|
|
||||||||
J Thorac Cardiovasc Surg 2009;137:160-166
© 2009 The American Association for Thoracic Surgery
Evolving Technology |
a Division of Thoracic Surgery, Massachusetts General Hospital, Harvard Medical School, Boston, Mass
b Division of Thoracic Radiology, Massachusetts General Hospital, Harvard Medical School, Boston, Mass
Received for publication May 2, 2008; revisions received August 17, 2008; accepted for publication August 21, 2008. * Address for reprints: Michael Lanuti, MD, Division of Thoracic Surgery, Massachusetts General Hospital, 55 Fruit St, Blake 1570, Boston, MA 02114. (Email: Mlanuti{at}partners.org).
| Abstract |
|---|
|
|
|---|
Methods: Thirty-one consecutive patients with biopsy-proven non–small cell lung cancer underwent 38 treatments of computed tomographically guided radiofrequency ablation in a 4.5-year period. All patients were carefully selected and deemed medically ineligible for resection by a multidisciplinary team. Radiofrequency ablation was performed with curative intent with a single or cluster cool-tip electrode. Patients were hospitalized for 23-hour observation.
Results: Treatment was complete in all cases, with no 30-day mortality. Local recurrence was confirmed radiographically by computed tomography, positron emission tomography, or both after 31.5% of treatments (12/38). Two patients were successfully retreated for technical failures related to pneumothorax; 3 underwent radiotherapy with stable disease. Mean maximal diameter of 38 tumors treated was 2.0 ± 1.0 cm (range 0.8–4.4 cm). After median follow-up of 17 ± 11 months, 74% of patients (23/31) were alive. Three patients died of metastatic disease; 5 died of pneumonia remote from treatment. The 2- and 4-year survivals were 78% and 47%, respectively. Median overall survival was 30 months. Pneumothorax (13%), pneumonia (16%), and pleural effusion (21%), were the most common complications.
Conclusions: Radiofrequency ablation of medically inoperable early–stage lung cancer in carefully selected patients yields encouraging midterm results without significant loss of pulmonary function. Local tumor progression appears related to lung tumors larger than 3 cm. Computed tomography and positron emission tomography need further validation for the early identification of local tumor progression following radiofrequency ablation.
| Introduction |
|---|
|
|
|---|
Because there have been no randomized, controlled trials assessing long-term outcome of newer radiation techniques versus RFA for lung cancer, these modalities must be carefully scrutinized. There is no standardization of lung RFA, and many of the reported series used different RFA systems. The clinical efficacy of tumor destruction with different RFA systems has not been formally compared. Regional disease progression is often poorly characterized in contemporary series, because hilar and mediastinal lymph nodes are evaluated by imaging studies and not pathologic verification.
In this study, we evaluated the outcomes of stage I NSCLC treated with a single RFA system and highlighted local control, patterns of failure, complications, and changes in pulmonary function. Positron emission tomography (PET) was also implemented in the detection of local tumor progression.
| Materials and Methods |
|---|
|
|
|---|
RFA Treatment
All lung RFA treatments were performed percutaneously with a single or cluster cool-tip electrode coupled to a generator and perfusion pump (Covidien, Mansfield, Mass). Procedures were performed with CT guidance (64-slice CT; Siemens Corp, New York, NY) with 2.5-mm collimation. Although some patients received general anesthesia because of severe cardiac or pulmonary compromise, conscious sedation with local anesthesia was the preferred method. The cluster electrode was implemented for lesions larger than 1 cm, whereas the single electrode was used for lesions 1 cm or smaller. Once in place, a 12-minute RFA treatment was initiated, during which the patient was closely monitored for adequate pain control, respiration, and oxygenation. After this time, temperature within the lesion was measured to achieve more than 60°C (usually about 90°C) for the ablation to be considered adequate. Before the electrode was removed, a CT scan was performed to assess for an adequate margin of treatment, which was seen as a ground-glass opacity surrounding the tumor, and to detect any immediate complications, such as hemothorax, pneumothorax, or pulmonary hemorrhage. If necessary, the electrode was repositioned and an additional treatment performed.
Patients were admitted overnight for 23-hour observation and prescribed oral analgesic narcotics to treat any pleuritic pain after the procedure. In addition, prophylactic antibiotics were administered to patients with prosthetic cardiac valves, pacemakers, mitral valve prolapse, or joint prostheses. Patients were instructed to watch for the onset of fever or sputum production, which could signal the development of pneumonia.
Pulmonary function was routinely measured a few months before scheduled RFA and then 3 to 6 months after treatment. Imaging follow-up was performed to assess for treatment efficacy and local tumor progression. PET with diagnostic quality CT was performed after 1, 6, 12, and 24 months to help define its role in the early identification of local tumor progression. Additional CT scans were obtained at 3, 9, and 18 months. If there was evidence of disease progression, repeated RFA was considered.
Local control was defined as lack of focal or diffuse enlargement of the ablated lesion on CT and no evidence of eccentric enhancement on PET at a minimum of 3 months of follow-up. Progression of fluorodeoxyglucose-avid regional lymphadenopathy (hilar or mediastinal) after 3 months was also considered a local failure.
Statistical Methods
Survival curves for overall, disease-free, and local tumor progression–free survivals were constructed with the Kaplan–Meier method and compared with the log rank test. Continuous variables were analyzed with the Student t test and confirmed with the Mann—Whitney U test. The statistical software package SPSS 14.0 (SPSS Inc, Chicago, Ill) was used for all survival analyses.
| Results |
|---|
|
|
|---|
|
Pneumonia, pneumothorax, minor hemoptysis, and pleural effusion were the most common complications, as listed in Table 2 . All pneumonias occurred within 4 weeks of ablation and resolved with a course of oral antibiotics. Peripheral cavities with air-fluid levels (representing limited bronchopleural fistula) were observed in 3 patients; these ultimately resolved with observation. One patient had transient right recurrent laryngeal nerve palsy as a result of thermal injury from treating a right upper lobe tumor in proximity to the anterior mediastinum. There was no major episodes of hemoptysis or pulmonary hemorrhage in this series.
|
|
|
|
|
|
|
|
| Discussion |
|---|
|
|
|---|
In a recent review of 17 studies examining case series of RFA for lung tumors, Zhu and associates15
highlighted many of the challenges that make comparisons and data analysis difficult. Aside from the implementation of different RFA systems, there is no uniformity in RFA treatment responses, patient populations and tumor characteristics are heterogeneous, the definitions of local tumor progression and radiologic assessment are variable, and patient follow-up remains short. Fernando and colleagues5
attempted to standardize treatment response by using modified response evaluation criteria in solid tumors (RECIST), but these criteria have not been widely accepted in RFA literature. In our experience, detection of local tumor progression was enhanced by comparing PET characteristics of the ablated lesion at 1, 6, and 12 months (the time at which 75% of the lesions in this study had local failure). Characteristics of local failure were defined by eccentric activity on PET or lack of expected contraction of the ablated zone on chest CT over time.
The American College of Surgeons is sponsoring a multi-institutional phase I study (ASOSOG Z4033) to address uniformity of RFA treatment in high-risk patients with stage IA NSCLC. Primary objectives are to establish 2-year survival, to assess freedom from local and regional recurrence, to assess the role of PET in predicting local control, and to explore short- and long-term effects of RFA on pulmonary function. Although we were unable to evaluate pulmonary function in our entire study cohort, this report is the first to demonstrate no statistically discernible difference in lung spirometry or diffusion capacity (N = 23) at 3 to 6 months after ablation. Despite higher than expected mean values for percentage predicted forced expiratory volume in 1 second (62%) and diffusion capacity of carbon monoxide (56%) in this patient series, many lesions were considered medically unresectable because of cardiac comorbidities (eg, critical aortic stenosis, coronary artery disease, cardiomyopathy), and 35% of patients refused surgery. Patients were selected for RFA on the basis of anatomic criteria, mostly influenced by peripheral location and size 3 cm or smaller. Central lesions were more likely to be referred for radiation therapy. The authors found that referrals for RFA were also partly patient driven, with high-risk patients favoring a single treatment modality rather than multiple treatments.
In studies that treated early–stage NSCLC with RFA, local tumor progression ranged from 3% to 42%.6,7,16,17
We observed a 31.5% local tumor progression rate at 17 months after RFA. Factors that have been reported to influence local failure include tumor size; proximity to pulmonary vessels, aorta, or azygous vein, which is associated with a "heat sink" effect; and a satisfactory margin of ablation at least 1 cm beyond the edge of tumor. The importance of ablation beyond the tumor margin was eloquently evaluated by separate investigators with porcine and rabbit models of lung RFA.18,19
These investigators demonstrated that there were three concentric histologic zones of ablation (central, intermediate, and peripheral) with mixed populations of cells. The central and intermediate zones contained nonviable cells, whereas the peripheral edge of ablation (2–5 mm) contained both necrotic and viable cells, suggesting the technical need for RFA well beyond the edge of tumor. These investigators also reported that the ground-glass attenuation observed on CT corresponded to all three zones and therefore would overestimate the diameter of complete necrosis. Limitations of RFA technology for lung neoplasm can potentially be addressed with combined modality treatment (RFA followed by XRT), because central tumor hypoxia responds better to ablation.20
There was no 30-day mortality in our patient series, and there were no complications requiring prolonged hospitalization or significant additional medical intervention. Although pleural effusions were relatively common (21%), none required drainage. Pneumothorax in this series was less common (13%) than in other published series, which is potentially attributable to pleural symphysis as a result of previous lung surgery in about 50% of the patients. Transient recurrent laryngeal nerve palsy, as observed in 1 patient after upper lobe RFA, has been previously described.21
Because many patients with medically unresectable stage I NSCLC have severe comorbidities, overall survival after lung RFA is an inappropriate surrogate for treatment success. Only 3 of the 8 patients in this study who died did so of complications related to malignant disease progression, whereas the remaining 5 of 8 died of complications of their medical comorbidities remote to the time of RFA ablation, emphasizing the compromised medical condition of these selected patients. Freedom from local and regional progression is a far superior method of analyzing RFA outcome. We observed estimated 3-year progression-free and disease-free survivals of 58% and 39%, respectively, at a median follow-up of 17 months (range 3–35 months). Although direct comparison is difficult with small patient numbers, limited or sublobar resection in compromised patients for T1N0 NSCLC in North America and Europe has a 3-year disease free survival of about 60%.11,22
RFA for stage I NSCLC is still considered a compromise and should not be substituted for anatomic surgical resection, which currently offers the best long-term survival.
Patterns of local or systemic recurrence after RFA of solid tumors have not been thoroughly evaluated in published series. Acceleration of metastatic disease after liver RFA has been considered,23
but this phenomenon has not been studied in the lung RFA experience. In an experience of more than 50 RFA treatments of solid tumors in lungs (even in patients with post-RFA pleural effusions), we have not documented any pleural dissemination of metastatic disease with the Covidien probe. The patterns of extraregional disease progression (listed in Table 6
) are commonly seen in the treatment of primary lung cancer, but no conclusions can be derived regarding acceleration of malignant disease in this study without a matched surgical cohort undergoing sublobar resection.
|
In summary, experience with RFA for stage I NSCLC is accumulating. Intermediate results are encouraging and rival the local control observed with conventional XRT. As newer noninvasive alternatives emerge for medically inoperable tumors, our ability to accurately stage cancers becomes even more critical. One should emphasize the need for thoracic surgeons to be involved in the clinical decisions for these difficult cases. Invasive staging modalities, such as mediastinoscopy and endobronchial or endoesophageal ultrasonography should be used for biopsy of suspect hilar or mediastinal lymph nodes. Determination of response to treatment and detection of local failure require careful evaluation with anatomic (diagnostic CT) and functional (PET) imaging by institutions experienced with RFA.
| Acknowledgments |
|---|
| Footnotes |
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
E. Vallieres, S. Peters, P. Van Houtte, P. Dalal, and E. Lim Therapeutic advances in non-small cell lung cancer Thorax, November 5, 2011; (2011) thoraxjnl-2011-201043v1. [Abstract] [Full Text] |
||||
![]() |
T. Hiraki, H. Gobara, H. Mimura, Y. Matsui, S. Toyooka, and S. Kanazawa Percutaneous radiofrequency ablation of clinical stage I non-small cell lung cancer J. Thorac. Cardiovasc. Surg., July 1, 2011; 142(1): 24 - 30. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Baisi, F. Raveglia, M. De Simone, and U. Cioffi Palliative role of percutaneous radiofrequency ablation for severe hemoptysis in an elderly patient with inoperable lung cancer J. Thorac. Cardiovasc. Surg., November 1, 2010; 140(5): 1196 - 1197. [Full Text] [PDF] |
||||
![]() |
T. T. Healey, M. D. Beland, C. W. Bowkley III, and D. E. Dupuy Stabilization of Mobile Pulmonary Nodules During Radiofrequency Ablation Am. J. Roentgenol., November 1, 2010; 195(5): 1238 - 1240. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. C. Fernando, M. Schuchert, R. Landreneau, and B. T. Daly Approaching the High-Risk Patient: Sublobar Resection, Stereotactic Body Radiation Therapy, or Radiofrequency Ablation Ann. Thorac. Surg., June 1, 2010; 89(6): S2123 - S2127. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Sharma, S. R. Digumarthy, M. K. Kalra, M. Lanuti, and J. A. O. Shepard Reversible Locoregional Lymph Node Enlargement After Radiofrequency Ablation of Lung Tumors Am. J. Roentgenol., May 1, 2010; 194(5): 1250 - 1256. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. D. Beland, E. J. Wasser, W. W. Mayo-Smith, and D. E. Dupuy Primary Non-Small Cell Lung Cancer: Review of Frequency, Location, and Time of Recurrence after Radiofrequency Ablation Radiology, January 1, 2010; 254(1): 301 - 307. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |