Fiducial marker for prostate radiotherapy: comparison of 0.35‑ and 0.5‑mm‑diameter computed tomography and magnetic resonance images Osamu Tanaka1 &middot

全文

(1)Author's personal copy Radiol med DOI 10.1007/s11547-016-0715-5. MAGNETIC RESONANCE IMAGING. Fiducial marker for prostate radiotherapy: comparison of 0.35‑ and 0.5‑mm‑diameter computed tomography and magnetic resonance images Osamu Tanaka1 · Hisao Komeda2 · Takayoshi Iida1 · Masayoshi Tamaki2 · Kensaku Seike2 · Daiki Kato2 · Shigeki Hirose1 · Daisuke Kawaguchi1 · Takamasa Yokoyama1. Received: 17 October 2016 / Accepted: 27 November 2016 © Italian Society of Medical Radiology 2016. Abstract Purpose When performing intensity-modulated radiotherapy for prostate cancer, a marker is inserted into the prostate to enable the recognition of its position using cone– beam computed tomography (CT). However, it is difficult to recognize the prostatic outline using CT alone. Magnetic resonance imaging (MRI) can depict the prostatic outline better than CT. In treatment plans using CT and MRI registration, various markers are used in institutions; however, the selection of an optimal marker size is difficult. Comparison of a different fiducial marker study was conducted using phantom, but no study in vivo was found. Therefore, we prospectively investigated the effects of different marker diameter sizes using CT and MR images. Methods Thirty-one consecutive patients were enrolled in this study. CT and MRI were performed 3 weeks after marker placement. The 0.35-mm-diameter marker was placed on the left side of the prostate, and the 0.5-mmdiameter marker was placed on the right side. The length of each marker was 10 mm. The better MRI image was selected between those obtained using T2*-two-dimensional weighted image (T2*2D) and T2*-three-dimensional weighted image (T2*3D). Two observers evaluated and scored the prostatic outline image quality as well as visualized the prostatic markers using CT and MRI. Results MRI was significantly superior to CT in depicting the prostatic outline. The CT artifacts were significantly. * Osamu Tanaka [email protected] 1. Department of Radiation Oncology, Gifu Municipal Hospital, 7‑1 Kashima‑cho, Gifu, Gifu 500‑8513, Japan. 2. Department of Urology, Gifu Municipal Hospital, 7‑1 Kashima‑cho, Gifu, Gifu 500‑8513, Japan. lesser for the 0.35-mm-diameter marker than for the 0.5-mm-diameter marker. The degree of marker recognition using MRI was significantly better with the 0.5-mm-diameter marker. Conclusion The 0.5-mm-diameter fiducial marker had significantly better visualization than the 0.35-mm-diameter marker. While CT artifacts were significantly worse with the 0.5-mm-diameter marker, the artifact level was tolerable for clinical practice. Therefore, we recommend the 0.5-mm-diameter diameter marker in terms of prostatic outline and marker visualization using MRI. Keywords Prostate radiotherapy · Fiducial marker · Marker size · Magnetic resonance imaging. Introduction Recently, the use of intensity-modulated radiotherapy (IMRT) for prostate cancer has gradually increased [1–4]. IMRT is less toxic than conventional three-dimensional conformal radiotherapy (3D-CRT). A fiducial gold marker (GM) is essential for increasing radiation dosage concentricity to the prostate and reducing dosage to the surrounding normal tissue. Real-time tracking can minimize the complications of image-guided radiotherapy (IGRT), because the position of the prostate moves with the body. However, detection of the prostatic outline is difficult using computed tomography (CT) alone; therefore, magnetic resonance imaging (MRI) is used [5–7]. CT and MRI registration is conducted by GM guidance. However, GM sizes vary. Comparison with a phantom study was conducted, but no comparison study was found for the same human body [8]. Therefore, we evaluated the visualization of different GM sizes using CT and MRI.. 13.

(2) Author's personal copy Radiol med. Materials and methods This clinical trial was approved by the internal review board, and the national University Hospital Information Network (UMIN) clinical trial registration number was 19402. Thirty-one patients were enrolled in this study from October 2015 to May 2016. All patients provided written informed consent. GM insertion was as follows: 0.5-mmdiameter and 10-mm-long GM was inserted on the right side of the prostate and 0.35-mm-diameter and 10-mm-long GM was inserted on the left side. Nineteen-gauge needles were used. GMs were VISICOIL (RadioMed Corporation, Bartlett, TN, USA). CT/MRI registration was performed 3 weeks after GM implantation. Image acquisition The patients drank 200 ml of water 30 min before CT (Optima CT580; GE Medical Systems, Milwaukee, WI, USA) and MRI (Intera 1.5 Nova; Philips Medical Systems, Eindhoven, The Netherlands) and underwent urine collection. MRI was performed within 20 min after CT. All the patients were given butylscopolamine to stop bowel movements. MRI was performed with 3-mm-thick sections, no intersection gaps, and a 16-cm field of view using a cardiac coil. • T1-weighted imaging (T1-WI): T1-weighted spin-echo. Repetition time (TR)/echo time (TE) range in milliseconds: 400–650/8, number of averages (NA): 4, number of phase-encoding steps (PESs): 192, number of frequency-encoding steps (FESs): 240, typical spatial resolutions (TPRs) of frequency/phase: 0.67/0.83. • T2-weighted imaging (T2-WI): T2-weighted fast spinecho. TR/TE: 4000/80, NA: 4, PESs: 205, FESs: 256, TPRs of frequency/phase: 0.63/0.80. • T2* two-dimensional-weighted imaging (T2*-2D): T2*weighted gradient echo. TR/TE: 700/18, NA: 2, PESs: 205, FESs: 256, TPRs of frequency/phase: 0.63/0.78. • T2* three-dimensional-weighted imaging (T2*3D): T2*-3D-weighted gradient echo. TR/TE1/delta TE: 37/14/7.3, NA: 2, PESs: 218, FESs: 272, TPRs of frequency/phase: 0.55/0.54. • Contrast-enhanced T1-weighted imaging (contrastenhanced T1-WI): contrast-enhanced T1-weighted spin-echo. TR/TE: range 400–650/8, NA: 4, PESs: 192, FESs: 240, TPRs of frequency/phase: 0.67/0.83. Among these five sequences, we evaluated the images obtained using T2-WI, T2*2D, and T2*3D, because these demonstrated the best visualization. Primarily, the degree of artifacts was examined using CT, and marker visualization was performed using MRI.. 13. The radiotherapy instrument at our hospital is a Novalis Tx system (Varian Medical Systems, Inc., Palo Alto, CA, USA). Before we began the present study, the 0.35mm × 10-mm and the 0.5-mm × 10-mm VIS GMs had been well-recognized visually using cone–beam CT in all cases. Evaluation of images The degree of prostatic outline recognition despite artifacts using CT was scored as follows: 1. poor, 2. slightly poor, 3. neutral, 4. marginally good, and 5. excellent. The degree of prostatic marker recognition using MRI was scored as follows: 1. poor, 2. slightly poor, 3. neutral, 4. marginally good, and 5. excellent. The degree of marker recognition and the degree of prostatic outline recognition using MRI were analyzed, and the best sequences among the T2-WI, T2*2D, and T2*3D sequences were adopted. The visibility of the outline of the prostate (CT and MRI) was evaluated in the hemisphere of the side, where each marker was placed. Urologists also evaluated the visibility of the marker and needle using transrectal echography. Comparisons were made by one radiation oncologist (15 years of experience) and one radiation therapy technician (20 years of experience). The observers were informed about the two markers being used, but the objective of this study and detailed information of was blinded. We used a standard two-sided t test. A p value <0.05 was considered significant (software BellCurve for Excel. Social Survey Research Information Co., Ltd. Tokyo Japan).. Results Table 1 shows that the 0.5-mm-diameter GM was significantly superior to the 0.35-mm-diameter GM for GM detection using CT and MRI (p < 0.05). With regard to visibility of the prostatic outline, there was no significant difference between the 0.5- and 0.35-mm-diameter GM (Figs. 1, 2). There were slightly more CT artifacts with 0.5-mm-diameter GM; however, there was no significant difference. Similarly, the evaluation of the echo by the urologist was good at both marker sizes.. Discussion The purpose of this study was to raise the precision of registration of initial planning CT and MRI using a marker. Furthermore, we aimed to raise the precision of the collation of cone–beam and planning CTs as well as reproducibility in everyday radiation. The 0.5-mm-diameter and.

(3) Author's personal copy Radiol med Table 1 Comparison of 0.35 mm in diameter marker and 0.5 mm in diameter marker. Observer 1 0.35 mm side 0.5 mm side p value Observer 2 0.35 mm side 0.5 mm side p value. Outline in CT. Outline in MRI. Artifact in CT. Signal void in MRI. 2.545 2.182. 4.194 4.194. 2.130 1.522. 2.903 4.000. 0.0023. None. 0.0005. <0.0001. 2.435 1.609. 4.871 4.290. 1.833 2.375. 3.000 4.258. 0.0002. 0.0071. 0.0448. 0.0002. The lengths of the marker were both 10 mm As outline of the prostate, higher score was regarded well depicted As artifact of CT, higher score was regarded lesser artifact As signal void in MRI, higher score was regarded well visualization of marker Statistical analysis was using paired t test CT computed tomography, MRI magnetic resonance imaging, Outline outline of the prostate. Fig. 1 MRI revealed that the 0.5-mm-diameter marker is well depicted, whereas as the 0.35-mm-diameter marker is not. CT revealed that the artifact of the 0.5-mm-diameter marker is slightly larger than that of the 0.35-mm-diameter marker. Fig. 2 Similar to Fig. 1, both the calcification and the 0.5-mm-diameter marker are well depicted. 1.0-mm-diameter GMs are frequently used, and the larger the GM diameter, the greater is the recognition precision using MRI. However, as the metal volume increases,. recognition precision decreases because of artifacts on CT. Furthermore, the prostate is a small organ and metal artifacts may influence the radiation dose distribution. The first problem is that it is hard to depict smalldiameter GMs using MRI. To overcome this problem, we researched and presented successful methods in GM recognition using T2*-2D and T2*3D MRI at the annual meeting of the European Society for Radiotherapy and Oncology (ESTRO35) in April 2016 [7]. Although 0.5-mm-diameter GM has become the mainstream larger size, smaller diameter GM (needle) is gradually being used through experience. To the best of our knowledge, there are no clinical trials on the insertion of different diameter GMs or comparison of CT/MRI images in the same human body. As GM size increases, the visibility on MRI increases, but adverse artifact and dose distribution effects occur on CT. Therefore, we selected the smallest diameter of the size of GM and compared it using the above-mentioned MRI and CT using both 0.35 and 0.5 mm this time. We discovered an MRI method with CT reproducibility that could be used at all institutions. Visibility was significantly better with 0.5-mm-diameter GM than with 0.35-mm-diameter GM using MRI. GM visibility using MRI is given priority over that using CT. However, it takes 5–7 min to obtain one MRI sequencing image; in this time, the rectum moves, and the movement is largely dependent on the nature of MRI. Thus, poor images were seen with both 0.35- and 0.5-mm-diameter GMs. We previously conducted a study to increase the degree of recognition on MRI [5], and we changed the MRI sequence parameters little by little and regulated them. Relevant to movement of the prostate, we recommend the use of butylscopolamine before MRI.. 13.

(4) Author's personal copy Radiol med Compliance with ethical standards Conflict of interest The authors declare that they have no conflict of interest. Research involving human participants Written informed consent was received from all patients. This study was approved by the Internal Review Board of our hospital and registered under National Clinical Investigation Network (UMIN). Funding This study was no funding.. Fig. 3 Placement of the 0.35-mm VISICOIL linear-shaped marker is difficult to depict on MRI. Calcification and iron-contain fiducial marker (Gold Anchor); the spherical shape well depicts the placement. In addition, the 0.5%-iron-containing GM [Gold Anchor (GA); Naslund Medical AB, Huddinge, Sweden] increases the recognition ability of MRI (Fig. 3) and is very useful for CT registration. Because the marker is 0.28 mm in diameter and a 25-gauge needle can be used, it is useful from the viewpoints of bleeding, pain, and tumor dissemination, and clinical trials are now being performed. We used 1.5 T; if we used 3 T, 0.35-mm-diameter GM would have been better visualized. We did not include GMs with larger size. If we had included those with 0.75- or 1.1mm diameters, the results may have been different.. Conclusion With regard to visibility on MRI, the 0.5-mm-diameter marker was more useful than the 0.35-mm-diameter marker. However, prostatic MRI is often influenced by rectal movement; without movement, even small-diameter GM would be precisely depicted. There were many artifacts on CT; however, the artifacts with the 0.5-mm-diameter marker were in a significantly more allowable range than those with the 0.35-mm-diameter marker using both CT and MRI.. 13. References 1. McLaughlin PW, Evans C, Feng M, Narayana V (2010) Radiographic and anatomic basis for prostate contouring errors and methods to improve prostate contouring accuracy. Int J Radiat Oncol Biol Phys 76:369–378 2. Susil RC, Ménard C, Krieger A, Coleman JA, Camphausen K, Choyke P, Fichtinger G, Whitcomb LL, Whitcomb CN, Atalar E (2006) Transrectal prostate biopsy and fiducial marker placement in a standard 1.5 T magnetic resonance imaging scanner. J Urol 175:113–120 3. Nichol AM, Brock KK, Lockwood GA, Moseley DJ, Rosewall T, Warde PR, Catton CN, Jaffray DA (2007) A magnetic resonance imaging study of prostate deformation relative to implanted gold fiducial markers. Int J Radiat Oncol Biol Phys 67:48–56 4. Schieda N, Avruch L, Shabana WM, Malone SC (2015) Multiecho gradient recalled echo imaging of the pelvis for improved depiction of brachytherapy seeds and fiducial markers facilitating radiotherapy planning and treatment of prostatic carcinoma. J Magn Reson Imaging 41:715–720 5. Tanaka O, Hayashi S, Matsuo M, Sakurai K, Nakano M, Maeda S, Kajita K, Deguchi T, Hoshi H (2006) Comparison of MRIbased and CT/MRI fusion-based post implant dosimetric analysis of prostate brachytherapy. Int J Radiat Oncol Biol Phys 66:597–602 6. Lim C, Malone SC, Avruch L, Breau RH, Flood TA, Lim M, Morash C, Quon JS, Walsh C, Schieda N (2015) Pictorial review. Magnetic resonance for radiotherapy management and treatment planning in prostatic carcinoma. Br J Radiol 88:20150507 7. Tanaka O, Hattori M, Hirose S, Iida T, Watanabe HT (2016) Comparison of the MRI sequences in ideal fiducial marker-based radiotherapy for prostate cancer. European Society of Therapeutic Radiation Oncology 35th annual meeting electric poster: ER-1830, 2016 April 29–May 3 8. Chan MF, Cohen GN, Deasy JO (2015) Qualitative evaluation of fiducial markers for radiotherapy imaging. Technol Cancer Res Treat 14:298–304.

(5)