Accepted Article
Radiofrequency Ablation versus Laser Ablation for the Treatment of Small Hepatocellular Carcinoma in Cirrhosis: a Randomized Trial1
Giovan Giuseppe Di Costanzo, MD, Raffaella Tortora, MD, Giuseppe D'Adamo, MD, Massimo De Luca, MD, Filippo Lampasi, MD, Luigi Addario, MD, Alfonso Galeota Lanza, MD, Francesco Paolo Picciotto, MD, Maria Teresa Tartaglione, MD, Gabriella Cordone, MD, Michele Imparato, MD, Silvana Mattera, MD, Claudio Maurizio Pacella, MD
From the Liver Unit, Cardarelli Hospital, Via A Cardarelli 9, 80131, Naples, Italy.
Corresponding author: Giovan Giuseppe Di Costanzo, Liver Unit, Cardarelli Hospital, Via A Cardarelli 9, 80131, Naples, Italy.
E-mail: [email protected] Fax: +39.0817472317
The study was presented in part as oral comunication at: -
ILCA Annual Conference, Hong Kong, september 2011;
-
AISF Annual Conference, Rome, february 2013;
-
ILCA Annual Conference, Washington, september 2013.
Running head: Radiofrequency versus Laser Ablation for HCC
This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as doi: 10.1111/jgh.12791 This article is protected by copyright. All rights reserved.
Accepted Article Abstract
Background and Aim In patients with cirrhosis and small hepatocellular carcinoma (HCC), thermal ablation is currently recognized as an effective local treatment. Among thermal procedures, radiofrequency ablation (RFA) is the most diffusely used and is the standard against which any new treatment should be compared. In retrospective studies, laser ablation (LA) resulted as safe and effective as RFA. Therefore, we performed a non-inferiority randomized trial comparing RFA with LA in patients with cirrhosis and HCC within Milan criteria. Methods
Overall 140 patients with 157 HCC nodules were randomly assigned to receive RFA or LA. The primary endpoint was the proportion of complete tumor ablation (CTA). Secondary endpoints were time to local tumor progression (TTLP) and overall survival (OS). Results
Per patient CTA rates after RFA and LA were 97.4% (95% CI, 91.0-99.3) and 95.7% (88.1-98.5),
respectively (difference = 1.4%, 95% CI from -6.0% to +9.0%). Per nodule CTA rates for RFA and LA were 97.4% (91.0-99.3) and 96.3% (89.6-98.7), respectively (difference = 1.1%, from -5.7% to +8.1%). The mean TTLP was comparable between RFA group (42.0 months; 95%CI, 36.83-47.3) and LA group (46.7 months; 95%CI, 41.5-51.9) (p=.591). The mean OS was 42 months in both groups and survival probability at one and three years was 94 and 89% in RFA group, 94 and 80% in LA group. Conclusion
This article is protected by copyright. All rights reserved.
Accepted Article
LA resulted not inferior to RFA in inducing the CTA of HCC nodules and therefore it should be considered as an evaluable alternative for thermal ablation of small HCC in cirrhotic patients.
Keywords: hepatocellular carcinoma, radiofrequency ablation, laser ablation
INTRODUCTION Hepatocellular carcinoma (HCC) is the sixth most frequent cancer worldwide and the leading cause of death among cirrhotic patients [1, 2]. In Western countries, only 30% of cases are diagnosed at an early stage when curative treatments can be applied [3]. A small number of these patients are candidates for resection and a more restricted number to liver transplantation [4]. According to internationally endorsed guidelines, thermal ablation is the mainstay of treatment in early stage unresectable HCC [4, 5]. Radiofrequency ablation (RFA) is the most widely employed technique
and is supported by major vendors. In retrospective studies, laser ablation (LA) resulted safe and effective in the treatment of HCC, but prospective trials are lacking [6-11]. Although both RFA and
LA are hyperthermic treatments, different heating mechanisms are involved. In RFA, resistive heating is generated by the agitation of ions in the tissue surrounding the electrodes due to the alternating directions of the electric field [12, 13]. In LA, thermal destruction of tissue is due to conversion into heat of adsorbed light that penetrates as a result of backward and forward scattering, reflection, and absorption [14-16]. These physical differences might have some effects on the
achievement of complete tumor ablation (CTA). For instance, there is evidence that laser light may be particularly efficient in causing death of endothelial cells, coagulation of microvessels, and progressive ischemic injury [17]. Therefore, we designed a study aimed to compare tumor response after RFA or LA of HCC within Milan criteria in cirrhotic patients.
This article is protected by copyright. All rights reserved.
Accepted Article
METHODS Patients
This is a randomized study approved by the local institutional review board and registered in ClinicalTrials.gov (identification number: NCT01096914). The study protocol conforms to the provisions of the Declaration of Helsinki (as revised in Tokyo 2004 and available at http://www.wma.net/en/30publications/10policies/b3/).
Inclusion criteria were as follows: (1) unresectable HCC (due to nodule location, multifocality,
presence of portal hypertension, age > 75 years, or comorbidity) or refusal of surgery; (2) solitary HCC ≤5.0 cm, or ≤3 lesions each ≤3.0 cm in diameter; (3) Child-Pugh class A or B; (4) a platelet count >40,000/µL or and INR < 2.0; (5) no previous HCC treatment. Exclusion criteria were as follows: (1) history of encephalopathy or refractory ascites; (2) vascular invasion or extrahepatic metastasis; (3) unfeasible percutaneous thermal ablation (incospicuous
nodules and tumors located within 5 mm of liver hilum or the main bile duct branches). From January 2009 to September 2012, 432 naïve HCCs in patients with cirrhosis were consecutively observed. Among these, 140 patients met the entry criteria, signed informed consent to study participation, and were enrolled into the study. Eighteen of these 140 (13%) patients were considered resectable but refused surgery. The diagnosis of HCC was histologically proven in 102 patients or established according to noninvasive criteria used by the European Association for the Study of the Liver [18]. On the basis of computer-generated random numbers, patients were assigned to either RFA (70 patients with 77 nodules) or LA (70 patients with 80 nodules) on the day
of hospitalization (Fig. 1).
Ablation procedures All procedures were performed percutaneously under conscious sedation and ultrasound guidance by two operators (GGDC, GDA) who had >5 years experience in performing an average of 20 This article is protected by copyright. All rights reserved.
Accepted Article
percutaneous ablations monthly. In agreement with the Italian Healthcare System guidelines,
patients were hospitalized for 2 days unless complications occurred. RFA was performed using a 17-gauge cooled-tip single electrode (Cool-Tip, Valleylab, Burlington, MA) with a 3-cm exposed portion. For each application, radiofrequency energy was delivered for 10-12 min even in tumors in high-risk location without the use of special methods to avoid unwanted complications (i.e. artificial pleural effusion, artificial ascites) [19]. For tumors >3.0 cm, multiple overlapping ablations and the pullback technique were used. The session ended when the hyperechoic ablation area was at least as large as the entire tumor and the still-activated needle was slowly withdrawn to avoid neoplastic seeding.
To perform LA, a commercially available system (Echolaser, Elesta Srl, Florence, Italy) composed of an ultrasound device and a diode laser unit was used. The beam was propagated through 300µm quartz-core bare fibers with flat tip that were inserted through 21-gauge needles. The details of LA procedure have been reported elsewhere [20, 21]. Depending on the size of the lesions, one to four fibers were positioned in the nodule, with spacing between fibers of 15mm. One fiber for nodules
up to 10 mm, two for nodules ranging between 10 and 20mm, and four fibers arranged in a square configuration for larger nodules were employed. In order to maximize the ablative effect and to make placement of multiple needles into tumors easy and precise, a needle guide allowing the insertion of two needles in a parallel fashion and at a prefixed distance was used. For a single illumination, laser light was applied for 4-6 minutes delivering 1200-1800 joules per fiber. For nodules >3.0 cm, multiple illuminations and the pullback technique were employed. We did not use specific methods for treating lesions in high-risk location [20, 22]. At the end of the treatment session, the still-illuminated fibers were withdrawn to induce hyperthermia along the needle tract to avoid neoplastic seeding. Complications of treatments were recorded according to the Society of Interventional Radiology (SIR) guidelines [23].
This article is protected by copyright. All rights reserved.
Accepted Article
Outcomes and assessments Four weeks after ablation, the treatment response was assessed by dynamic contrast-enhanced computed tomography (CT) or magnetic resonance imaging (MRI). In case of residual enhancing tumor tissue an additional session of ablation was given. The design of the study scheduled up to three ablation procedures during a six-month period. Primary outcome measure was CTA defined as absence of any contrast enhancement within or at the periphery of the HCC nodule [24]. Treatment failure was defined when imaging showed a residual nodule activity after the three scheduled ablations. For these patients, transarterial chemoembolization was recommended. Patients with CTA underwent follow-up studies, including -fetoprotein measurement and ultrasonography every 3 months, CT or MRI every 6 months and in any case of suspected tumor recurrence.
Secondary outcome measures were time to local progression (TTLP) and overall survival (OS). TTLP was defined as the time from CTA to reappearance of arterial enhancement on CT or MRI either within a treated tumor or near its borders, as previously defined by other groups [25, 26]. The
OS was defined as the interval between the first treatment and either death or last follow-up visit
before November 30, 2013. Statistical analyses This is a non-inferiority randomized control trial aiming to test if the efficacy of a not still validated treatment is not worse than a standard treatment minus a prefixed non-inferiority margin. The selection of this margin was based on literature data and clinical judgement. According to the published studies [7, 10, 27-31], the rate of CTA after RFA and LA is 96% and 86%, respectively. We evaluated that this might be a plausible difference from a clinical point of view and therefore we choose a non-inferiority margin of 10%. A dedicated statistical software (Study size, ver. 3.0.0, CreoStatHB, 2001-2013) was used to calculate the dimension of the two samples depending on our requirements: non-inferiority limit -0.10, power 80%, alpha error 2.5%, standard treatment event’s proportion 96%. A sample size per group of 70 patients was estimated. Continuity-corrected chiThis article is protected by copyright. All rights reserved.
Accepted Article
square and ninety-five percent confidence intervals (95% CI) with continuity correction for absolute difference in percentages between groups were used to reject the null hypothesis of difference between groups. Categorical variables were compared with the chi-square test and continuous variables with the Mann-Whitney test. A P value 3.0 cm or in high-risk location were included in LA arm, but the differences were not statistically significant.
Treatment response The CTA was achieved in 97.1% (95% CI, 90.2-99.2) of patients and in 97.4% (91.0-99.3) of nodules among RFA group. In LA group, CTA was observed in 95.7% (95% CI, 88.1-98.5) of patients and in 96.3% (89.6-98.7) of nodules. Therefore, the rate of CTA was comparable between the two techniques with a difference per patient of 1.4% (95% CI from -6.0% to +9.0%) and per nodule of 1.1% (from -5.7% to +8.1%) (P =.5). Overall treatment sessions to obtain CTA were 72
in RFA group and 82 in LA group (P =.058) (Table 2). In two patients of LA group treatment was This article is protected by copyright. All rights reserved.
Accepted Article
not repeated: in one due to the distant cancer progression and in the other for liver failure. Three patients in LA group and 1 patient in RFA group underwent to liver transplantation; evaluation of explanted livers showed complete necrosis of treated nodules.
HCC recurrence and survival Local tumor progression occurred in 25.7% of RFA patients and in 22.9% of LA patients. The mean TTLP was comparable between RFA (42.0 months; 95%CI, 36.83-47.3) and LA groups (46.7 months; 95%CI, 41.5-51.9) (p=.591). The mean local progression-free survival was 35.7 (95% CI, 30.6-40.9) months and 35.5 (30.1-40.8) months in RFA and LA groups, respectively (Fig. 2). In RFA group, recurrences were managed with TACE (n=6), iterative RFA (n=11), or were untreated for liver failure (n=1) and vascular cancer invasion (n=1). In LA group, recurrences were treated with TACE (n=5), iterative LA (n=6), liver transplantation (n=1), or were untreated due to liver decompensation (n=3) or extrahepatic metastasis (n=1). During the study period, 18 patients in the RFA group and 24 patients in the LA group had died. The causes of death are shown in table 3. Eight patients in the RFA group and 12 in the LA group died for liver failure (p=.47). In all cases, liver failure occurred > 4 weeks after ablation and was unrelated to the ablation procedures. Among these patients, 3/8 (37%) of RFA and 10/12 (83%) of LA group showed no evidence of cancer progression. The mean OS was 42 months in both groups, the 1- and 3-year survival probability was 94 and 89% in RFA group, 94 and 80% in LA group (Fig. 2).
Complications
There were no treatment-related deaths. After treatment, moderate pain (SIR class A) was recorded in 36% and 33% of RFA and LA patients, respectively. Self-limiting fever lasting 90% as reported with RFA in previous trials [30-32]. A slightly higher number of treatments for obtaining the CTA were needed in LA group than in RFA group. This finding might be due to the incorrect positioning of the multiple laser fibers, but the higher proportion of patients with larger nodules in LA group than in RFA group likely contributed to this result. Local tumor progression occurred in 26% and 23% of RFA and LA patients, respectively. The local progression rates after RFA vary widely across published studies from 2% and 53% [30-38] and several factors have been investigated as potential causes [39]. The achievement of large ablative margins to kill occult foci of cancer cells might decrease this progression rate [30, 40, 41]. However, cancer variables as cancer malignancy, size and location of nodules, and growth pattern have significant impact on recurrence. In this study we performed percutaneous ablation also in
patients with high-risk located lesions and this strategy had certainly some influence on local progression rate observed by us.
Differences between the two techniques are summarized in the table 4. For the LA with the multifiber technique, the cons include the need of positioning multiple needles into the tumor. However, after a learning periode of 3-6 months it is possible to have enough experience with this technique [42]. Furthermore, the use of multiple thin needles may theoretically increase the
This article is protected by copyright. All rights reserved.
Accepted Article
applicability of thermal ablation: nodules irregularly shaped or located in difficult areas as behind large vessels may be treated more easily and safely using thin needles [20, 22]. For treating a 3.0 cm nodule, ablation time was 12 minutes for RFA and 6 minutes for LA. The shorter duration of LA may be relevant when multiple ablations for large or multifocal HCC are required: in this case the procedural time becomes considerably longer with the use of RFA. This study was not formally designed to compare the costs of procedures, however considering only the price of the devices used for the trial, LA resulted cheaper than RFA (90,000 vs 144,000 euros). In patients with nodules > 2.0 cm in size, the cost of laser devices was 40% less than RFA, whereas for smaller nodules only two laser kits were employed with a cost that was 70% less than RFA. Both RFA and LA were safe procedures, self-limiting fever and moderate pain being the more frequent adverse events. The only major complication was subcutaneous seeding that was observed in two cases (1.4%), one per group. This frequency is comparable to that reported in large retrospective series ranging from 0.1 to 12.6% [36-38]. Seeding was surgically treated and it had no impact on survival or cancer progression. There are some limitations of this study. First, the number of patients is relatively small and therefore speculations about the efficacy of RFA or LA in nodules of different size cannot be done. Second, the two groups were not entirely balanced due to the random inclusion of more large and high-risk located nodules in LA group. Third, this is a single-institution study. The results may not be generalizable to patients with HCC in other centers or countries, therefore they will need to be externally confirmed. In conclusion, this is the first study that validates the use of LA for the treatment of HCC. LA resulted not inferior to RFA in achieving the CTA and therefore it should be considered as an evaluable alternative for thermal ablation of small HCC in cirrhotic patients.
This article is protected by copyright. All rights reserved.
Accepted Article
Potential conflict of interest: The authors indicated no potential conflicts of interest. Aknowledgments: We thank Mrs. Maria Barbato and Mr. Luca Iodice for technical assistance in performing the study.
REFERENCES
Figure 1. CONSORT diagram of the trial. BCLC, Barcelona Clinic Liver Cancer; RFA, radiofrequency ablation; LA, laser ablation. Figure 2. Overall (A) and local progression-free (B) survival curves for the RFA and LA groups.
This article is protected by copyright. All rights reserved.
Accepted Article
Table 1. Baseline characteristics of patients RFA
LA
n=70
n=70
Pvalue
Males, n (%)
53 (75.7)
47 (67.1)
0.350
Median age (range)
70 (50-83)
70 (36-84)
0.588
HCV, n (%)
53 (75.7)
60 (85.7)
0.198
Child A, n (%)
63 (90.0)
67 (95.7)
0.325
Median MELD score (range)
9.0 (6-17)
8.5 (6-21)
0.750
F1 varices, n (%)
16/64 (25.0)
16/62 (25.8)
1.0
F2-F3 varices, n (%)
20/64 (31.3)
15/62 (24.2)
0.429
41 (58.6)
37 (52.9)
0.610
63/70 (90.0)
61/70 (87.1)
0.791
Diameter mm, mean ± SD
25.5 ± 6.6
26.2 ± 10.4
0.965
Diameter ≤20 mm, n (%) Diameter 21-30 mm, n (%)
25/77 (32.5) 38/77 (49.3)
32/80 (40.0) 23/80 (28.7)
0.051
Diameter 31-40 mm, n (%)
11/77 (14.3)
18/80 (22.5)
Diameter 41-50 mm, n (%)
3/77 (3.9)
7/80 (8.8)
9 2
13 8
0.063
Edmondson grade I-II, n (%)
27/48 (56.3)
36/54 (66.7)
0.312
-fetoprotein ng/mL, median (range)
9 (1-1025)
20.5 (1-1526)
0.057
27.7 ± 12.7 (9-56)
0.790
Platelet count 2 cm as
This article is protected by copyright. All rights reserved.
Accepted Article
-for nodules
Euros
Euros
2,000
1,200
Euros
Euros
compared to RFA device
> 2 cm
This article is protected by copyright. All rights reserved.
Accepted Article
JGH_12791_F1.tiff
Accepted Article JGH_12791_F2.tiff
Radiofrequency Ablation versus Laser Ablation for the Treatment of Small Hepatocellular Carcinoma in Cirrhosis: a Randomized Trial1
Giovan Giuseppe Di Costanzo, MD, Raffaella Tortora, MD, Giuseppe D'Adamo, MD, Massimo De Luca, MD, Filippo Lampasi, MD, Luigi Addario, MD, Alfonso Galeota Lanza, MD, Francesco Paolo Picciotto, MD, Maria Teresa Tartaglione, MD, Gabriella Cordone, MD, Michele Imparato, MD, Silvana Mattera, MD, Claudio Maurizio Pacella, MD
From the Liver Unit, Cardarelli Hospital, Via A Cardarelli 9, 80131, Naples, Italy.
Corresponding author: Giovan Giuseppe Di Costanzo, Liver Unit, Cardarelli Hospital, Via A Cardarelli 9, 80131, Naples, Italy.
E-mail: [email protected] Fax: +39.0817472317
The study was presented in part as oral comunication at: -
ILCA Annual Conference, Hong Kong, september 2011;
-
AISF Annual Conference, Rome, february 2013;
-
ILCA Annual Conference, Washington, september 2013.
Running head: Radiofrequency versus Laser Ablation for HCC
This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as doi: 10.1111/jgh.12791 This article is protected by copyright. All rights reserved.
Accepted Article Abstract
Background and Aim In patients with cirrhosis and small hepatocellular carcinoma (HCC), thermal ablation is currently recognized as an effective local treatment. Among thermal procedures, radiofrequency ablation (RFA) is the most diffusely used and is the standard against which any new treatment should be compared. In retrospective studies, laser ablation (LA) resulted as safe and effective as RFA. Therefore, we performed a non-inferiority randomized trial comparing RFA with LA in patients with cirrhosis and HCC within Milan criteria. Methods
Overall 140 patients with 157 HCC nodules were randomly assigned to receive RFA or LA. The primary endpoint was the proportion of complete tumor ablation (CTA). Secondary endpoints were time to local tumor progression (TTLP) and overall survival (OS). Results
Per patient CTA rates after RFA and LA were 97.4% (95% CI, 91.0-99.3) and 95.7% (88.1-98.5),
respectively (difference = 1.4%, 95% CI from -6.0% to +9.0%). Per nodule CTA rates for RFA and LA were 97.4% (91.0-99.3) and 96.3% (89.6-98.7), respectively (difference = 1.1%, from -5.7% to +8.1%). The mean TTLP was comparable between RFA group (42.0 months; 95%CI, 36.83-47.3) and LA group (46.7 months; 95%CI, 41.5-51.9) (p=.591). The mean OS was 42 months in both groups and survival probability at one and three years was 94 and 89% in RFA group, 94 and 80% in LA group. Conclusion
This article is protected by copyright. All rights reserved.
Accepted Article
LA resulted not inferior to RFA in inducing the CTA of HCC nodules and therefore it should be considered as an evaluable alternative for thermal ablation of small HCC in cirrhotic patients.
Keywords: hepatocellular carcinoma, radiofrequency ablation, laser ablation
INTRODUCTION Hepatocellular carcinoma (HCC) is the sixth most frequent cancer worldwide and the leading cause of death among cirrhotic patients [1, 2]. In Western countries, only 30% of cases are diagnosed at an early stage when curative treatments can be applied [3]. A small number of these patients are candidates for resection and a more restricted number to liver transplantation [4]. According to internationally endorsed guidelines, thermal ablation is the mainstay of treatment in early stage unresectable HCC [4, 5]. Radiofrequency ablation (RFA) is the most widely employed technique
and is supported by major vendors. In retrospective studies, laser ablation (LA) resulted safe and effective in the treatment of HCC, but prospective trials are lacking [6-11]. Although both RFA and
LA are hyperthermic treatments, different heating mechanisms are involved. In RFA, resistive heating is generated by the agitation of ions in the tissue surrounding the electrodes due to the alternating directions of the electric field [12, 13]. In LA, thermal destruction of tissue is due to conversion into heat of adsorbed light that penetrates as a result of backward and forward scattering, reflection, and absorption [14-16]. These physical differences might have some effects on the
achievement of complete tumor ablation (CTA). For instance, there is evidence that laser light may be particularly efficient in causing death of endothelial cells, coagulation of microvessels, and progressive ischemic injury [17]. Therefore, we designed a study aimed to compare tumor response after RFA or LA of HCC within Milan criteria in cirrhotic patients.
This article is protected by copyright. All rights reserved.
Accepted Article
METHODS Patients
This is a randomized study approved by the local institutional review board and registered in ClinicalTrials.gov (identification number: NCT01096914). The study protocol conforms to the provisions of the Declaration of Helsinki (as revised in Tokyo 2004 and available at http://www.wma.net/en/30publications/10policies/b3/).
Inclusion criteria were as follows: (1) unresectable HCC (due to nodule location, multifocality,
presence of portal hypertension, age > 75 years, or comorbidity) or refusal of surgery; (2) solitary HCC ≤5.0 cm, or ≤3 lesions each ≤3.0 cm in diameter; (3) Child-Pugh class A or B; (4) a platelet count >40,000/µL or and INR < 2.0; (5) no previous HCC treatment. Exclusion criteria were as follows: (1) history of encephalopathy or refractory ascites; (2) vascular invasion or extrahepatic metastasis; (3) unfeasible percutaneous thermal ablation (incospicuous
nodules and tumors located within 5 mm of liver hilum or the main bile duct branches). From January 2009 to September 2012, 432 naïve HCCs in patients with cirrhosis were consecutively observed. Among these, 140 patients met the entry criteria, signed informed consent to study participation, and were enrolled into the study. Eighteen of these 140 (13%) patients were considered resectable but refused surgery. The diagnosis of HCC was histologically proven in 102 patients or established according to noninvasive criteria used by the European Association for the Study of the Liver [18]. On the basis of computer-generated random numbers, patients were assigned to either RFA (70 patients with 77 nodules) or LA (70 patients with 80 nodules) on the day
of hospitalization (Fig. 1).
Ablation procedures All procedures were performed percutaneously under conscious sedation and ultrasound guidance by two operators (GGDC, GDA) who had >5 years experience in performing an average of 20 This article is protected by copyright. All rights reserved.
Accepted Article
percutaneous ablations monthly. In agreement with the Italian Healthcare System guidelines,
patients were hospitalized for 2 days unless complications occurred. RFA was performed using a 17-gauge cooled-tip single electrode (Cool-Tip, Valleylab, Burlington, MA) with a 3-cm exposed portion. For each application, radiofrequency energy was delivered for 10-12 min even in tumors in high-risk location without the use of special methods to avoid unwanted complications (i.e. artificial pleural effusion, artificial ascites) [19]. For tumors >3.0 cm, multiple overlapping ablations and the pullback technique were used. The session ended when the hyperechoic ablation area was at least as large as the entire tumor and the still-activated needle was slowly withdrawn to avoid neoplastic seeding.
To perform LA, a commercially available system (Echolaser, Elesta Srl, Florence, Italy) composed of an ultrasound device and a diode laser unit was used. The beam was propagated through 300µm quartz-core bare fibers with flat tip that were inserted through 21-gauge needles. The details of LA procedure have been reported elsewhere [20, 21]. Depending on the size of the lesions, one to four fibers were positioned in the nodule, with spacing between fibers of 15mm. One fiber for nodules
up to 10 mm, two for nodules ranging between 10 and 20mm, and four fibers arranged in a square configuration for larger nodules were employed. In order to maximize the ablative effect and to make placement of multiple needles into tumors easy and precise, a needle guide allowing the insertion of two needles in a parallel fashion and at a prefixed distance was used. For a single illumination, laser light was applied for 4-6 minutes delivering 1200-1800 joules per fiber. For nodules >3.0 cm, multiple illuminations and the pullback technique were employed. We did not use specific methods for treating lesions in high-risk location [20, 22]. At the end of the treatment session, the still-illuminated fibers were withdrawn to induce hyperthermia along the needle tract to avoid neoplastic seeding. Complications of treatments were recorded according to the Society of Interventional Radiology (SIR) guidelines [23].
This article is protected by copyright. All rights reserved.
Accepted Article
Outcomes and assessments Four weeks after ablation, the treatment response was assessed by dynamic contrast-enhanced computed tomography (CT) or magnetic resonance imaging (MRI). In case of residual enhancing tumor tissue an additional session of ablation was given. The design of the study scheduled up to three ablation procedures during a six-month period. Primary outcome measure was CTA defined as absence of any contrast enhancement within or at the periphery of the HCC nodule [24]. Treatment failure was defined when imaging showed a residual nodule activity after the three scheduled ablations. For these patients, transarterial chemoembolization was recommended. Patients with CTA underwent follow-up studies, including -fetoprotein measurement and ultrasonography every 3 months, CT or MRI every 6 months and in any case of suspected tumor recurrence.
Secondary outcome measures were time to local progression (TTLP) and overall survival (OS). TTLP was defined as the time from CTA to reappearance of arterial enhancement on CT or MRI either within a treated tumor or near its borders, as previously defined by other groups [25, 26]. The
OS was defined as the interval between the first treatment and either death or last follow-up visit
before November 30, 2013. Statistical analyses This is a non-inferiority randomized control trial aiming to test if the efficacy of a not still validated treatment is not worse than a standard treatment minus a prefixed non-inferiority margin. The selection of this margin was based on literature data and clinical judgement. According to the published studies [7, 10, 27-31], the rate of CTA after RFA and LA is 96% and 86%, respectively. We evaluated that this might be a plausible difference from a clinical point of view and therefore we choose a non-inferiority margin of 10%. A dedicated statistical software (Study size, ver. 3.0.0, CreoStatHB, 2001-2013) was used to calculate the dimension of the two samples depending on our requirements: non-inferiority limit -0.10, power 80%, alpha error 2.5%, standard treatment event’s proportion 96%. A sample size per group of 70 patients was estimated. Continuity-corrected chiThis article is protected by copyright. All rights reserved.
Accepted Article
square and ninety-five percent confidence intervals (95% CI) with continuity correction for absolute difference in percentages between groups were used to reject the null hypothesis of difference between groups. Categorical variables were compared with the chi-square test and continuous variables with the Mann-Whitney test. A P value 3.0 cm or in high-risk location were included in LA arm, but the differences were not statistically significant.
Treatment response The CTA was achieved in 97.1% (95% CI, 90.2-99.2) of patients and in 97.4% (91.0-99.3) of nodules among RFA group. In LA group, CTA was observed in 95.7% (95% CI, 88.1-98.5) of patients and in 96.3% (89.6-98.7) of nodules. Therefore, the rate of CTA was comparable between the two techniques with a difference per patient of 1.4% (95% CI from -6.0% to +9.0%) and per nodule of 1.1% (from -5.7% to +8.1%) (P =.5). Overall treatment sessions to obtain CTA were 72
in RFA group and 82 in LA group (P =.058) (Table 2). In two patients of LA group treatment was This article is protected by copyright. All rights reserved.
Accepted Article
not repeated: in one due to the distant cancer progression and in the other for liver failure. Three patients in LA group and 1 patient in RFA group underwent to liver transplantation; evaluation of explanted livers showed complete necrosis of treated nodules.
HCC recurrence and survival Local tumor progression occurred in 25.7% of RFA patients and in 22.9% of LA patients. The mean TTLP was comparable between RFA (42.0 months; 95%CI, 36.83-47.3) and LA groups (46.7 months; 95%CI, 41.5-51.9) (p=.591). The mean local progression-free survival was 35.7 (95% CI, 30.6-40.9) months and 35.5 (30.1-40.8) months in RFA and LA groups, respectively (Fig. 2). In RFA group, recurrences were managed with TACE (n=6), iterative RFA (n=11), or were untreated for liver failure (n=1) and vascular cancer invasion (n=1). In LA group, recurrences were treated with TACE (n=5), iterative LA (n=6), liver transplantation (n=1), or were untreated due to liver decompensation (n=3) or extrahepatic metastasis (n=1). During the study period, 18 patients in the RFA group and 24 patients in the LA group had died. The causes of death are shown in table 3. Eight patients in the RFA group and 12 in the LA group died for liver failure (p=.47). In all cases, liver failure occurred > 4 weeks after ablation and was unrelated to the ablation procedures. Among these patients, 3/8 (37%) of RFA and 10/12 (83%) of LA group showed no evidence of cancer progression. The mean OS was 42 months in both groups, the 1- and 3-year survival probability was 94 and 89% in RFA group, 94 and 80% in LA group (Fig. 2).
Complications
There were no treatment-related deaths. After treatment, moderate pain (SIR class A) was recorded in 36% and 33% of RFA and LA patients, respectively. Self-limiting fever lasting 90% as reported with RFA in previous trials [30-32]. A slightly higher number of treatments for obtaining the CTA were needed in LA group than in RFA group. This finding might be due to the incorrect positioning of the multiple laser fibers, but the higher proportion of patients with larger nodules in LA group than in RFA group likely contributed to this result. Local tumor progression occurred in 26% and 23% of RFA and LA patients, respectively. The local progression rates after RFA vary widely across published studies from 2% and 53% [30-38] and several factors have been investigated as potential causes [39]. The achievement of large ablative margins to kill occult foci of cancer cells might decrease this progression rate [30, 40, 41]. However, cancer variables as cancer malignancy, size and location of nodules, and growth pattern have significant impact on recurrence. In this study we performed percutaneous ablation also in
patients with high-risk located lesions and this strategy had certainly some influence on local progression rate observed by us.
Differences between the two techniques are summarized in the table 4. For the LA with the multifiber technique, the cons include the need of positioning multiple needles into the tumor. However, after a learning periode of 3-6 months it is possible to have enough experience with this technique [42]. Furthermore, the use of multiple thin needles may theoretically increase the
This article is protected by copyright. All rights reserved.
Accepted Article
applicability of thermal ablation: nodules irregularly shaped or located in difficult areas as behind large vessels may be treated more easily and safely using thin needles [20, 22]. For treating a 3.0 cm nodule, ablation time was 12 minutes for RFA and 6 minutes for LA. The shorter duration of LA may be relevant when multiple ablations for large or multifocal HCC are required: in this case the procedural time becomes considerably longer with the use of RFA. This study was not formally designed to compare the costs of procedures, however considering only the price of the devices used for the trial, LA resulted cheaper than RFA (90,000 vs 144,000 euros). In patients with nodules > 2.0 cm in size, the cost of laser devices was 40% less than RFA, whereas for smaller nodules only two laser kits were employed with a cost that was 70% less than RFA. Both RFA and LA were safe procedures, self-limiting fever and moderate pain being the more frequent adverse events. The only major complication was subcutaneous seeding that was observed in two cases (1.4%), one per group. This frequency is comparable to that reported in large retrospective series ranging from 0.1 to 12.6% [36-38]. Seeding was surgically treated and it had no impact on survival or cancer progression. There are some limitations of this study. First, the number of patients is relatively small and therefore speculations about the efficacy of RFA or LA in nodules of different size cannot be done. Second, the two groups were not entirely balanced due to the random inclusion of more large and high-risk located nodules in LA group. Third, this is a single-institution study. The results may not be generalizable to patients with HCC in other centers or countries, therefore they will need to be externally confirmed. In conclusion, this is the first study that validates the use of LA for the treatment of HCC. LA resulted not inferior to RFA in achieving the CTA and therefore it should be considered as an evaluable alternative for thermal ablation of small HCC in cirrhotic patients.
This article is protected by copyright. All rights reserved.
Accepted Article
Potential conflict of interest: The authors indicated no potential conflicts of interest. Aknowledgments: We thank Mrs. Maria Barbato and Mr. Luca Iodice for technical assistance in performing the study.
REFERENCES
Figure 1. CONSORT diagram of the trial. BCLC, Barcelona Clinic Liver Cancer; RFA, radiofrequency ablation; LA, laser ablation. Figure 2. Overall (A) and local progression-free (B) survival curves for the RFA and LA groups.
This article is protected by copyright. All rights reserved.
Accepted Article
Table 1. Baseline characteristics of patients RFA
LA
n=70
n=70
Pvalue
Males, n (%)
53 (75.7)
47 (67.1)
0.350
Median age (range)
70 (50-83)
70 (36-84)
0.588
HCV, n (%)
53 (75.7)
60 (85.7)
0.198
Child A, n (%)
63 (90.0)
67 (95.7)
0.325
Median MELD score (range)
9.0 (6-17)
8.5 (6-21)
0.750
F1 varices, n (%)
16/64 (25.0)
16/62 (25.8)
1.0
F2-F3 varices, n (%)
20/64 (31.3)
15/62 (24.2)
0.429
41 (58.6)
37 (52.9)
0.610
63/70 (90.0)
61/70 (87.1)
0.791
Diameter mm, mean ± SD
25.5 ± 6.6
26.2 ± 10.4
0.965
Diameter ≤20 mm, n (%) Diameter 21-30 mm, n (%)
25/77 (32.5) 38/77 (49.3)
32/80 (40.0) 23/80 (28.7)
0.051
Diameter 31-40 mm, n (%)
11/77 (14.3)
18/80 (22.5)
Diameter 41-50 mm, n (%)
3/77 (3.9)
7/80 (8.8)
9 2
13 8
0.063
Edmondson grade I-II, n (%)
27/48 (56.3)
36/54 (66.7)
0.312
-fetoprotein ng/mL, median (range)
9 (1-1025)
20.5 (1-1526)
0.057
27.7 ± 12.7 (9-56)
0.790
Platelet count 2 cm as
This article is protected by copyright. All rights reserved.
Accepted Article
-for nodules
Euros
Euros
2,000
1,200
Euros
Euros
compared to RFA device
> 2 cm
This article is protected by copyright. All rights reserved.
Accepted Article
JGH_12791_F1.tiff
Accepted Article JGH_12791_F2.tiff
