Abstract
During fractionated radiotherapy of tumors, especially those located in the head and neck regions, the patient must be accurately and reproducibly positioned which requires use of immobilization devices. In this retrospective study, a comparison between two commercially available immobilization devices (“BrainLab” and “Orfit”) for the treatment of head and neck lesions using “ExacTrac” localization system was done. The data were extracted and analyzed from 28 patients that underwent 5 or more fractions of external beam radiation therapy (EBRT) for head-and-neck tumors. Two groups of the patients (14 for each group) were created for comparison purposes. The study looked at how reliable these two masks were at reproducing the patient position based on the amount of shift in different directions one will have over the other. Group A patients had been planned and treated using “BrainLab” immobilization device while group B patients had been planned and treated using “Orfit” immobilization mask. Results showed that the overall patient displacement during treatment was significantly lower by “BrainLab” device (mean = 0.625, sd = 0.16) compared to “Orfit” device (mean = 0.987, sd = 0.136), p = 0.0023, paired t-test. However, the control of the displacements was significant in the lateral, yaw, roll and pitch dimensions (p< 0.05), and not in the vertical and longitudinal dimensions (p> 0.05) paired t-test. Isocenter offsets (displacements) for patients treatments using BrainLAB immobilization device was of narrower range (appeared less scattered on a line graph) compared to isocenter offsets from patients treated using the Orfit immobilization device, and the overall isocenter offsets (displacements) were more to the positive (+) dimensions for the Orfit device compared to those of the BrainLAB device. The “BrainLab” device therefore significantly reproduces the treatment set up compared to the “Orfit” immobilization device.
Introduction
It is important to accurately and reproducibly position a patient during fractional radiotherapy of tumors especially those located in the head and neck regions. This is particularly important because treatment setup inaccuracies can result into irradiation of non-target healthy tissues and organs (e.g. eyes, salivary glands, optic chiasm, and brain stem) which can lead to treatment failure, morbidity and increased risks to patients (Bentel et al., 1997; Gilbean et al., 2001; Rotundo et al., 2008).
Immobilization devices such as thermoplastic face masks have been used during radiotherapeutic treatments of brain and head and neck tumors to reduce setup inaccuracies (Bentel et al., 1995, 1997; Niewald et al., 1988; Weltens et al., 1995) but with limited successes because other parameters including local infrastructure (table couch, laser beams for alignments, linac) and even the medical, nursing and technical staff can influence the accuracy and reproducibility of patient positioning during radiotherapy (Gibbeau et al., 2001), which calls for stringent quality control guidelines (Mitina et al., 1991) or use of computer controlled repositioning systems (Rotondo et al., 2008) with advanced image acquisition softwares.
Currently, BrainLAB software capabilities can allow patients’ treatment planning and radiation delivery to be completed in a fraction of the time while exponentially increasing delivery accuracy compared to earlier possibilities in radiotherapy and treatments o cancer with a better treatment outcome. The BrainLABs advanced software solutions, i.e. the ExacTrac system, utilize diagnostic images and innovative planning and targeting tools (Ahlswede et al., 2001; Bin et al., 2007; Clark et al., 2002; Langen et al., 2005; Verllen et al., 2003; Willner et al., 1997; Wurn et al., 2008) combined with comprehensive quality assurance standards which result in enormous increase in accuracy and efficiency of patient during radiotherapy resulting in millimeter accuracy for tumor targeting in extracranial locations. BrainLAB offers a versatile and complete portfolio of stereotactic immobilization solutions to ensure maximum safety and precision during treatment. Also, BrainLAB’s systems are compatible with all major linear accelerators and stereotactic headings from other manufacturers.
“The BrainLABs ExacTrac’s high-resolution x-ray imaging module features clinically-implemented Image Guided Targeting as an upgrade for existing Linacs and the high quality imaging of internal structures and organs results in an extremely accurate set-up of the target volume – and overcomes the limitations of conventional patient positioning methods, which are based on external skin markers. During patient set-up, the x-ray imaging system calculates the discrepancy between the actual and the planned target positions with sub-millimeter accuracy. These differences are then compensated for with automatic table movement. The fast and streamlined ExacTrac X-Ray procedure can be carried out every day, providing millimeter accuracy for patient set-up within standard 15-minute treatment intervals” (BrainLAB Oncology Solutions, IGRT With ExacTracR.
In this study, “BrainLab” immobilization device which incorporates “ExacTrac” system capable of calculating any required patient shift and the remote coach aligns the patient to the correct isocenter point, as compared with the “Orfit®” immobilization masks which have in use for a longer time.
Materials and Methods
Study design, patients’ population, and immobilization devices
This was a retrospective study which was designed to compare two immobilization devices (masks) that were used in the planning and treatment of patients that underwent external beam radiotherapy (EBRT) for head-and–neck tumors. Data collection for the study was done at Huntsman Cancer Hospital located in salt Lake City, Utah. The study looked at how reliable the two immobilization devices (masks) were in reproducing the patients’ position based on the amount of shift or displacements of the patients’ treatment setup.
The sampling was taken from the treatment records of patients who were treated using the “BrainLab” or the “Orfit” masks. The BrainLab immobilization device incorporated the image-guided verification ExacTrac® system, BrainLab Inc. which calculates any required patient shift and the remote couch aligns the patient to the correct isocenter or point, while the “Orfit” immobilization device did not.
The data was extracted from 28 previously diagnosed head-and-neck tumor patients that underwent 5 or more fractions of EBRT for head and tumors at the Stereostatic Radiation Surgery (SRC) room of the hospital. Two groups of patients were created, 14 in each group A or B. Group A patients had been planned and treated using the BrainLab immobilization device with ExacTrac® system while group B patients had been planned and treated using “Orfit” immobilization device (mask).
Regarding ethical concerns, permission to obtain data from the hospital was granted from the clinical institution before data extraction. Confidentiality and anonymity of the patients were maintained. Specifically, no names, birth data, medical record numbers, social security numbers or form of information linking a patient to the study were included in the retrospective comparison and evaluation of patient information. Only group data was used for summarizing the results of the study. Since this was a retrospective study, there were therefore no physical risks to the patients and only data pertinent to the study was collected.
Treatment set-up and image acquisition
As detailed in the BrainLAB brochure:
“The patients are first pre-positioned on the treatment couch based on conventional skin marks and wall lasers, or with the ExacTrac Body Markers. ExacTrac X-Ray automatically moves the patient’s target volume into the beam-path of the first x-ray source and the first x-ray image is taken. A further x-ray is taken after the target volume has been automatically moved into the beam-path of the second x-ray source. The entire procedure is guided by the ExacTrac X-Ray software from outside the treatment room. The precise stereoscopic imaging of the target area allows users to pinpoint the anatomical target seconds before radiation is delivered. The software uses the initial CT scan, which was also used during planning, to automatically calculate Digitally Reconstructed Radiographs (DRR) that show the desired set-up. An automatic algorithm fuses each x-ray image to its corresponding DRR. The software then calculates the table movement necessary to optimally align the patient’s target volume with the Linac isocenter in longitudinal, lateral, vertical and other directions or dimensions. This process ensures that the set-up, as defined in the CT-based treatment plan, is aligned with high-quality x-rays immediately before treatment. Automatic fusion is possible both for implanted markers, as well as for bony structures. Should a positioning error be detected, ExacTrac X-Ray automatically calculates the couch movement necessary for accurate repositioning of the patient. ExacTrac X-Ray will then automatically and quickly move the patient to the correct treatment position as defined by the image fusion procedure. As this procedure is completely automated, and there is no need for the user to enter the treatment room”.
As explained in the article by Weltens et al. (1995), with the “Orfit” technique, “the patient is first placed in an adequate irradiation position on the simulator and an optimal head support used. Lasers (indicating the field center) and the light field are drawn on the mask during simulation and are used for daily positioning on the treatment couch. Orfit masks are made of a thermoplastic material, about 1.5 mm thick, and are molded directly on the patient’s face, just before simulation. On the simulator, the patient is prepared by a nursing aid who puts a protection stocking over the patient’s face. Subsequently two nursing aids warm up the thermoplastic material in a water heater. Then the material is molded immediately on the patient’s face by two nursing aids and one medical doctor or a treatment technician. To assess the reproducibility of the treated volume, repetitive portal images are taken on the patients. Distances between anatomical structures and field edges are measured and the mean distances are calculated for each patient. The deviations from these mean distances are then calculated to assess the day-to-day variability in the patients’ position”.
Statistical analysis
Reproducibility of the treatment set-up was assessed by calculating the deviations from the mean value for the repeated treatments ranging from 6 to 25 for each patient (Weltens et al., 1995) using either the “BrainLab” or “Orfit” device. Five different displacement (shift) positions or dimensions were considered: vertical, longitudinal, lateral, yaw, roll and pitch. Mean and standard deviations of the displacements (shifts) in the various five dimensions were recorded and paired t-test were used to compare the means and assess the significance (with p< 0.05 considered statistically significantly).
Results
“BrainLab” performed better in controlling patients displacements (shifts) compared to the “Orfit” immobilization device (Figure 2). “BrainLab” device significantly controlled more of the lateral (p= 0.018), yaw (p= 0.009), roll (p= 0.02) and pitch (p= 0.011) displacement but not the vertical (p= 0.508) and longitudinal (p= 0.196) displacements compared to “Orfit” device as statistically determined by paired t-test.
Considering the mean displacements in all the 5 dimensions combined, “BrainLab” performed better (mean= 0.625, sd= 0.16) compared to “Orfit” (mean = 0.987, sd = 0.136), Table 1. These differences were significant by the paired t-test p=0.0023).
Isocenter offsets for patients treatments using BrainLAB immobilization device also appeared less scattered on a line graph (Figure 3) compared to offsets treated using the Orfit immobilization device (Figure 4). The overall isocenter offsets (displacements) were more to the positive (+) dimensions for the Orfit device compared to those of the BrainLAB device (Figure 5).
Discussion
This study has found that the “BrainLab” immobilization device performs better in reproducing patients positions during EBRT treatment of neck-and-head tumors compared to the “Orfit” immobilization device. The overall patient displacement by “BrainLab” was lower (mean = 0.625, sd= 0.16) compared to “Orfit” (mean = 0.987, sd = 0.136). However, all these displacements are still lower (>1.00 mm) and within patients safety margins of less than 2mm standard deviations (Gibbean et al., 2001; Rotondo et al., 2008). This study also found that the isocenter offsets for patients treatments using BrainLAB immobilization device were of a narrow range (appeared less scattered on a line graph) compared to isocenter offsets for patients treated using the Orfit immobilization device, and the overall isocenter offsets (displacements) were more to the positive (+) dimensions for the Orfit device compared to those of the BrainLAB device the reasons for which needs interpretation or further studies.
In this study, “BrainLab” device significantly controlled the lateral, yaw, roll and pitch shifts and not the vertical and longitudinal shifts compared to “Orfit”. This is consistent within previous studies which have also reported treatment set-up reproducibility to differ at different levels and directions of patient movements (Gilbeau et al., 2001; Bentel et al., 1997; Rotondo et al., 2008). These data are similar or even slightly better than those from published studies which reported a total displacement in the range of 2±5 mm (1 SD) (Bel et al., 1995; Hess et al.,1995; Mitine et al., 1991; Welters et al., 1995; Willner et al., 1997). These studies only included data from head and neck patients and comparison of the results from this study with published series on thermoplastic masks needs to be done carefully. The earlier studies assessed the setup accuracy in a single region and used manual detection of displacements on X-ray ® films (El-Gayed et al., 1995; Hunt et al., 1993; Mitine et al., 1991; Rabinovitz et al., 1985; Welters et al., 1995; Willner et al., 1997). In two other studies which compared portal films with digitized simulator films using image matching software, total displacements in the order of 2.1±2.6 mm (1SD) were reported for the three main directions (Bel et al.,1995; Yan et al., 1997).
Conclusions
“BrainLab” device significantly reproduces the treatment set-up compared to the “Orfit” immobilization device, especially the lateral, yaw, roll and pitch shifts and not the vertical and longitudinal shifts.
References
Ahlswede, J., et. al. (2001). “Improvement of Patient Positioning in a Thermoplastic Mask using a Bite Block” Medical Physics 28 (6), 1200, June 2001, SU-FF-EXH C-42.
Bel, A., Keus, R., Vijibrie, R. E., Lebesque, J. V. (1995). Setup deviations in wedged pair irradiation of parotid gland and tonsillar tumors, measured with an electronic portal imaging device. Radiother Oncol, 37:153-159.
Bentel, G. C., Marks, L. B., Sherouse, G. W., Spencer, D. P. (1995). A customized head and neck support system. Int J Radiat Oncol Biol Phys, 32, 245-248.
Bentel, G. C., Marks, L. B., Hendren, K., Brizel, D. (1997). Comparison of two head and neck immobilization systems. Int J Radiat Oncol Biol Phy; 38, 867-873.
Bin, S. Teh et.al. (2007). Versatility of the Novalis® System to Deliver Image-Guided Stereotactic Body Radiation Therapy (SBRT) for Various Anatomical Sites; Technology in Cancer Research and Treatment, Vol. 6, No. 4.
BrainLAB Image Guided Targeting. 2010.
BrainLAB Oncology Solutions, IGRT With ExacTracR 2010. Web.
Clark, B., et. al.(2002). “Immobilization in Stereotactic Radiotherapy: Comparison of Thermoplastic Mask Systems” Med. Phys. 29(6), 1369, TH-C-517B-10.
El-Gayed, A. A. H., Bel, A., Vijlbrief, R., Bartelink, H., Lebesque, J. (1993). Time
trend of patient setup deviations during pelvic irradiation using electronic portal imaging. Radiother Oncol, 26: 162-171.
Gilbeau, L., Octave-Prignot, M., Loncol, T., et al. (2001). Comparison of setup accuracy of three different thermoplastic masks for the treatment of brain and head and neck tumors. Radiother Oncol, 58: 155–162.
Hess, C. F., Kortmann, R. D., Jany, R., Hamberger, A., Bamberg, M. (1995). Accuracy of ®eld alignment in radiotherapy of head and neck cancer utilizing face mask immobilization: a retrospective analysis of clinical practice. Radiother Oncol, 34: 69-72.
Huizenga, H., Levendag, P.C., De Porre, P. M. Z. R., Visser, A. G. (1988). Accuracy in radiation field alignment in head and neck cancer: a prospective study.
Radiother Oncol,11:181-187.
Hunt, M., Kutcher, G., Burman, C., et al. (1993). The effect of uncertainties on the treatment of nasopharynx cancer. Int J Radiat Oncol Biol Phys, 27: 437-447.
Langen, M. et al. (2005). Initial Experience with Mv Ct Guidance for Daily Prostate Alignments; IJROBP, Vol. 62, (5): 1517-1524.
Mitine, C., Leunens, G., Verstraete, J., et al. (1991). Is it necessary to repeat quality control procedures for head and neck patients? Radiother Oncol, 21: 201-210.
Niewald, M., Lehmann, W., Uhlmann, U., Schnabel, K., Leetz, H. (1988). Plastic material used to optimize radiotherapy of head and neck tumors and the mammary carcinoma. Radiother Oncol,11: 55-63.
Rabinovitz, I., Broomberg, J., Goitein, M., McCarthy, K., Leong, J. (1985). Accuracy of radiation field alignment in clinical practice. Int J Radiat Oncol Biol Phys, 11:1857-1867.
Rotondo, R. L., Sultanem, K., Lavoie, I., Skelly, J., Raymond, L. (2008). Comparison of repositioning accuracy of two commercially available immobilization systems for treatment of head-and-neck tumors using simulation computed tomography imaging. Montreal, Quebec, Canada Int. J. Radiation Oncology Biol. Phys, 70: (5),1389-1396.
Verllen, D. et. al. (2003). AQ of a system for improved target localization and patient setup; RT Onc, Vol. 67, pp. 129-141.
Weltens, C., Kestloot, K., Vandevelde, G., VandenBogaert, W. (1995). Comparison of plastic and Orfit® masks for patient head fixation during radiotherapy: precision and costs. Int J Radiat Oncol Biol Phys, 33, 499-507.
Willner, J., Hadinger, U., Neumann, M., Schawb, F. J., Bratengeier, K., Flentje, M. (1997). Three dimensional variability in patient positioning using bite block immobilization in 3D-conformal radiation treatment for ENT-tumors. Radiother Oncol, 43: 315-321.
Willner, J. et. al. (1997). “CT simulation in stereotactic brain radiotherapy – analysis of isocenter reproducibility with mask fixation”, Radiotherapy and Oncology 45: 83-88.
Wurm, R. et al. (2008). Novalis® non-invasive frameless Image-Guided Radiosurgery: Initial Experience; Neurosurgery, Vol. 62, No. 6 (Suppl).
Yan, D., Wong, J., Vicini, F., et al. (1997). Adaptive modification of treatment planning to minimize the deleterious effects of treatment setup errors. Int J Radiat Onc Biol Phys, 38, 197-206.