ORIGINAL ARTICLE
Surgical Resection Versus Radiofrequency Ablation for Single Hepatocellular Carcinoma 2 cm in a Propensity Score Model Po-Hong Liu, MD, y Chia-Yang Hsu, MD, MPH, yz Cheng-Yuan Hsia, MD, § Yun-Hsuan Lee, MD, y Yi-Hsiang Huang, MD, PhD,yjj Yi-You Chiou, MD, ô Han-Chieh Lin, MD, y and Teh-Ia Huo, MD y
Objectives: To evaluate the efficacy of surgical resection (SR) and radiofrequency ablation (RFA) for single hepatocellular carcinoma (HCC) 2 cm or less. Background: The optimal management for Barcelona Clı´nic Liver Cancer (BCLC) very early-stage HCC is undetermined. Methods: Between 2002 and 2013, a total of 237 (SR, 109; RFA, 128) patients with BCLC very early-stage HCC were enrolled. Their overall survival (OS) and recurrence-free survival (RFS) were compared. Propensity score matching analysis identified 79 matched pairs of patients to compare outcomes. Results: At baseline, patients with SR were younger and had larger tumors (both P < 0.05). The 5-year OS rates were 81% versus 76% (P ¼ 0.136), whereas 5-year RFS rates were 49% versus 24% (P < 0.001) for SR and RFA groups, respectively. In the propensity model, the baseline variables were well balanced between 2 groups. Surgical resection was significantly associated with better OS and RFS compared with RFA; the 5-year OS rates were 80% versus 66% (P ¼ 0.034), and 5-year RFS rates were 48% versus 18% (P < 0.001) for SR and RFA groups, respectively. The Cox proportional hazards model identified RFA as an independent predictor for mortality and tumor recurrence in the propensity model (hazard ratio, 2.120 and 2.421, respectively; both P < 0.05). Patients with recurrent HCC had inferior prognosis compared with patients without recurrence (P ¼ 0.001). However, the survival after recurrence was similar between patients initially treated with SR or RFA (P ¼ 0.415). Conclusions: Surgical resection provides better long-term OS and RFS compared with RFA in patients with BCLC very early-stage HCC. Surgical resection should be considered as the first-line treatment for these patients. Keywords: Barcelona Clı´nic Liver Cancer stage, hepatocellular carcinoma, overall survival, radiofrequency ablation, recurrence-free survival, surgical resection
(Ann Surg 2016;263:538–545)
H
patients with chronic liver disease combined with improved imaging technologies had led to increased percentage of patients with HCC diagnosed at Barcelona Clı´nic Liver Cancer (BCLC) very early stage (BCLC stage 0, single tumor 2 cm).2 According to the current European Association for the Study of Liver (EASL) and American Association for the Study of Liver Diseases (AASLD) management guidelines, the recommended treatment modalities for very earlystage HCC include surgical resection (SR), radiofrequency ablation (RFA), and liver transplantation, on the basis of the presence or absence of portal hypertension and associated diseases.3,4 With 5-year survival rates around 70%, SR is generally regarded as the preferred treatment for patients with very early-stage HCC.5 Surgical resection offers therapeutic possibilities of complete eradication of satellite tumor lesions and microscopic tumor emboli in adjacent vasculature at the expense of more destruction to nontumor liver parenchyma.6 Alternatively, RFA is generally considered the most effective treatment among various percutaneous ablative therapies.7 Radiofrequency ablation was shown to achieve comparable overall survival (OS) and better tolerability for early HCC, and is recommended as the priority treatment for very early-stage HCC with impaired liver functional reserve.8,9 Because of limited availability of liver transplantation, SR and RFA are frequently employed in patients with very early-stage HCC.10,11 However, whether SR or RFA should be the choice of treatment in terms of OS and recurrencefree survival (RFS) is still under intense debate without universally accepted treatment algorithm.12 In this study, we aimed to investigate the impact of treatment selection on HCC recurrence and long-term survival in a prospectively followed-up cohort of patients with very early-stage HCC receiving SR or RFA as the primary treatment. A propensity score matching analysis was employed to overcome potential confounding bias at baseline; prognostic predictors affecting the outcome were determined.
PATIENTS AND METHODS
epatocellular carcinoma (HCC) is one of the most common malignancies worldwide, with rising incidence in both Eastern and Western countries.1 Wide application of surveillance programs in
Patients
From the Faculty of Medicine, National Yang-Ming University School of Medicine, Taipei, Taiwan; yDepartment of Medicine, Taipei Veterans General Hospital, Taipei, Taiwan; zDepartment of Biostatistics, UCLA, Los Angeles, CA; §Department of Surgery, Taipei Veterans General Hospital, Taipei, Taiwan; ôInstitute of Clinical Medicine, National Yang-Ming University School of Medicine, Taipei, Taiwan; jjDepartment of Radiology, Taipei Veterans General Hospital, Taipei, Taiwan; and Institute of Pharmacology, National Yang-Ming University School of Medicine, Taipei, Taiwan. Supported by grants from the Center of Excellence for Cancer Research at Taipei Veterans General Hospital (MOHW103-TD-B-111-02), Taiwan, from Taipei Veterans General Hospital (V104C-008), Taipei, Taiwan, and from the Ministry of Education, Aiming for the Top University Plan (103AC-P618), Taiwan. Disclosure: The authors declare no conflicts of interest. Reprints: Teh-Ia Huo, MD, Professor of Medicine, Division of Gastroenterology, Department of Medicine, Taipei Veterans General Hospital, No. 201, Sec. 2, Shipai Rd, Taipei 11217, Taiwan. E-mail: [email protected]. Copyright ß 2015 Wolters Kluwer Health, Inc. All rights reserved. ISSN: 0003-4932/14/26105-0821 DOI: 10.1097/SLA.0000000000001178
We retrospectively analyzed 3117 patients with newly diagnosed HCC admitted to Taipei Veterans General Hospital in more than a decade’s period from 2002 to 2013. Patients with single HCC 2 cm or smaller, with no evidence of vascular invasion and extrahepatic disease, and with good performance status (status 0 or 1) receiving SR or RFA as the primary treatment for HCC were identified. Comprehensive baseline information, including patient demographics, etiology of underlying liver disease, tumor characteristics, serum biochemistries, severity of cirrhosis, and performance status, was recorded at the time of diagnosis. All patients were followed up with imaging studies and serum a-fetoprotein (AFP) level every 3 to 6 months until death or dropout from the follow-up program. Tumor recurrence, subsequent therapy, and OS were recorded. Specific patient information was de-identified before statistical analysis. This study was approved by the institutional review board and complied with the standards of the Declaration of Helsinki and current ethical guidelines.
538 | www.annalsofsurgery.com
Annals of Surgery Volume 263, Number 3, March 2016
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
Annals of Surgery Volume 263, Number 3, March 2016
Diagnosis and Definitions The diagnosis of HCC was pathologically confirmed for patients receiving SR. For patients undergoing RFA, the diagnosis of HCC was confirmed by biopsy during the RFA procedure or according to the criteria from EASL and AASLD HCC management guidelines in force.3,4,13,14 Performance status was assessed at the time of diagnosis by the Eastern Cooperative Oncology Group performance scale.15 The Child-Turcotte-Pugh classification and the model for end-stage liver disease (MELD) score were used to define severity of chronic liver disease.16 Total tumor volume (TTV) was calculated on the basis of tumor diameter as previously reported.17 Local recurrence was defined as residual disease within or adjacent to the previously treated tumor site, whereas nonlocal recurrence was defined as emergence of the tumor elsewhere in or outside the liver.18
Treatment The criteria for resection and surgical techniques were described previously.19–21 Surgical resection was performed by an experienced surgical team. During parenchymal resection, attempts were made to keep an adequate surgical margin at least 1 cm.22 Anatomical resection was defined as removal of the entire Couinaud segment involved with the tumor. The extent of hepatectomy was determined by the retention rate of indocyanine green at 15 minutes after injection.19,23 Radiofrequency ablation was performed under sonographic guidance.7,24,25 The tumor was ablated in the automatic impedance control mode by a single 17-gauge electrode with the Cool-Tip Radiofrequency System (Radionics, Burlington, MA). Post-RFA sonography was performed to confirm that there were no immediate complications including hemorrhage or hematoma. For patients with local recurrent tumor(s) initially treated with RFA, repeated RFA was exclusively attempted whenever possible. Other recurrent tumors were treated with SR, RFA, transarterial chemoembolization (TACE), liver transplantation, percutaneous ethanol injection, sorafenib, or best supportive care, depending on clinical presentation and tumor status.
Process of Treatment Selection Patients were presented to a multidisciplinary HCC team of Taipei Veterans General Hospital for treatment guidance. Information of therapeutic benefits, risks, and long-term outcomes of different treatment modalities was comprehensively provided to individual patients with HCC. Shared decisions were made between patients and clinicians after detail counseling. Written informed consent was obtained before initiation of any definite treatment.
Propensity Score Matching Analysis To investigate the association between treatment selection and clinical outcomes in an observational, non-randomized study, a propensity score matching analysis was employed to reduce bias in patient selection and to generate matched pairs of patients to compare OS and RFS between patients receiving SR or RFA.25,26 Possible variables associated with the selection of treatment, including age, sex, serum bilirubin and AFP level, platelet count, TTV, and performance status, were comprehensively included in the generation of propensity scores. Binary logistic regression with selected variables was used to generate continuous propensity scores from 0 to 1. A one-to-one nearest-neighbor match without replacement between patients receiving SR and RFA was performed to select patients into subsequent analyses.27
Statistics The Mann-Whitney U test was used to compare continuous variables between 2 patient groups. The x2 and 2-tailed Fisher exact ß
2015 Wolters Kluwer Health, Inc. All rights reserved.
SR vs RFA in BCLC Very Early-Stage HCC
tests were employed to compare categorical data. Overall survival and RFS were examined by the Kaplan-Meier method with log-rank tests. Prognostic factors that were possibly linked to survival, including sex, etiology and severity of chronic liver disease, serum biochemistries, platelet count, TTV, treatment selection, and performance status, were comprehensively included in survival analyses. Factors with P value less than 0.1 in univariate analyses were introduced into the multivariate Cox proportional hazards model to determine the adjusted hazard ratios (HRs) and 95% confidence intervals (CIs). All statistical analyses were conducted with SPSS for Windows version 20.0 (IBM, NY). A 2-tailed P value less than 0.05 was considered statistically significant.
RESULTS Patients During the study period, a total of 237 patients with single HCC 2 cm or smaller were identified. Among these patients, 109 received SR and 128 underwent RFA as the primary anticancer treatment. Their median follow-up duration was 43 and 44 months, respectively. Anatomical resection was achieved in 100 patients in the SR group. The surgical margin was 1.1 1.1 cm. Fifteen patients in the SR group had tumors located near major hepatic vessels, whereas 13 patients had protruding tumors or had tumors at liver dome, near the gallbladder or near the kidney. For the RFA group, 7 patients had tumors located near major hepatic vasculature, and 6 patients had tumors at subcapsular region, liver dome, or near the gallbladder. The propensity score matching analysis identified 79 matched pairs of patients from each treatment arm to compare OS and RFS.
Characteristics, Recurrence, and Survival of All Patients Patients with very early-stage HCC receiving SR were significantly younger and had larger TTV compared with patients receiving RFA (both P < 0.05; Table 1). The SR group was also associated with higher serum albumin level, lower serum alanine transaminase (ALT) level, lower international normalized ratio (INR) of prothrombin time (PT), and higher platelet count (all P < 0.05). The OS was similar between SR and RFA groups (P ¼ 0.136; Fig. 1A). The 1-, 3-, and 5-year OS rates were 98%, 97%, and 81% versus 98%, 88%, and 76%, for SR and RFA groups, respectively. Patients receiving SR had a significantly better RFS compared with patients receiving RFA (P < 0.001; Fig. 2A). The estimated 1-, 3-, and 5-year RFS rates were 91%, 64%, and 49% for the SR group and 72%, 38%, and 24% for the RFA group. There was no treatmentrelated mortality in both treatment arms.
Characteristics, Recurrence, and Survival of Patients Selected in the Propensity Model The characteristics of patients with very early-stage HCC receiving SR or RFA selected in the propensity model are shown in Table 2. The 2 treatment groups had well-matched baseline characteristics including age, sex, etiology of underlying liver disease, serum biochemistry and AFP level, severity of chronic liver disease, tumor characteristics, and performance status (all P > 0.05). Patients in the propensity model receiving SR had significantly better OS when compared with those receiving RFA (P ¼ 0.034; Fig. 1B). The estimated 1-, 3-, and 5-year OS rates were 97%, 97%, and 80% for the SR group and 97%, 83%, and 66% for the RFA group. Patients undergoing SR also had better RFS compared with patients receiving RFA in the propensity model (P < 0.001; Fig. 2B). The estimated 1-, 3-, and 5-year RFS rates were 92%, 65%, and 48% for the SR group and 68%, 36%, and 18% for the RFA group. www.annalsofsurgery.com | 539
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
Annals of Surgery Volume 263, Number 3, March 2016
Liu et al
TABLE 1. Baseline Demographics in Patients With BCLC Very Early-Stage HCC Receiving SR or RFA SR (n ¼ 109)
Variables Age, yr, mean (SD) Male, n (%) Positive for HBsAg, n (%) Positive for anti-HCV, n (%) Alcoholism, n (%) Performance status 1, n (%) Serum biochemistry, mean (SD) Albumin, g/dL Bilirubin, mg/dL Creatinine, mg/dL ALT, U/L Sodium, mmol/L INR of PT, mean (SD) Platelets, 1000/mL, mean (SD) AFP, ng/mL, mean (SD) CTP score, mean (SD) MELD score, mean (SD) TTV, cm3, mean (SD)
RFA (n ¼ 128)
60 78 65 37 12 13
(13) (72) (60) (34) (11) (12)
64 84 62 55 13 15
(12) (66) (48) (43) (10) (12)
4.0 0.9 1.0 57 140 1.0 154 145 5.2 7.8 2.6
(0.5) (0.4) (0.6) (56) (2.8) (0.1) (68) (262) (0.4) (1.3) (1.3)
3.9 0.8 1.0 68 139 1.1 124 92 5.3 8.4 2.0
(0.5) (0.4) (0.9) (57) (3.5) (0.1) (62) (247) (0.5) (2.5) (1.1)
P 0.038 0.401 0.091 0.182 0.836 1.000 0.032 0.312 0.161 0.009 0.528 0.028
Surgical Resection Versus Radiofrequency Ablation for Single Hepatocellular Carcinoma 2 cm in a Propensity Score Model Po-Hong Liu, MD, y Chia-Yang Hsu, MD, MPH, yz Cheng-Yuan Hsia, MD, § Yun-Hsuan Lee, MD, y Yi-Hsiang Huang, MD, PhD,yjj Yi-You Chiou, MD, ô Han-Chieh Lin, MD, y and Teh-Ia Huo, MD y
Objectives: To evaluate the efficacy of surgical resection (SR) and radiofrequency ablation (RFA) for single hepatocellular carcinoma (HCC) 2 cm or less. Background: The optimal management for Barcelona Clı´nic Liver Cancer (BCLC) very early-stage HCC is undetermined. Methods: Between 2002 and 2013, a total of 237 (SR, 109; RFA, 128) patients with BCLC very early-stage HCC were enrolled. Their overall survival (OS) and recurrence-free survival (RFS) were compared. Propensity score matching analysis identified 79 matched pairs of patients to compare outcomes. Results: At baseline, patients with SR were younger and had larger tumors (both P < 0.05). The 5-year OS rates were 81% versus 76% (P ¼ 0.136), whereas 5-year RFS rates were 49% versus 24% (P < 0.001) for SR and RFA groups, respectively. In the propensity model, the baseline variables were well balanced between 2 groups. Surgical resection was significantly associated with better OS and RFS compared with RFA; the 5-year OS rates were 80% versus 66% (P ¼ 0.034), and 5-year RFS rates were 48% versus 18% (P < 0.001) for SR and RFA groups, respectively. The Cox proportional hazards model identified RFA as an independent predictor for mortality and tumor recurrence in the propensity model (hazard ratio, 2.120 and 2.421, respectively; both P < 0.05). Patients with recurrent HCC had inferior prognosis compared with patients without recurrence (P ¼ 0.001). However, the survival after recurrence was similar between patients initially treated with SR or RFA (P ¼ 0.415). Conclusions: Surgical resection provides better long-term OS and RFS compared with RFA in patients with BCLC very early-stage HCC. Surgical resection should be considered as the first-line treatment for these patients. Keywords: Barcelona Clı´nic Liver Cancer stage, hepatocellular carcinoma, overall survival, radiofrequency ablation, recurrence-free survival, surgical resection
(Ann Surg 2016;263:538–545)
H
patients with chronic liver disease combined with improved imaging technologies had led to increased percentage of patients with HCC diagnosed at Barcelona Clı´nic Liver Cancer (BCLC) very early stage (BCLC stage 0, single tumor 2 cm).2 According to the current European Association for the Study of Liver (EASL) and American Association for the Study of Liver Diseases (AASLD) management guidelines, the recommended treatment modalities for very earlystage HCC include surgical resection (SR), radiofrequency ablation (RFA), and liver transplantation, on the basis of the presence or absence of portal hypertension and associated diseases.3,4 With 5-year survival rates around 70%, SR is generally regarded as the preferred treatment for patients with very early-stage HCC.5 Surgical resection offers therapeutic possibilities of complete eradication of satellite tumor lesions and microscopic tumor emboli in adjacent vasculature at the expense of more destruction to nontumor liver parenchyma.6 Alternatively, RFA is generally considered the most effective treatment among various percutaneous ablative therapies.7 Radiofrequency ablation was shown to achieve comparable overall survival (OS) and better tolerability for early HCC, and is recommended as the priority treatment for very early-stage HCC with impaired liver functional reserve.8,9 Because of limited availability of liver transplantation, SR and RFA are frequently employed in patients with very early-stage HCC.10,11 However, whether SR or RFA should be the choice of treatment in terms of OS and recurrencefree survival (RFS) is still under intense debate without universally accepted treatment algorithm.12 In this study, we aimed to investigate the impact of treatment selection on HCC recurrence and long-term survival in a prospectively followed-up cohort of patients with very early-stage HCC receiving SR or RFA as the primary treatment. A propensity score matching analysis was employed to overcome potential confounding bias at baseline; prognostic predictors affecting the outcome were determined.
PATIENTS AND METHODS
epatocellular carcinoma (HCC) is one of the most common malignancies worldwide, with rising incidence in both Eastern and Western countries.1 Wide application of surveillance programs in
Patients
From the Faculty of Medicine, National Yang-Ming University School of Medicine, Taipei, Taiwan; yDepartment of Medicine, Taipei Veterans General Hospital, Taipei, Taiwan; zDepartment of Biostatistics, UCLA, Los Angeles, CA; §Department of Surgery, Taipei Veterans General Hospital, Taipei, Taiwan; ôInstitute of Clinical Medicine, National Yang-Ming University School of Medicine, Taipei, Taiwan; jjDepartment of Radiology, Taipei Veterans General Hospital, Taipei, Taiwan; and Institute of Pharmacology, National Yang-Ming University School of Medicine, Taipei, Taiwan. Supported by grants from the Center of Excellence for Cancer Research at Taipei Veterans General Hospital (MOHW103-TD-B-111-02), Taiwan, from Taipei Veterans General Hospital (V104C-008), Taipei, Taiwan, and from the Ministry of Education, Aiming for the Top University Plan (103AC-P618), Taiwan. Disclosure: The authors declare no conflicts of interest. Reprints: Teh-Ia Huo, MD, Professor of Medicine, Division of Gastroenterology, Department of Medicine, Taipei Veterans General Hospital, No. 201, Sec. 2, Shipai Rd, Taipei 11217, Taiwan. E-mail: [email protected]. Copyright ß 2015 Wolters Kluwer Health, Inc. All rights reserved. ISSN: 0003-4932/14/26105-0821 DOI: 10.1097/SLA.0000000000001178
We retrospectively analyzed 3117 patients with newly diagnosed HCC admitted to Taipei Veterans General Hospital in more than a decade’s period from 2002 to 2013. Patients with single HCC 2 cm or smaller, with no evidence of vascular invasion and extrahepatic disease, and with good performance status (status 0 or 1) receiving SR or RFA as the primary treatment for HCC were identified. Comprehensive baseline information, including patient demographics, etiology of underlying liver disease, tumor characteristics, serum biochemistries, severity of cirrhosis, and performance status, was recorded at the time of diagnosis. All patients were followed up with imaging studies and serum a-fetoprotein (AFP) level every 3 to 6 months until death or dropout from the follow-up program. Tumor recurrence, subsequent therapy, and OS were recorded. Specific patient information was de-identified before statistical analysis. This study was approved by the institutional review board and complied with the standards of the Declaration of Helsinki and current ethical guidelines.
538 | www.annalsofsurgery.com
Annals of Surgery Volume 263, Number 3, March 2016
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
Annals of Surgery Volume 263, Number 3, March 2016
Diagnosis and Definitions The diagnosis of HCC was pathologically confirmed for patients receiving SR. For patients undergoing RFA, the diagnosis of HCC was confirmed by biopsy during the RFA procedure or according to the criteria from EASL and AASLD HCC management guidelines in force.3,4,13,14 Performance status was assessed at the time of diagnosis by the Eastern Cooperative Oncology Group performance scale.15 The Child-Turcotte-Pugh classification and the model for end-stage liver disease (MELD) score were used to define severity of chronic liver disease.16 Total tumor volume (TTV) was calculated on the basis of tumor diameter as previously reported.17 Local recurrence was defined as residual disease within or adjacent to the previously treated tumor site, whereas nonlocal recurrence was defined as emergence of the tumor elsewhere in or outside the liver.18
Treatment The criteria for resection and surgical techniques were described previously.19–21 Surgical resection was performed by an experienced surgical team. During parenchymal resection, attempts were made to keep an adequate surgical margin at least 1 cm.22 Anatomical resection was defined as removal of the entire Couinaud segment involved with the tumor. The extent of hepatectomy was determined by the retention rate of indocyanine green at 15 minutes after injection.19,23 Radiofrequency ablation was performed under sonographic guidance.7,24,25 The tumor was ablated in the automatic impedance control mode by a single 17-gauge electrode with the Cool-Tip Radiofrequency System (Radionics, Burlington, MA). Post-RFA sonography was performed to confirm that there were no immediate complications including hemorrhage or hematoma. For patients with local recurrent tumor(s) initially treated with RFA, repeated RFA was exclusively attempted whenever possible. Other recurrent tumors were treated with SR, RFA, transarterial chemoembolization (TACE), liver transplantation, percutaneous ethanol injection, sorafenib, or best supportive care, depending on clinical presentation and tumor status.
Process of Treatment Selection Patients were presented to a multidisciplinary HCC team of Taipei Veterans General Hospital for treatment guidance. Information of therapeutic benefits, risks, and long-term outcomes of different treatment modalities was comprehensively provided to individual patients with HCC. Shared decisions were made between patients and clinicians after detail counseling. Written informed consent was obtained before initiation of any definite treatment.
Propensity Score Matching Analysis To investigate the association between treatment selection and clinical outcomes in an observational, non-randomized study, a propensity score matching analysis was employed to reduce bias in patient selection and to generate matched pairs of patients to compare OS and RFS between patients receiving SR or RFA.25,26 Possible variables associated with the selection of treatment, including age, sex, serum bilirubin and AFP level, platelet count, TTV, and performance status, were comprehensively included in the generation of propensity scores. Binary logistic regression with selected variables was used to generate continuous propensity scores from 0 to 1. A one-to-one nearest-neighbor match without replacement between patients receiving SR and RFA was performed to select patients into subsequent analyses.27
Statistics The Mann-Whitney U test was used to compare continuous variables between 2 patient groups. The x2 and 2-tailed Fisher exact ß
2015 Wolters Kluwer Health, Inc. All rights reserved.
SR vs RFA in BCLC Very Early-Stage HCC
tests were employed to compare categorical data. Overall survival and RFS were examined by the Kaplan-Meier method with log-rank tests. Prognostic factors that were possibly linked to survival, including sex, etiology and severity of chronic liver disease, serum biochemistries, platelet count, TTV, treatment selection, and performance status, were comprehensively included in survival analyses. Factors with P value less than 0.1 in univariate analyses were introduced into the multivariate Cox proportional hazards model to determine the adjusted hazard ratios (HRs) and 95% confidence intervals (CIs). All statistical analyses were conducted with SPSS for Windows version 20.0 (IBM, NY). A 2-tailed P value less than 0.05 was considered statistically significant.
RESULTS Patients During the study period, a total of 237 patients with single HCC 2 cm or smaller were identified. Among these patients, 109 received SR and 128 underwent RFA as the primary anticancer treatment. Their median follow-up duration was 43 and 44 months, respectively. Anatomical resection was achieved in 100 patients in the SR group. The surgical margin was 1.1 1.1 cm. Fifteen patients in the SR group had tumors located near major hepatic vessels, whereas 13 patients had protruding tumors or had tumors at liver dome, near the gallbladder or near the kidney. For the RFA group, 7 patients had tumors located near major hepatic vasculature, and 6 patients had tumors at subcapsular region, liver dome, or near the gallbladder. The propensity score matching analysis identified 79 matched pairs of patients from each treatment arm to compare OS and RFS.
Characteristics, Recurrence, and Survival of All Patients Patients with very early-stage HCC receiving SR were significantly younger and had larger TTV compared with patients receiving RFA (both P < 0.05; Table 1). The SR group was also associated with higher serum albumin level, lower serum alanine transaminase (ALT) level, lower international normalized ratio (INR) of prothrombin time (PT), and higher platelet count (all P < 0.05). The OS was similar between SR and RFA groups (P ¼ 0.136; Fig. 1A). The 1-, 3-, and 5-year OS rates were 98%, 97%, and 81% versus 98%, 88%, and 76%, for SR and RFA groups, respectively. Patients receiving SR had a significantly better RFS compared with patients receiving RFA (P < 0.001; Fig. 2A). The estimated 1-, 3-, and 5-year RFS rates were 91%, 64%, and 49% for the SR group and 72%, 38%, and 24% for the RFA group. There was no treatmentrelated mortality in both treatment arms.
Characteristics, Recurrence, and Survival of Patients Selected in the Propensity Model The characteristics of patients with very early-stage HCC receiving SR or RFA selected in the propensity model are shown in Table 2. The 2 treatment groups had well-matched baseline characteristics including age, sex, etiology of underlying liver disease, serum biochemistry and AFP level, severity of chronic liver disease, tumor characteristics, and performance status (all P > 0.05). Patients in the propensity model receiving SR had significantly better OS when compared with those receiving RFA (P ¼ 0.034; Fig. 1B). The estimated 1-, 3-, and 5-year OS rates were 97%, 97%, and 80% for the SR group and 97%, 83%, and 66% for the RFA group. Patients undergoing SR also had better RFS compared with patients receiving RFA in the propensity model (P < 0.001; Fig. 2B). The estimated 1-, 3-, and 5-year RFS rates were 92%, 65%, and 48% for the SR group and 68%, 36%, and 18% for the RFA group. www.annalsofsurgery.com | 539
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
Annals of Surgery Volume 263, Number 3, March 2016
Liu et al
TABLE 1. Baseline Demographics in Patients With BCLC Very Early-Stage HCC Receiving SR or RFA SR (n ¼ 109)
Variables Age, yr, mean (SD) Male, n (%) Positive for HBsAg, n (%) Positive for anti-HCV, n (%) Alcoholism, n (%) Performance status 1, n (%) Serum biochemistry, mean (SD) Albumin, g/dL Bilirubin, mg/dL Creatinine, mg/dL ALT, U/L Sodium, mmol/L INR of PT, mean (SD) Platelets, 1000/mL, mean (SD) AFP, ng/mL, mean (SD) CTP score, mean (SD) MELD score, mean (SD) TTV, cm3, mean (SD)
RFA (n ¼ 128)
60 78 65 37 12 13
(13) (72) (60) (34) (11) (12)
64 84 62 55 13 15
(12) (66) (48) (43) (10) (12)
4.0 0.9 1.0 57 140 1.0 154 145 5.2 7.8 2.6
(0.5) (0.4) (0.6) (56) (2.8) (0.1) (68) (262) (0.4) (1.3) (1.3)
3.9 0.8 1.0 68 139 1.1 124 92 5.3 8.4 2.0
(0.5) (0.4) (0.9) (57) (3.5) (0.1) (62) (247) (0.5) (2.5) (1.1)
P 0.038 0.401 0.091 0.182 0.836 1.000 0.032 0.312 0.161 0.009 0.528 0.028
