| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2007;83:1820-1825
© 2007 The Society of Thoracic Surgeons
a Heart, Lung, and Esophageal Surgery Institute, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania
b Department of Surgery, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania
c Department of Radiation Oncology, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania
d The University of Pittsburgh Cancer Institute Biostatistics Facility, Pittsburgh, Pennsylvania
e Boston Medical Center, Boston, Massachusetts
Accepted for publication November 13, 2006.
* Address correspondence to Dr Luketich, Heart, Lung, and Esophageal Surgery Institute, University of Pittsburgh, Pittsburgh PA 15213 (Email: luketichjd{at}upmc.edu).
Presented at the Fifty-second Annual Meeting of the Southern Thoracic Surgical Association, Orlando, FL, Nov 10–12, 2005.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsCommentDiscussionFootnotesReferences |
|---|
Methods: Patients who were medically inoperable were offered SRS. Thoracic surgeons evaluated all patients, placed fiducials, and planned treatment in collaboration with radiation oncologists. A median dose of 20 Gy prescribed to the 80% isodose line was administered in a single fraction. The initial response rate, time to progression, and overall survival were evaluated.
Results: During a 2-year period, 32 patients, 27 with NSCLC and 5 with pulmonary metastases, underwent SRS. Fiducial placement resulted in a pneumothorax requiring a pigtail catheter in 9 patients (28%). An initial complete response was observed in 7 patients (22%) and partial response in 10 (31%). Disease was stable in 9 (28%) and progression occurred in 5 patients (16%), with a median time to local progression of 11 months. The median overall survival was 26 months. The probability of 1-year overall survival was 78% (95% confidence interval [CI], 65% to 94%) for the entire group and 91% (95% CI, 75% to 100%) for stage I patients.
Conclusions: Our preliminary experience indicates that SRS has reasonable results in this high-risk group of patients, with pneumothorax being a significant morbidity. Surgery continues to offer the best chance of cure for resectable patients; however, SRS offers an alternative to high-risk patients.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionFootnotesReferences |
|---|
Studies have demonstrated that higher radiation doses appear to increase local control and cancer-specific survival [4]. Increasing the dose during conventional radiotherapy also increases the dose to normal tissue, thus resulting in increased toxicity. Stereotactic radiosurgery (SRS) was a term coined by Leksell to describe an approach using multiple convergent beams, precise localization with a stereotactic coordinate system, rigid immobilization, and single fraction treatment. SRS provides precise delivery of beams from multiple collimated paths, which maximizes the delivery to the tumor and minimizes the exposure of normal tissue [7].
SRS is well established for the treatment of intracranial malignancies, and its use in extracranial malignancies is now being explored. In 1994, Lax and colleagues [8] from the Karolinska institute described a technique of extracranial SRS using a custom-made body cast and stereotactic coordinates. Few reports of the application of SRS in the treatment of lung malignancies from the United States have been published, and this treatment method is evolving [9, 10]. In this article we present our preliminary results with the use of SRS in the treatment of lung neoplasm in medically inoperable patients.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionFootnotesReferences |
|---|
Selection of Patients
The selection criteria for pulmonary SRS included (1) patients who were considered medically inoperable owing to poor pulmonary function, high cardiac risk, or other comorbidities; (2) patients who had had failure of previous therapies, including patients who had had prior lung resection and chemoradiation, and (3) patient refusal to have a surgical procedure. All patients were evaluated by a thoracic surgeon and a radiation oncologist before treatment.
Treatment was delivered using the CyberKnife (Accuray, Sunnyvale, CA) system. This is a frameless system that consists of a 6-MV linear accelerator mounted on a computer-controlled robotic arm. The technology and the protocol used with this system have previously been well described by Whyte and colleagues [8].
Treatment Protocol
Placement of fiducials
Fiducials, which are small tumor markers of 1 to 2 mm, were placed percutaneously with computed tomography (CT) guidance. A total of 1 to 4 fiducials were placed in and around the tumor for tumor tracking.
Computed tomography, immobilization, and planning
An immobilization device (Alpha Cradle, Smithers Medical Products, North Canton, OH), which partially immobilizes the patient to decrease the motion and provide a reproducible setup, was custom made for each patient. A week to 10 days after placement of the fiducials, a contrast-enhanced CT scan of the chest and upper abdomen with 1.25-mm sections was performed.
The thoracic surgeon and radiation oncologist then evaluated the CT scan and jointly formulated a treatment plan. The tumor size, location, and range of motion were assessed, and the ability of the patient to tolerate the planned immobilization was also evaluated. Volumetric considerations for normal tissue complications in the surrounding normal lung, spinal cord, heart, aorta, liver, and stomach were addressed during the treatment planning. This plan was developed using a nonisocentric, inverse-planning algorithm (Accuray Inc, Sunnyvale, CA). The treatment-planning software was used to outline the gross tumor volume and add a 5-mm to 10-mm margin, preferably, which was limited by adjacent critical structure dose limitation, to create the planning target volume. The adjoining critical structures were identified to limit the radiation dose.
Repositioning, relocalization, and treatment delivery
On the day of the treatment, the patient was repositioned accurately to simulate the original planning setup. The tumor and the planned isocenter of the treatment field were identified. Tumor motion during respiration was minimized by breath-holding technique, and treatment was delivered.
Targeting was coupled with a real-time image guidance system that used two ceiling-mounted diagnostic x-ray fluoroscopes and table-mounted flat panel detectors. For precise localization, the percutaneously placed markers (fiducials) were used. Data from the two oblique images were digitally processed using real-time image processing and combined with data from the planning CT scan. The cameras frequently updated the body position and accommodated for the patient’s motion. This information was then used to direct the radiation source, which delivered a highly collimated beam. The device has 6 degrees of freedom, allowing for volumes with more complex shapes.
The accuracy by the breath-hold technique was studied by Ozhasoglu and colleagues [11] in patients treated with this system. The tumor position and orientation was reproducible to within 1 to 2 mm in all degrees of freedom and to within 1 to 2 degrees for all three rotational movements. The current device allows up to 12 different beam directions from 110 robot arm locations, for a total of more than 1200 possible beam paths. The median dose prescribed was 20 Gy in a single fraction, prescribed to the 80% isodose line.
The dose limits to the surrounding critical structures were maximum spinal dose, 500 cGy; brachial plexus, 800 cGy; two-thirds of lung volume to receive a maximum dose of 500 cGy; heart left ventricle, 500 cGy; heart outside left ventricle, 800 cGy; esophagus, 800 cGy; and liver, 700 cGy.
Patient Follow-Up and Assessment of Response
Patients attended follow-up in 3-month intervals that included clinical examinations, CT scans, and positron emission tomography (PET) scans. A modified Response Evaluation Criteria in Solid Tumors (RECIST) criterion was used to assess response to treatment at 3 months (Table 1) [12, 13]. Patients were evaluated for initial response rate, time to progression, and overall survival.
|
2 test. |
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionFootnotesReferences |
|---|
|
Initial Response
The initial response was assessed at 3 months by the modified RECIST criterion (Table 1). Response could not be evaluated in 1 patient because of reported mortality at one month due to drug resistant pneumonia. Of the 32 patients treated, the response was complete in 7 (22%), partial in 10 (31%), stable disease in 9 (28%), and progressive disease in 5 (16%).
Time to Progression
Six patients remain progression free at all sites at a median follow-up of 9 months (range, 7 to 15 months; mean, 10 months). The median overall time to progression at all sites was 8 months (95% confidence interval [CI], 4 to 11 months). The predominant pattern of progression was local progression, which occurred in 17 patients. The median time to local progression was 11 months (95% CI, 8 months to NR [not reached]). Local progression was treated with SRS in most patients, and distal progression was treated with systemic therapy.
Survival
The median overall survival for the entire group of patients was 26 months (95% CI, 14 months to NR) at a median follow-up of 15 months (range, 9 to 36 months). The median overall survival was 26 months (95% CI, 13 months to NR) for patients with primary NSCLC and 17 months (95% CI, 12 months-NR) months for recurrent NSCLC. The median survival for 5 patients with metastatic disease was not reached.
The estimated 1-year overall survival for the entire group was 78% (95% CI, 65% to 94%), and the 1-year overall survival was 61% (95% CI, 40% to 91%) for patients with primary NSCLC, 89% (95% CI, 71% to 100%) for recurrent NSCLC, and 80% (95% CI, 52% to 100%) for metastatic disease. The median overall survival in stage I NSCLC patients was 26 months (95% CI, 26 months to NR) versus 10 months (95% CI, 4 months to NR; p = 0.03) for stage II-IV patients. The probability of overall 1-year survival for stage I NSCLC patients was 91% (95% CI, 75% to 100%).
Other variables analyzed were the association with tumor size (<3 cm versus >3 cm), and tumor volume (<40 cm3 versus >40 cm3) with outcome, and these did not show a significant difference.
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionFootnotesReferences |
|---|
Radiofrequency ablation is a thermal ablative technique and is a relatively new modality of treatment, which may also be applicable in high-risk patients with lung cancer. We have previously described our experience with RFA in the treatment of both primary and metastatic lung neoplasm [12, 13]. The principal finding of the initial report of primary NSCLC and metastatic lung neoplasm were that RFA was more effective for smaller (
5 cm) tumors with better early response to treatment [12]. Fernando and colleagues subsequently reported their experience in 18 patients with NSCLC encompassing all stages. At a median follow-up of 14 months, mean survival was 21 months and the median survival was not reached. The mean and median progression free interval was 16.8 and 18 months respectively [13]. The assessment of response is challenging after ablative therapies such as SRS and RFA because many of these patients have a residual scar. We have therefore used a modified RECIST criterion, which incorporates not only lesion size but also lesion quality, and assessment of metabolic activity by PET scanning [12, 13].
This study is a report of our initial experience with SRS using a frameless stereotactic system in patients with primary and metastatic lung cancer. Whyte and colleagues [9] provided the first report from the United States using a similar system for frameless SRS in 23 patients treated with a single fraction of 15Gy. The median follow-up was 7 months, and the reported response rates were similar to the current study.
Dose escalation may allow better local control, and the potential advantage of dose escalation has been reported [10, 16]. Timmerman and colleagues [10] reported the results in 37 patients with stage I NSCLC treated with SRS. In their study, the dose was escalated from 24 Gy to 60 Gy in three fractions. The results reported were encouraging, with a complete response obtained in 27% and a partial response in 60%. At a median follow-up of 15 months, 13 patients had recurrences, of which 6 experienced local failure; all these patients had received less than 18 Gy per fraction. The median time to local progression was 13 months. The disease free and overall survival at a median follow-up of 15 months was 50% and 64%, respectively [10].
This response rate reported is higher and the local recurrence lower than that obtained in our preliminary study, where a lower dose of 20 Gy in a single fraction was used. The overall 1-year survival of 91% for stage I reported in our study compares well with their results. SRS has also been reported in several other countries, including Japan, where an early study by Uematsu and colleagues was reported [16–19].
Unique radiobiology applies to SRS by virtue of a high, single-focused dose that provides a higher biologically effective dose [20], which in turn is determined by the linear quadratic equation. Onishi and colleagues [18] evaluated the clinical outcomes in 245 patients with stage I NSCLC from 13 Japanese institutions and reported a lower local recurrence rate when the biologically effective dose of greater than 100 Gy was used. In our current study, the estimated median biologically effective dose was about 60 to 70 Gy, and the median time to local progression was 11 months.
One possible reason for local progression is the lower dose that was used in this preliminary series; other factors that may have contributed include the margins obtained around the tumor and the technique of tumor tracking during respiration. Further, several patients had failed other therapy before SRS, including conventional radiation. Thus, this group of treated patients may represent a group with biologically more aggressive tumors. Our data on wedge resections suggests a significant increase in the local recurrence rates when the margins were less than 1 cm [21]. The margins applied in this preliminary study with SRS were 5 to 10 mm. In the future, a higher dose, wider margin with dose limitation to critical structures, and dynamic respiratory tracking may improve the local recurrence rates.
The current study has the limitations, such as selection bias, that are inherent to retrospective studies. The group of patients treated in this study was very heterogeneous, encompassing not only early stage disease but also patients with more advanced stages and those with metastatic disease. Further, these patients had significant associated comorbidities, with a median CCI of 3 (mean, 4.5). In fact, in a study of resected patients with NSCLC, multivariate analysis showed that CCI 3 to 4 was the only predictive factor of increased risk of major complications (odds ratio, 9.8; 95% CI, 2.1 to 45.9) [14].
In this preliminary study, we used a 20-Gy dose in a single fraction and minimal toxicity was observed. In the future, however, we plan to increase the dose up to 60 Gy in three fractions to evaluate the efficacy and toxicity of such a regimen. We now have a prospective IRB-approved protocol in place to study this regimen.
Finally, the follow-up for this cohort is rather short, and full evaluation of survival end points will require greater maturity of time-to-event data.
In summary, this study is a preliminary report on the use of SRS for both primary NSCLC and metastatic lung neoplasm in medically inoperable patients. Although the time to progression was short, our 91% 1-year overall survival results at a median follow-up of 15 months seem reasonable, particularly in patients with stage I NSCLC.
Several factors merit further investigation, including optimal patient selection, appropriate dose, fractionation, and balancing the efficacy of the treatment with toxicity. In addition, the role of adjuvant therapy to SRS remains to be determined. Surgical resection clearly remains the best treatment for resectable lung cancer; however, SRS may have a role in patients who are medically inoperable. Prospective studies are necessary to address these issues and to define the role of SRS in the treatment of lung neoplasm.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionFootnotesReferences |
|---|
DR PENNATHUR: This patient was admitted to a different hospital in a different city, and the information we have is from the primary physician who reported that the patient had panresistant pneumonia. He essentially ended up in respiratory failure, got intubated, and it was not thought to be a classical radiation pneumonitis, which occurs in a little bit different fashion with stereotactic radio surgery than conventional fashion.
DR KEITH S. NAUNHEIM (St Louis, MO): Doctor, first of all, that was a very nice presentation, and you are to be congratulated for leading the way with CyberKnife therapy. I have actually a few questions. First of all, were your patients treated utilizing the Synchrony respiratory synchronization system, the software that allows for respiratory synchronization of the CyberKnife?
Secondly, you stated you gave an "adequate margin." What you do you consider an adequate margin? Is it 1 cm, is it 15 mm, and why did you choose that?
Number three, have you noticed a decreased incidence of radiation pneumonitis? We have only done 18 such patients but have noticed no incidence yet of radiation pneumonitis, which, considering the doses we are giving, is really quite extraordinary.
That brings up the next question of doses. How did you really figure out that it is a biological equivalent of 7000 rads? We are not using 2000 but rather using 4800 rads in a four-dose therapy, and surprisingly, you’d think you would burn a hole in the lung, but people do really quite well.
Lastly, have any of your patients had concomitant chemotherapy, and if they did, did that change the site of progression? You talked about the time to progression, but you didn’t mention whether or not that was local progression or distant progression, and that is going to be critical in determining the success of CyberKnife. Thank you.
DR PENNATHUR: Thank you very much for your kind comments and some excellent questions. The first issue about using Synchrony, this patient series represents before we obtained Synchrony and respiratory tracking, but now we are using the Synchrony equipment and the software.
The second question raised—margins—is a tricky issue. We strive for margins of 1 cm, but that has to be limited by the dose constraints to protect adjacent critical structures. So our goal is 1 cm, but it is not always 1 cm; sometimes it is 5 mm, and I think that does compromise the margins to some degree. In fact, we presented a paper on wedge resections last year at The Society of Surgical Oncology wherein we analyzed patients of less than a 1-cm margin or more than a 1-cm margin and demonstrated that if we have less than a 1-cm margin, the local recurrence rate was higher. So we strive for that, but it is not always possible in every case because of the dosing constraints.
The third question was regarding the incidence of pneumonitis. We have not noticed any pneumonitis in this early follow-up, but at higher doses, discussing this with radiation oncologists, radiation toxicity can occur several months later, and there is some incidence of delayed fibrosis, bronchial stenosis, et cetera, at long-term follow-up in other series, but we have not noticed any evidence of radiation pneumonitis.
And next, figuring out the dose and the biological effective dose: it is determined by the linear quadratic equation and it may not be very accurate for SRS. Our radiation oncologists have estimated that 20 Gy in a single fraction represents about 60 to 70 Gy. The literature I have read estimates that 12 Gy times four doses corresponds to more than 100 Gy of biological effective dose.
And finally, use of concomitant chemo. Yes, some of the patients did have chemotherapy, particularly metastatic patients, they were on chemotherapy, but the numbers were small enough that we couldn’t analyze them separately.
DR THOMAS M. EGAN (Chapel Hill, NC): I concede the Tiki Award to Dr Naunheim. A couple of questions. Nice presentation. Do you have any data on pulmonary function tests? Did you do PFTs after the therapy to look at DLCO? Where do you think this fits in, in terms of other options like RFA, like 3D conformal external beam radiation therapy? And finally, is this off-label?
DR. PENNATHUR: First, the pulmonary function tests. We have now designed a prospective study where we are doing pulmonary function tests at routine intervals to see the progression. This particular cohort we didn’t do at 3-month intervals. The FEV1 mean was .6 in the compromised pulmonary group. And I don’t know how much lower it can go, but I guess it could, and so we are going to look at it.
The second question was regarding the RFA and the 3D conformal radiation. For more peripheral lesions, RFA can potentially be comparable; for more central lesions, we prefer the CyberKnife at this point at this dose; and the 3D conformal radiation is not thought to be as conformal as the stereotactic radio surgery.
This is an FDA-approved therapy, but these patients are not under a protocol. We were not doing a dose-escalation study. At present, we have a prospective protocol, which is under IRB consideration. And thank you for your comments.
DR RICHARD E. WOOD (Dallas, TX): I just wanted to comment—a lot of the things have already been discussed here. We have only been doing this for about 6 or 7 months now. We have got 17 patients. About half of them have primary lung cancer, and they were refused surgery, or they refused surgery, or they had comorbidities and therefore we felt that they were too sick to undergo standard therapy.
One of the things that I am having trouble with is contouring the outline of the cancer. In other words, we get a CT scan that is every 1.25 mm, and then we will find that of that, there are about 45 different cuts that we need to contour the tumor. I have a hard time in deciding what is atelectasis or what has changed, or what is really the cancer and the like. That is one of the things I wanted to ask you.
The other thing is we are looking at the pulmonary functions. We are looking at the quality of life after this, and I need to see more follow-up in what we are doing. So I will be watching your cases very carefully. Thank you.
DR PENNATHUR: Absolutely. I will be glad to come back next year to present more follow-up if the Society allows me to.
In terms of contouring, the software package is actually pretty good, very user-friendly, easy to do, and so we have not had much trouble. But sometimes it takes an hour for large tumors, and you really have to go through a 100 sections and do it. It can be a time-consuming process. And we will follow up on the PFTs and quality of life. Thank you for your comments.
DR RICK SCHMIDT (Safety Harbor, FL): In follow-up of a previous question, are you comparing the conformal radiation therapy results with this?
DR PENNATHUR: The thing is, I think that there is a paucity of data in terms of comparison of 3D conformal or 4D IMRT* and SRS, some of the newer modalities. We have discussed this with some of our radiation oncologists, and we feel that the stereotactic radiosurgery is actually more conformal. I am not aware of a direct head-to-head comparison between these two modalities, and thank you for the question.
|
Footnotes |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionFootnotesReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionFootnotesReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. A. Cesaretti, A. Pennathur, B. S. Rosenstein, S. J. Swanson, and H. C. Fernando Stereotactic Radiosurgery for Thoracic Malignancies Ann. Thorac. Surg., February 1, 2008; 85(2): S785 - S791. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Pennathur, J. D. Luketich, G. Abbas, M. Chen, H. C. Fernando, W. E. Gooding, M. J. Schuchert, S. Gilbert, N. A. Christie, and R. J. Landreneau Radiofrequency ablation for the treatment of stage I non-small cell lung cancer in high-risk patients. J. Thorac. Cardiovasc. Surg., October 1, 2007; 134(4): 857 - 864. [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 |
