- Email: [email protected]
Practical Assessment of Bronchoscopically Inserted Fiducial Markers for Image Guidance in Stereotactic Lung Radiotherapy
BRIEF REPORT
Practical Assessment of Bronchoscopically Inserted Fiducial Markers for Image Guidance in Stereotactic Lung Radiotherapy Price Jackson, PhD,a,* Daniel P. Steinfort, PhD,b,c,d Tomas Kron, PhD,e,f Shankar Siva, PhDe,f a
Department of Physical Science, Peter MacCallum Cancer Centre, East Melbourne, Australia Department of Respiratory Medicine, Royal Melbourne Hospital, Parkville, Australia c Department of Cancer Medicine, Peter MacCallum Cancer Centre, East Melbourne, Australia d Department of Medicine, University of Melbourne, Parkville, Australia e Division of Radiation Oncology and Cancer Imaging, 3002, Peter MacCallum Cancer Centre, East Melbourne, Australia f Sir Peter MacCallum Department of Oncology, University of Melbourne, Parkville, Australia b
Received 28 December 2015; revised 21 March 2016; accepted 2 April 2016 Available online - 26 April 2016
ABSTRACT Introduction: Stereotactic radiotherapy is a high-dose precision technique necessitating accurate target visualization through either cone beam computed tomography (CBCT) or planar imaging with implanted fiducial markers. We have investigated the properties for image guidance using fiducial markers implanted through minimally invasive bronchoscopy. Methods: Two fiducial marker types were implanted endobronchially in 10 patients undergoing radical radiation treatment for non–small cell lung cancer (eight using Visicoil linear fiducial markers [IBA Dosimetry GmbH, Schwarzenbruck Germany] and two using superDimension and superLock two-band markers [Covidien Inc., Minneapolis, MN]). Patients underwent four-dimensional computed tomography imaging for treatment planning and after completion of treatment to investigate marker movement. As part of the image guidance assessment, megavolt electronic portal images (EPIs) were acquired in addition to kilovolt planar and CBCT (Varian Medical Systems, Palo Alto, CA) images. Results: In two of 10 patients (both receiving Visicoil markers), marker migration was observed before treatment. In patients with stable markers, both types were clearly visible in planar kilovolt imaging; however, in EPIs the markers could be detected only in selected beam directions in which bony interference was minimal. Diagnostic computed tomography scanning was able to demonstrate the markers with clarity, but significant starring artifacts were observed in CBCT. This was particularly problematic in patients with some lateral component of tumor motion during breathing.
Conclusions: The potential for fiducial migration must be considered and investigated if bronchoscopic implantation of fiducial markers is performed. The choice of marker is a compromise between trying to minimize CBCT artifacts while enabling visualization in EPI imaging, which is an ideal tool to verify gated radiotherapy delivery. Crown Copyright 2016 Published by Elsevier Inc. on behalf of International Association for the Study of Lung Cancer. All rights reserved. Keywords: Fiducial; Endobronchial; Image guidance; SBRT; SABR; EBUS
Introduction Image guidance of radiotherapy in lung tumors is challenging because of both respiratory and cardiac induced motion. These motion-induced uncertainties result in an increased risk for geographic tumor miss and consequent potential of reduced tumor control. Fiducial markers have been investigated both as a
*Corresponding author. Disclosure: The authors declare no conflict of interest. Address for correspondence: Price Jackson, PhD, Department of Physical Science, Peter MacCallum Cancer Centre, 2 St. Andrews Pl., East Melbourne 3002, Australia. E-mail: [email protected] Crown Copyright ª 2016 Published by Elsevier Inc. on behalf of International Association for the Study of Lung Cancer. All rights reserved. ISSN: 1556-0864 http://dx.doi.org/10.1016/j.jtho.2016.04.016
Journal of Thoracic Oncology
Vol.
-
No.
-: ---
2 Jackson et al
surrogate of tumor position and to inform tumor motion in this context. However, fiducial markers are typically inserted using percutaneous techniques that are associated with the potential for significant toxicities. These include a pneumothorax rate that ranges from 13% to more than 60%.1–3 Subsequent hospital admission and intercostal drain tube insertion to manage this complication is required in 3% to 44% of cases,1,3–5 whereas this severity of complication is not reported in any studies describing endobronchial fiducial insertion.6–9 Bronchoscopy offers an appealing alternative to percutaneous fiducial marker insertion given its superior safety profile. Our group has previously reported bronchoscopic implantation of gold fiducials using radial probe endobronchial ultrasound with virtual bronchoscopy and fluoroscopic or electromagnetic navigation guidance to achieve tumor localization and placement within/adjacent to peripheral lung tumors. Importantly, no procedural complications were sustained.7 In this study, we investigated the utility of bronchoscopically implanted fiducial markers for radiotherapy image guidance. In particular, we focused on the imaging implications of the composition of the fiducials, impact of geometric alignment with the tumor, and the capacity to visualize these fiducials using onboard imaging techniques.
Materials and Methods This study was designed to assess the feasibility of improved image guidance by noninvasive implantation of fiducial markers in patients receiving radical radiotherapy for lung cancer.
Fiducials In this ethics board–approved prospective study two types of fiducial markers were implanted endobronchially in 10 patients undergoing radical radiation treatment for non–small cell lung cancer. Eight patients received a single Visicoil (IBA Dosimetry GmbH, Schwarzenbruck Germany) linear fiducial 10 0.75-mm marker. Two were implanted with superDimension and superLock (Covidien Inc., Minneapolis, MN) two-band 13 0.9-mm markers (one receiving two markers to assess volumetric definition capabilities). The bronchoscopic implantation was performed under conscious sedation as previously described10 using radial probe endobronchial ultrasound11,12 and fluoroscopic guidance to achieve tumor localization and placement within or adjacent to peripheral lung tumors.
Radiographic Imaging Patients had a time-resolved, 10-phase fourdimensional (4D) computed tomography (CT) scan (using the Philips Brilliance Big Bore system with
Journal of Thoracic Oncology
Vol.
-
No.
-
bellows [Philips Healthcare, DA Best, the Netherlands]) for treatment planning and after completion of treatment to investigate marker movement. On the 4D series, mean breathing rate and motion range in each orthogonal plane were recorded using fiducial location. One patient’s therapy was planned with only a three-dimensional positron emission tomography/CT image series with no time-resolved imaging. Onboard planar kilovoltage images were acquired for lateral and anterior-posterior views during patient setup for treatment. Cone beam CT (CBCT) scans were recorded for each fraction for stereotactic ablative body radiotherapy and used to validate patient positioning. During treatment delivery cine electronic portal imaging (EPI) series were acquired for selected gantry angles in six of the patients.
Results Fiducial Placement Fiducial markers were implanted as previously described7 in all 10 patients without complication. In two of the eight patients implanted with Visicoil markers, migration or complete loss was observed in the period between insertion and first fraction of treatment delivery (identified on imaging at days 10 and 13 after placement, respectively). This time line of displacement is consistent with previous reports identifying migration during or immediately after insertion.5 No positional change was observable in the two patients receiving superDimension-type fiducials. In those with stable fiducial markers, good correspondence between positioning on planning CT and treatment CBCT was noted (see Fig. 1). All patients with stable markers at the time of treatment had persistent fiducial location on followup (6–18 months after radiotherapy).
Radiographic Imaging Markers were clearly visualized on all 10 phases of the planning 4D CT image series. The averaged threedimensional volume (as used for planning) displayed the effect of motion primarily along the inferior-superior axis. Imaging with 4D CT included assessment of patient motion ranges in each orthogonal plane. Both fiducial types were clearly visualized in each of the 10 gated breathing phases. Tumor motion ranged from 2 to 16 mm and was predominantly along the superior-inferior axis (Table 1). Median breathing rate was 16 breaths per minute (range 12–26). Visualization of fiducial markers was not influenced by breathing rate either on 4D CT or with onboard imaging. Both marker types displayed characteristic metal streak artifacts on CBCT imaging. When combined with the effects of motion averaging during the cone beam acquisition, 2 to 3 mm of spatial uncertainty was
---
2016
Fiducial Markers in Lung Radiotherapy
3
Figure 1. Coregistered computed tomography (CT) (top) and cone beam CT (bottom) images for a patient with a right upper lobe lesion. Note the severity of streak/star artifact on cone beam CT.
introduced in attempting to align based solely on fiducial location. This was most pronounced in patients with some component of lateral breathing motion (notably in patient 4), and in these instances the presence of gold markers offered no improvement over alignment using soft-tissue boundaries. The effects of lateral motion effectively resulted in two separate starring foci owing to the bias for dwell time near the limits of inhalation and exhalation. Multiple intersecting star patterns could confound the perception of fiducial at the location of the inhale/exhale limits or at a nearby intersection point. Both marker types were clearly visible with planar kilovoltage imaging. Even when overlaid with osseous
structures, the in-plane location was resolved with submillimeter accuracy (Fig. 2). Fiducial markers exhibited limited utility in megavolt portal imaging (Fig. 3). Visualization was restricted primarily to anterior-posterior fields and was hindered if overlaid with osseous structures such as sternum, ribs, or spine. In patients with some component of lateral motion, CBCT star artifacts introduced considerable uncertainty in the true in-plane position. Figure 4B shows a patient with 5 mm of motion in the axial plane. As there is a bias for breathing duration in the maximum inhale and exhale phases, the reconstructed view appears to represent two separate star patterns. It is not evident whether tumor position is best described by either one
Table 1. Summary of Marker Types, Breathing Rate, and Motion Ranges as Assessed by 4D CT (mm) Motion Range (mm) Patient No. 1 2 3 4 5 6 7 8 9 10 a
Marker Type
Mean Breathing Rate (breaths/min) Lung
Lobe
Superior-Inferior
Anterior-Posterior Left-Right
Total
Superdimension Superdimension Visicoil Visicoil Visicoila Visicoil Visicoilb Visicoil Visicoil Visicoilb
26 14 12 13 — 16 — 21 22 —
Lower Upper Upper Upper Upper Upper Upper Upper Middle Lower
15 3 2 7 — 2 — 4 1 —
5 0 1 2 — 2 — 3 1 —
15.8 3.0 3.0 8.8 N/A 3.5 N/A 5.0 1.7 N/A
Left Right Right Right Right Right Left Right Right Right
No planning four-dimensional computed tomography scan was available for patient 5. Marker loss or migration was observed in patients 7 and 10. 4D, four dimensional; CT, computed tomography; No., number; N/A, not applicable. b
0 0 2 5 — 2 — 0 1 —
4 Jackson et al
Journal of Thoracic Oncology
Vol.
-
No.
-
Figure 2. Implanted fiducials were clearly resolved using onboard kilovoltage imaging for image-guided radiotherapy.
of the extents of the artifact cross pattern or some central intersection point.
Discussion Both types of markers are well visualized in all lowenergy radiographic modalities. In planning CT images and with onboard kilovoltage exposures, the margins of marker are clearly defined and relatively free of motion. Portal imaging during treatment displayed limited utility, even when acquired in cine mode.13 Fiducials could be visualized only when beam path was not obscured by other high-contrast features, generally, in instances in which the tumor margin was natively visible without the aid of implanted material. Orientation of the fiducials potentially affects their appearance. A marker aligned
with the superior-inferior axis should result in less lateral positional uncertainty in patients with normal respiratory motion. The characteristic distribution of upper lobe tumors is noted in this study. Compared with lower lobe markers, upper lobe fiducial motion artifact on CBCT is expected to be less severe and in EPI imaging there is a reduced likelihood of obfuscation by mediastinal soft tissue. The type of motion has demonstrated an effect on the ability to clearly resolve the fiducial position. A lateral component of respiratory motion may result in multiple star artifacts for a single marker for CBCT. The true extents of motion can produce two separate foci that when combined, can indicate the appearance of one or more potential marker locations outside of the axis of motion.
Figure 3. Digitally reconstructed radiograph (A) and the associated electronic portal image obtained during treatment (B). This gantry angle (320 ) in this case provided best visibility.
---
2016
Fiducial Markers in Lung Radiotherapy
5
Figure 4. Star artifact for patient with motion predominantly along the superior-inferior axis (A) and for a patient with some component of lateral motion (B). In the worst-case scenario (a patient with a small lesion and lateral respiratory motion), the implanted fiducial offers very limited information to improve image-guided localization.
The spatial uncertainty in localizing fiducial markers is most pronounced in CBCT. In this modality, the combination of respiratory motion and metallic streak artifact are shown to be problematic. In some instances, the severity of the artifact alone is less troublesome than the positioning relative to structures of interest. Artifacts may obscure the edge of a target structure. Even if the marker is well localized, the potential for fiducial migration over small distances suggests that positional accuracy should be confirmed before treatment. The insertion of multiple markers may enable confirmation of stable implantation on kilovolt imaging by quickly assessing their relative spatial alignment. In cases exhibiting lateral motion due either to lateral compression of the lung during breathing or to markers implanted at an angle experiencing motion along the superior-inferior axis, a triangulated crossing of streak artifacts can indicate an apparent third marker location (Fig. 4). One solution to reduce artifacts that affect tumor visualization is to position fiducial markers in planes just above or below the tumor, thereby enabling localization of the adjacent tissue without obscuring the appearance of low-contrast features that would indicate the true tumor margin. With continued development in adaptive reconstruction algorithms for CBCT, the appearance of metal and beam hardening artifacts will be reduced considerably.
Conclusion In stereotactic ablative body radiotherapy we expect extremely sharp dose drop-off outside of the target
volume and the small margins for patient setup and tumor position error that are typically on the order of 5 mm or less. On the basis of the observed localization precision with on-board imaging techniques, fiducial marker insertion in stereotactic ablative body radiotherapy may be effectively used as a surrogate of tumor position and facilitates the use of gated radiotherapy. The potential for fiducial migration must be considered and investigated if bronchoscopic implantation of fiducial markers is performed. The choice of marker is a compromise between trying to minimize CBCT artifacts and enabling visualization in EPI imaging, which is an ideal tool to verify gated radiotherapy delivery.
References 1. Yousefi S, Collins BT, Reichner CA, et al. Complications of thoracic computed tomography–guided fiducial placement for the purpose of stereotactic body radiation therapy. Clin Lung Cancer. 2007;8:252–256. 2. Bhagat N, Fidelman N, Durack JC, et al. Complications associated with the percutaneous insertion of fiducial markers in the thorax. Cardiovasc Intervent Radiol. 2010;33:1186–1191. 3. Ohta K, Shimohira M, Iwata H, et al. Percutaneous fiducial marker placement under CT fluoroscopic guidance for stereotactic body radiotherapy of the lung: an initial experience. J Radiat Res. 2013;54:957–961. 4. 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. 5. Hong JC, Yu Y, Rao AK, et al. High retention and safety of percutaneously implanted endovascular embolization coils as fiducial markers for image-guided stereotactic ablative radiotherapy of pulmonary tumors. Int J Radiat Oncol Biol Phys. 2011;81:85–90.
6 Jackson et al 6. Nabavizadeh N, Zhang J, Elliott DA, et al. Electromagnetic navigational bronchoscopy-guided fiducial markers for lung stereotactic body radiation therapy: analysis of safety, feasibility, and interfraction stability. J Bronchology Interv Pulmonol. 2014;21:123–130. 7. Steinfort DP, Siva S, Kron T, et al. Multimodality guidance for accurate bronchoscopic insertion of fiducial markers. J Thorac Oncol. 2015;10:324–330. 8. Kupelian PA, Willoughby TR, Reddy CA, Klein EA, Mahadevan A. Hypofractionated intensity-modulated radiotherapy (70 Gy at 2.5 Gy per fraction) for localized prostate cancer: Cleveland Clinic experience. Int J Radiat Oncol Biol Phys. 2007;68:1424–1430. 9. Anantham D, Feller-Kopman D, Shanmugham LN, et al. Electromagnetic navigation bronchoscopy-guided fiducial placement for robotic stereotactic radiosurgery of lung tumors: a feasibility study. Chest. 2007;132: 930–935.
Journal of Thoracic Oncology
Vol.
-
No.
-
10. Steinfort DP, Irving LB. Patient satisfaction during endobronchial ultrasound-guided transbronchial needle aspiration performed under conscious sedation. Respir Care. 2010;55:702–706. 11. Steinfort DP, Vincent J, Heinze S, Antippa P, Irving LB. Comparative effectiveness of radial probe endobronchial ultrasound versus CT-guided needle biopsy for evaluation of peripheral pulmonary lesions: a randomized pragmatic trial. Respir Med. 2011;105: 1704–1711. 12. Steinfort DP, Bonney A, See K, Irving LB. Sequential multimodality bronchoscopic investigation of peripheral pulmonary lesions. Eur Respir J. 2015;47: 607–614. 13. Kron T, Lee C, Perera F, Yu E. Evaluation of intra-and inter-fraction motion in breast radiotherapy using electronic portal cine imaging. Technol Cancer Res Treat. 2004;3:443–449.
Practical Assessment of Bronchoscopically Inserted Fiducial Markers for Image Guidance in Stereotactic Lung Radiotherapy Price Jackson, PhD,a,* Daniel P. Steinfort, PhD,b,c,d Tomas Kron, PhD,e,f Shankar Siva, PhDe,f a
Department of Physical Science, Peter MacCallum Cancer Centre, East Melbourne, Australia Department of Respiratory Medicine, Royal Melbourne Hospital, Parkville, Australia c Department of Cancer Medicine, Peter MacCallum Cancer Centre, East Melbourne, Australia d Department of Medicine, University of Melbourne, Parkville, Australia e Division of Radiation Oncology and Cancer Imaging, 3002, Peter MacCallum Cancer Centre, East Melbourne, Australia f Sir Peter MacCallum Department of Oncology, University of Melbourne, Parkville, Australia b
Received 28 December 2015; revised 21 March 2016; accepted 2 April 2016 Available online - 26 April 2016
ABSTRACT Introduction: Stereotactic radiotherapy is a high-dose precision technique necessitating accurate target visualization through either cone beam computed tomography (CBCT) or planar imaging with implanted fiducial markers. We have investigated the properties for image guidance using fiducial markers implanted through minimally invasive bronchoscopy. Methods: Two fiducial marker types were implanted endobronchially in 10 patients undergoing radical radiation treatment for non–small cell lung cancer (eight using Visicoil linear fiducial markers [IBA Dosimetry GmbH, Schwarzenbruck Germany] and two using superDimension and superLock two-band markers [Covidien Inc., Minneapolis, MN]). Patients underwent four-dimensional computed tomography imaging for treatment planning and after completion of treatment to investigate marker movement. As part of the image guidance assessment, megavolt electronic portal images (EPIs) were acquired in addition to kilovolt planar and CBCT (Varian Medical Systems, Palo Alto, CA) images. Results: In two of 10 patients (both receiving Visicoil markers), marker migration was observed before treatment. In patients with stable markers, both types were clearly visible in planar kilovolt imaging; however, in EPIs the markers could be detected only in selected beam directions in which bony interference was minimal. Diagnostic computed tomography scanning was able to demonstrate the markers with clarity, but significant starring artifacts were observed in CBCT. This was particularly problematic in patients with some lateral component of tumor motion during breathing.
Conclusions: The potential for fiducial migration must be considered and investigated if bronchoscopic implantation of fiducial markers is performed. The choice of marker is a compromise between trying to minimize CBCT artifacts while enabling visualization in EPI imaging, which is an ideal tool to verify gated radiotherapy delivery. Crown Copyright 2016 Published by Elsevier Inc. on behalf of International Association for the Study of Lung Cancer. All rights reserved. Keywords: Fiducial; Endobronchial; Image guidance; SBRT; SABR; EBUS
Introduction Image guidance of radiotherapy in lung tumors is challenging because of both respiratory and cardiac induced motion. These motion-induced uncertainties result in an increased risk for geographic tumor miss and consequent potential of reduced tumor control. Fiducial markers have been investigated both as a
*Corresponding author. Disclosure: The authors declare no conflict of interest. Address for correspondence: Price Jackson, PhD, Department of Physical Science, Peter MacCallum Cancer Centre, 2 St. Andrews Pl., East Melbourne 3002, Australia. E-mail: [email protected] Crown Copyright ª 2016 Published by Elsevier Inc. on behalf of International Association for the Study of Lung Cancer. All rights reserved. ISSN: 1556-0864 http://dx.doi.org/10.1016/j.jtho.2016.04.016
Journal of Thoracic Oncology
Vol.
-
No.
-: ---
2 Jackson et al
surrogate of tumor position and to inform tumor motion in this context. However, fiducial markers are typically inserted using percutaneous techniques that are associated with the potential for significant toxicities. These include a pneumothorax rate that ranges from 13% to more than 60%.1–3 Subsequent hospital admission and intercostal drain tube insertion to manage this complication is required in 3% to 44% of cases,1,3–5 whereas this severity of complication is not reported in any studies describing endobronchial fiducial insertion.6–9 Bronchoscopy offers an appealing alternative to percutaneous fiducial marker insertion given its superior safety profile. Our group has previously reported bronchoscopic implantation of gold fiducials using radial probe endobronchial ultrasound with virtual bronchoscopy and fluoroscopic or electromagnetic navigation guidance to achieve tumor localization and placement within/adjacent to peripheral lung tumors. Importantly, no procedural complications were sustained.7 In this study, we investigated the utility of bronchoscopically implanted fiducial markers for radiotherapy image guidance. In particular, we focused on the imaging implications of the composition of the fiducials, impact of geometric alignment with the tumor, and the capacity to visualize these fiducials using onboard imaging techniques.
Materials and Methods This study was designed to assess the feasibility of improved image guidance by noninvasive implantation of fiducial markers in patients receiving radical radiotherapy for lung cancer.
Fiducials In this ethics board–approved prospective study two types of fiducial markers were implanted endobronchially in 10 patients undergoing radical radiation treatment for non–small cell lung cancer. Eight patients received a single Visicoil (IBA Dosimetry GmbH, Schwarzenbruck Germany) linear fiducial 10 0.75-mm marker. Two were implanted with superDimension and superLock (Covidien Inc., Minneapolis, MN) two-band 13 0.9-mm markers (one receiving two markers to assess volumetric definition capabilities). The bronchoscopic implantation was performed under conscious sedation as previously described10 using radial probe endobronchial ultrasound11,12 and fluoroscopic guidance to achieve tumor localization and placement within or adjacent to peripheral lung tumors.
Radiographic Imaging Patients had a time-resolved, 10-phase fourdimensional (4D) computed tomography (CT) scan (using the Philips Brilliance Big Bore system with
Journal of Thoracic Oncology
Vol.
-
No.
-
bellows [Philips Healthcare, DA Best, the Netherlands]) for treatment planning and after completion of treatment to investigate marker movement. On the 4D series, mean breathing rate and motion range in each orthogonal plane were recorded using fiducial location. One patient’s therapy was planned with only a three-dimensional positron emission tomography/CT image series with no time-resolved imaging. Onboard planar kilovoltage images were acquired for lateral and anterior-posterior views during patient setup for treatment. Cone beam CT (CBCT) scans were recorded for each fraction for stereotactic ablative body radiotherapy and used to validate patient positioning. During treatment delivery cine electronic portal imaging (EPI) series were acquired for selected gantry angles in six of the patients.
Results Fiducial Placement Fiducial markers were implanted as previously described7 in all 10 patients without complication. In two of the eight patients implanted with Visicoil markers, migration or complete loss was observed in the period between insertion and first fraction of treatment delivery (identified on imaging at days 10 and 13 after placement, respectively). This time line of displacement is consistent with previous reports identifying migration during or immediately after insertion.5 No positional change was observable in the two patients receiving superDimension-type fiducials. In those with stable fiducial markers, good correspondence between positioning on planning CT and treatment CBCT was noted (see Fig. 1). All patients with stable markers at the time of treatment had persistent fiducial location on followup (6–18 months after radiotherapy).
Radiographic Imaging Markers were clearly visualized on all 10 phases of the planning 4D CT image series. The averaged threedimensional volume (as used for planning) displayed the effect of motion primarily along the inferior-superior axis. Imaging with 4D CT included assessment of patient motion ranges in each orthogonal plane. Both fiducial types were clearly visualized in each of the 10 gated breathing phases. Tumor motion ranged from 2 to 16 mm and was predominantly along the superior-inferior axis (Table 1). Median breathing rate was 16 breaths per minute (range 12–26). Visualization of fiducial markers was not influenced by breathing rate either on 4D CT or with onboard imaging. Both marker types displayed characteristic metal streak artifacts on CBCT imaging. When combined with the effects of motion averaging during the cone beam acquisition, 2 to 3 mm of spatial uncertainty was
---
2016
Fiducial Markers in Lung Radiotherapy
3
Figure 1. Coregistered computed tomography (CT) (top) and cone beam CT (bottom) images for a patient with a right upper lobe lesion. Note the severity of streak/star artifact on cone beam CT.
introduced in attempting to align based solely on fiducial location. This was most pronounced in patients with some component of lateral breathing motion (notably in patient 4), and in these instances the presence of gold markers offered no improvement over alignment using soft-tissue boundaries. The effects of lateral motion effectively resulted in two separate starring foci owing to the bias for dwell time near the limits of inhalation and exhalation. Multiple intersecting star patterns could confound the perception of fiducial at the location of the inhale/exhale limits or at a nearby intersection point. Both marker types were clearly visible with planar kilovoltage imaging. Even when overlaid with osseous
structures, the in-plane location was resolved with submillimeter accuracy (Fig. 2). Fiducial markers exhibited limited utility in megavolt portal imaging (Fig. 3). Visualization was restricted primarily to anterior-posterior fields and was hindered if overlaid with osseous structures such as sternum, ribs, or spine. In patients with some component of lateral motion, CBCT star artifacts introduced considerable uncertainty in the true in-plane position. Figure 4B shows a patient with 5 mm of motion in the axial plane. As there is a bias for breathing duration in the maximum inhale and exhale phases, the reconstructed view appears to represent two separate star patterns. It is not evident whether tumor position is best described by either one
Table 1. Summary of Marker Types, Breathing Rate, and Motion Ranges as Assessed by 4D CT (mm) Motion Range (mm) Patient No. 1 2 3 4 5 6 7 8 9 10 a
Marker Type
Mean Breathing Rate (breaths/min) Lung
Lobe
Superior-Inferior
Anterior-Posterior Left-Right
Total
Superdimension Superdimension Visicoil Visicoil Visicoila Visicoil Visicoilb Visicoil Visicoil Visicoilb
26 14 12 13 — 16 — 21 22 —
Lower Upper Upper Upper Upper Upper Upper Upper Middle Lower
15 3 2 7 — 2 — 4 1 —
5 0 1 2 — 2 — 3 1 —
15.8 3.0 3.0 8.8 N/A 3.5 N/A 5.0 1.7 N/A
Left Right Right Right Right Right Left Right Right Right
No planning four-dimensional computed tomography scan was available for patient 5. Marker loss or migration was observed in patients 7 and 10. 4D, four dimensional; CT, computed tomography; No., number; N/A, not applicable. b
0 0 2 5 — 2 — 0 1 —
4 Jackson et al
Journal of Thoracic Oncology
Vol.
-
No.
-
Figure 2. Implanted fiducials were clearly resolved using onboard kilovoltage imaging for image-guided radiotherapy.
of the extents of the artifact cross pattern or some central intersection point.
Discussion Both types of markers are well visualized in all lowenergy radiographic modalities. In planning CT images and with onboard kilovoltage exposures, the margins of marker are clearly defined and relatively free of motion. Portal imaging during treatment displayed limited utility, even when acquired in cine mode.13 Fiducials could be visualized only when beam path was not obscured by other high-contrast features, generally, in instances in which the tumor margin was natively visible without the aid of implanted material. Orientation of the fiducials potentially affects their appearance. A marker aligned
with the superior-inferior axis should result in less lateral positional uncertainty in patients with normal respiratory motion. The characteristic distribution of upper lobe tumors is noted in this study. Compared with lower lobe markers, upper lobe fiducial motion artifact on CBCT is expected to be less severe and in EPI imaging there is a reduced likelihood of obfuscation by mediastinal soft tissue. The type of motion has demonstrated an effect on the ability to clearly resolve the fiducial position. A lateral component of respiratory motion may result in multiple star artifacts for a single marker for CBCT. The true extents of motion can produce two separate foci that when combined, can indicate the appearance of one or more potential marker locations outside of the axis of motion.
Figure 3. Digitally reconstructed radiograph (A) and the associated electronic portal image obtained during treatment (B). This gantry angle (320 ) in this case provided best visibility.
---
2016
Fiducial Markers in Lung Radiotherapy
5
Figure 4. Star artifact for patient with motion predominantly along the superior-inferior axis (A) and for a patient with some component of lateral motion (B). In the worst-case scenario (a patient with a small lesion and lateral respiratory motion), the implanted fiducial offers very limited information to improve image-guided localization.
The spatial uncertainty in localizing fiducial markers is most pronounced in CBCT. In this modality, the combination of respiratory motion and metallic streak artifact are shown to be problematic. In some instances, the severity of the artifact alone is less troublesome than the positioning relative to structures of interest. Artifacts may obscure the edge of a target structure. Even if the marker is well localized, the potential for fiducial migration over small distances suggests that positional accuracy should be confirmed before treatment. The insertion of multiple markers may enable confirmation of stable implantation on kilovolt imaging by quickly assessing their relative spatial alignment. In cases exhibiting lateral motion due either to lateral compression of the lung during breathing or to markers implanted at an angle experiencing motion along the superior-inferior axis, a triangulated crossing of streak artifacts can indicate an apparent third marker location (Fig. 4). One solution to reduce artifacts that affect tumor visualization is to position fiducial markers in planes just above or below the tumor, thereby enabling localization of the adjacent tissue without obscuring the appearance of low-contrast features that would indicate the true tumor margin. With continued development in adaptive reconstruction algorithms for CBCT, the appearance of metal and beam hardening artifacts will be reduced considerably.
Conclusion In stereotactic ablative body radiotherapy we expect extremely sharp dose drop-off outside of the target
volume and the small margins for patient setup and tumor position error that are typically on the order of 5 mm or less. On the basis of the observed localization precision with on-board imaging techniques, fiducial marker insertion in stereotactic ablative body radiotherapy may be effectively used as a surrogate of tumor position and facilitates the use of gated radiotherapy. The potential for fiducial migration must be considered and investigated if bronchoscopic implantation of fiducial markers is performed. The choice of marker is a compromise between trying to minimize CBCT artifacts and enabling visualization in EPI imaging, which is an ideal tool to verify gated radiotherapy delivery.
References 1. Yousefi S, Collins BT, Reichner CA, et al. Complications of thoracic computed tomography–guided fiducial placement for the purpose of stereotactic body radiation therapy. Clin Lung Cancer. 2007;8:252–256. 2. Bhagat N, Fidelman N, Durack JC, et al. Complications associated with the percutaneous insertion of fiducial markers in the thorax. Cardiovasc Intervent Radiol. 2010;33:1186–1191. 3. Ohta K, Shimohira M, Iwata H, et al. Percutaneous fiducial marker placement under CT fluoroscopic guidance for stereotactic body radiotherapy of the lung: an initial experience. J Radiat Res. 2013;54:957–961. 4. 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. 5. Hong JC, Yu Y, Rao AK, et al. High retention and safety of percutaneously implanted endovascular embolization coils as fiducial markers for image-guided stereotactic ablative radiotherapy of pulmonary tumors. Int J Radiat Oncol Biol Phys. 2011;81:85–90.
6 Jackson et al 6. Nabavizadeh N, Zhang J, Elliott DA, et al. Electromagnetic navigational bronchoscopy-guided fiducial markers for lung stereotactic body radiation therapy: analysis of safety, feasibility, and interfraction stability. J Bronchology Interv Pulmonol. 2014;21:123–130. 7. Steinfort DP, Siva S, Kron T, et al. Multimodality guidance for accurate bronchoscopic insertion of fiducial markers. J Thorac Oncol. 2015;10:324–330. 8. Kupelian PA, Willoughby TR, Reddy CA, Klein EA, Mahadevan A. Hypofractionated intensity-modulated radiotherapy (70 Gy at 2.5 Gy per fraction) for localized prostate cancer: Cleveland Clinic experience. Int J Radiat Oncol Biol Phys. 2007;68:1424–1430. 9. Anantham D, Feller-Kopman D, Shanmugham LN, et al. Electromagnetic navigation bronchoscopy-guided fiducial placement for robotic stereotactic radiosurgery of lung tumors: a feasibility study. Chest. 2007;132: 930–935.
Journal of Thoracic Oncology
Vol.
-
No.
-
10. Steinfort DP, Irving LB. Patient satisfaction during endobronchial ultrasound-guided transbronchial needle aspiration performed under conscious sedation. Respir Care. 2010;55:702–706. 11. Steinfort DP, Vincent J, Heinze S, Antippa P, Irving LB. Comparative effectiveness of radial probe endobronchial ultrasound versus CT-guided needle biopsy for evaluation of peripheral pulmonary lesions: a randomized pragmatic trial. Respir Med. 2011;105: 1704–1711. 12. Steinfort DP, Bonney A, See K, Irving LB. Sequential multimodality bronchoscopic investigation of peripheral pulmonary lesions. Eur Respir J. 2015;47: 607–614. 13. Kron T, Lee C, Perera F, Yu E. Evaluation of intra-and inter-fraction motion in breast radiotherapy using electronic portal cine imaging. Technol Cancer Res Treat. 2004;3:443–449.