Ann Thorac Surg 2003;75:1097-1101
© 2003 The Society of Thoracic Surgeons
Original article: general thoracic
Stereotactic radiosurgery for lung tumors: preliminary report of a phase I trial
Richard I. Whyte, MDa*,
Richard Crownover, MD, PhDb,
Martin J. Murphy, PhDa,
David P. Martin, MDa,
Thomas W. Rice, MDb,
Malcolm M. DeCamp, Jr, MDa,
Raymond Rodebaugh, PhDb,
Martin S. Weinhous, PhDb,
Quynh-Thu Le, MDa
a Departments of Cardiothoracic Surgery, Neurosurgery, and Radiation Oncology, Stanford University, Stanford, California, USA
b Departments of Thoracic Surgery and Radiation Oncology, The Cleveland Clinic Medical Foundation, Cleveland, Ohio, USA
Accepted for publication October 29, 2002.
* Address reprint requests to Dr Whyte, Department of Cardiothoracic Surgery, CVRB 205, 300 Pasteur Dr, Stanford, CA94305-5407, USA
e-mail: riwhyte{at}stanford.edu
Presented at the Poster Session of the Thirty-eighth Annual Meeting of The Society of Thoracic Surgeons, Fort Lauderdale, FL, Jan 28–30, 2002.
 |
Abstract
|
|---|
BACKGROUND: Stereotactic radiosurgery is well established for the treatment of intracranial neoplasms but its use for lung tumors is novel.
METHODS: Twenty-three patients with biopsy-proven lung tumors were recruited into a two-institution, dose-escalation, phase I clinical trial using a frameless stereotactic radiosurgery system (CyberKnife). Fifteen patients had primary lung tumors and 8 had metastatic tumors. The age range was 23 to 87 years (mean, 63 years). After undergoing computed tomography–guided percutaneous placement of two to four small metal fiducials directly into the tumor, patients received 1,500 cGY of radiation in a single fraction using a linear accelerator mounted on a computer-controlled robotic arm. Safety, feasibility, and efficacy were studied.
RESULTS: Nine patients were treated with a breath-holding technique, and 14 with a respiratory-gating, automated, robotic technique. Tumor size ranged from 1 to 5 cm in maximal diameter. There were four complications related to fiducial placement: three pneumothoraces requiring chest tube insertion and one emphysema exacerbation. There were no grade 3 to 5 radiation-related complications. Follow-up ranged from 1 to 26 months (mean, 7.0 months). Radiographic response was scored as complete in 2 patients, partial in 15, stable in 4, and progressive in 2. Four patients died of non–treatment-related causes at 1, 5, 9, and 11 months after radiation.
CONCLUSIONS: Single-fraction stereotactic radiosurgery is safe and feasible for the treatment of selected lung tumors. Additional studies are planned to investigate the optimal radiation dose, best motion-suppression technique, and overall treatment efficacy.
|
Introduction
|
|---|
Pulmonary resection remains the standard of care for early-stage lung tumors as well as certain pulmonary metastases. In patients of extremely advanced age, poor underlying lung function, or multiple comorbidities, such operations may be associated with significant morbidity and mortality. In such patients, external beam radiation therapy is often the chosen treatment; however, it is accompanied with drawbacks, including damage to adjacent lung and nearby structures and a local failure rate that is greater than that seen with resection [1]. To deliver a high dose to the tumor without excessive collateral injury, multiple doses of radiation are administered, often over a several-week period of time. Stereotactic radiosurgery attempts to solve these problems with a highly collimated beam, delivered through multiple paths intersecting in the tumor volume, in a single fraction [2]. This maximizes radiation to the tumor and minimizes radiation exposure to the surrounding normal lung and adjacent structures. Although the total dose of ionizing radiation is less than that delivered using conventional fractionation, the injury to the tumor is theoretically equivalent.
Stereotactic radiosurgery is well established in the treatment of intracranial lesions, especially brain metastases and arterio-venous malformations; however, its use in extracranial neoplasms is less well defined [3]. This paper presents the preliminary results of a phase I trial of the CyberKnife stereotactic radiosurgery system (Accuray, Inc, Sunnyvale, CA) in lung tumors.
|
Patients and methods
|
|---|
The CyberKnife radiosurgery system (Fig 1)
involves a 6-MV x-band linear accelerator (LINAC) mounted on a computer-controlled robotic arm. Targeting solutions for the LINAC are based on image guidance through using two ceiling-mounted diagnostic x-ray sources with table-mounted flat-panel detectors [4]. For precise image localization, percutaneously placed metal markers (fiducials) are inserted in, or adjacent to, the tumor. Using "real-time" image processing, data from the two oblique images are digitally processed and combined with data from the planning computed tomography (CT) scan. This information is then used to direct the highly mobile radiation source, which delivers a highly collimated beam to be delivered through multiple paths over a relatively short period of time.
View larger version (91K):
[in this window]
[in a new window]
|
Fig 1. CyberKnife stereotactic radiation system demonstrating a robotically mounted linear accelerator, one (of two) ceiling-mounted x-ray localizing sources, and two table-mounted image localizing detectors. The simulated patient head is immobilized with a custom-fitted face mask, while a plastic body cradle is used for stabilization of the torso when lung tumors are treated.
|
|
Tumor motion due to respiration was addressed either using a breath-holding technique (Fig 2)
or by tracking light-emitting diodes (LEDs) placed on the patients’ skin. At Stanford, patients were coached in consistent breathing techniques and taught to hold their breath at the same phase of the respiratory cycle such that radiation could be delivered in bursts of approximately 20 seconds. At The Cleveland Clinic, LED were placed on the patients’ chest wall, and the movement of the LEDs was correlated with tumor movement using a ceiling-mounted light detector. In this fashion, the robotic arm and linear accelerator were made to track respiratory motion of the tumor [5].
View larger version (54K):
[in this window]
[in a new window]
|
Fig 2. Variation in position of fiducials during single stereotactic radiosurgery treatment. In any given direction, the targets move a maximum of 3.8 mm.
|
|
Treatment planning was done with a proprietary system allowing inverse planning and nonisocentric radiation delivery. After determining the total radiation dose to be delivered to the tumor and defining the boundaries to protect the adjacent critical structures, the software determines the targeting positions and dose to be delivered from the targeting positions. Radiation is then delivered for a calculated period of time from specific orientations and directions, each of which is called a node. The current CyberKnife device (hardware and software) allows for up to 12 different beam directions from up to 110 robot arm locations, for a total of 1,320 possible beam paths.
Protocol
After obtaining institutional review board approval from both institutions, a phase I clinical trial of the CyberKnife radiosurgery system was instituted. The study was designed to establish the safety and feasibility of treating unresectable primary and secondary tumors of the lung using the CyberKnife system. Criteria for entry included: (1) histologic confirmation of malignant tumor (primary lung or metastatic); (2) a maximum tumor diameter of 5.0 cm or less; (3) age greater than 18 years; (4) the tumor must be deemed unresectable either by radiographic criteria (such as direct invasion of the mediastinum, heart, great vessels, or trachea) or by virtue of excessive risk to patient, patient refusal to undergo surgery, or prior operative findings; (5) patients must be of Eastern Clinical Oncology Group performance status 0, 1, or 2; and (6) no prior radiation therapy to the site of radiosurgery.
Upon entry into the protocol, all patients were evaluated by both a surgeon and a radiation oncologist. Each patient’s CT scan and clinical information were reviewed by a radiologist for the purpose of ensuring correct fiducial placement. Patients then underwent percutaneous placement of two to four cylindrical gold metal fiducials (1 mm in diameter by 3 mm in length) into, or adjacent to, the tumor. This was accomplished using a 18- to 19-gauge needle under CT guidance and local anesthesia.
Within 7 days of the fiducial placement, a radiation therapy immobilization device (Alpha Cradle; Smithers Medical Products, North Canton, OH) was custom made for each patient. This partially immobilizes patients in order to minimize nonrespiratory motion during treatment. Patients then underwent a contrast-enhanced CT scan through the entire thoracic cavity using 1.25-mm-thick slices. The treating surgeon and radiation oncologist identified the location of the pulmonary tumor, and a radiosurgical treatment plan was developed using a nonisocentric, inverse-planning algorithm (Accuray, Inc) based on tumor geometry and location. The tumor was outlined in sequential axial CT images and the gross tumor volume (GTV) was calculated. Adjacent normal structures, including the heart, aorta, liver, and stomach, within 5 cm of the GTV were identified for the purpose of limiting incidental radiation to these structures. The radiation to be delivered to the tumor was prescribed to the maximum isodose line that completely covers the GTV, and the imaging set was processed for radiosurgery, using a Food and Drug Adminstration–approved proprietary treatment planning system developed for the CyberKnife. This computerized treatment planning system is a modified form of the Accuracy treatment planning system that is presently used for frameless intracranial stereotactic radiosurgery with small fields. The system coordinates the radiation treatment plan with the mechanical delivery of therapy by dividing the dose into approximately 100 beam directions (called nodes).
Radiation was delivered in a single fraction of 1,500 cGy, with the following dose limits for surrounding critical structures: maximal spinal cord dose, 800 cGy; maximal brachial plexus dose, 1,000 cGy; two-thirds of total lung volume to receive a maximum of 500 cGy; 50% of the heart to receive a maximum of 1,000 cGy; 50% of the esophagus volume should be kept under 1,000 cGy; and 50% of the liver volume should be kept under 750 cGy.
Patients were followed clinically and radiographically after radiosurgery. Detailed medical and physical examinations were performed every 4 weeks for 3 months, and again at 6 months and 1 year. A complete blood count (CBC) and comprehensive chemistry panel were performed at 4-week intervals for 3 months, and continuing at intervals of every 6 months until death. Chest CTs were performed at 2, 3, and 6 months after radiosurgery, and annually thereafter.
Toxicity beyond 3 months was scored according to the Radiation Therapy Oncology Group (RTOG) late Radiation Toxicity Scale. Radiographic response was determined by the CT appearance at 3 months according to the following definitions: complete tumor regression (CR), no measurable tumor is visible in the anterior to posterior, lateral, and inferior to superior dimensions; partial response (PR), radiographic partial regression of greater than or equal to 50% in overall tumor volume; stable disease (SD), radiographic regression less than 50% in tumor volume or less than 25% increase in tumor volume; and progressive disease (PD), greater than 25% increase in tumor volume or reappearance of any lesion that had previously disappeared.
|
Results
|
|---|
Patients
Twenty-three patients were enrolled at the two institutions. There were 13 men and 9 women. The age range was 23 to 87 years (mean, 63 years). Fifteen patients had primary lung tumors and 8 had metastatic tumors; 2 patients underwent radiosurgery to two lesions. Tumor size ranged from 1 to 5 cm in maximal diameter. Nine patients were treated with a breath-holding technique, and 14 with a respiratory-tracking, automated, robotic technique. The median treatment time was approximately 4 hours (range, 2 to 6 hours), and the radiation dose was delivered using 57 to 225 nodes.
Complications
Four patients had complications related to fiducial placement. Three patients had pneumothoraces; 2 were managed expectantly, but 1 required urgent chest tube placement (the patient had had a prior contralateral pneumonectomy). One patient had exacerbation of his underlying chronic obstructive pulmonary disease. There were no significant (RTOG grade 3 to 5) radiation-related complications. No patients had leukopenia, radiation esophagitis, or clinically apparent radiation pneumonitis.
Tumor response and follow-up
Follow-up ranged from 1 to 26 months (mean, 7.0 months). Radiographic response at 1 to 3 months was scored as complete in 2 patients, partial in 15, stable in 4, and progressive in 2. Four patients died of non–treatment-related causes at 1, 5, 9, and 11 months after treatment.
Sample patient
A 60-year-old woman with metastatic pancreatic adenocarcinoma was referred with a new right upper lobe lung metastasis. She had previously undergone right lower lobectomy with chest wall resection for a large metastasis that had invaded the chest wall. Her underlying lung function was poor and she refused further surgery. Four gold fiducials were percutaneously inserted under CT guidance into or adjacent to the tumor (Fig 3).
The tumor was outlined on a treatment-planning CT, and the software calculated that 1,500 cGy would be delivered to the tumor using 132 beam paths (Fig 4)
with the isodose curves shown in Figure 5. Her posttreatment course was uncomplicated and she experienced no
side effects. At 3 months, the right upper lobe tumor was unchanged in size, but the patient developed a new left lower lobe lesion, for which she received chemotherapy. She died of progressive metastatic disease 11 months after stereotactic radiosurgery
View larger version (112K):
[in this window]
[in a new window]
|
Fig 3. Chest radiograph of patient showing a right upper lobe lung nodule with four fiducials placed into or near the tumor (arrow points to one of the fiducials).
|
|
View larger version (114K):
[in this window]
[in a new window]
|
Fig 4. Computed tomography–generated isodose curves for right upper lobe tumor demonstrated in Figure 3. The dark red curve marks the 1,500-cGy curve; each curve represents an increment of 500 cGy.
|
|
View larger version (85K):
[in this window]
[in a new window]
|
Fig 5. Three-dimensional representation of beam paths directed to right upper lobe tumor depicted in Figure 3. The point of view is from the lower, right side of the patient.
|
|
|
Comment
|
|---|
This is the first report of clinical outcomes that follows the use of single-fraction frameless stereotactic radiosurgery being used to treat tumors of the lung. It presents the initial results of a three-phase, dose-escalation study with the CyberKnife stereotactic radiosurgery system. Other reports of stereotactic radiation to lung tumors have involved either multiple fractions or frame-based systems, and previous reports of the CyberKnife system have focused on intracranial, spinal, and pancreatic applications [6–11]. The reported study represents a coordination of efforts at two institutions using similar, but not identical, systems.
Thoracic surgeons may wonder why the term "surgery" is used in this form of treatment and why it should not simply be considered as another form of radiation therapy. The reason for this partly lies in the history of radiosurgery. Radiosurgery was initially developed as a collaboration between neurosurgeons, radiation oncologists, and radiation physicists who were concerned with the potential treatment-related injury on normal brain tissues with conventionally delivered wide-field external beam radiation. Brain tissue is highly sensitive to radiation, and injury to surrounding tissue limited the effectiveness of convention external beam radiation in the treatment of lesions such as gliomas, metastases, arteriovenous malformations, and meningiomas. Extremely fine spatial tolerances combined with the limited ability to focus the radiation led to unacceptably high rates of complication, such as blindness for lesions near the optic chiasm and hearing loss with acoustic neuromas. This motivated a search for better ways to deliver the radiation. The neurosurgeons’ precise knowledge of neuro-anatomy, coupled with developments in radiation delivery, led to the popularization of frame-based stereotactic radiosurgery systems, one of which is the Gamma Knife (Elekta Radiosurgery, Atlanta, GA). These systems, which require a frame that is affixed to the patient’s head for stabilization, are not applicable to other parts of the body, particularly those that move.
A more precise beam-localizing mechanism allows for increased radiation delivery to the target with less exposure to the surrounding structures. In the chest, the sensitive structures include the surrounding lung, esophagus, spinal cord, and heart. The low radiation tolerance of these structures limits the overall dose being delivered. Conventionally fractionated external beam radiation therapy exploits the differences between neoplastic and normal cells in terms of growth rates, DNA repair mechanisms, and reoxygenation. This led to a strategy whereby the radiation is delivered in multiple fractions, thereby maximizing injury to the target and minimizing collateral injury [12]. In stereotactic radiosurgery, all of the radiation is delivered in one dose, albeit at a smaller total dose than is administered using conventional external beam radiation.
The greatest limitation to the more widespread use of external beam radiation in treating solitary lung tumors at the present time is its lower overall efficacy when compared with resection. In a review of radiation therapy as "curative management" of lung cancers, Wagner reports local failure rates as high as 30% for T1 lesions and 70% for T2 lesions [1]. These data should be compared with locoregional failure rates as low as 2% per year after lobectomy for T1 tumors, as reported by the Lung Cancer Study Group [13]. Because there are differences in sensitivity to radiation between cells within a tumor, the local failure rate associated with external beam radiation is presumably due to inadequate radiation being delivered to all of the neoplastic cells. By increasing the dose to the target, while reducing collateral tissue damage, this drawback to radiation could be minimized.
This report focuses on the clinical results of the first dose increment, 1,500 cGy, of a three-increment, dose-escalation study. As planned, the trial involves subsequent doses of 2,000 and 2,500 cGy, with patients currently being recruited to the lower of these two dose levels. As this study demonstrates, the lowest dose has resulted in no radiation side effects, but long-term response has not been optimal.
The use of stereotactic radiosurgery in treating lung tumors is in its infancy. Numerous treatment variables remain unknown. These include patient selection criteria, the optimal dose and fractionation, treatment planning algorithms, the best way to compensate for respiratory motion, and whether radiosurgery should be combined with other treatment modalities, such as chemotherapy or radiosensitizing agents. Furthermore, comparisons in dosimetry between stereotactic radiosurgery and other forms of external beam radiation, such as intensity-modulated radiation (IMRT) and three-dimensional conformal radiation, remain to be performed. Nonetheless, even if stereotactic radiosurgery is equal to, but no more effective than, other forms of external beam radiation, its single dose requirement offers great advantages in patient ease and compliance.
In conclusion, single-fraction stereotactic radiosurgery is safe and feasible for the treatment of selected lung tumors. Additional studies are planned to investigate the optimal radiation dose, best motion-compensation technique, and overall treatment efficacy.
|
References
|
|---|
- Wagner H., Jr Resectable non-small cell lung cancer in the medically inoperable patient: curative management with radiation therapy. In: Pass H.I., Johnson D.H., Turrisi A.T., Minna J.D., eds. Lung cancer: principles and practice. Philadelphia: Lippincott, Williams, and Wilkins, 2000:829-842.
- Leksell L. The stereotactic method and radiosurgery of the brain. Acta Chir Scand 1951;102:316-319.[Medline]
- Adler J.R., Jr, Chang S.D., Murphy M.J., et al. The Cyberknife: a frameless robotic system for radiosurgery. Stereotact Funct Neurosurg 1997;69(1–4 Pt 2):124-128.[Medline]
- Antonuk L.E., Yorkston J., Huang W., et al. A real-time, flat-panel, amorphous silicon, digital x-ray imager. Radiographics 1995;15:993-1000.[Abstract]
- Schweikard A., Glosser G., Bodduluri M., Murphy M.J., Adler J.R. Robotic motion compensation for respiratory movement during radiosurgery. Comp Aid Surg 2000;5:263-277.
- Klaus H., Jurgen D., Frank L., et al. Extracranial stereotactic conformal radiation therapy of tumors in the liver and the lung. Int J Radiat Oncol Biol Phys 1998;42(Suppl 1):214.
- Blomgren H., Lax I., Naslund I., Svanstrom R. Stereotactic high dose fraction radiation therapy of extracranial tumors using an accelerator: clinical experience of the first thirty-one patients. Acta Oncol 1995;34:861-870.[Medline]
- Uematsu M., Shioda A., Suda A., et al. Computed tomography-guided frameless stereotactic radiotherapy for stage I non-small cell lung cancer: a 5-year experience. Int J Radiat Oncol Biol Phys 2001;51:666-670.[Medline]
- Chang S.D., Murphy M., Geis P., et al. Clinical experience with image-guided robotic radiosurgery (the Cyberknife) in the treatment of brain and spinal cord tumors. Neurol Med Chir (Tokyo) 1998;38:780-783.
- Uematsu M., Shioda A., Tahara K., et al. Focal, high dose, and fractionated modified stereotactic radiation therapy for lung carcinoma patients: a preliminary experience. Cancer 1998;82:1062-1070.[Medline]
- Takai Y., Mituya M., Nemoto K., et al. Simple method of stereotactic radiotherapy without stereotactic body frame for extracranial tumors. Nippon Igaku Hoshasen Gakkai Zasshi 2001;61:403-407.[Medline]
- Withers H.R. Biologic basis for altered fractionation schemes. Cancer 1985;55(Suppl 9):2086-2095.[Medline]
- Ginsberg R.J., Rubinstein L.V. Randomized trial of lobectomy versus limited resection for T1 N0 non-small cell lung cancer: Lung Cancer Study Group. Ann Thorac Surg 1995;60:615-623.[Abstract/Free Full Text]
Top
Abstract
Introduction
