- Email: [email protected]
Insertion and fixation of fiducial markers for setup and tracking of lung tumors in radiotherapy
Int. J. Radiation Oncology Biol. Phys., Vol. 63, No. 5, pp. 1442–1447, 2005 Copyright © 2005 Elsevier Inc. Printed in the USA. All rights reserved 0360-3016/05/$–see front matter
doi:10.1016/j.ijrobp.2005.04.024
CLINICAL INVESTIGATION
Lung
INSERTION AND FIXATION OF FIDUCIAL MARKERS FOR SETUP AND TRACKING OF LUNG TUMORS IN RADIOTHERAPY MIKADO IMURA, M.D.,* KOICHI YAMAZAKI, M.D.,* HIROKI SHIRATO, M.D.,† RIKIYA ONIMARU, M.D.,† MASAHARU FUJINO, M.D.,† SHINICHI SHIMIZU, M.D.,† TOSHIYUKI HARADA, M.D.,‡ SHIGEAKI OGURA, M.D.,§ HIROTOSHI DOSAKA-AKITA, M.D.,储 KAZUO MIYASAKA, M.D.,† AND MASAHARU NISHIMURA, M.D.* *First Department of Medicine and †Department of Radiology, Hokkaido University School of Medicine, Sapporo, Japan; ‡ Department of Medicine, Iwamizawa Municipal Hospital, Iwamizawa, Japan; §Department of Respiratory Diseases, Sapporo City General Hospital, Sapporo, Japan; 储Department of Medical Oncology, Hokkaido University Graduate School of Medicine, Sapporo, Japan Purpose: Internal 1.5-mm fiducial markers were used in real-time tumor-tracking radiotherapy (RT) for lung cancer. The fixation rate of the markers using the bronchial insertion technique, reliability of the setup using markers around the target volume, dislocation of the markers after real-time tumor-tracking RT, and long-term toxicity of marker insertion were investigated. Methods and Materials: Between July 2000 and April 2004, 154 gold markers were inserted into 57 patients with peripheral lung cancer. The distances between the implanted markers in 198 measurements in 71 setups in 11 patients were measured using two sets of orthogonal diagnostic X-ray images of the real-time tumor-tracking RT system. The distance between the markers and the chest wall was also measured in a transaxial CT image on 186 occasions in 48 patients during treatment planning and during follow-up. The median treatment time was 6 days (range, 4 –14 days). Results: In 115 (75%) of the 154 inserted markers, the gold marker was detected throughout the treatment period. In 122 markers detected at CT planning, 115 (94%) were detected until the end of treatment. The variation in the distances between the implanted markers was within ⴞ2 mm in 95% and ⴞ1 mm in 80% during treatment. The variation in the distances between the implanted markers was >2 mm in at least one direction in 9% of the setups for which reexamination with a CT scan was indicated. The fixation rate in the left upper lobe was lower than in the other lobes. A statistically significant relationship was found between a shorter distance between the markers and the chest wall and the fixation rate, suggesting that the markers in the smaller bronchial lumens fixed better than those in the larger lumens. A learning curve among the endoscopists was suggested in the fixation rate. The distance between the markers and the chest wall changed significantly within a median of 44 days (range, 16 –181 days) after treatment. Conclusion: The fixation of markers into the bronchial tree was useful for the setup for peripheral lung cancer and had an accuracy of ⴞ2 mm during the 1–2-week treatment period. The relationship between the markers and tumor can change significantly after 2 weeks, suggesting that adaptive four-dimensional RT is required. © 2005 Elsevier Inc. Real-time tumor-tracking radiotherapy, Four-dimensional radiotherapy, Fiducial marker, Bronchoscopy.
INTRODUCTION Interest in four-dimensional (4D) planning (1, 2) and the 4D delivery of radiotherapy (RT) (3–5) to improve the temporal accuracy of beam delivery for tumors in motion, such as lung tumors, has been great. We previously reported the existence of intrafractional and interfractional changes in tumor position due, not only to the normal respiratory cycle, but also to unpredictable baseline shifts and variable amplitude and respiration rates (6, 7). To track the changes in tumor position, fiducial markers near the lung tumor are
useful for daily setup and real-time tumor tracking of the tumor position (8). Insertion of the marker through the skin surface was not recommended in a recent study because of the high frequency of pneumothorax after the procedure (9). The pneumothorax that resulted made treatment planning difficult, and sometimes adversely affected the general condition of patients with poor respiratory function. The endoscopists at our institution have developed equipment to insert markers through a bronchial fiberscope in conjunction with a virtual bronchoscopic navigation system (10). Reports
Reprint requests to: Koichi Yamazaki, M.D., First Department of Medicine, Hokkaido University School of Medicine, North 15, West 7, Kita-ku, Sapporo 060-8638 Japan. Tel: (⫹81) 11-706-5911; Fax:
(⫹81) 11-706-7899; E-mail: [email protected] Received Feb 9, 2005, and in revised form April 7, 2005. Accepted for publication April 16, 2005. 1442
Fiducial markers in RT
have been published describing the feasibility of this equipment, and showing that marker insertion into the lungs is as safe as it is for other organs (11, 12). The fixation of the marker relative to the isocenter of the target volume was shown to be reliable for various organs, including the prostate, liver, and paraspinal region (12–14). The real-time tumor-tracking RT (RTRT) system consists of four sets of diagnostic X-ray television systems (two of which offer an unobstructed view of the patient at any time), an image processor unit, a gating control unit, and an image display unit (8, 12). The position of the patient can be corrected by adjusting the actual marker position to the planned marker position, which has been transferred from the three-dimensional treatment planning system and superimposed on the fluoroscopic image on the display unit of the RTRT system. The system recognizes the position of a 2.0-mm gold marker in the human body 30 times/s using two X-ray television systems. The position of the markers can be visualized during RT and after treatment delivery to verify the accuracy of the localization. Setup of the target volume in solid organs has been shown to be improved with the use of three markers and the RTRT system (8, 12, 14 –16). However, the accuracy of the setup in lung tissue with three markers has not been reported. In the present study, we investigated the fixation rate of markers placed using the bronchial insertion technique, the reliability of the setup using markers around the target volume, and the dislocation of the markers after RTRT during the follow-up period. The long-term toxicity of the inserted markers in patients with lung cancer was also investigated. METHODS AND MATERIALS Details on the technique for inserting gold markers with a diameter of 1.5 mm into the lung have been previously reported (11, 12). In brief, special equipment for the insertion of gold markers through bronchoscopy was developed and used to insert the markers into small bronchi with a diameter of ⱕ1.5 mm (Olympus, Tokyo, Japan). Insertion of the markers during the fiberscopic examination took 20 –30 min for each patient. This technique was used for peripheral lung tumors ⬍6 cm in diameter in patients scheduled to undergo hypofractionated high-dose radiation using RTRT. The patients in this study were selected from those who were enrolled in a prospective feasibility study (12) or dose-finding study (unpublished data). Between July 2000 and April 2004, 154 markers were inserted in 57 patients (42 men, 15 women, average age 70.1 years, standard deviation, 10.4; range, 43– 86 years), with peripheral lung cancer. Of the 57 patients, 51 had primary lung cancer (adenocarcinoma in 26, squamous cell carcinoma in 19, unclassified non– small-cell lung cancer in 1, and unknown in 5) and 6 had metastatic lung cancer (lung in 1, head and neck in 4, and malignant melanoma in 1). The Eastern Cooperative Oncology Group performance status of the 57 patients enrolled in this study was Grade 1 in 39, Grade 2 in 9, Grade 3 in 1, Grade 4 in 0, and unknown in 8. In the initial 2 years of the study, only one marker was implanted. Beginning in 2002, two or more markers were implanted in patients who were able to tolerate the procedure to account for
● M. IMURA et al.
1443
migration, rotation, and deformation. One, two, three, four, five, six, seven, and nine markers were inserted into the peripheral bronchial small lumen in 22, 6, 12, 10, 3, 1, 2, and 1 patient, respectively. The patients who received more than five markers had two tumors and were treated using RTRT for both. Computed tomography for planning was performed on the same day or up to 5 days after the insertion. The patients underwent CT by a multidetector CT scanner at the end of the expiration phase in normal breathing with a slice thickness of 1 mm. RT was begun 6.4 ⫾ 3.1 days (mean ⫾ SD) after marker insertion. As we have previously noted, judging whether the marker is fixed to the same location in moving soft organs such as lung parenchyma is difficult (12, 13). Thus, in the present study, we measured the distances between the implanted markers (DIMs) using the RTRT system and compared them with the DIMs measured in CT planning. If the discrepancy in any of the DIM measurements was ⬎2 mm, we determined the dislocation of the marker to be beyond the tolerance level, and reexamination of the marker position using CT was indicated before RT to recalculate the distance between the marker and the isocenter. The new data on the relationship between the marker and the isocenter were again transferred to the RTRT system, and the patient was repositioned using the new data. The treatment period from the initial RT session to the last was 4 –14 days (median, 6 days). If the markers dropped off the lung after insertion and before CT planning, the patients were treated with RTRT using the remaining markers. If no marker was left at the start of RTRT, conventional setup techniques were applied using orthogonal portal images at the isocenter (17, 18). In the present study, the following two measurements were used for analysis: 1. Measurement of the three-dimensional distance between two markers, DIM, using two sets of orthogonal diagnostic X-ray images of the RTRT system to judge the intermarker dislocation. The measurement was performed during the setup of the patient in each treatment session. Eleven patients who agreed to the use of their data for research were enrolled in this part of the study. No more than one measurement for each DIM in the same day was used for the analysis. Because a 1-mm slice thickness was used in the CT examination, the uncertainty in the measurement was not better than 0.5 mm. The diameter of the fiducial marker was 1.5 mm; thus, the uncertainty in the measurement of the location of the marker was about 1–2 mm at best. 2. Measurement of the distance between the marker and the chest wall (DMC) in the transaxial CT image to analyze the relationship between the location (peripheral or central) of the marker and the fixation rate. The DMC was selected for analysis because endoscopists routinely measure the DMC using fluoroscopic guidance to prevent pneumothorax during the insertion of markers. In the present study, the distance was also measured to judge the dislocation of the marker relative to the normal structure, the chest wall, in the follow-up CT scans. Patients underwent CT at planning and at follow-up after treatment with the same multidetector CT scanner, with a slice thickness of ⱕ5 mm, every 1–2 months during the initial year and then every 3– 6 months afterward. Forty-eight patients were enrolled in this part of the study, and the data of the distance were used for the analysis. To determine the accuracy of the DMC measurements, we selected 12 benign pulmonary nodules in 9 patients who were examined more than twice. Because they had not been treated for, or had, any other
1444
I. J. Radiation Oncology
● Biology ● Physics
Table 1. Number of fiducial markers at insertion, CT planning, at start of RTRT, at end of RTRT, and at last follow-up of patient
Volume 63, Number 5, 2005
Table 2. Relationship between location of marker and fixation rate of marker at last follow-up Right lung lobes*
Days after insertion*
Insertion Detected Detected Detected Detected
performed at CT planning at start of RTRT at end of RTRT at last follow-up
n
Minimum
Maximum
154 122 117 115 104
— 0 6 6 16
— 5 6 15 181
Abbreviations: CT ⫽ computed tomography; RTRT ⫽ real-time tumor tracking radiotherapy. * Minimal and maximal number of elapsed days after insertion.
pulmonary disease during the period between the two CT scans, the measurement of the distance between the nodule and the chest wall was regarded as the physically accurate DMC measurement. The discrepancy was 0 ⫾ 1 mm (mean ⫾ SD) within the follow-up period of 294 ⫾ 190 days. However, the CT scans for these 9 patients and for the 48 patients in the following study were performed at the end of inspiration, and not at the end of expiration, which may have influenced the accuracy of the measurements. Moreover, because the CT slice thickness was 5 mm, the uncertainty in the DMC measurement in the transaxial image was estimated to be not better than 3 mm. The following statistical analyses were used in this study. The Kruskal-Wallis test was used for the relationship analysis between the doctor’s experience of insertion and the success rate of the fixation of implanted markers, as well as for the relationship analysis between the DMC at insertion and the eventual fixation rate. The Mann-Whitney U test was used to compare the DMC between the fixed markers and the dropped markers. The Wilcoxon signed-rank test was used to compare the DMC before and after RTRT. The chi-square test for independence or Fisher’s exact probability test was used to compare the fixation rate among the different lobes. The Kaplan-Meier method was used to calculate the fixation rate of the markers after the date of insertion.
Fixation Dropped Total
Upper
Middle/lower
Total
26 9 35
22 6 28
48 15 63
Left lung lobes†
Fixation Dropped Total
Upper
Lower
28 26 54
28 9 37
56 35 91
* p ⫽ 0.5 (Fisher’s exact probability test). p ⫽ 0.04 (chi-square test for independence).
†
fixation rate was 68%. The other remaining 50 markers were defined as “dropped” markers in the present study. A significant difference in the fixation rate was noted among the positions of the markers. Markers in the left upper lobe had a lower fixation rate than those in the left lower lobe (p ⫽ 0.04, chi-square test for independence; Table 2). No difference was observed between the right upper lobe, right middle/lower lobe, and left lower lobe. A Kaplan-Meier curve was plotted for the fixation rate of the markers from the date of insertion (Fig. 1), according to which lobe the markers were inserted into. The curve of the fixation rate in the left upper lobe was significantly lower than that of the others by the log–rank test (p ⫽ 0.01). A statistically significant difference was found in the fixation rate of the markers among markers with different DMCs at insertion and before RT. A shorter DMC correlated with a better fixation rate (p ⫽ 0.01, Kruskal-Wallis test; Table 3). The average ⫾ SD of the DMC was 17 ⫾ 12 mm in the fixed markers and 24 ⫾ 12 mm in the dropped ones (p ⫽
RESULTS Table 1 shows the performance of the fixation technique of the gold markers. Of the 154 markers, 122 (79%) were detected at CT planning, which was performed 0 –5 days after insertion (median, 2 days). Of 122 markers seen on CT, 117 (96%) were ready to be used at the start of RTRT (or 76% of the 154 inserted markers). Of the 117 markers, 115 (98%) were detected throughout the treatment period (median 10 days, range, 6 –15 days). The marker was detected at the last follow-up date (range, 16 –181 days; median, 44 days) after insertion in 104 (68%) of the 154 implanted markers. Thus, 5.3 markers/d were dropped within the initial 6 days, 0.7 marker/d in the following 6 –15 days, and 0.07 marker/d in the following 16 –181 days. When we defined “the fixation rate” as the ratio of the fiducial markers detected on CT or X-ray images at the end of the follow-up period to all 154 markers inserted, the
Fig. 1. Kaplan-Meier curve of fixation rate of markers from date of insertion according to location of each marker.
Fiducial markers in RT
● M. IMURA et al.
1445
Table 3. Relationship between distance from marker to chest wall and the fixation rate of marker at last follow-up Distance between marker and chest wall (mm)
Fixation (n ⫽ 104) Dropped (n ⫽ 17*)
ⱕ10
⬎10–20
⬎20
40 1
33 7
31 9
p ⫽ 0.01 (Kruskal-Wallis test). * Of 154 markers, 122 were detected at CT planning, performed 0 –5 days after insertion. Afterward, 18 markers were dropped; 1 patient experienced drop of marker before X-ray examination. We were able to investigate the distances between the marker and chest wall in 17 of those 18 dropped markers.
0.01, Mann-Whitney U test). Table 4 shows the relationship between the number of times the doctor had performed insertions and the success rate of the fixation of the marker at the last follow-up date. These data suggest the possibility of a significant difference among doctors who had performed insertions ⬍20, 20 – 49, and ⱖ50 times and the fixation rate at the last follow-up date. The fixation rate at the last follow-up date was 58%, 64%, and 75% for doctors who had performed insertions ⬍20, 20 – 49, and ⱖ50 times, respectively (p ⫽ 0.05, Kruskal-Wallis test). Figure 2 shows the DIM, plotted according to the days after the start of RTRT in 71 treatment setups in 11 patients. Of the 198 measurements, 189 (95%) were within 2 mm, and 158 (80%) were within ⫾1 mm. In 9 (5%) of 198 measurements in 71 setups, the discrepancy was ⬎2 mm. In 6 (8%) of 71 setups in which the discrepancy was ⬎2 mm, a CT scan for recalculation of the marker position relative to the isocenter was indicated. The discrepancy was ⱕ2 mm in the remaining 198 measurements in 64 (90%) of 71 setups in 11 patients. Figure 3 shows the change in DMC according to the days after the start of RTRT in 186 measurements in 93 fixed markers that were measurable on CT in 48 patients. Eleven markers were not measurable by CT because the follow-up on these markers was conducted at other institutes. Large dislocation of the markers, often ⬎5–15 mm from the original position, was noted during the follow-up period in 93 markers. A significant difference was found in the DMC before (mean ⫾ SD 16 ⫾ 11 mm) and after (18 ⫾ 12 mm) RTRT in these
Fig. 2. Discrepancy in distance between markers and days passed after start of real-time tumor-tracking radiotherapy (RTRT). Open circles indicate discrepancy for which computed tomography scan was required to recalculate position of marker relative to isocenter.
93 markers (p ⫽ 0.0002, Wilcoxon signed-rank test). The average ⫾ SD of the discrepancy from the original DMC to the DMC at the last follow-up date was 1 ⫾ 4 mm. This deviation was significantly greater than the deviation in the measurement of benign pulmonary nodules (0 ⫾ 1 mm, as described above; p ⫽ 0.02, Mann-Whitney U test). All the patients tolerated the procedure. Only 1 patient experienced pneumothorax after insertion of the marker, which resolved within 1 week merely by bed rest. Most of the dropped markers were detected on the abdominal X-ray in the intestinal tract. Several patients coughed the markers up. None of these patients suffered any symptomatic adverse effects from the dropped markers, although most were disappointed that the markers had fallen out. DISCUSSION Markers dropped within the first week after insertion at a high rate (38 markers, 76% of total dropped markers) and at a much lower rate 1 week after insertion (12 markers, 24%; p ⬍0.001, Mann-Whitney U test).
Table 4. Relationship between number of times doctors had performed fixation and fixation rate of markers Doctor’s experience (n)
Fixation Dropped Total Fixation rate
⬍20
20–50
⬎50
Total
25 18 43 0.58
27 15 42 0.64
52 17 69 0.75
104 50 154 —
p ⫽ 0.05 (Kruskal-Wallis test).
Fig. 3. Changes in distance between chest wall and marker in transaxial computed tomography slice showing marker, during days after start of real-time tumor-tracking radiotherapy (RTRT).
1446
I. J. Radiation Oncology
● Biology ● Physics
The diameter of the bronchial lumina is most likely grossly related to the DMC, because the diameter of the bronchial lumina is large at the central part of the lung, where the DMC is long, and is small at the periphery of the lung, where the DMC is short. The statistical correlation between the fixation rate and the DMC suggests that the small diameter of a bronchial lumen is an important factor for the success of marker fixation. Markers may simply fix in the bronchial lumen, which has a diameter ⬍1.5 mm, and drop if the bronchial lumen is enlarged after forced insertion of the marker. In addition, there may be some biologic reaction that fixes the marker. Our speculation is that the biologic reaction to the insertion of the marker consists of an inflammatory change secondary to local traumatic injury of the bronchial mucosal surface and surrounding normal tissue. After insertion of the marker, we speculate that an edematous reaction occurs and that subsequent fibrotic changes around the marker occur, causing the marker to be fixed in the pulmonary parenchyma or stenotic lumen of the bronchus. In an autopsy case, a fiducial marker had been implanted for a lung tumor 3 months before the patient’s death. The gold marker was trapped in the bronchial tree, which was surrounded by fibrotic tissue that developed as result of high-dose radiation. The safety of the procedure for implanting fiducial markers with bronchofiberscopy was reconfirmed (11, 12). An apparent learning curve on fixation of the marker and a lower fixation rate at the left upper lobe were suggested. These markers are pushed by a device through the bronchial lumen, which means that the anatomic distribution of the bronchial tree influences the difficulty of the procedure. When the target lesions are in the upper lobe or superior segment, both of which have sharply curved bronchial trees, endoscopists experience difficulty pushing the marker in correctly. Our data revealed that the left upper lobe is the most difficult portion in which to fix the markers. A relationship between fixation rate and the DMC was also demonstrated. On the basis of these results, we concluded that markers should be inserted as close to the peripheral parietal pleura as possible. This idea is consistent with the anatomic fact that the peripheral bronchi consist of tapering and curving air lumens with smaller diameters than the central bronchi. Thus, training endoscopists would be an important factor in increasing the success rate of fixation and the safety of the procedure. We have produced a free videostreaming website for endoscopists, illustrating the technical aspects of the insertion procedure (19). This material will help new endoscopists increase the fixation rate of the marker within a shorter learning period. After about 2 weeks, apparent dislocation of the markers from the planned position in the CT measurements occurred. These results contradicted the results of the measurement of the DIM, which showed less dislocation even after 2 weeks. This discrepancy between the DMC on CT and the fluoroscopic measurement of the distance between the three markers suggests that the change in the DMC is at least partly due to the dislocation of lung tissue from the
Volume 63, Number 5, 2005
original position, rather than migration of the markers. This speculation does not contradict our clinical experience showing that lung tissue moves during the treatment period as a result of shrinkage of the tumor mass and radiation pneumonitis. Recently, Ebe et al. (20), at Yamaguchi University Hospital, developed a new system consisting of an RTRT system with CT on a rail in the treatment room. Their treatment machine provides a quick check of the dislocation between the marker and the tumor. The operator can immediately correct the coordinates of the marker to the isocenter before each treatment. If the dislocation is due to shrinkage of the tumor, adaptive RT, changing the shape of the field may be indicated, although this procedure is beyond our capability. The DIM was ⱕ2 mm in 95% of the measurements, and 1 mm or better in 80%. Provided that plural markers do not move the same distance in the same direction, these values can be regarded as the translational setup accuracy of our system for lung cancer. In conjunction with the use of CT verification of the marker position relative to the isocenter of the target volume, three markers and two fluoroscopic systems can provide excellent set-up accuracy for patients with peripheral lung cancer. In practice, having to perform reexamination by CT in a different room is quite inconvenient for both patient and staff; thus, using both an RTRT system and CT on a rail is probably the best solution for 4D RT of lung cancer. In summary, 1.5-mm gold markers inserted using bronchoscopy near lung tumors were detected throughout RTRT in 75% (115 of 154) of inserted markers and in 94% (115 of 122) of markers used in CT planning. The distance between the two markers was within ⫾2 mm in 95% of measurements and ⫾1 mm in 80% during RTRT. The distance was ⬎2 mm at least in one direction in 5% of the setups for which reexamination by CT was indicated. A statistically significant relationship was found between a shorter DMC and the fixation rate, suggesting that markers in smaller bronchial lumina fixed better than those in larger lumina. A learning curve among endoscopists was also suggested to affect the fixation rate of the markers. A large change in the DMC 2 weeks after insertion suggested dislocation of the tumor and surrounding lung tissue took place owing to shrinkage or deformation of the tumor mass by RT. CONCLUSION The relationship between fiducial markers and gross tumor volume can change during RT owing to changes in the volume and location of the tumor mass, as well as possible migration of the marker. Nevertheless, in combination with CT measurement for recalculation between the marker and tumor volume in ⬍10% of the setups, implantation of three markers using our technique was useful for setup, with an accuracy of ⫾2 mm. For RT lasting ⬎2 weeks, three gold markers and the RTRT system would play an additional role in realizing individually based, precise, adaptive RT for moving, shrinking, and deforming lung cancer.
Fiducial markers in RT
● M. IMURA et al.
1447
REFERENCES 1. Keall PJ, Starkschall G, Shukla H, et al. Acquiring 4D thoracic CT scans using a multislice helical method. Phys Med Biol 2004;49:2053–2067. 2. Mageras GS, Pevsner A, Yorke ED, et al. Measurement of lung tumor motion using respiration-correlated CT. Int J Radiat Oncol Biol Phys 2004;60:933–941. 3. Berbeco RI, Jiang SB, Sharp GC, et al. Integrated radiotherapy imaging system (IRIS): Design considerations of tumour tracking with LINAC gantry-mounted diagnostic x-ray systems with flat-panel detectors. Phys Med Biol 2004;49:243– 255. 4. Shimizu S, Shirato H, Ogura S, et al. Detection of lung tumor movement in real-time tumor-tracking radiotherapy. Int J Radiat Oncol Biol Phys 2001;51:304 –310. 5. Shirato H, Shimizu S, Kitamura K, et al. Four-dimensional treatment planning and fluoroscopic real-time tumor tracking radiotherapy for moving tumor. Int J Radiat Oncol Biol Phys 2000;48:435– 442. 6. Shirato H, Seppenwoolde Y, Kitamura K, et al. Intrafractional tumor motion: Lung and liver. Semin Radiat Oncol 2004;14: 10 –18. 7. Seppenwoolde Y, Shirato H, Kitamura K, et al. Precise and real-time measurement of 3D tumor motion in lung due to breathing and heartbeat, measured during radiotherapy. Int J Radiat Oncol Biol Phys 2002;53:822– 834. 8. Shirato H, Shimizu S, Kunieda T, et al. Physical aspects of a real-time tumor-tracking system for gated radiotherapy. Int J Radiat Oncol Biol Phys 2000;48:1187–1195. 9. Whyte RI, Crownover R, Murphy MJ, et al. Stereotactic radiosurgery for lung tumors: Preliminary report of a phase I trial. Ann Thorac Surg 2003;75:1097–1101. 10. Shinagawa N, Yamazaki K, Onodera Y, et al. CT-guided transbronchial biopsy using an ultrathin bronchoscope with virtual bronchoscopic navigation. Chest 2004;125:1138 – 1143. 11. Harada T, Shirato H, Ogura S, et al. Real-time tumor-tracking radiation therapy for lung carcinoma by the aid of insertion of a gold marker using bronchofiberscopy. Cancer 2002;95: 1720 –1727.
12. Shirato H, Harada T, Harabayashi T, et al. Feasibility of insertion/implantation of 2.0-mm-diameter gold internal fiducial markers for precise setup and real-time tumor tracking in radiotherapy. Int J Radiat Oncol Biol Phys 2003;56:240 –247. 13. Kitamura K, Shirato H, Shimizu S, et al. Registration accuracy and possible migration of internal fiducial gold marker implanted in prostate and liver treated with real-time tumortracking radiation therapy (RTRT). Radiother Oncol 2002;62: 275–281. 14. Onimaru R, Shirato H, Aoyama H, et al. Calculation of rotational setup error using the real-time tracking radiation therapy (RTRT) system and its application to the treatment of spinal schwannoma. Int J Radiat Oncol Biol Phys 2002;54: 939 –947. 15. Yamamoto R, Yonesaka A, Nishioka S, et al. High dose three-dimensional conformal boost (3DCB) using an orthogonal diagnostic X-ray set-up for patients with gynecological malignancy: A new application of real-time tumor-tracking system. Radiother Oncol 2004;73:219 –222. 16. Shirato H, Oita M, Fujita K, et al. Three-dimensional conformal setup (3D-CSU) of patients using the coordinate system provided by three internal fiducial markers and two orthogonal diagnostic X-ray systems in the treatment room. Int J Radiat Oncol Biol Phys 2004;60:607– 612. 17. Fukumoto S, Shirato H, Shimzu S, et al. Small-volume imageguided radiotherapy using hypofractionated, coplanar, and noncoplanar multiple fields for patients with inoperable Stage I nonsmall cell lung carcinomas. Cancer 2002;95:1546 –1553. 18. Onimaru R, Shirato H, Shimizu S, et al. Tolerance of organs at risk in small-volume, hypofractionated, image-guided radiotherapy for primary and metastatic lung cancers. Int J Radiat Oncol Biol Phys 2003;56:126 –135. 19. Real-time tumor-tracking radiotherapy video streaming site. http://hokudai.systemk.co.jp/. 20. Ebe K, Shirato H, Hiyama A, et al. Integration of fluoroscopic real-time tumor-tracking system and tomographic scanner on the rail in the treatment room [Abstract]. Int J Radiat Oncol Biol Phys 2004;60(Suppl.):S604.
doi:10.1016/j.ijrobp.2005.04.024
CLINICAL INVESTIGATION
Lung
INSERTION AND FIXATION OF FIDUCIAL MARKERS FOR SETUP AND TRACKING OF LUNG TUMORS IN RADIOTHERAPY MIKADO IMURA, M.D.,* KOICHI YAMAZAKI, M.D.,* HIROKI SHIRATO, M.D.,† RIKIYA ONIMARU, M.D.,† MASAHARU FUJINO, M.D.,† SHINICHI SHIMIZU, M.D.,† TOSHIYUKI HARADA, M.D.,‡ SHIGEAKI OGURA, M.D.,§ HIROTOSHI DOSAKA-AKITA, M.D.,储 KAZUO MIYASAKA, M.D.,† AND MASAHARU NISHIMURA, M.D.* *First Department of Medicine and †Department of Radiology, Hokkaido University School of Medicine, Sapporo, Japan; ‡ Department of Medicine, Iwamizawa Municipal Hospital, Iwamizawa, Japan; §Department of Respiratory Diseases, Sapporo City General Hospital, Sapporo, Japan; 储Department of Medical Oncology, Hokkaido University Graduate School of Medicine, Sapporo, Japan Purpose: Internal 1.5-mm fiducial markers were used in real-time tumor-tracking radiotherapy (RT) for lung cancer. The fixation rate of the markers using the bronchial insertion technique, reliability of the setup using markers around the target volume, dislocation of the markers after real-time tumor-tracking RT, and long-term toxicity of marker insertion were investigated. Methods and Materials: Between July 2000 and April 2004, 154 gold markers were inserted into 57 patients with peripheral lung cancer. The distances between the implanted markers in 198 measurements in 71 setups in 11 patients were measured using two sets of orthogonal diagnostic X-ray images of the real-time tumor-tracking RT system. The distance between the markers and the chest wall was also measured in a transaxial CT image on 186 occasions in 48 patients during treatment planning and during follow-up. The median treatment time was 6 days (range, 4 –14 days). Results: In 115 (75%) of the 154 inserted markers, the gold marker was detected throughout the treatment period. In 122 markers detected at CT planning, 115 (94%) were detected until the end of treatment. The variation in the distances between the implanted markers was within ⴞ2 mm in 95% and ⴞ1 mm in 80% during treatment. The variation in the distances between the implanted markers was >2 mm in at least one direction in 9% of the setups for which reexamination with a CT scan was indicated. The fixation rate in the left upper lobe was lower than in the other lobes. A statistically significant relationship was found between a shorter distance between the markers and the chest wall and the fixation rate, suggesting that the markers in the smaller bronchial lumens fixed better than those in the larger lumens. A learning curve among the endoscopists was suggested in the fixation rate. The distance between the markers and the chest wall changed significantly within a median of 44 days (range, 16 –181 days) after treatment. Conclusion: The fixation of markers into the bronchial tree was useful for the setup for peripheral lung cancer and had an accuracy of ⴞ2 mm during the 1–2-week treatment period. The relationship between the markers and tumor can change significantly after 2 weeks, suggesting that adaptive four-dimensional RT is required. © 2005 Elsevier Inc. Real-time tumor-tracking radiotherapy, Four-dimensional radiotherapy, Fiducial marker, Bronchoscopy.
INTRODUCTION Interest in four-dimensional (4D) planning (1, 2) and the 4D delivery of radiotherapy (RT) (3–5) to improve the temporal accuracy of beam delivery for tumors in motion, such as lung tumors, has been great. We previously reported the existence of intrafractional and interfractional changes in tumor position due, not only to the normal respiratory cycle, but also to unpredictable baseline shifts and variable amplitude and respiration rates (6, 7). To track the changes in tumor position, fiducial markers near the lung tumor are
useful for daily setup and real-time tumor tracking of the tumor position (8). Insertion of the marker through the skin surface was not recommended in a recent study because of the high frequency of pneumothorax after the procedure (9). The pneumothorax that resulted made treatment planning difficult, and sometimes adversely affected the general condition of patients with poor respiratory function. The endoscopists at our institution have developed equipment to insert markers through a bronchial fiberscope in conjunction with a virtual bronchoscopic navigation system (10). Reports
Reprint requests to: Koichi Yamazaki, M.D., First Department of Medicine, Hokkaido University School of Medicine, North 15, West 7, Kita-ku, Sapporo 060-8638 Japan. Tel: (⫹81) 11-706-5911; Fax:
(⫹81) 11-706-7899; E-mail: [email protected] Received Feb 9, 2005, and in revised form April 7, 2005. Accepted for publication April 16, 2005. 1442
Fiducial markers in RT
have been published describing the feasibility of this equipment, and showing that marker insertion into the lungs is as safe as it is for other organs (11, 12). The fixation of the marker relative to the isocenter of the target volume was shown to be reliable for various organs, including the prostate, liver, and paraspinal region (12–14). The real-time tumor-tracking RT (RTRT) system consists of four sets of diagnostic X-ray television systems (two of which offer an unobstructed view of the patient at any time), an image processor unit, a gating control unit, and an image display unit (8, 12). The position of the patient can be corrected by adjusting the actual marker position to the planned marker position, which has been transferred from the three-dimensional treatment planning system and superimposed on the fluoroscopic image on the display unit of the RTRT system. The system recognizes the position of a 2.0-mm gold marker in the human body 30 times/s using two X-ray television systems. The position of the markers can be visualized during RT and after treatment delivery to verify the accuracy of the localization. Setup of the target volume in solid organs has been shown to be improved with the use of three markers and the RTRT system (8, 12, 14 –16). However, the accuracy of the setup in lung tissue with three markers has not been reported. In the present study, we investigated the fixation rate of markers placed using the bronchial insertion technique, the reliability of the setup using markers around the target volume, and the dislocation of the markers after RTRT during the follow-up period. The long-term toxicity of the inserted markers in patients with lung cancer was also investigated. METHODS AND MATERIALS Details on the technique for inserting gold markers with a diameter of 1.5 mm into the lung have been previously reported (11, 12). In brief, special equipment for the insertion of gold markers through bronchoscopy was developed and used to insert the markers into small bronchi with a diameter of ⱕ1.5 mm (Olympus, Tokyo, Japan). Insertion of the markers during the fiberscopic examination took 20 –30 min for each patient. This technique was used for peripheral lung tumors ⬍6 cm in diameter in patients scheduled to undergo hypofractionated high-dose radiation using RTRT. The patients in this study were selected from those who were enrolled in a prospective feasibility study (12) or dose-finding study (unpublished data). Between July 2000 and April 2004, 154 markers were inserted in 57 patients (42 men, 15 women, average age 70.1 years, standard deviation, 10.4; range, 43– 86 years), with peripheral lung cancer. Of the 57 patients, 51 had primary lung cancer (adenocarcinoma in 26, squamous cell carcinoma in 19, unclassified non– small-cell lung cancer in 1, and unknown in 5) and 6 had metastatic lung cancer (lung in 1, head and neck in 4, and malignant melanoma in 1). The Eastern Cooperative Oncology Group performance status of the 57 patients enrolled in this study was Grade 1 in 39, Grade 2 in 9, Grade 3 in 1, Grade 4 in 0, and unknown in 8. In the initial 2 years of the study, only one marker was implanted. Beginning in 2002, two or more markers were implanted in patients who were able to tolerate the procedure to account for
● M. IMURA et al.
1443
migration, rotation, and deformation. One, two, three, four, five, six, seven, and nine markers were inserted into the peripheral bronchial small lumen in 22, 6, 12, 10, 3, 1, 2, and 1 patient, respectively. The patients who received more than five markers had two tumors and were treated using RTRT for both. Computed tomography for planning was performed on the same day or up to 5 days after the insertion. The patients underwent CT by a multidetector CT scanner at the end of the expiration phase in normal breathing with a slice thickness of 1 mm. RT was begun 6.4 ⫾ 3.1 days (mean ⫾ SD) after marker insertion. As we have previously noted, judging whether the marker is fixed to the same location in moving soft organs such as lung parenchyma is difficult (12, 13). Thus, in the present study, we measured the distances between the implanted markers (DIMs) using the RTRT system and compared them with the DIMs measured in CT planning. If the discrepancy in any of the DIM measurements was ⬎2 mm, we determined the dislocation of the marker to be beyond the tolerance level, and reexamination of the marker position using CT was indicated before RT to recalculate the distance between the marker and the isocenter. The new data on the relationship between the marker and the isocenter were again transferred to the RTRT system, and the patient was repositioned using the new data. The treatment period from the initial RT session to the last was 4 –14 days (median, 6 days). If the markers dropped off the lung after insertion and before CT planning, the patients were treated with RTRT using the remaining markers. If no marker was left at the start of RTRT, conventional setup techniques were applied using orthogonal portal images at the isocenter (17, 18). In the present study, the following two measurements were used for analysis: 1. Measurement of the three-dimensional distance between two markers, DIM, using two sets of orthogonal diagnostic X-ray images of the RTRT system to judge the intermarker dislocation. The measurement was performed during the setup of the patient in each treatment session. Eleven patients who agreed to the use of their data for research were enrolled in this part of the study. No more than one measurement for each DIM in the same day was used for the analysis. Because a 1-mm slice thickness was used in the CT examination, the uncertainty in the measurement was not better than 0.5 mm. The diameter of the fiducial marker was 1.5 mm; thus, the uncertainty in the measurement of the location of the marker was about 1–2 mm at best. 2. Measurement of the distance between the marker and the chest wall (DMC) in the transaxial CT image to analyze the relationship between the location (peripheral or central) of the marker and the fixation rate. The DMC was selected for analysis because endoscopists routinely measure the DMC using fluoroscopic guidance to prevent pneumothorax during the insertion of markers. In the present study, the distance was also measured to judge the dislocation of the marker relative to the normal structure, the chest wall, in the follow-up CT scans. Patients underwent CT at planning and at follow-up after treatment with the same multidetector CT scanner, with a slice thickness of ⱕ5 mm, every 1–2 months during the initial year and then every 3– 6 months afterward. Forty-eight patients were enrolled in this part of the study, and the data of the distance were used for the analysis. To determine the accuracy of the DMC measurements, we selected 12 benign pulmonary nodules in 9 patients who were examined more than twice. Because they had not been treated for, or had, any other
1444
I. J. Radiation Oncology
● Biology ● Physics
Table 1. Number of fiducial markers at insertion, CT planning, at start of RTRT, at end of RTRT, and at last follow-up of patient
Volume 63, Number 5, 2005
Table 2. Relationship between location of marker and fixation rate of marker at last follow-up Right lung lobes*
Days after insertion*
Insertion Detected Detected Detected Detected
performed at CT planning at start of RTRT at end of RTRT at last follow-up
n
Minimum
Maximum
154 122 117 115 104
— 0 6 6 16
— 5 6 15 181
Abbreviations: CT ⫽ computed tomography; RTRT ⫽ real-time tumor tracking radiotherapy. * Minimal and maximal number of elapsed days after insertion.
pulmonary disease during the period between the two CT scans, the measurement of the distance between the nodule and the chest wall was regarded as the physically accurate DMC measurement. The discrepancy was 0 ⫾ 1 mm (mean ⫾ SD) within the follow-up period of 294 ⫾ 190 days. However, the CT scans for these 9 patients and for the 48 patients in the following study were performed at the end of inspiration, and not at the end of expiration, which may have influenced the accuracy of the measurements. Moreover, because the CT slice thickness was 5 mm, the uncertainty in the DMC measurement in the transaxial image was estimated to be not better than 3 mm. The following statistical analyses were used in this study. The Kruskal-Wallis test was used for the relationship analysis between the doctor’s experience of insertion and the success rate of the fixation of implanted markers, as well as for the relationship analysis between the DMC at insertion and the eventual fixation rate. The Mann-Whitney U test was used to compare the DMC between the fixed markers and the dropped markers. The Wilcoxon signed-rank test was used to compare the DMC before and after RTRT. The chi-square test for independence or Fisher’s exact probability test was used to compare the fixation rate among the different lobes. The Kaplan-Meier method was used to calculate the fixation rate of the markers after the date of insertion.
Fixation Dropped Total
Upper
Middle/lower
Total
26 9 35
22 6 28
48 15 63
Left lung lobes†
Fixation Dropped Total
Upper
Lower
28 26 54
28 9 37
56 35 91
* p ⫽ 0.5 (Fisher’s exact probability test). p ⫽ 0.04 (chi-square test for independence).
†
fixation rate was 68%. The other remaining 50 markers were defined as “dropped” markers in the present study. A significant difference in the fixation rate was noted among the positions of the markers. Markers in the left upper lobe had a lower fixation rate than those in the left lower lobe (p ⫽ 0.04, chi-square test for independence; Table 2). No difference was observed between the right upper lobe, right middle/lower lobe, and left lower lobe. A Kaplan-Meier curve was plotted for the fixation rate of the markers from the date of insertion (Fig. 1), according to which lobe the markers were inserted into. The curve of the fixation rate in the left upper lobe was significantly lower than that of the others by the log–rank test (p ⫽ 0.01). A statistically significant difference was found in the fixation rate of the markers among markers with different DMCs at insertion and before RT. A shorter DMC correlated with a better fixation rate (p ⫽ 0.01, Kruskal-Wallis test; Table 3). The average ⫾ SD of the DMC was 17 ⫾ 12 mm in the fixed markers and 24 ⫾ 12 mm in the dropped ones (p ⫽
RESULTS Table 1 shows the performance of the fixation technique of the gold markers. Of the 154 markers, 122 (79%) were detected at CT planning, which was performed 0 –5 days after insertion (median, 2 days). Of 122 markers seen on CT, 117 (96%) were ready to be used at the start of RTRT (or 76% of the 154 inserted markers). Of the 117 markers, 115 (98%) were detected throughout the treatment period (median 10 days, range, 6 –15 days). The marker was detected at the last follow-up date (range, 16 –181 days; median, 44 days) after insertion in 104 (68%) of the 154 implanted markers. Thus, 5.3 markers/d were dropped within the initial 6 days, 0.7 marker/d in the following 6 –15 days, and 0.07 marker/d in the following 16 –181 days. When we defined “the fixation rate” as the ratio of the fiducial markers detected on CT or X-ray images at the end of the follow-up period to all 154 markers inserted, the
Fig. 1. Kaplan-Meier curve of fixation rate of markers from date of insertion according to location of each marker.
Fiducial markers in RT
● M. IMURA et al.
1445
Table 3. Relationship between distance from marker to chest wall and the fixation rate of marker at last follow-up Distance between marker and chest wall (mm)
Fixation (n ⫽ 104) Dropped (n ⫽ 17*)
ⱕ10
⬎10–20
⬎20
40 1
33 7
31 9
p ⫽ 0.01 (Kruskal-Wallis test). * Of 154 markers, 122 were detected at CT planning, performed 0 –5 days after insertion. Afterward, 18 markers were dropped; 1 patient experienced drop of marker before X-ray examination. We were able to investigate the distances between the marker and chest wall in 17 of those 18 dropped markers.
0.01, Mann-Whitney U test). Table 4 shows the relationship between the number of times the doctor had performed insertions and the success rate of the fixation of the marker at the last follow-up date. These data suggest the possibility of a significant difference among doctors who had performed insertions ⬍20, 20 – 49, and ⱖ50 times and the fixation rate at the last follow-up date. The fixation rate at the last follow-up date was 58%, 64%, and 75% for doctors who had performed insertions ⬍20, 20 – 49, and ⱖ50 times, respectively (p ⫽ 0.05, Kruskal-Wallis test). Figure 2 shows the DIM, plotted according to the days after the start of RTRT in 71 treatment setups in 11 patients. Of the 198 measurements, 189 (95%) were within 2 mm, and 158 (80%) were within ⫾1 mm. In 9 (5%) of 198 measurements in 71 setups, the discrepancy was ⬎2 mm. In 6 (8%) of 71 setups in which the discrepancy was ⬎2 mm, a CT scan for recalculation of the marker position relative to the isocenter was indicated. The discrepancy was ⱕ2 mm in the remaining 198 measurements in 64 (90%) of 71 setups in 11 patients. Figure 3 shows the change in DMC according to the days after the start of RTRT in 186 measurements in 93 fixed markers that were measurable on CT in 48 patients. Eleven markers were not measurable by CT because the follow-up on these markers was conducted at other institutes. Large dislocation of the markers, often ⬎5–15 mm from the original position, was noted during the follow-up period in 93 markers. A significant difference was found in the DMC before (mean ⫾ SD 16 ⫾ 11 mm) and after (18 ⫾ 12 mm) RTRT in these
Fig. 2. Discrepancy in distance between markers and days passed after start of real-time tumor-tracking radiotherapy (RTRT). Open circles indicate discrepancy for which computed tomography scan was required to recalculate position of marker relative to isocenter.
93 markers (p ⫽ 0.0002, Wilcoxon signed-rank test). The average ⫾ SD of the discrepancy from the original DMC to the DMC at the last follow-up date was 1 ⫾ 4 mm. This deviation was significantly greater than the deviation in the measurement of benign pulmonary nodules (0 ⫾ 1 mm, as described above; p ⫽ 0.02, Mann-Whitney U test). All the patients tolerated the procedure. Only 1 patient experienced pneumothorax after insertion of the marker, which resolved within 1 week merely by bed rest. Most of the dropped markers were detected on the abdominal X-ray in the intestinal tract. Several patients coughed the markers up. None of these patients suffered any symptomatic adverse effects from the dropped markers, although most were disappointed that the markers had fallen out. DISCUSSION Markers dropped within the first week after insertion at a high rate (38 markers, 76% of total dropped markers) and at a much lower rate 1 week after insertion (12 markers, 24%; p ⬍0.001, Mann-Whitney U test).
Table 4. Relationship between number of times doctors had performed fixation and fixation rate of markers Doctor’s experience (n)
Fixation Dropped Total Fixation rate
⬍20
20–50
⬎50
Total
25 18 43 0.58
27 15 42 0.64
52 17 69 0.75
104 50 154 —
p ⫽ 0.05 (Kruskal-Wallis test).
Fig. 3. Changes in distance between chest wall and marker in transaxial computed tomography slice showing marker, during days after start of real-time tumor-tracking radiotherapy (RTRT).
1446
I. J. Radiation Oncology
● Biology ● Physics
The diameter of the bronchial lumina is most likely grossly related to the DMC, because the diameter of the bronchial lumina is large at the central part of the lung, where the DMC is long, and is small at the periphery of the lung, where the DMC is short. The statistical correlation between the fixation rate and the DMC suggests that the small diameter of a bronchial lumen is an important factor for the success of marker fixation. Markers may simply fix in the bronchial lumen, which has a diameter ⬍1.5 mm, and drop if the bronchial lumen is enlarged after forced insertion of the marker. In addition, there may be some biologic reaction that fixes the marker. Our speculation is that the biologic reaction to the insertion of the marker consists of an inflammatory change secondary to local traumatic injury of the bronchial mucosal surface and surrounding normal tissue. After insertion of the marker, we speculate that an edematous reaction occurs and that subsequent fibrotic changes around the marker occur, causing the marker to be fixed in the pulmonary parenchyma or stenotic lumen of the bronchus. In an autopsy case, a fiducial marker had been implanted for a lung tumor 3 months before the patient’s death. The gold marker was trapped in the bronchial tree, which was surrounded by fibrotic tissue that developed as result of high-dose radiation. The safety of the procedure for implanting fiducial markers with bronchofiberscopy was reconfirmed (11, 12). An apparent learning curve on fixation of the marker and a lower fixation rate at the left upper lobe were suggested. These markers are pushed by a device through the bronchial lumen, which means that the anatomic distribution of the bronchial tree influences the difficulty of the procedure. When the target lesions are in the upper lobe or superior segment, both of which have sharply curved bronchial trees, endoscopists experience difficulty pushing the marker in correctly. Our data revealed that the left upper lobe is the most difficult portion in which to fix the markers. A relationship between fixation rate and the DMC was also demonstrated. On the basis of these results, we concluded that markers should be inserted as close to the peripheral parietal pleura as possible. This idea is consistent with the anatomic fact that the peripheral bronchi consist of tapering and curving air lumens with smaller diameters than the central bronchi. Thus, training endoscopists would be an important factor in increasing the success rate of fixation and the safety of the procedure. We have produced a free videostreaming website for endoscopists, illustrating the technical aspects of the insertion procedure (19). This material will help new endoscopists increase the fixation rate of the marker within a shorter learning period. After about 2 weeks, apparent dislocation of the markers from the planned position in the CT measurements occurred. These results contradicted the results of the measurement of the DIM, which showed less dislocation even after 2 weeks. This discrepancy between the DMC on CT and the fluoroscopic measurement of the distance between the three markers suggests that the change in the DMC is at least partly due to the dislocation of lung tissue from the
Volume 63, Number 5, 2005
original position, rather than migration of the markers. This speculation does not contradict our clinical experience showing that lung tissue moves during the treatment period as a result of shrinkage of the tumor mass and radiation pneumonitis. Recently, Ebe et al. (20), at Yamaguchi University Hospital, developed a new system consisting of an RTRT system with CT on a rail in the treatment room. Their treatment machine provides a quick check of the dislocation between the marker and the tumor. The operator can immediately correct the coordinates of the marker to the isocenter before each treatment. If the dislocation is due to shrinkage of the tumor, adaptive RT, changing the shape of the field may be indicated, although this procedure is beyond our capability. The DIM was ⱕ2 mm in 95% of the measurements, and 1 mm or better in 80%. Provided that plural markers do not move the same distance in the same direction, these values can be regarded as the translational setup accuracy of our system for lung cancer. In conjunction with the use of CT verification of the marker position relative to the isocenter of the target volume, three markers and two fluoroscopic systems can provide excellent set-up accuracy for patients with peripheral lung cancer. In practice, having to perform reexamination by CT in a different room is quite inconvenient for both patient and staff; thus, using both an RTRT system and CT on a rail is probably the best solution for 4D RT of lung cancer. In summary, 1.5-mm gold markers inserted using bronchoscopy near lung tumors were detected throughout RTRT in 75% (115 of 154) of inserted markers and in 94% (115 of 122) of markers used in CT planning. The distance between the two markers was within ⫾2 mm in 95% of measurements and ⫾1 mm in 80% during RTRT. The distance was ⬎2 mm at least in one direction in 5% of the setups for which reexamination by CT was indicated. A statistically significant relationship was found between a shorter DMC and the fixation rate, suggesting that markers in smaller bronchial lumina fixed better than those in larger lumina. A learning curve among endoscopists was also suggested to affect the fixation rate of the markers. A large change in the DMC 2 weeks after insertion suggested dislocation of the tumor and surrounding lung tissue took place owing to shrinkage or deformation of the tumor mass by RT. CONCLUSION The relationship between fiducial markers and gross tumor volume can change during RT owing to changes in the volume and location of the tumor mass, as well as possible migration of the marker. Nevertheless, in combination with CT measurement for recalculation between the marker and tumor volume in ⬍10% of the setups, implantation of three markers using our technique was useful for setup, with an accuracy of ⫾2 mm. For RT lasting ⬎2 weeks, three gold markers and the RTRT system would play an additional role in realizing individually based, precise, adaptive RT for moving, shrinking, and deforming lung cancer.
Fiducial markers in RT
● M. IMURA et al.
1447
REFERENCES 1. Keall PJ, Starkschall G, Shukla H, et al. Acquiring 4D thoracic CT scans using a multislice helical method. Phys Med Biol 2004;49:2053–2067. 2. Mageras GS, Pevsner A, Yorke ED, et al. Measurement of lung tumor motion using respiration-correlated CT. Int J Radiat Oncol Biol Phys 2004;60:933–941. 3. Berbeco RI, Jiang SB, Sharp GC, et al. Integrated radiotherapy imaging system (IRIS): Design considerations of tumour tracking with LINAC gantry-mounted diagnostic x-ray systems with flat-panel detectors. Phys Med Biol 2004;49:243– 255. 4. Shimizu S, Shirato H, Ogura S, et al. Detection of lung tumor movement in real-time tumor-tracking radiotherapy. Int J Radiat Oncol Biol Phys 2001;51:304 –310. 5. Shirato H, Shimizu S, Kitamura K, et al. Four-dimensional treatment planning and fluoroscopic real-time tumor tracking radiotherapy for moving tumor. Int J Radiat Oncol Biol Phys 2000;48:435– 442. 6. Shirato H, Seppenwoolde Y, Kitamura K, et al. Intrafractional tumor motion: Lung and liver. Semin Radiat Oncol 2004;14: 10 –18. 7. Seppenwoolde Y, Shirato H, Kitamura K, et al. Precise and real-time measurement of 3D tumor motion in lung due to breathing and heartbeat, measured during radiotherapy. Int J Radiat Oncol Biol Phys 2002;53:822– 834. 8. Shirato H, Shimizu S, Kunieda T, et al. Physical aspects of a real-time tumor-tracking system for gated radiotherapy. Int J Radiat Oncol Biol Phys 2000;48:1187–1195. 9. Whyte RI, Crownover R, Murphy MJ, et al. Stereotactic radiosurgery for lung tumors: Preliminary report of a phase I trial. Ann Thorac Surg 2003;75:1097–1101. 10. Shinagawa N, Yamazaki K, Onodera Y, et al. CT-guided transbronchial biopsy using an ultrathin bronchoscope with virtual bronchoscopic navigation. Chest 2004;125:1138 – 1143. 11. Harada T, Shirato H, Ogura S, et al. Real-time tumor-tracking radiation therapy for lung carcinoma by the aid of insertion of a gold marker using bronchofiberscopy. Cancer 2002;95: 1720 –1727.
12. Shirato H, Harada T, Harabayashi T, et al. Feasibility of insertion/implantation of 2.0-mm-diameter gold internal fiducial markers for precise setup and real-time tumor tracking in radiotherapy. Int J Radiat Oncol Biol Phys 2003;56:240 –247. 13. Kitamura K, Shirato H, Shimizu S, et al. Registration accuracy and possible migration of internal fiducial gold marker implanted in prostate and liver treated with real-time tumortracking radiation therapy (RTRT). Radiother Oncol 2002;62: 275–281. 14. Onimaru R, Shirato H, Aoyama H, et al. Calculation of rotational setup error using the real-time tracking radiation therapy (RTRT) system and its application to the treatment of spinal schwannoma. Int J Radiat Oncol Biol Phys 2002;54: 939 –947. 15. Yamamoto R, Yonesaka A, Nishioka S, et al. High dose three-dimensional conformal boost (3DCB) using an orthogonal diagnostic X-ray set-up for patients with gynecological malignancy: A new application of real-time tumor-tracking system. Radiother Oncol 2004;73:219 –222. 16. Shirato H, Oita M, Fujita K, et al. Three-dimensional conformal setup (3D-CSU) of patients using the coordinate system provided by three internal fiducial markers and two orthogonal diagnostic X-ray systems in the treatment room. Int J Radiat Oncol Biol Phys 2004;60:607– 612. 17. Fukumoto S, Shirato H, Shimzu S, et al. Small-volume imageguided radiotherapy using hypofractionated, coplanar, and noncoplanar multiple fields for patients with inoperable Stage I nonsmall cell lung carcinomas. Cancer 2002;95:1546 –1553. 18. Onimaru R, Shirato H, Shimizu S, et al. Tolerance of organs at risk in small-volume, hypofractionated, image-guided radiotherapy for primary and metastatic lung cancers. Int J Radiat Oncol Biol Phys 2003;56:126 –135. 19. Real-time tumor-tracking radiotherapy video streaming site. http://hokudai.systemk.co.jp/. 20. Ebe K, Shirato H, Hiyama A, et al. Integration of fluoroscopic real-time tumor-tracking system and tomographic scanner on the rail in the treatment room [Abstract]. Int J Radiat Oncol Biol Phys 2004;60(Suppl.):S604.