Ablation of Perivascular Hepatic Malignant Tumors with Irreversible Electroporation T Peter Kingham, MD, Ami M Karkar, MD, Michael I D’Angelica, MD, FACS, Peter J Allen, MD, FACS, Ronald P DeMatteo, MD, FACS, George I Getrajdman, MD, Constantinos T Sofocleous, MD, Stephen B Solomon, MD, William R Jarnagin, MD, FACS, Yuman Fong, MD, FACS Ablation is increasingly used to treat primary and secondary liver cancer. Ablation near portal pedicles and hepatic veins is challenging. Irreversible electroporation (IRE) is a new ablation technique that does not rely on heat and, in animals, appears to be safe and effective when applied near hepatic veins and portal pedicles. This study evaluated the safety and short-term outcomes of IRE to ablate perivascular malignant liver tumors. STUDY DESIGN: A retrospective review of patients treated with IRE between January 1, 2011 and November 2, 2011 was performed. Patients were selected for IRE when resection or thermal ablation was not indicated due to tumor location. Treatment outcomes were classified by local, regional, and systemic recurrence and complications. Local failure was defined as abnormal enhancement at the periphery of an ablation defect on post-procedure contrast imaging. RESULTS: Twenty-eight patients had 65 tumors treated. Twenty-two patients (79%) were treated via an open approach and 6 (21%) were treated percutaneously. Median tumor size was 1 cm (range 0.5 to 5 cm). Twenty-five tumors were ⬍1 cm from a major hepatic vein; 16 were ⬍1 cm from a major portal pedicle. Complications included 1 intraoperative arrhythmia and 1 postoperative portal vein thrombosis. Overall morbidity was 3%. There were no treatment-associated mortalities. At median follow-up of 6 months, there was 1 tumor with persistent disease (1.9%) and 3 tumors recurred locally (5.7%). CONCLUSIONS: This early analysis of IRE treatment of perivascular malignant hepatic tumors demonstrates safety for treating liver malignancies. Larger studies and longer follow-up are necessary to determine long-term efficacy. (J Am Coll Surg 2012;215:379–387. © 2012 by the American College of Surgeons) BACKGROUND:

dates for resection due to the location of the tumor, comorbidities, or earlier hepatic resection.3 There is, however, a high recurrence rate with ablative techniques, ranging from 4% to 43% in patients with hepatocellular carcinoma (HCC)4,5 or metastatic colorectal cancer.6-8 Some tumors cannot be effectively or safely ablated with RFA or MWA techniques. Proximity to major portal pedicles or hepatic veins has been associated with suboptimal ablation and high recurrence rates.9 In addition, complications can be caused by thermal damage to major bile ducts or blood vessels. Results of animal studies suggest that irreversible electroporation (IRE) technology can be useful in treating patients with perivascular tumors that cannot be treated safely or effectively by RFA or MWA.10 Electroporation is the permeabilization of the cell membrane via application of electrical pulses across the cell, which causes cell death.11 Its advantage lies in its avoidance of injury to the bile ducts and vessels within an organ, while

Ablative therapies are being used increasingly to treat primary and metastatic cancer in the liver. This treatment is relevant clinically because half of all patients with colorectal cancer will develop liver metastases, and hepatocellular carcinoma is the 5th most common cancer worldwide.1,2 The most common techniques in current use include radiofrequency ablation (RFA) and microwave ablation (MWA). These techniques are used in patients who are poor candiDisclosure Information: Dr Solomon is the principal investigator on research grants funded by Angiodynamics and Johnson & Johnson, and he is a research consultant for and is paid a consultancy fee by Covidien. All other authors have nothing to disclose. Received February 11, 2012; Revised April 10, 2012; Accepted April 11, 2012. From the Departments of Surgery (Kingham, Karkar, D’Angelica, Allen, DeMatteo, Jarnagin, Fong) and Radiology (Getrajdman, Sofocleous, Solomon), Memorial Sloan-Kettering Cancer Center, New York, NY. Correspondence address: T Peter Kingham, MD, Department of Surgery, Memorial Sloan-Kettering Cancer Center, 1275 York Ave, New York, NY 10065. email: [email protected]

© 2012 by the American College of Surgeons Published by Elsevier Inc.

379

ISSN 1072-7515/12/$36.00 http://dx.doi.org/10.1016/j.jamcollsurg.2012.04.029

380

Kingham et al

Hepatic Irreversible Electroporation

Abbreviations and Acronyms

AST HCC IRE MWA RFA

⫽ ⫽ ⫽ ⫽ ⫽

aspartate aminotransferase hepatocellular carcinoma irreversible electroporation microwave ablation radiofrequency ablation

still causing tumor (and normal parenchymal) cell death.10,12 The exact mechanism by which electrical impulses permeabilize the cell membrane is not fully understood. Unlike thermal ablation techniques that destroy all normal and pathologic tissue and blood vessels in the ablation zone, IRE is discriminate. Normal and pathologic tissue are destroyed by IRE, but vessels and bile ducts are not.10 The endothelium of these structures is affected by IRE, but it repopulates shortly after treatment.13,14 The effect on normal and pathologic tissue might be caused by small nanopores that open due to IRE, leading to cell membrane disruption. An alternative theory is that gap junctions in vessel walls allow electrical pulses to travel through the walls rather than form nanopores, therefore protecting them from IRE-induced cell death. Possible advantages of IRE include its short duration of application, lack of a heat-sink effect, absence of thermal damage to surrounding structures, and conservation of connective tissue and large blood vessels. There are limited data on the safety and efficacy of IRE for treating liver tumors in humans.15 Although safety of IRE for treating liver tumors has been shown in porcine models, it is unclear how reliably this translates to humans. In addition, there are few data on the efficacy of IRE in treating human liver tumors. This article reports on the safety and short-term efficacy of IRE for liver tumors in humans, including a subset analysis of tumors ⬍1 cm from major hepatic veins or portal pedicles.

METHODS Patients with tumors that were not appropriate for resection because of pathologic subtype and disease stage, tumor location, and/or disease extent, and that were suboptimally located for RFA or MWA as determined by the treating physician, were selected for IRE. Tumors meeting these criteria were ⬍2 cm from third-order or larger hepatic veins or portal pedicles. Tumors ⬎2 cm from a major biliary or vascular structure were treated in patients when IRE was already being used concurrently on a perivascular tumor in that patient’s liver. The NanoKnife (Angiodynamics) IRE device was used in all patients. The device is approved by the US Food and Drug Administration for use in ablating soft tissue. This retrospective review was approved

J Am Coll Surg

by the Institutional Review Board at Memorial SloanKettering Cancer Center. The IRE device comprises a generator, a foot pedal, and 15- or 25-cm⫺long electrodes.15 The electrodes used to treat these patients have a tip length ranging from 5 to 40 mm. This distance represents the active tip of the electrode and the remainder of the needle is insulated. The IRE device generates 1,500 to 3,000 V. Voltage was determined by a standard algorithm (Angiodynamics) that uses factors such as the intended size of the ablation zone, the number of probes, the distance between probes, and the length of the active electrode tip.15 The size or shape of the tumor determined the number of electrodes used. All patients were staged preoperatively with contrastenhanced CT scan and/or MRI scan of the chest, abdomen, and pelvis. Contraindications to using IRE included location of lesions in the vicinity of defibrillators or pacemakers, and history of cardiac arrhythmia or recent myocardial infarction. In the operating room, a laparotomy was performed and IRE was performed either alone or in combination with a liver resection or thermal ablation method, or with implantation of an hepatic artery infusion pump. In the interventional radiology suite, probes were placed with CT scan guidance in patients who were deemed non⫺surgical candidates. All patients were managed with general anesthesia, mechanical ventilation, and neuromuscular blockade to ensure complete paralysis. The delivery of electrical impulses to the liver tumors was automatically synchronized to each patient’s cardiac cycle.16 The electrodes were placed at a predetermined distance to bracket the tumor as described by the standard algorithm, with the assistance of ultrasound or CT guidance. The ablation zone was calculated to include a 1-cm margin around the entire tumor. Before delivering the 90 therapeutic pulses, a test pulse at 270 V was delivered. After the test pulse confirmed adequate conductivity, 90 pulses were delivered in less than 2 minutes. When the current generated by the electrodes exceeded 48 amps, those electrodes were withdrawn from the therapeutic algorithm and pulses between those electrodes were aborted. More than 90 pulses were used when a tumor was re-treated. Re-treatment occurred if the first treatment was aborted before 90 pulses. After the procedure, patients were followed with contrast-enhanced CT scans or MRI in the immediate perioperative period, at 1 to 3 months, and at 6 months. In addition, blood tests to measure hepatic function were examined serially. Tumor response was defined as a loss of enhancement for hypervascular tumors and as a lack of persistent tumor rim enhancement for hypovascular tumors on contrast imaging studies. Persistent disease was

Vol. 215, No. 3, September 2012

Kingham et al

Table 1. Patient Demographics Variable

Total patients, n Median follow-up, d (range) Treated tumors, n Procedures, n Tumors treated per patient, n, median (range) Age, y, median (range) Sex, n % Male Diabetes mellitus, n (%) Coronary artery disease, n (%) Absent Tumor type, n (%) Metastatic colorectal cancer Hepatocellular carcinoma Metastatic pancreatic neuroendocrine tumor Metastatic ampullary carcinoma Hemangiopericytoma Leiomyosarcoma metastasis Patients treated with preoperative chemotherapy, n (%) Patients treated with postoperative chemotherapy, n (%) Patients treated with perioperative pump chemotherapy, n (%)

Data

28 181 (11⫺272) 65 31 1.5 (1⫺6) 51 (32⫺81) 17 (61) 0 27 (96) 21 (75) 2 (7) 2 1 1 1

(7) (3.7) (3.7) (3.7)

24 (86) 20 (71) 2 (7)

defined as residual tumor enhancement on the first posttreatment imaging study, and local recurrence was defined as an enhancing tumor within 1 cm of an ablation zone. Complications were graded on a scale from 1 to 5.17

RESULTS From January 1, 2011 to November 2, 2011, twenty-eight patients had hepatic tumors that were treated with IRE (Table 1). These 28 patients had a total of 31 separate procedures to treat 65 tumors. Most tumors did not have a tissue diagnosis to confirm their malignancy, but the imaging characteristics and clinical scenario were consistent with a malignant tumor. The large majority of patients (75%) had colorectal cancer liver metastases. Other tumor types treated with IRE included HCC (7%), metastatic pancreatic neuroendocrine tumor (7%), ampullary carcinoma metastasis (3.7%), hemangiopericytoma (3.7%), and leiomyosarcoma metastasis (3.7%). Most patients were treated with preoperative chemotherapy (86%) and postoperative chemotherapy (71%). The majority of tumors treated with IRE were small. Median tumor size was 1 cm (range 0.5 to 5 cm), with 60% of tumors ⱕ1 cm in greatest diameter (Table 2). Only 2%

Hepatic Irreversible Electroporation

381

of tumors had diameter ⬎3 cm. On a per-lesion basis, most tumors were treated with both preoperative (89%) and postoperative (67%) chemotherapy. The liver was divided into 4 anatomic areas that have similar ablation approaches (Fig. 1). Fifty-five percent of tumors were located in segments 4a, 7, or 8. Tumors were located in the anterior and lateral segments of the left liver (segments 2, 3, or 4b) in 19%, and 17% of tumors were present in segments 5 or 6. Percutaneous IRE was performed in 6 patients for recurrences after a resection. Laparotomy was performed in 22 patients. The 6 patients treated percutaneously had 12 tumors in total. Median largest diameter of the percutaneous tumors was 1 cm (range 0.5 to 5 cm). Distance from the tumor margin to the closest major hepatic vein or portal pedicle was measured on preoperative imaging. Tumors were ⱕ1 cm from a major hepatic vein in 57% of cases, and 40% of tumors were ⱕ1 cm from a portal pedicle. Five tumors were within 1 cm of both a major hepatic vein and a portal pedicle. Many tumors were Table 2. Tumor Characteristics Variable

Tumor size (largest diameter), (n ⫽ 60), cm ⱕ1, n (%) 1.1⫺2, n (%) 2.1⫺3, n (%) ⬎3, n (%) Median largest diameter, cm (range) Ablation size, (largest diameter), (n ⫽ 43), cm, median (range) Ablation size, (n ⫽ 41), cm/tumor size, cm, median (range) Location of tumor, n (%) Caudate lobe Segments 4a/7/8 Segments 2/3/4b Segments 5/6 Tumors treated with preoperative chemotherapy, n (%) Tumors treated with postoperative chemotherapy, n (%) Proximity to major hepatic vein, cm 0-0.5, n (%) 0.6-1, n (%) ⬎1 n (%) Proximity to major bile ducts, cm 0-0.5, n (%) 0.6-1, n (%) ⬎1, n (%)

Data

36/60 11/60 12/60 1/60

(60) (18) (20) (2)

1 (0.5⫺5) 3.25 (0.8⫺9.4) 2.5 (0.6⫺9.4) 5/59 33/59 11/59 10/59

(8) (56) (19) (17)

58/65 (89) 37/55 (67) 19/43 (44) 6/43 (14) 18/43 (42) 12/40 (30) 4/40 (10) 24/40 (60)

Unless otherwise indicated, n refers to number of tumors.

382

Kingham et al

Hepatic Irreversible Electroporation

Figure 1. The liver was divided into 4 anatomic zones. Each zone has a similar ablation approach. The percentage of tumors located in each zone is noted. The caudate lobe is not visualized in this figure.

⬍0.5 cm from major venous structures (44%) or major portal pedicles (29%). Most tumors were treated by bracketing with 2 electrodes (Table 3). A third probe was required in 20% of cases. Median tip exposure was 2 cm (range 1 to 2.5 cm). Median number of pulses delivered per tumor was 90 (range 10 to 530 pulses). More than 90 pulses were used when a tumor was re-treated. Re-treatment was required when a procedure was aborted. Reasons for aborting the procedure included high current (18%), need to reposition the probes (12%), failed delivery for an unknown reason (6%), and arrhythmia (2%). Sixty percent required a repeat treatment of the tumor due to uncertainty about the adequacy of the size of the original ablation as predicted by the computer-generated ablation zone model. The patient with an arrhythmia had a supraventricular tachycardia, followed by an asystolic pause that occurred only when the Nanoknife was activated. There were no postoperative sequelae of this intraoperative arrhythmia. Postoperative imaging was used to assess the patency rate of hepatic veins and portal pedicles that were adjacent to tumors treated with IRE (Table 4). Of the 25 tumors near hepatic veins, all hepatic veins were patent on postoperative imaging studies. This group includes 11 tumors that were within 0.5 cm of a hepatic vein. All 8 portal pedicles adjacent to (within 1 cm) treated tumors were also patent on follow-up imaging, without evidence of biliary injury or stricture. The only post-procedure major vessel occlusion occurred in 1 of the 6 tumors that was ⬍0.5 cm from a major portal pedicle. This patient had a history of metastatic colorectal cancer and multiple earlier liver resections.

J Am Coll Surg

One of the recurrent tumors was adherent to the right portal pedicle. This tumor was treated with percutaneous IRE at 2 different settings. The post-procedure imaging showed a thrombus in the segment-6 portal vein branch, with no biliary dilation associated with this grade 1 complication. In the 12 patients who underwent IRE treatment only (ie, no other ablation procedures or hepatic resections), there were several changes in serum values of total bilirubin, aspartate aminotransferase (AST), and alkaline phosphatase (Fig. 2). Median preprocedural total bilirubin was 0.7 mg/dL (range 0.4 to 1.2 mg/dL). This rose to a median of 1 mg/dL (range 0.5 to 2.2 mg/dL). The bilirubin returned to baseline with a median of 0.7 mg/dL (range 0.4 to 1.6 mg/dL) at a median of 79 days post procedure. Similarly, AST serum levels rose from normal levels to a median peak of 923 U/L (range 55 to 1,533 U/L) on postprocedure day 1. This returned to normal at a median of post-procedure day 79 with a serum AST median value of 33 U/L (range 15 to 52 U/L). The appearance of the ablation zones after IRE treatTable 3. Irreversible Electroporation Settings and Complications Variable

No. of electrodes 2, n (%) 3, n (%) 4, n (%) 5, n (%) Tip exposure, (n ⫽ 48 tumors), cm, median (range) Maximum voltage (n ⫽ 52 tumors), median (range) Electric field strength (n ⫽ 50 tumors), V, median (range) No. of pulses delivered to tumor (n ⫽ 50 tumors), median (range) Intraoperative complications, n (%) High current Failed delivery other reason Arrhythmia Needed to reposition probes Repeat treatment of same tumor, n (%) Yes No Patients receiving ablation technique, n (%) Percutaneous IRE Laparotomy

Data

36/56 11/56 6/56 3/56

(64) (20) (11) (5)

2 (1⫺2.5) 2700 (1500⫺3000) 1781 (1303⫺3000)

90 (10⫺530) 9/50 3/50 1/50 6/50

(18) (6) (2) (12)

32/53 (60) 21/53 (40)

6/28 (21) 22/28 (79)

Unless indicated otherwise, n refers to number of tumors. IRE, irreversible electroporation.

Vol. 215, No. 3, September 2012

Kingham et al

Hepatic Irreversible Electroporation

383

Table 4. Postoperative Imaging Results Tumors, n

Variable

Hepatic vein status 1 mo post-IRE with ablation ⬍1 cm from hepatic vein, n (%) Vein patent Hepatic vein status 1 mo post-IRE with ablation ⬍0.5 cm from hepatic vein, n (%) Vein patent Unknown Bile duct status 1 mo post-IRE with ablation ⬍1 cm from portal in-flow structures, n (%) Bile duct patent Bile duct status 1 mo post-IRE with ablation ⬍0.5 cm from portal in-flow structures, n (%) Bile duct patent Bile duct occluded Ablation zone area at 1 mo, cm2, median (range) Ablation zone area at 3 mo cm2, median (range) Ablation zone area at 6 mo cm2, median (range) Area at 1 mo to area at 3 mo cm2, median (range) Disease persistence/recurrence Local recurrence, n (%) Time to recurrence, d Patient 1 Patient 2 Patient 3 Persistent disease, n (%)

Data

25 25 (100) 19 11 (58) 8 (42) 16 15 (94) 12

25 32 26 24 54

11 1 9 2.3 2.3 ⫺2.33

(92) (8) (1.3⫺32.2) (0⫺17.6) (0⫺17.2) (⫺17.7 to ⫹4.9cm2)

3 (5.6) 66 164 230 1 (2)

IRE, irreversible electroporation.

ment was variable (Fig. 3). On scans performed at 1-month post procedure, the median ablation zone area using the greatest cross-sectional area was 9 cm2 (range 1.3 to 32.2 cm2). This decreased to 2.3 cm2 (range 0 to 17.6 cm2) by 3 months. The area of the ablation zone was similar between 3 months and 6 months post procedure. Median decrease in ablation zone area at 3 months post procedure was 58% (range 11% to 100%). Five ablation zones that were visible on the 1-month post-procedure scan were not visible on the 3-month scan (Fig. 4). There were 3 local recurrences (5.6%) and 1 tumor with persistent disease (1.9%) after IRE for a combined local failure rate of 7.5% (Table 4). The local recurrences were noted at 66 days, 164 days, and 230 days. Local recurrence at 230 days was in the patient with the portal vein thrombosis treated with a percutaneous IRE described previously. One local recurrence was noted in a patient intolerant of IRE due to cardiac rhythm abnormalities, and another in a patient with a 2.9-cm tumor adjacent to the inferior vena cava. The patient with persistent disease had a segment-8 colorectal cancer liver metastasis, which was 5 ⫻ 5 ⫻ 5 cm, abutting the inferior vena cava and right portal vein and was treated percutaneously.

DISCUSSION As ablation is used more frequently to treat liver tumors, the limitations of RFA and MWA have become more apparent. Local recurrences, associated with large tumors and tumors close to vessels, are reported for both HCC and colorectal cancer liver metastases.9 In addition, there are tumors that cannot be treated safely by MWA because of their proximity to important biliary or vascular structures. Similarly, tumors in these positions are often not treated effectively by RFA due to a heatsink effect. It is the patient population with tumors near hepatic veins and/or portal pedicles that was examined in this study. The ability to ablate tumors on vascular pedicles often changes the therapeutic options for combination resection and ablation procedures. There are 2 important clinical questions associated with the use of IRE to treat liver tumors. First, are vascular and biliary structures adjacent to treated tumors at risk for injury? Second, if IRE technology can avoid damaging major vessels and bile ducts, can it also effectively treat liver tumors? Definitive answers to both questions are as yet unknown outside of animal models.

Hepatic Irreversible Electroporation

0.5

ce du re

1

Median time pre/post IRE

AST (Units/L)

2000 1500 1000 500

B

oc ed ur e pr po st

70

1

da y

s

da y

Pr e

po st

pr o

ce

pr oc ed u

du re

re

0

Median time pre/post IRE 800 600 400 200

tp po s ys da 70

1

da y

po s

Pr e

tp

ro

ro

pr oc

ce

ed u

ce du re

du re

0 re

Alkaline Phosphatase (Units/L)

0

Figure 3. Irreversible electroporation (IRE) ablation zone size.

70

1

A

5

Time post ablation

da ys

da y

Pr

po

e

st

po st p

ro

pr o

pr oc ed ur e

ce du re

0.0

10

M on th

1.0

15

6

1.5

20

M on th

2.0

25

3

Total Bilirubin (mg/dL)

2.5

C

J Am Coll Surg

Area of ablation zone (cm2)

Kingham et al

M on th

384

Median time pre/post IRE

Figure 2. (A) Serum total bilirubin, (B) aspartate aminotransferase (AST), and (C) alkaline phosphatase pre and post irreversible electroporation (IRE).

Potential complications of IRE include blood vessel thrombosis, bile duct injury, intestinal perforation, hemorrhage, hematoma, infection, and arrhythmia. Complications of IRE need to be placed in context with those of RFA and MWA. Although the morbidity of RFA and MWA is less than that for liver resections, it does exist. One multicenter Italian study reported a 0.3% mortality rate and a 2.2% major complication rate with percutaneous RFA of liver tumors.18 These authors showed a similar major complication rate (2.9%) with the use of MWA.19 A Korean study reported a major complication rate of 2.3%, with reported complications including hepatic abscess, biloma, hemorrhage, and pneumothorax.20 Given the high recurrence rates that have been reported with perivascular tumors and the complications that can be caused by thermal damage, there is a need for alternative ablative techniques such as IRE. The safety of IRE was first examined in porcine models. Lee and colleagues performed percutaneous IRE in 16 swine, with survival times as long as 14 days.11 Ultrasound, CT, and MRI were used to image the animals post treatment. The largest diameter ablation zone was 60 mm, and the bile ducts appeared intact. In a follow-up study, hematoxylin and eosin stains showed extensive cell death, with no microscopically viable cells in the ablation zone by 24 hours and minimal endothelial damage to blood vessels. Irreversible electroporation appeared to cause ablation up to the wall of large vessels and traversed them without damaging the endothelium.11 Charpentier and colleagues10 reported on 8 swine that underwent 20 ablations of the liver and liver hilum. Irreversible electroporation appeared to be safe when used near bile ducts, portal veins, and hepatic arteries. The safety of IRE has been reported in 1 human study. Thomson and colleagues15 treated 38 patients with advanced malignancies of the liver, kidney, or lung. Two patients had transient ventricular arrhythmias before a change in protocol to include ECG-synchronized

Vol. 215, No. 3, September 2012

Figure 4. (A) Preoperative MRI demonstrating colorectal cancer liver metastasis between the middle and left hepatic veins. Arrow points to the location of the middle hepatic vein. Circle encompasses the tumor. (B) CT scan on post-procedure day 4. Arrow points to the patent middle hepatic vein. (C) MRI at 4 months post procedure. Arrow points to the middle hepatic vein that remains patent. The ablation zone is difficult to visualize.

Kingham et al

Hepatic Irreversible Electroporation

385

delivery of the IRE pulse. After this change, 1 patient had a supraventricular tachycardia and 1 had atrial fibrillation. There were increases in serum values for total bilirubin, AST, and alkaline phosphatase on post-procedure day 1. These values all normalized by approximately 3 months post procedure. The current study shows that the use of IRE in human liver tumors, some of which were adjacent to portal pedicles or hepatic veins, was safe. None of the hepatic veins within 1 cm of treated tumors was occluded on postprocedure imaging. There were similar findings with ablations performed within 1 cm of portal in-flow structures, except for 1 patient with a segmental pedicular injury. The patient has no clinical sequelae related to these imaging findings, but it is unclear if similar pedicle injuries can happen in more central structures that could affect hepatic biliary drainage. In addition, it is unclear if this thrombosis was caused by electrical effects of IRE or the thermal effect that be created in the tissue immediately adjacent to the IRE electrode. One patient had a supraventricular arrhythmia despite ECG synchronization. This arrhythmia was triggered only when the device was actively administering IRE pulses, and the patient returned to normal sinus rhythm when the pulses ceased. This arrhythmia altered the treatment plan and therefore might have contributed to the patient’s local recurrence. It is important to note that the majority of IRE treatments were performed during a laparotomy, as the safety profile of different ablative approaches can differ. Overall, complication rates are reported to be similar between percutaneous, laparoscopic, and open ablations, and this might be similar for IRE treatments when a larger number of percutaneous IRE treatments can be analyzed.21 The effects of IRE on tumors have been shown in in vitro and in vivo models. Miller and colleagues12 determined that in vitro cancer cell lines were ablated with IRE due to the electrical effect, not thermal damage, induced by IRE. Multiple parameters, including pulse amplitude, length, and number were all factors in determining the ultimate effect of IRE. Al-Sakere and colleagues22 reported on the first application of IRE for cutaneous murine tumors. Using a sarcoma cell line, 6 mice were inoculated and treated with IRE. The mice were sacrificed between 1 and 72 hours after IRE treatment. Tumors were examined with hematoxylin and eosin and immunohistochemistry for CD31 (antiplatelet endothelial cell adhesion molecule). A terminal deoxynucleotidyl transferase⫺mediated dUTP nick end-labeling kit was used to determine cell apoptosis. Changes in the tumor architecture and morphology were minimal until 6 hours after the application of IRE. By 24 hours, the tumor had a homogeneous appearance. At 48

386

Kingham et al

Hepatic Irreversible Electroporation

hours, tissue necrosis was even more evident. Terminal deoxynucleotidyl transferase⫺mediated dUTP nick endlabeling staining increased beginning 1 hour after IRE. Changes in the tumor vascularity were evident with CD31 staining at 2 hours, when vascular congestion was initiated. In Thomson and colleagues’15 publication on the IRE treatment in humans, the authors reported tumor response as a loss of hypervascularity on 1-month and 3-month post-procedure imaging. Overall response rate for liver metastases was 50%. The current study, however, found that the initial tumor response rate was 98.1% (53 of 54 tumors responded). In the 53 tumors that responded initially, there were 6 local recurrences (5.7%) at a median follow-up of 6 months. These low rates of persistent disease and recurrent disease at 6 months post procedure serve as the first suggestion that IRE might be effective for the treatment of perivascular tumors, but longer follow-up will be required to prove its durability. The difference in response rates between Thomson and colleagues’15 study and the current study can be partially attributed to the difference in approaches, as the percutaneous approach used in Thomson and colleagues’ study is associated with a higher local recurrence rate, or to larger tumor sizes (median tumor size 2.3 cm; range 1 to 6.5 cm).9 Two of the 6 patients in the current study treated with percutaneous IRE had recurrent or persistent disease. Although this failure rate is higher than with open IRE, the patients are not comparable, as the patients referred for percutaneous IRE had recurrences after surgery and were not surgical candidates. The different mechanism of parenchymal and tumor cell destruction with IRE is evident in the radiographic evaluation of IRE ablation zones. The initial ablation zone 1 month post procedure had an area of 9 cm2 (range 1.3 to 32.2 cm2). This was reduced by more than two thirds by 3 months post procedure, when the area was, on average, 2.3 cm2 (range 0 to 17.6 cm2). Five of the ablation zones visible on the 1-month scan were not visible on the 3-month scans. Both MRI and CT scans have been used to evaluate IRE ablation zones in animal models.23,24 In the current study, there was no difference in the ability to assess ablation zones with CT or MRI scans (data not shown). Rubinsky and colleagues,25 in a study with 14 pigs, sacrificed at 24 hours, 3 days, 7 days, and 14 days showed that at day 7, there was regeneration in most ablation zones, with large veins demonstrating necrotic endothelium but lumenal patency. By day 14, it was difficult to identify the ablation zones. A similar time course was reported in a study that examined 44 rats after electroporation was used on HCC tumors. The majority of the ablation sites finished evolving by day 7.26

J Am Coll Surg

There are multiple limitations to this study. This is a heterogeneous population with multiple tumor types and short follow-up. Despite the limitations, however, safety and efficacy of IRE are possible to assess with these data.

CONCLUSIONS Although the indications for using IRE to treat liver tumors are still not clearly defined, given the low morbidity rate the current study demonstrates, IRE appears to be safe in treating perivascular hepatic tumors. In addition, the 7.5% treatment failure rate with short follow-up is low enough to justify studying IRE in a greater number of patients. The ability to treat these perivascular tumors that were extremely challenging to ablate previously might expand the therapeutic options to treat patients with hepatic tumors. Author Contributions

Study conception and design: Kingham, Karkar, Fong Acquisition of data: Kingham, Karkar, Solomon, Fong Analysis and interpretation of data: Kingham, Karkar, D’Angelica, Allen, DeMatteo, Getrajdman, Sofocleous, Solomon, Jarnagin, Fong Drafting of manuscript: Kingham, Karkar Critical revision: Kingham, Karkar, D’Angelica, Allen, DeMatteo, Getrajdman, Sofocleous, Solomon, Jarnagin, Fong

REFERENCES 1. Bozzetti F, Doci R, Bignami P, et al. Patterns of failure following surgical resection of colorectal cancer liver metastases. Rationale for a multimodal approach. Ann Surg 1987;205: 264–270. 2. Jemal A, Siegel R, Xu J, Ward E. Cancer statistics, 2010. CA Cancer J Clin 2010;60:277–300. 3. Sofocleous CT, Petre EN, Gonen M, et al. CT-guided radiofrequency ablation as a salvage treatment of colorectal cancer hepatic metastases developing after hepatectomy. J Vasc Interv Radiol 2011;22:755–761. 4. Lu DS, Yu NC, Raman SS, et al. Radiofrequency ablation of hepatocellular carcinoma: treatment success as defined by histologic examination of the explanted liver. Radiology 2005;234: 954–960. 5. Raut CP, Izzo F, Marra P, et al. Significant long-term survival after radiofrequency ablation of unresectable hepatocellular carcinoma in patients with cirrhosis. Ann Surg Oncol 2005;12: 616–628. 6. Solbiati L, Ierace T, Tonolini M, et al. Radiofrequency thermal ablation of hepatic metastases. Eur J Ultrasound 2001;13:149– 158. 7. Abitabile P, Hartl U, Lange J, Maurer CA. Radiofrequency ablation permits an effective treatment for colorectal liver metastasis. Eur J Surg Oncol 2007;33:67–71. 8. Kingham TP, Tanoue M, Eaton A, et al. Patterns of recurrence

Vol. 215, No. 3, September 2012

9. 10. 11. 12. 13. 14. 15. 16.

17. 18.

after ablation of colorectal cancer liver metastases. Ann Surg Oncol 2012;19:834–841. Mulier S, Ni Y, Jamart J, et al. Local recurrence after hepatic radiofrequency coagulation: multivariate meta-analysis and review of contributing factors. Ann Surg 2005;242:158–171. Charpentier KP, Wolf F, Noble L, et al. Irreversible electroporation of the liver and liver hilum in swine. HPB (Oxford) 2011; 13:168–173. Lee EW, Chen C, Prieto VE, et al. Advanced hepatic ablation technique for creating complete cell death: irreversible electroporation. Radiology 2010;255:426–433. Miller L, Leor J, Rubinsky B. Cancer cells ablation with irreversible electroporation. Technol Cancer Res Treat 2005;4:699– 705. Davalos RV, Mir IL, Rubinsky B. Tissue ablation with irreversible electroporation. Ann Biomed Eng 2005;33:223–231. Daniels C, Rubinsky B. Electrical field and temperature model of nonthermal irreversible electroporation in heterogeneous tissues. J Biomech Eng 2009;131:071006. Thomson KR, Cheung W, Ellis SJ, et al. Investigation of the safety of irreversible electroporation in humans. J Vasc Interv Radiol 2011;22:611–621. Deodhar A, Dickfeld T, Single GW, et al. Irreversible electroporation near the heart: ventricular arrhythmias can be prevented with ECG synchronization. AJR Am J Roentgenol 2011;196: W330–W335. Martin RC 2nd, Jaques DP, Brennan MF, Karpeh M. Achieving RO resection for locally advanced gastric cancer: is it worth the risk of multiorgan resection? J Am Coll Surg 2002;194:568–577. Livraghi T, Solbiati L, Meloni MF, et al. Treatment of focal liver tumors with percutaneous radio-frequency ablation: complica-

Kingham et al

19.

20.

21. 22. 23.

24.

25. 26.

Hepatic Irreversible Electroporation

387

tions encountered in a multicenter study. Radiology 2003; 226:441–451. Livraghi T, Meloni F, Solbiati L, et al. Complications of microwave ablation for liver tumors: results of a multicenter study. Cardiovasc Interv Radiol 2011 Aug 11. [Epub ahead of print]. Rhim H, Yoon KH, Lee JM, et al. Major complications after radio-frequency thermal ablation of hepatic tumors: spectrum of imaging findings. Radiographics 2003;23:123–134; discussion 134⫺136. Mulier S, Mulier P, Ni Y, et al. Complications of radiofrequency coagulation of liver tumours. Br J Surg 2002;89: 1206–1222. Al-Sakere B, André F, Bernat C, et al. Tumor ablation with irreversible electroporation. PLoS One 2007;2(11):e1135. Zhang Y, Guo Y, Ragin AB, et al. MR imaging to assess immediate response to irreversible electroporation for targeted ablation of liver tissues: preclinical feasibility studies in a rodent model. Radiology 2010;256:424–432. Guo Y, Zhang Y, Nijm GM, et al. Irreversible electroporation in the liver: contrast-enhanced inversion-recovery MR imaging approaches to differentiate reversibly electroporated penumbra from irreversibly electroporated ablation zones. Radiology 2011;258: 461–468. Rubinsky B, Onik G, Mikus P. Irreversible electroporation: a new ablation modality—clinical implications. Technol Cancer Res Treat 2007;6:37–48. Guo Y, Zhang Y, Klein R, et al. Irreversible electroporation therapy in the liver: longitudinal efficacy studies in a rat model of hepatocellular carcinoma. Cancer Res 2010;70:1555–1563.