutic algorithm was published [7]), surgical resection was proposed according to BCLC and AASLD guidelines: patients who had a single lesion were offered surgical resection if they had cirrhosis with preserved liver function. Portal hypertension was not considered a contraindication in all cases [23]. When surgery was not feasible or warranted, patients were evaluated for percutaneous or laparoscopic ablation. Percutaneous RFA was considered in patients with favourable target tumour for percutaneous access (conspicuous nodule on pre-operative ultrasound), at higher surgical risk (more than 2 segments with a postoperative remnant liver size less than 40–50%), impaired liver function (Child B) or severe comorbidities; laparoscopic approaches were considered for tumours in the dome of the liver or in other dangerous locations (due to proximity with visceral structures such as gallbladder, colon and stomach) or locations difficult to accurately target percutaneously, as confirmed by expert radiologists skilled in interventional procedures. Patients selected for the present cohort analysis fulfilled all of the following criteria at presentation: single lesion, tumour size less than 2 cm and Child–Pugh class A were submitted to liver resection (if they required resection of less than 2 segments); in other cases they were treated by LATs (percutaneous ablation or surgery not indicated). Based on these criteria, 679 patients were excluded and 176 small HCC patients were included in the study: 84 patients who underwent resection and 92 who underwent LATs. The residual liver function was classified according to the Child–Pugh classification and by MELD (model for end-stage liver disease) score (which was retrospectively re-calculated for patients included before 2000) [19,24,25]. We also retrospectively applied an s-BCLC staging system [26]. Comorbidity was assessed using

the Charlson’s index [27]: according to this score, patients were categorized as having slight (<2) or severe comorbidities (>3). Furthermore, the diagnosis and staging of HCC was achieved by a sequential contrast-enhanced imaging technique as the percutaneous ultrasound, triple phase helical computed tomography (CT), and contrast-enhanced magnetic resonance imaging (MRI). Prior to the establishment of the criteria for non-invasive diagnosis of HCC at the European Association of the Study of the Liver in 2000, the diagnosis was established by liver biopsy. As suggested by current guidelines [1,2,7], a preoperative ultrasound-guided percutaneous liver biopsy was performed only in patients with uncertain diagnosis. Based on the BCLC group criteria, the presence of preoperative portal hypertension was defined by esophageal varices detected with endoscopy or by enlarged spleen (major diameter >12 cm) with a platelet count <100,000 mm−3 [22,25]. Direct measurements of venous pressure were not routinely performed in the current series.

2.2. Treatment All surgical procedures included intra-operative ultrasonography (IOUS) examinations by using finger or laparoscopic probes equipped with a multi-frequency linear-array transducer. Similarly to the histological criteria described by Yamashita et al. [28], IOUS definition of micro-invasive HCC in the Italian subgroup was based according to the presence of portal vein, hepatic vein, bile duct infiltration, and/or intra-hepatic metastasis, as previously described [29] (Fig. 1). Thermo-ablation therapies are described in Appendix A: (Supplementary methods). Laparotomy for liver resection was carried out following a standardized technique [19,24], while laparoscopic HR was considered in patients with HCC lesions limited to the left lobe or segments IV through VI.

2.3. Assessment and follow-up Postoperative mortality was defined as occurrence of death within 30 days after treatment. Severity of postoperative morbidity was defined according to the Dindo-Clavien classification of surgical complications [31]. Postoperative hepatic insufficiency was graded according to International Study Group of Liver Surgery (ISGLS) [32]. Liver US and CT (and/or MRI) were performed within one month after treatment to assess the response to either liver resection or ablation. Post-treatment follow-up evaluation was performed by spiral CT (and/or MRI) after 3 months and every 6 months thereafter.

2.4. Technical evaluation Technical outcome and oncologic response were defined using the International Working Group on Image-Guided Tumour Ablation [33] standardized definitions and according to the guidelines of the International Union Against Cancer (UICC) [34] (Appendix A: Supplementary methods). Local tumour progression was diagnosed when a follow-up exam showed findings of interval development/growth of the tumour along the margin of the ablation or resected zone where the procedures had been considered to be technically effective. Recurrence was categorized as either intrasegmental (including local tumour progression) or extra-segmental based on the segment of the original nodule. HCC recurrence was classified as early or late, using a cut-off of 12 months. Experienced radiologists reviewed all CT scans.

R. Santambrogio et al. / Digestive and Liver Disease 48 (2016) 189–196

191

Fig. 1. Intraoperative patterns of micronvasive hepatocellular carcinoma. Panel (A) confluent multinodular type (black arrow) with satellite (white arrow); panel (B) single nodular type with satellites (arrow); panel (C) portal venous micro-infiltration (arrow); panel (D) upstream biliary invasion (arrow) with dilatation of biliary ducts. HCC, hepatocellular carcinoma; T, tumour.

2.5. Statistical analyses The primary endpoint of the study was overall survival (OS). Secondary endpoints were post-treatment complications, HCC recurrences, and disease-free survival. To minimize the effects of potential confounders from selection bias, propensity score matching was carried out. The psmatch2 macro in Stata 11.0 software for Macintosh (StataCorp LP, College Station, TX, USA) was used for propensity score match, and oneto-one matching between the groups was accomplished by using the nearest-neighbour matching method. A multivariable logistic regression model, including the entered variables of age, MELD score, BCLC classification, maximum diameter of nodules, platelet

count, serum albumin and serum AST levels was developed to calculate the propensity score. Propensity scores ranging from 0 to 1 were generated using binary logistic regression to estimate the probability that a patient would undergo HR or RFA. Distribution of propensity scores was evaluated for the sufficient overlap between two groups to ensure comparability. Results from the propensity score match were also reported as effect size: values <0.20 indicated small differences; between 0.20 and 0.50 indicated moderate differences, between 0.50 and 0.80 indicated large differences and >0.80 indicated very large differences [35]. After adjustment with propensity score match, cumulative actuarial curves (survival, recurrence-free survival and recurrences) were analyzed by the Kaplan–Meier method and were compared

192

R. Santambrogio et al. / Digestive and Liver Disease 48 (2016) 189–196

by the log-rank test. Mann–Whitney U test and Wilcoxon matched pairs test were used to compare continuous variables between and within groups. Comparison of proportions was done by the Fisher exact probability test. Data following a normal distribution were expressed as mean ± standard deviation, if data were nonparametric, median and interquartile range values were reported. The association of each parameter with survival and recurrence rates was estimated by univariate analysis comparing actuarial curves: only parameters with P values <0.05 were then included in the multivariate analysis. For each parameter analyzed in the multivariate analyses, t-values (Hazard ratio) and 95% confidence intervals (CI) were obtained. Initial evaluation and subsequent follow up data were collected in a dedicated database (FileMaker Pro, FileMaker Inc., Santa Clara, CA). The Ethics Committees at both hospitals approved this study. Our institutional review boards approved this retrospective study, and informed consent was not required. 3. Results The baseline characteristics of patients allocated to therapy groups (resection or LATs) are illustrated in Table 1 and in Supplementary Table S1. The 84 patients who underwent resection and the 92 receiving LATs were significantly different in the distribution of HCC maximum diameter, platelet count and serum

albumin levels. Furthermore, effect size analysis shows large differences for other variables such as age, MELD score, BCLC classification, and serum AST levels. Propensity score matching method was applied to minimize these confounding factors in survival and HCC recurrence analyses: 76 patients were matched in each group (in liver resection group there were 61 patients belonging to Milan and 15 patients to Creteil group) and the previously mentioned variables appeared to be well-matched between these two groups (see Table 1). Furthermore, the two groups showed no difference in term of severe comorbidities (defined as Charlson’s index >3). With regard to the characteristic of treatments, 41 patients (54%) underwent an anatomical resection of one segment, 28 patients (37%) a non-anatomical resection and only 7 patients (9%) underwent a left lateral resection. In 29 cases (38%) a laparoscopic resection was attempted (10% of cases were converted to laparotomy due to bleeding or oncological criteria). Pringle manoeuvre was used in 33 cases (43%). Regarding indications in the ablation group, the reasons for selecting the laparoscopic approach instead of either percutaneous RFA or HR are shown in Supplementary Table S2. Thermal ablation alone was performed in 61/76 patients (80%) using a cooled-tip needle, while 11/76 (14%) patients also received an intra-hepatic vascular occlusion. MWA technology was used in only 15/76 (20%) patients. IOUS identified a microinvasive HCC pattern in 41 patients (27%): 22 cases in the surgical group (29%) and 19 in LATs group (25%; p = 0.584). Furthermore, IOUS

Table 1 Demographic and clinical characteristics of patients after propensity match analysis.

Male sex Age (years) (median; IR) Cirrhosis aetiology HCV HBV Other Child–Pugh class A MELD score (median; IR) BCLC [7] A0 A2 A3 s-BCLC [26] AA AB Esophageal varices Portal hypertension [23] HCC segment localization II/III/IV/V VI/VII/VIII HCC lesion diameter (mm) (median, IR) IOUS MI HCC Haemoglobin (g/dl) (median, IR) Platelet count (×103 /mm3 ) (median, IR) Total bilirubin (mg/dl) (median, IR) Serum albumin (g/l) (median, IR) Prothrombin time (INR) (median, IR) AST (U/L) (median, IR) ␣-Fetoprotein (ng/ml) (median, IR) Charlson’s index ≥3

HR (n = 76)

LATs (n = 76)

p value

Effect size

52 (68%) 66 ± 9 (66; 61–72)

59 (78%) 68 ± 8 (69; 63–73)

0.201 0.101

0.2086 0.2349

0.874

0.0844

52 (68%) 10 (13%) 14 (19%) 84 (100%) 8.5 ± 1.4 (8; 7–9)

49 (65%) 11 (14%) 16 (21%) 92 (100%) 8.9 ± 2.2 (8; 7–10)

1.000 0.2226

0.2169

0.602

0.1640

51 (67%) 16 (21%) 9 (12%)

46 (61%) 17 (22%) 13 (17%) 0.501

0.1092

50 (66%) 26 (34%) 17 (23%) 24 (32%)

46 (61%) 30 (39%) 19 (25%) 29 (38%)

0.597 0.428 0.104

0.1658 0.1292 0.2663

40 (53%) 36 (47%) 17.5 ± 2.8 (19; 15–20) 22 (29%) 13.8 ± 1.4 (14, 12.9–14.9) 126 ± 66 (116, 83–156) 1.04 ± 0.54 (0.94, 0.7–1.22) 3.99 ± 0.41 (4.09, 3.7–4.23) 1.12 ± 0.08 (1.13, 1.06–1.16) 69.2 ± 45.7 (48, 30–99) 58.9 ± 183 8 (3.35–24.6) 25%

30 (39%) 46 (61%) 16.6 ± 3.6 (17.5; 15–20) 19 (25%) 13.7 ± 1.7 (13.9, 12.5–14.9) 116 ± 57 (106.5, 80–139) 1.09 ± 0.54 (1, 0.7–1.4) 4.00 ± 0.44 (3.95, 3.7–4.34) 1.14 ± 0.24 (1.09, 1.04–1.18) 71.9 ± 57 (51, 34–89) 101.6 ± 498 6.15 (2.9–24) 33%

0.1066

0.2790

0.584 0.634

0.0890 0.0642

0.2966

0.1622

0.481

0.0926

0.788

0.0235

0.497

0.1118

0.753

0.0523

0.484

0.1138

0.283

0.1880

HR, hepatic resection; LATs, laparoscopic ablation therapies; BCLC, Barcelona Clinic Liver Cancer; s-BCLC, simplified BCLC according to Santambrogio et al. classification [26]; IOUS, intraoperative ultrasound; IR, interquartile range; MI HCC, micro-invasive hepatocellular carcinoma; HCV, hepatitis C virus; HBV, hepatitis B virus; MELD, model for end-stage liver disease; HCC, hepatocellular carcinoma; AST, aspartate aminotransferase.

R. Santambrogio et al. / Digestive and Liver Disease 48 (2016) 189–196

193

detected new HCC nodules in 5 patients (7%) in the HR group and in 10 patients (13%) in the LATs group (p = 0.174). Treatments for new nodules were: 8 additional RFAs, 4 ethanol injections and 3 additional surgical resections. Based on these intraoperative findings, we established a modified intraoperative staging re-classification by BCLC staging system: the 5 patients in the HR group and the 10 cases in LATS group were upgraded to BCLC A4 staging class for long-term evaluation. 3.1. HCC recurrence pattern during follow-up The mean follow-up period was 50.4 ± 37.9 months (median 39.9 months; range: 2–197 months). The detailed pattern of HCC recurrences is shown in Supplementary Table S3. Briefly, local tumour progression (resection, 3%; LATs, 16%; p = 0.005), intrasegmental (resection, 7%; LATs, 33%; p = 0.0001) and intra-hepatic HCC recurrences (resection, 57%; LATs, 74%; p = 0.027) showed a significant difference between the 2 treatment groups. On the other hand, patients with micro-invasive HCC showed a slight difference for intra-segmental recurrences (12/41; 29%) in comparison to the patients without micro-invasive HCC (18/111; 16%; p = 0.073). However, analysing the treatment subgroups, micro-invasive HCC patients showed a higher incidence of intra-segmental recurrences in ablation group (10 cases; 53%) than in resection group (2 cases; 9%; p = 0.002). On the other hand, the subgroup with intraoperative detection of new HCC nodules (intraoperative BCLC A4 restaging) showed a higher incidence of intra-segmental recurrences (6/15; 40%) than patients with single tumours (24/137; 17%; p = 0.038) In the ablation group, cumulative intra-hepatic tumour recurrence rates at 1-, 3- and 5-years were 30%, 67% and 80%, respectively, whereas in the surgical group, they were 15%, 40% and 60%, respectively (p = 0.0014) (Fig. 2A). Patients with microinvasive HCC showed higher rates of intra-hepatic recurrences (Fig. 2B) than patients without micro-invasive HCC (p = 0.0111). Patients with intraoperative detection of new HCC nodules showed higher rates of intra-hepatic recurrences (Fig. 2C) than patients with single HCC tumours (p = 0.0037). Among patients with HCC recurrences, it was not possible to retreat HCC recurrences in 14/56 patients (25%) in the ablation group (8 cases for multifocal HCC recurrence, 2 for patient’s refusal, 3 for liver insufficiency and 1 for extra-hepatic tumours) and in 11/43 HCC recurrences (26%) in the resection group (9 cases for multifocal HCC recurrence, 2 for liver insufficiency, p = 0.947). Seven surgical patients (6 belonging to Creteil) and 1 ablation patient were submitted to liver transplantation. 3.2. Survival analysis During the follow-up period, 74/152 patients died: 46 patients (60%) in the LATs group and 28 patients (37%) in the resection group. Causes of death in the ablation group were multifocal HCC recurrence in 24 patients (52%), liver failure in 9 patients (20%), sepsis in 3 patients, primitive tumours arising in other organs in 5 patients, and cardiovascular complications in 5 patients. In the surgical group, 10 patients died of multifocal HCC recurrence (33%; 2 from Creteil), 8 of liver failure (25%; 2 from Creteil), 2 of variceal bleeding, 2 of intestinal occlusion, 4 of primitive tumours arising in other organs, and 2 of cardiovascular complications (p = 0.155). In the ablation group, the cumulative 1-, 3-, and 5-year OS rates were 91%, 64%, and 48%, respectively, whereas OS rates in the surgical group were 97%, 86%, and 69%, respectively. These differences were statistically significant (p = 0.0006) (Fig. 3A). No statistical differences were found for micro-invasive HCC (p = 0.1075) (Fig. 3B), while the intra-operative detection of new HCC nodules significantly worsened the survival rates (Fig. 3C; p = 0.0063).

Fig. 2. Hepatocellular carcinoma recurrence rate. Cumulative recurrence rate according to: panel (A) hepatic resection or laparoscopic ablation therapies (p = 0.0014); panel (B) hepatocellular carcinoma with or without microinvasive pattern (p = 0.0111); panel (C) single hepatocellular carcinoma or new hepatocellular carcinoma nodules detected by intraoperative ultrasound (p = 0.0037). HR, hepatic resection; LATs, laparoscopic ablation therapies; MI-HCC, microinvasive hepatocellular carcinoma; No-MI HCC, hepatocellular carcinoma without microinvasive pattern; IOUS, intraoperative ultrasound.

Different survival rates were shown between LATs and resection if two recruitment periods were considered (before 2008: p = 0.0077 and after 2008: p = 0.0478) (Supplementary Fig. S1). On the other hand, recurrence-free survival rates after LATs were significantly worse than those after resection (p = 0.0007; Supplementary Fig. S2). The results of univariate (Supplementary Table S4) and multivariate analyses (Table 2) among the 152 patients with small HCC showed that MELD score, AFP value, operation type and intraoperative restaging were independent predictors of survival, while

194

R. Santambrogio et al. / Digestive and Liver Disease 48 (2016) 189–196

Fig. 3. Overall survival. Overall survival curves of patients with hepatocellular carcinoma <2 cm according to: panel (A) hepatic resection or laparoscopic ablation therapies (p = 0.0006); panel (B) hepatocellular carcinoma with or without microinvasive pattern (p = 0.1075); panel (C) single hepatocellular carcinoma or new hepatocellular carcinoma nodules detected by intraoperative ultrasound (p = 0.0063). HR, hepatic resection; LATs, laparoscopic ablation therapies; MI-HCC, microinvasive hepatocellular carcinoma; No-MI HCC, hepatocellular carcinoma without microinvasive pattern; IOUS, intraoperative ultrasound.

the surgical procedure approach was the only independent predictor of intra-hepatic recurrence. 4. Discussion Small HCC defined as a single HCC nodule ≤2 cm is considered to be the earliest clinical detectable stage of HCC with the highest rate of cure [6,36,37]. It is characterized not only by its biological nature but also by an extremely favourable long-term outcome after liver resection [8]. Nonetheless, despite a recent meta-analysis confirmed that resection led to a higher long-term survival rate and

a lower recurrence rate compared to percutaneous thermal ablation [38,39], the BCLC therapeutic algorithm recommended RFA as the first-line treatment-option, while surgical approach should be considered in patients with failure or contraindications to ablation therapies [6–8]. However, it is well known that in several cases both procedures could not be used on the basis of nodule localization, concurrent poor liver function and concomitant severe co-morbidities [11–15]. In the present study we found that, in patients with small HCC, when percutaneous RFA is not viable, resection yielded higher overall survival and lower recurrence rates compared to LATs, without significant differences in severe complications (see Appendix A: Supplementary methods). We acknowledge that the lack of randomization and partial retrospective design of the study may have led to incorrect results and spurious associations, therefore our results must be interpreted cautiously. The best candidates submitted to a liver resection, in our study, were the subset of patients with a single peripheral lesion, which permitted a minimal surgical resection, a maximal sparing of liver parenchyma and a radical removal of the lesion [40]. To minimize selection bias, we used the propensity score matching analysis. An important reason which may justify better results after liver resection is the routine use of IOUS guidance which allows the surgeon to accomplish a radical but conservative resection [40], leading to decrease operative mortality and maintaining liver function in the long-term [26,40]. Another reason for better survival obtained in resection group is justified by the fact that HCC recurrences were treated in a similar manner after resection and LATs, while in previous comparative studies a higher proportion of patients treated by surgery had received less effective local treatments compared to those previously submitted to RFA [2–5]. At multivariate analysis, resection was confirmed to be prognostic factor for OS as well as liver function (by MELD score in Child–Pugh class A) and the biological characteristics of HCC (by AFP level and BCLC intra-operative restaging). Furthermore, local HCC progression, intra-segmental recurrences and early recurrences (<12 months) were more frequent after LATs than resection and this finding could not be attributed to a selection bias. It is well established that HCC nodules with a diameter less than 2 cm may contain zones of less differentiated tissue with more intense proliferative activity, thus justifying the finding of hidden portal micro invasion or microsatellites [28,36,41]. The IOUS approach allows the surgeon to accomplish a complete eradication of local satellites occurring via local vascular invasion [30,40,41] and this approach may further explain the observed results. On the other hand, the routine use of IOUS should permit to obtain a complete intra operative restaging in both groups of patients: in 15 patients missed HCC nodules were identified by IOUS and patients were reclassified as BCLC A4 class, permitting a proper intraoperative restaging of the disease. Despite the proper methodology used to balance the baseline characteristics of patients included in the analysis, when attempting to compare the efficacy of two therapies, several variables specific to each treatment and patient, which may have an impact on outcomes, are often unable to be accounted for in this analysis. Lesion location (peripheral vs. central), its proximity to major bile ducts, a patient’s body habitus with or without comorbidities, or the provider’s experience have the potential to influence the outcome of treatment [11–13,21]. In addition, while assessing the provided results, one should keep in mind that patients with deeply located and difficult-to-treat lesions were associated with a worse treatment outcome, irrespective of the procedure used [20,42]. On the other hand, it is conceivable that multifocal HCC recurrence was higher in the LATs group because ablation sometimes failed to control the primary tumour and this is a clear negative prognostic factor [43,44].

R. Santambrogio et al. / Digestive and Liver Disease 48 (2016) 189–196

195

Table 2 Overall survival and tumour recurrence prognostic factors. Overall survival

Univariate analysis (OS at 5-year)

p value

Multivariate analysis Haz. ratio (95% CI)

p value

MELD score (≤9, >9) BCLC stage (A0; A2 + A3) [7] AFP: normal vs. abnormal Operation (HR; LATs) Intraoperative restaging: A0–A3; A4

60%; 40% 63%; 49% 64%; 44% 69%; 48% 61%; 38%

0.0014 0.0439 0.0495 0.0006 0.0063

1.781 (1.053–3.011) 1.637 (0.967–2.769) 1.802 (1.078–3.012) 1.913 (1.153–3.177) 2.015 (1.055–3.849)

0.031 0.066 0.025 0.012 0.034

HCC recurrence

Univariate analysis (at 5-year)

p value

Multivariate analysis Haz. ratio (95% CI)

p value

s-BCLC stage (AA; AB) [26] Operation (HR; LATs) Microinvasive HCC: not/yes Intraoperative restaging: A0–A3; A4

63%; 79% 60%; 80% 63%; 82% 67%; 87%

0.0206 0.0014 0.0111 0.0037

1.501 (0.993–2.269) 1.827 (1.215–2.746) 1.322 (0.803–2.177) 1.537 (0.774–3.049)

0.054 0.004 0.272 0.219

Univariate and multivariate analyses of the influence of preoperative and intraoperative factors on overall survival rates among 152 patients with HCC: treatment option and variables with p value either <0.05 at univariate analysis were retained for multivariate analysis (s-BCLC staging system was excluded in survival analysis because it included AFP value). IOUS, intraoperative ultrasound findings; s-BCLC, simplified BCLC according to Santambrogio et al. classification [26]; AFP, alpha-fetoprotein; HR, hepatic resection; LATs, laparoscopic ablation therapies; BCLC, Barcelona Clinic Liver Cancer; MELD, model for end-stage liver disease; HCC, hepatocellular carcinoma.

In conclusion, when percutaneous RFA is not feasible, liver resection is the first-line option for patients with small HCC, particularly in the presence of peripheral tumours, preserved liver function and no severe comorbidities. Furthermore, resection allows eradication of the primary tumour in addition to any micrometastases or microsatellite lesions along the portal tributaries, and this might play a role in preventing recurrence caused by a micro-invasive pattern. LATs are a valid alternative in patients with severe portal hypertension and/or with deeply located tumours, tumours located bordering the gallbladder, the main biliary ducts, the bowel loops, or nearby the big vessels, in which percutaneous RFA may be less effective and more dangerous. Conflict of interest None declared. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.dld.2015.11.010. References [1] Li L, Zhang J, Liu X, et al. Clinical outcomes of radiofrequency ablation and surgical resection for small hepatocellular carcinoma: a meta-analysis. Journal of Gastroenterology and Hepatology 2012;27:51–8. [2] Huang J, Yan L, Cheng Z, et al. A randomized trial comparing radiofrequency ablation and surgical resection for HCC conforming to Milan criteria. Annals of Surgery 2010;252:903–12. [3] Cho YC, Kim JK, Kim WT, et al. Hepatic resection versus radiofrequency ablation for very early stage hepatocellular carcinoma: a Markov model analysis. Hepatology 2010;51:1284–90. [4] Kuang M, Xie XY, Huang C, et al. Long-term outcome of percutaneous ablation in very early-stage hepatocellular carcinoma. Journal of Gastrointestinal Surgery 2011;15:2165–71. [5] Wang JH, Wang CC, Hung CH, et al. Survival comparison between surgical resection and radiofrequency ablation for patients in BCLC very early/early stage hepatocellular carcinoma. Journal of Hepatology 2012;56:412–8. [6] Livraghi T, Meloni F, Di Stasi M, et al. Sustained complete response and complications rates after radiofrequency ablation of very early hepatocellular carcinoma in cirrhosis: is resection still the treatment of choice? Hepatology 2008;47:82–9. [7] Forner A, Llovet JM, Bruix J. Hepatocellular carcinoma. Lancet 2012;379: 1245–55. [8] De Lope CR, Tremosini S, Forner A, et al. Management of HCC. Journal of Hepatology 2012;56(Suppl. 1):S75–87. [9] Shiina S, Tateishi R, Arano T, et al. Radiofrequency ablation for hepatocellular carcinoma: 10-year outcome and prognostic factors. American Journal of Gastroenterology 2012;107:569–77.

[10] Lencioni R, Cioni D, Crocetti L, et al. Early-stage hepatocellular carcinoma in patients with cirrhosis: long-term results of percutaneous image-guide radiofrequency ablation. Radiology 2005;234:961–7. [11] Kim PN, Choi D, Rhim H, et al. Planning ultrasound for percutaneous radiofrequency ablation to treat (<3 cm) hepatocellular carcinomas detected on computed tomography or magnetic resonance imaging: a multicenter prospective study to assess factors affecting ultrasound visibility. Journal of Vascular and Interventional Radiology 2012;23:627–34. [12] Kim JE, Kim YS, Rhim H, et al. Outcomes of patients with hepatocellular carcinoma referred for percutaneous radiofrequency ablation at a tertiary center: analysis focused on the feasibility with the use of ultrasonography guidance. European Journal of Radiology 2011;79:e80–4. [13] Rhim H, Lee MH, Kim YS, et al. Planning sonography to assess the feasibility of percutaneous radiofrequency ablation of hepatocellular carcinomas. American Journal of Roentgenology 2008;190:1324–30. [14] Farinati F, Sergio A, Baldan A, et al. Early and very early hepatocellular carcinoma: when and how much do staging and choice of treatment really matter? A multicenter study. BMC Cancer 2009;9:33–44. [15] Borzio M, Fornari F, De Slo I, et al. Adherence to American Association for the Study of the Liver Diseases guidelines for the management of hepatocellular carcinoma: results of a an Italian field practice multicenter study. Future Oncology 2013;9:283–94. [16] Chung MH, Wood TF, Tsioulias GJ, et al. Laparoscopic radiofrequency ablation of unresectable hepatic malignancies. A phase 2 trial. Surgery Endoscopy 2001;15:1020–6. [17] Santambrogio R, Opocher E, Costa M, et al. Survival and intra-hepatic recurrences after laparoscopic radiofrequency of hepatocellular carcinoma in patients with liver cirrhosis. Journal of Surgical Oncology 2005;89: 218–26. [18] Ballem N, Berber E, Pitt T, et al. Laparoscopic radiofrequency ablation of unresectable hepatocellular carcinoma: long-term follow-up. HPB 2008;10:315–20. [19] Santambrogio R, Opocher E, Zuin M, et al. Surgical resection versus laparoscopic radiofrequency ablation in patients with hepatocellular carcinoma and Child-Pugh class: a liver cirrhosis. Annals of Surgical Oncology 2009;16:3289–98. [20] De la Serna S, Vilana R, Sanchez-Cabus S, et al. Results of laparoscopic radiofrequency ablation of HCC. Could the location of the tumour influence a complete response to treatment? A single European experience. HPB 2015;17:387–93. [21] Santambrogio R, Barabino M, Bruno S, et al. Long-term outcome of laparoscopic ablation therapies for unresectable hepatocellular carcinoma: a single European center experience of 426 patients. Surgical Endoscopy 2015 (Pub online). [22] Bruix J, Sherman M. Management of hepatocellular carcinoma: an update. Hepatology 2011;53:1020–2. [23] Santambrogio R, Kluger MD, Costa M, et al. Hepatic resection for hepatocellular carcinoma in patients with Child-Pugh’s A cirrhosis: is clinical evidence of portal hypertension a contraindication? HPB 2013;15:78–84. [24] Cherqui D, Laurent A, Mocellin N, et al. Liver resection for transplantable hepatocellular carcinoma. Long-term survival and role of secondary liver transplantation. Annals of Surgery 2009;250:738–46. [25] Llovet JM, Bru’ C, Bruix J. Prognosis of hepatocellular carcinoma: the BCLC staging classification. Seminars in Liver Disease 1999;19:329–38. [26] Santambrogio R, Opocher E, Costa M, et al. Hepatic resection for “BCLC stage A” hepatocellular carcinoma. The prognostic role of alpha-fetoprotein. Annals of Surgical Oncology 2012;19:426–34. [27] Simons JP, Chau S, Hill JS, et al. In-hospital mortality from liver resection for hepatocellular carcinoma. Cancer 2010;116:1733–8.

196

R. Santambrogio et al. / Digestive and Liver Disease 48 (2016) 189–196

[28] Yamashita Y, Tsuijita E, Takeishi K, et al. Predictors for microinvasion of small hepatocellular carcinoma ≤ 2 cm. Annals of Surgical Oncology 2012;19:2027–34. [29] Santambrogio R, Costa M, Strada D, et al. Intraoperative ultrasound score to predict recurrent hepatocellular carcinoma after radical treatments. Ultrasound in Medicine and Biology 2011;37:7–15. [30] Santambrogio R, Costa M, Barabino M, et al. Laparoscopic radiofrequency of hepatocellular carcinoma using ultrasound-guided selective intrahepatic vascular occlusion. Surgical Endoscopy 2008;22:2051–5. [31] Dindo D, Demartines N, Clavien PA. Classification of surgical complications. A new proposal with evaluation in a cohort of 633 patients and results of a survey. Annals of Surgery 2004;240:205–13. [32] Rahbari NN, Garden J, Padbury R, et al. Posthepatectomy liver failure: a definition and grading by the international Study Group of Liver Surgery (ISGLS). Surgery 2011;149:713–24. [33] Ahmed M. Image-guided tumour ablation: standardization of terminology and reporting criteria – a 10-year update. Radiology 2014;273:241–60. [34] Wittekind C, Compton CC, Greene FL, et al. TNM residual tumour classification revisited. Cancer 2002;94:2511–9. [35] Cohen J. Statistical power analysis for the behavioral science. 2nd ed. Hillsdale: Lawrence Erlbaum; 1988. [36] Roayaie S, Obeidat K, Sposito C, et al. Resection of hepatocellular cancer ≤ 2 cm: results from two Western Centers. Hepatology 2013;57:1426–35. [37] Lee JI, Lee JW, Kim YS, et al. Analysis of survival in very early hepatocellular carcinoma after the resection. Journal of Clinical Gastroenterology 2011;45:366–71.

[38] Fu C, Liu N, Deng Q, et al. Radiofrequency ablation vs. surgical resection on the treatment of patients with small hepatocellular carcinoma: a system review and meta-analysis of five randomized controlled trials. Hepato-gastroenterology 2014;61:1722–9. [39] Feng Q, Chi Y, Liu Y, et al. Efficacy and safety of percutaneous radiofrequency ablation versus surgical resection for small hepatocellular carcinoma: a meta-analysis of 23 studies. Journal of Cancer Research and Clinical Oncology 2015;141:1–9. [40] Torzilli G, Montorsi M, Donadon M, et al. Radical but conservative” is the main goal for ultrasonography-guided liver resection: prospective validation of this approach. Journal of the American College of Surgeons 2005;201: 517–28. [41] Sasaki A, Kai S, Iwashita Y, et al. Microsatellite distribution and indication for locoregional therapy in small hepatocellular carcinoma. Cancer 2005;103:299–306. [42] Cillo U, Vitale A, Dupuis D, et al. Laparoscopic ablation of hepatocellular carcinoma in cirrhotic patients unsuitable for liver resection or percutaneous treatment: a cohort study. PLOS ONE 2013;8:e57249. [43] Sala M, Llovet JM, Vilana R, et al. Initial response to percutaneous ablation predicts survival in patients with hepatocellular carcinoma. Hepatology 2004;40:1352–60. [44] Takahashi S, Kudo M, Chung H, et al. Initial treatment response is essential to improve survival in patients with hepatocellular carcinoma who underwent curative radiofrequency ablation therapy. Oncology 2007;72(Suppl. 1): 98–103.