| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Experimental Studies |
1 From the Department of Medicine I (J.H., M.W., T.B., T.S., M.F., S.S., E.G.H., D.S.), Institute of Pathology (D.N.), and Institute of Diagnostic Radiology (R.J.), Friedrich-Alexander-University of Erlangen-Nuremberg, Ulmenweg 18, 91054 Erlangen, Germany; Institute of Biomedical Engineering, University Furthwangen, Germany (W.M.); and Department of Medicine, Hospital of Eckernfoerde, Germany (D.B.). Received September 3, 2002; revision requested November 18; final revision received April 15, 2003; accepted April 30. Supported by grants from the Bavarian Economics Ministry (High-Tech Projekt: Leitprojekte Medizintechnik) and the Hans-Löwel-Foundation, Bamberg. Address correspondence to J.H. (e-mail: johannes.haensler@med1.med.uni-erlangen.de).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: VX2 liver tumors were implanted in 31 rabbits according to a standardized protocol. After 21 days, 16 animals were treated percutaneously with RF ablation. Four animals died of complications related to anesthesia, and 12 animals were evaluated. All animals were followed for 110 days and monitored with computed tomography (CT) and US at 1, 7, and 30 days. A control group of 15 animals did not receive treatment. Autopsy was performed at the end of scheduled follow-up or immediately after death. For survival analysis, the Kaplan-Meier method was used; for nominal data, the Fisher exact test was used.
RESULTS: In comparison to controls, animals in the treatment group showed significantly prolonged survival (P < .001). Eight of 12 animals (67%) treated with RF ablation survived to 110 days, while none of the controls did so. Metastases developed in all controls (100%) and in eight of 12 treated animals (67%) (P = .001). In comparison with controls, animals that developed metastases despite treatment also showed significantly prolonged survival (P = .02). Local recurrence was observed in three of 12 animals (25%) in the treatment group. CT and US performed 1 week after treatment did not allow prediction of local recurrence.
CONCLUSION: RF ablation of liver tumors with perfused needle applicators prolongs survival in the VX2 rabbit liver tumor model, regardless of whether complete remission is achieved. In comparison with controls, RF ablation results in a lower frequency of metastases.
© RSNA, 2003
Index terms: Animals • Experimental study • Liver neoplasms, therapeutic radiology, 761.1269 • Radiofrequency (RF) ablation
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
To our knowledge, there is no conclusive evidence from prospective randomized trials that RF ablation has a positive effect on long-term survival, irrespective of the technique used. Nevertheless, a report dealing with long-term survival showed promising results (11).
Furthermore, uncertainty remains as to whether postinterventional imaging is capable of depicting early recurrent disease, thus allowing prediction of treatment outcome.Therefore, the purpose of our study was to evaluate survival, metastases, tumor necrosis, and prediction of local recurrence after percutaneous US-guided RF ablation by using electrodes perfused continuously with isotonic saline.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
The tumor tissue required for implantation was obtained from tumor-bearing rabbits (New Zealand Whites; Charles River, Kisslegg, Germany). After laparotomy was performed, a fragment of tumor tissue (1 x 1 x 1 mm) was implanted beneath the liver capsule of the recipient, 2 cm distal to the fundus of the gallbladder. All surgical procedures were performed by one author (J.H., assisted by M.W., T.B., M.F., S.S.). The abdominal wall was closed in double layers. For ethical reasons, moribund animals that failed to eat and drink for more than 4 days, had weight loss of more than 20%, and displayed apathetic behavior were sacrificed by means of intravenous injection (J.H.) of methohexital sodium (Brevimytal; Lilly, Giessen, Germany).
Experimental Design
In a preliminary experiment involving 24 animals, all procedures were tested and standardized, and optimal equipment settings and application times were established. A definitive study comprising 31 animals was then performed. The animals were assigned randomly either to the RF ablation group (n = 16) or to the control group (n = 15). Animal weight was similar in the two groups, with a mean weight of 3,348 g (range, 2,400–4,050 g) in the treatment group and 3,458 g (range, 2,300–5,100 g) in the control group.
Twenty-one days after implantation, when a tumor diameter of at least 5 mm was observed (determined by two blinded observers [T.B., D.S.] in consensus by using US) with typical imaging patterns (early arterial vascularization, absence of contrast medium in the portal venous phase), the animals (n = 16) in the RF ablation group were treated with US-guided percutaneous RF ablation while the controls (n = 15) received no RF ablation. The follow-up regimen was designed analogously to the regimen used for follow-up in human patients. The animals were examined (a) with US immediately after treatment and 1 week later and (b) with computed tomography (CT) 1 week and 1 month after treatment. All animals that survived to the defined end point of 110 days after implantation were sacrificed. Autopsy was performed, and tissue specimens were investigated histologically. Animals that did not survive to the defined end point were investigated in the same way at the time of death.
RF Ablation
RF ablation was performed by one author (J.H., assisted by M.W., T.B., M.F., S.S.) by using an RF generator (Elektrotom HF 106; Berchtold, Tuttlingen, Germany) and a miniature perfused RF needle applicator with an outside diameter of 1.1 mm and a 10-mm-long active electrode. This miniature probe was used instead of the regular 2-mm-diameter probe with 20-mm active length that is used for treatment in human patients, bearing in mind the possibility of extrapolating the outcomes to a human population.
A self-adhesive grounding pad was applied to the animal’s shaven back. A nick was made before the needle was advanced through the skin. With US guidance, the needle applicator was advanced into the center of the tumor. Isotonic saline was delivered by an infusion pump (Pilot C; Fresenius Medical Care, Alzenau, Germany) through openings in the tip of the needle into the coagulation zone at a rate of 40 mL/h. All treatments were performed as single-needle applications at the same application site. Treatment was performed with an energy output of 20 W and an application time of 4 minutes.
US Examination
All US examinations and treatments were performed by using a unit with a 7.5-MHz linear array (Elegra Advanced; Siemens, Erlangen, Germany). Every examination was documented on videotape, which was subsequently evaluated by two blinded readers (T.B., D.S.).
For contrast material–enhanced US, a 1-mL bolus of contrast material (Optison; Mallinckrodt, St Louis, Mo) was injected into an ear vein, followed by a 5-mL saline flush. Phase-inversion US was performed immediately after injection of the contrast agent at a low mechanical index (<0.4). Enhancement within the tumor during the arterial phase was considered to indicate vital tumor tissue. The absence of enhancement during the arterial phase on posttreatment scans, which is associated with absence of portal venous perfusion in the tumor and a peripheral safety margin measuring at least 5 mm, indicated complete remission.
CT Examination
CT was performed with a multisection spiral CT scanner (Somatom Volume Zoom; Siemens). All images were evaluated by two blinded readers (R.J., M.F.)
For lesion detection, a native scan of the entire liver was obtained. This was followed by a contrast-enhanced scan of the lesion 9, 14.5, and 23 seconds after intravenous administration of contrast medium. To exclude the possibility of multiple lesions, the entire liver was scanned 39 seconds after administration of contrast medium. The scanning parameters were collimation of 4.00 x 1.25 mm, rotation time of 0.5 second, 200 mAs, 120 kV, and reconstruction section thickness of 1 mm.
Contrast medium (Imeron 300; Byk Gulden, Konstanz, Germany) was injected into an ear vein by using a power injector (Medrad, Pittsburgh, Pa) (7 mL at 1 mL/sec). Peripheral hyperperfusion of the lesion in the arterial phase (14.5 seconds) was considered to indicate vital tumor tissue. The absence of early arterial hyperperfusion at the margins of the tumor with no lesions in the remaining liver was interpreted as complete remission.
Histologic Findings
Complete histologic analysis was performed by a pathologist (D.N.). In accordance with the protocol, the animals were sacrificed by means of injection of 50 mg of methohexital sodium, and autopsy was performed within 5 minutes (except for the brain) as proposed by Rokitansky. Animals that died before the end of the study underwent autopsy within 8 hours. The liver lobe bearing the tumor was investigated completely to detect residual tumor, estimate vitality of residual tumor, and evaluate histologic reactions of RF ablation and, ultimately, the effectiveness of RF ablation treatment.
Specimens were obtained from the heart, lungs, liver, and metastases (in the lymph nodes, pleura, and peritoneum) and stored in 10% buffered formalin. After the 5-µm-thick sections were embedded in paraffin, they were prepared and stored at room temperature. Routine histologic examination with hematoxylin-eosin stain was performed to evaluate basic histomorphology of the specimens. Intra- and extracellular iron deposits were evaluated by using Prussian blue stain, which is used to detect iron-loaded hemosiderin, the major long-term storage protein of iron. In addition, for detection of residual tumor and determination of the proliferation rate of residual tumor, respectively, levels of cytokeratin 5/6 (monoclonal mouse antibodies with microwave pretreatment and dilution of 1:50; Zymed Laboratories, San Francisco, Calif) and proliferating cell nuclear antigen (PCNA) (monoclonal mouse antibodies with microwave pretreatment and dilution of 1:500; Dako, Glostrup, Denmark) were determined immunohistochemically as described by Hall et al (13). PCNA proliferation analysis of residual tumor was determined with an image analysis system (analySIS; Soft Imaging System, Münster, Germany) by investigating 10 high-power fields (original magnification, x400).
To differentiate local recurrence from residual tumor, we defined local recurrence as nodular tumor masses not altered morphologically by RF ablation that were located at the periphery of a circumscribed fibrosis with morphology typical of that with RF ablation. In contrast, residual tumor cells had to be embedded within the coagulation zone.
Statistical Evaluation
Statistical analysis was performed (D.N., M.W.) with SPSS D 11.0 software (SPSS, Chicago, Ill). The results in the RF ablation group were compared statistically with those in the control group: for nominal data (number of metastases), the Fisher exact test was used. Survival analysis was performed with the Kaplan-Meier method, with comparison of survival curves by using the log-rank test. P values less than .05 were considered to indicate a statistically significant difference. Sensitivity, specificity, and positive and negative predictive values were calculated for CT and US findings.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Survival
In comparison to the controls, animals in the RF ablation group survived significantly longer (Kaplan-Meier method with log-rank test, P < .001) (Fig 1). This effect was observed even in those animals in the RF group that developed metastases (Kaplan-Meier method with log-rank test, P < .02). Eight of the 12 (67%) animals in the RF ablation group survived to the end point of 110 days, whereas none of the animals in the control group reached the end point (Tables 1 and 2).
|
|
|
US and CT Findings
With regard to the evaluation of local results of treatment, contrast-enhanced US findings obtained immediately after ablation had a sensitivity of 0, specificity of 1.00, positive predictive value of 0, and negative predictive value of 0.75. US and CT findings obtained 1 week after RF ablation each had a sensitivity of 0, specificity of 0.89, positive predictive value of 0, and negative predictive value of 0.73. Neither modality enabled early prediction of local recurrence. In addition, CT was performed 1 month after RF ablation to enable reliable detection of local recurrence (sensitivity of 0.33, specificity of 1.00, positive predictive value of 1.00, and negative predictive value of 0.82) (Table 2).
Histologic Analysis
Animals in complete remission showed chronic inflammation with circumscribed scarring and small aggregations of macrophages in the periphery of the coagulation (Fig 2). In addition, focal areas of calcification with solitary giant cells of the foreign body type were seen (Fig 2). By using Prussian blue stain for visualization of hemosiderin-bound iron, deposits were detected in mainly mononuclear cells and in a few giant cells (Fig 2). Mild bile duct proliferations were observed (Fig 2). No remnants of the implanted tumor were found, even when cytokeratin 5/6 or PCNA immunohistochemistry was applied.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Four of our 12 (33%) animals in the RF ablation group achieved complete remission at the 110-day end point. In comparison with the control group, even treated animals that developed metastases despite RF ablation (n = 8) also demonstrated a statistically significant survival benefit. Whether this effect can be explained by the perifocal and often chronic inflammatory reaction seen at histologic examination and the partial necrosis of metastatic lesions (signifying an antitumoral immune response) remains unclear and needs further investigation. The reduction in tumor mass brought about by RF ablation might also be considered responsible for the survival benefit, as the concept of debulking in tumors remains controversial.
The 67% metastasis rate in the RF ablation group, however, should certainly prompt critical discussion. One cause could be metastasis of the tumor prior to ablation, as the significantly lower metastasis rate in comparison to that of the control group would appear to indicate. However, treatment-induced seeding of tumor cells cannot be excluded.
In the animals in the RF ablation group that died before the end point (range, 65–73 days after treatment), thermally altered "vital" residual tumor was found in the center of the coagulation zones. These so-called ghost cells were reported to be nonvital by other authors (1,14,16). The immunohistochemical determination of PCNA done in our study clearly showed a low mean proliferation rate of about 5% of all vital tumor cells, which indicated that most of the vital tumor cells were noncycling, possibly damaged irreversibly, and subject to apoptosis (1). Other investigators showed absence of cytosolic and mitochondrial enzyme activity after RF ablation, which was considered to be evidence of irreversible cellular damage (1).
Our apparently contradictory findings of PCNA-positive tumor cells within the coagulation zone could be explained by the fact that the PCNA antigen could be detected not only in proliferating cells but also sometimes in resting cells and in cells with DNA repair processes without evidence of proliferation (17). Nevertheless, these cells, which show morphologic signs of cellular damage, should be regarded as severely damaged and not yet apoptotic. After 3 months, however, vital tumor cells can no longer be seen. Therefore, biopsy specimens should be acquired at 3 months or later to assess tumor viability.
Among the treated animals, 25% developed local tumor recurrence or were treated incompletely. These treatment failures must be assumed to be caused by incomplete coagulation of tumor tissue. Neither contrast-enhanced US nor CT performed after 1 and 7 days allowed prediction of any of the three local recurrences. Even on CT scans obtained 1 month after RF ablation, only one of the local recurrences was predicted correctly, despite the fact that the VX2 tumor investigated is known to have a high proliferation rate. There were, however, no false-positive findings. This underscores the importance of performing follow-up at short intervals.
Practical application: The findings of this study demonstrate that percutaneous US-guided RF ablation with wet electrodes significantly improves survival in the VX2 tumor model and produces complete destruction of the tumor in most treated animals. Whether this positive effect of RF treatment on survival in the VX2 rabbit tumor model can be reproduced in clinical practice remains unclear. Success in reproducing the survival benefit in clinical practice, especially as shown for animals with metastatic disease, would mean that patients with multiple metastases could benefit from RF ablation treatment. In addition, further studies on the investigation of the presumed immunostimulatory effect of RF ablation and histologic assessment of tumor vitality are required.
|
FOOTNOTES |
|---|
Author contributions: Guarantors of integrity of entire study, D.S., E.G.H.; study concepts, W.M., D.B., J.H., T.S.; study design, D.B., J.H., T.S.; literature research, J.H., M.W.; experimental studies, T.S., J.H., T.B., M.F., S.S., M.W.; data acquisition, M.W., R.J.; data analysis/interpretation, M.W.; statistical analysis, M.W., D.N.; manuscript preparation, J.H., D.N., D.S.; manuscript definition of intellectual content, J.H., D.S.; manuscript editing, D.S., J.H., E.G.H.; manuscript revision/review, D.S., T.S., D.B., R.J., M.W., W.M.; manuscript final version approval, E.G.H., R.S., M.W., W.M.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  D. Sudheendra, M. M. Barth, U. Hegde, W. H. Wilson, and B. J. Wood Radiofrequency ablation of lymphoma Blood, February 15, 2006; 107(4): 1624 - 1626. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
