skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: SU-F-J-157: Effect of Contouring Uncertainty in Post Implant Dosimetry of Low-Dose-Rate Prostate Permanent Seed Brachytherapy

Abstract

Purpose: There is strong evidence relating post-implant dosimetry for low-dose-rate (LDR) prostate seed brachytherapy to local control rates. The delineation of the prostate on CT images, however, represents a challenge due to the lack of soft tissue contrast in order to identify the prostate borders. This study aims at quantifying the sensitivity of clinically relevant dosimetric parameters to uncertainty in the contouring of prostate. Methods: CT images, post-op plans and contours of a cohort of patients (n=43) (low risk=55.8%, intermediate risk=39.5%, high risk=4.7%), who had received prostate seed brachytherapy, were imported into MIM Symphony treatment planning system. The prostate contours in post-implant CT images were expanded/contracted uniformly for margins of ±1.00 mm, ±2.00 mm, ±3.00 mm, ±4.00 mm and ±5.00 mm. The values for V100 and D90 were extracted from Dose Volume Histograms for each contour and compared. Results: Significant changes were observed in the values of D90 and V100 as well as the number of suboptimal plans for expansion or contraction margins of only few millimeters. Evaluation of coverage based on D90 was found to be less sensitive to expansion errors compared to V100. D90 led to a lower number of implants incorrectly identified with insufficient coverage for expandedmore » contours which increases the accuracy of post-implant QA using CT images compared to V100. Conclusion: In order to establish a successful post implant QA for LDR prostate seed brachytherapy, it is necessary to identify the low and high thresholds of important dose metrics of the target volume such as D90 and V100. Since these parameters are sensitive to target volume definition, accurate identification of prostate borders would help to improve accuracy and predictive value of the post-implant QA process. In this respect, use of imaging modalities such as MRI where prostate is well delineated should prove useful.« less

Authors:
; ; ; ;  [1];  [2]; ;  [3]
  1. Sunnybrook Health Sciences Center, Odette Cancer Centre, Toronto, ON (Canada)
  2. (Canada)
  3. Sunnybrook Research Institute, Toronto, ON (Canada)
Publication Date:
OSTI Identifier:
22634758
Resource Type:
Journal Article
Resource Relation:
Journal Name: Medical Physics; Journal Volume: 43; Journal Issue: 6; Other Information: (c) 2016 American Association of Physicists in Medicine; Country of input: International Atomic Energy Agency (IAEA)
Country of Publication:
United States
Language:
English
Subject:
60 APPLIED LIFE SCIENCES; 61 RADIATION PROTECTION AND DOSIMETRY; ACCURACY; ANIMAL TISSUES; BIOMEDICAL RADIOGRAPHY; BRACHYTHERAPY; COMPUTERIZED TOMOGRAPHY; DOSE RATES; DOSIMETRY; IMAGE PROCESSING; NMR IMAGING; PROSTATE; RADIATION DOSES; RADIATION SOURCE IMPLANTS

Citation Formats

Mashouf, S, Merino, T, Ravi, A, Morton, G, Song, W, University of Toronto, Dept. of Radiation Oncology, Toronto, ON, Safigholi, H, and Soliman, A. SU-F-J-157: Effect of Contouring Uncertainty in Post Implant Dosimetry of Low-Dose-Rate Prostate Permanent Seed Brachytherapy. United States: N. p., 2016. Web. doi:10.1118/1.4956065.
Mashouf, S, Merino, T, Ravi, A, Morton, G, Song, W, University of Toronto, Dept. of Radiation Oncology, Toronto, ON, Safigholi, H, & Soliman, A. SU-F-J-157: Effect of Contouring Uncertainty in Post Implant Dosimetry of Low-Dose-Rate Prostate Permanent Seed Brachytherapy. United States. doi:10.1118/1.4956065.
Mashouf, S, Merino, T, Ravi, A, Morton, G, Song, W, University of Toronto, Dept. of Radiation Oncology, Toronto, ON, Safigholi, H, and Soliman, A. Wed . "SU-F-J-157: Effect of Contouring Uncertainty in Post Implant Dosimetry of Low-Dose-Rate Prostate Permanent Seed Brachytherapy". United States. doi:10.1118/1.4956065.
@article{osti_22634758,
title = {SU-F-J-157: Effect of Contouring Uncertainty in Post Implant Dosimetry of Low-Dose-Rate Prostate Permanent Seed Brachytherapy},
author = {Mashouf, S and Merino, T and Ravi, A and Morton, G and Song, W and University of Toronto, Dept. of Radiation Oncology, Toronto, ON and Safigholi, H and Soliman, A},
abstractNote = {Purpose: There is strong evidence relating post-implant dosimetry for low-dose-rate (LDR) prostate seed brachytherapy to local control rates. The delineation of the prostate on CT images, however, represents a challenge due to the lack of soft tissue contrast in order to identify the prostate borders. This study aims at quantifying the sensitivity of clinically relevant dosimetric parameters to uncertainty in the contouring of prostate. Methods: CT images, post-op plans and contours of a cohort of patients (n=43) (low risk=55.8%, intermediate risk=39.5%, high risk=4.7%), who had received prostate seed brachytherapy, were imported into MIM Symphony treatment planning system. The prostate contours in post-implant CT images were expanded/contracted uniformly for margins of ±1.00 mm, ±2.00 mm, ±3.00 mm, ±4.00 mm and ±5.00 mm. The values for V100 and D90 were extracted from Dose Volume Histograms for each contour and compared. Results: Significant changes were observed in the values of D90 and V100 as well as the number of suboptimal plans for expansion or contraction margins of only few millimeters. Evaluation of coverage based on D90 was found to be less sensitive to expansion errors compared to V100. D90 led to a lower number of implants incorrectly identified with insufficient coverage for expanded contours which increases the accuracy of post-implant QA using CT images compared to V100. Conclusion: In order to establish a successful post implant QA for LDR prostate seed brachytherapy, it is necessary to identify the low and high thresholds of important dose metrics of the target volume such as D90 and V100. Since these parameters are sensitive to target volume definition, accurate identification of prostate borders would help to improve accuracy and predictive value of the post-implant QA process. In this respect, use of imaging modalities such as MRI where prostate is well delineated should prove useful.},
doi = {10.1118/1.4956065},
journal = {Medical Physics},
number = 6,
volume = 43,
place = {United States},
year = {Wed Jun 15 00:00:00 EDT 2016},
month = {Wed Jun 15 00:00:00 EDT 2016}
}
  • Purpose: There is a strong evidence relating post-implant dosimetry for permanent seed prostate brachytherpy to local control rates. The delineation of the prostate on CT images, however, represents a challenge as it is difficult to confidently identify the prostate borders from soft tissue surrounding it. This study aims at quantifying the sensitivity of clinically relevant dosimetric parameters to prostate contouring uncertainty. Methods: The post-implant CT images and plans for a cohort of 43 patients, who have received I–125 permanent prostate seed implant in our centre, were exported to MIM Symphony LDR brachytherapy treatment planning system (MIM Software Inc., Cleveland, OH).more » The prostate contours in post-implant CT images were expanded/contracted uniformly for margins of ±1.00mm, ±2.00mm, ±3.00mm, ±4.00mm and ±5.00mm (±0.01mm). The values for V100 and D90 were extracted from Dose Volume Histograms for each contour and compared. Results: The mean value of V100 and D90 was obtained as 92.3±8.4% and 108.4±12.3% respectively (Rx=145Gy). V100 was reduced by −3.2±1.5%, −7.2±3.0%, −12.8±4.0%, −19.0±4.8%, − 25.5±5.4% for expanded contours of prostate with margins of +1mm, +2mm, +3mm, +4mm, and +5mm, respectively, while it was increased by 1.6±1.2%, 2.4±2.4%, 2.7±3.2%, 2.9±4.2%, 2.9±5.1% for the contracted contours. D90 was reduced by −6.9±3.5%, −14.5±6.1%, −23.8±7.1%, − 33.6±8.5%, −40.6±8.7% and increased by 4.1±2.6%, 6.1±5.0%, 7.2±5.7%, 8.1±7.3% and 8.1±7.3% for the same set of contours. Conclusion: Systematic expansion errors of more than 1mm may likely render a plan sub-optimal. Conversely contraction errors may Result in labeling a plan likely as optimal. The use of MRI images to contour the prostate should results in better delineation of prostate organ which increases the predictive value of post-op plans. Since observers tend to overestimate the prostate volume on CT, compared with MRI, the impact of the contouring uncertainty on V100 and D90 fortunately, has a conservative effect of underestimating the prostate coverage.« less
  • Purpose: To validate in a water phantom the use of plastic scintillation detectors to measure dose to the urethra and the rectal wall during a clinically realistic low dose rate (LDR) brachytherapy implant. Methods: A template was designed to replicate a clinically realistic LDR brachytherapy prostate implant inside a water phantom. Twenty-two catheters were inserted, including one mimicking the urethra and another the rectal wall. The needles inserted in the remaining 20 catheters were composed of thin-walled nylon tubes in which I-125 radioactive seeds (Air Kerma Strengths of (0.328±0.020)U) were abutted together with plastic spacers to replicate a typical loading.more » A plastic scintillation detector (PSD) with a 5-mm long × 1-mm diameter sensitive element was first placed inside the urethra and 1-second measurements were performed for 60s after each needle implant. Measurements were also performed at multiple positions along the urethra once all the needles were inserted. The procedure was then repeated with the PSD placed at the rectal wall. Results: Individual dose-rates ranging from 0.07µGy/s to 1.5µGy/s were measured after each needle implant. The average absolute relative differences were (6.2±3.6)% and (6.9±6.5)% to the values calculated with the TG-43 formalism, for the urethra and rectal wall respectively. These results are within expectations from the error uncertainty budget once accounting for uncertainties in seeds’ strength and positioning. Interestingly, the PSD allowed for unplanned error detection as the study was performed. Finally, the measured dose after the full implant at different positions along the mimicked organs at risk were in agreement with TG-43 values for all of the positions tested. Conclusion: Plastic scintillation detectors could be used as in vivo detectors for LDR brachytherapy as they would provide accurate dose information after each needle implant as well as along the organs at risk at the end of the implant.« less
  • Purpose: The current CT-based post-implant dosimetry allows precise seed localization but limited anatomical delineation. Switching to MR-based post-implant dosimetry is confounded by imprecise seed localization. One approach is to place positive-contrast markers (Sirius) adjacent to the negative-contrast seeds. This patient study aims to assess the utility of a 3D fast spoiled gradient-recalled echo (FSPGR) sequence to visualize Sirius markers for post-implant dosimetry. Methods: MRI images were acquired in prostate implant patients (n=10) on Day 0 (day-of-implant) and Day 30. The post-implant MR protocol consisted of 3D T2-weighted fast-spin-echo (FSE), T2-weighted 2D-FSE (axial) and T1-weighted 2D-FSE (axial/sagittal/coronal). We incorporated a 3D-FSPGRmore » sequence into the post-implant MR protocol to visualize the Sirius markers. Patients were scanned with different number-of-excitations (6, 8, 10), field-of-view (10cm, 14cm, 18cm), slice thickness (1mm, 0.8mm), flip angle (14 degrees, 20 degrees), bandwidth (122.070 Hz/pixel, 325.508 Hz/pixel, 390.625 Hz/pixel), phase encoding steps (160, 192, 224, 256), frequency-encoding direction (right/left, anterior/posterior), echo-time type (minimum-full, out-of-phase), field strength (1.5T, 3T), contrast (with, without), scanner vendor (Siemens, GE), coil (endorectal-coil only, endorectal-and-torso-coil, torsocoil only), endorectal-coil filling (30cc, 50cc) and endorectal-coil filling type (air, perfluorocarbon [PFC]). For post-implant dosimetric evaluation with greater anatomical detail, 3D-FSE images were fused with 3D-FSPGR images. For comparison with CT-based post-implant dosimetry, CT images were fused with 3D-FSPGR images. Results: The 3D-FSPGR sequence facilitated visualization of markers in patients. Marker visualization helped distinguish signal voids as seeds versus needle tracks for more definitive MR-based post-implant dosimetry. On the CT-MR fused images, the distance between the seed on CT to MR images was 3.2±1.6mm in patients with no endorectal coil, 2.3±0.8mm in patients with 30cc-PFC-filled endorectal-coil and 5.0±1.8mm in patients with 50cc-PFC-filled endorectal-coil. Conclusion: An MR protocol to visualize positive-contrast Sirius markers to assist in the identification of negative-contrast seeds was demonstrated. S Frank is a co-founder of C4 Imaging LLC, the manufacturer of the MRI markers.« less
  • To investigate the dosimetric and radiobiological differences among volumetric modulated arc therapy (VMAT), high-dose rate (HDR) brachytherapy, and low-dose rate (LDR) permanent seeds implant for localized prostate cancer. A total of 10 patients with localized prostate cancer were selected for this study. VMAT, HDR brachytherapy, and LDR permanent seeds implant plans were created for each patient. For VMAT, planning target volume (PTV) was defined as the clinical target volume plus a margin of 5 mm. Rectum, bladder, urethra, and femoral heads were considered as organs at risk. A 78 Gy in 39 fractions were prescribed for PTV. For HDR andmore » LDR plans, the dose prescription was D{sub 90} of 34 Gy in 8.5 Gy per fraction, and 145 Gy to clinical target volume, respectively. The dose and dose volume parameters were evaluated for target, organs at risk, and normal tissue. Physical dose was converted to dose based on 2-Gy fractions (equivalent dose in 2 Gy per fraction, EQD{sub 2}) for comparison of 3 techniques. HDR and LDR significantly reduced the dose to rectum and bladder compared with VMAT. The D{sub mean} (EQD{sub 2}) of rectum decreased 22.36 Gy in HDR and 17.01 Gy in LDR from 30.24 Gy in VMAT, respectively. The D{sub mean} (EQD{sub 2}) of bladder decreased 6.91 Gy in HDR and 2.53 Gy in LDR from 13.46 Gy in VMAT. For the femoral heads and normal tissue, the mean doses were also significantly reduced in both HDR and LDR compared with VMAT. For the urethra, the mean dose (EQD{sub 2}) was 80.26, 70.23, and 104.91 Gy in VMAT, HDR, and LDR brachytherapy, respectively. For localized prostate cancer, both HDR and LDR brachytherapy were clearly superior in the sparing of rectum, bladder, femoral heads, and normal tissue compared with VMAT. HDR provided the advantage in sparing of urethra compared with VMAT and LDR.« less
  • Purpose: Permanent prostate implant brachytherapy (PPI), three-dimensional conformal radiotherapy (3D-CRT), and conformal proton beam radiotherapy (CPBRT) are used in the treatment of localized prostate cancer, although no head-to-head trials have compared these modalities. We studied the biochemical control (biochemical no evidence of disease [bNED]) and prostate-specific antigen (PSA) nadir achieved with contemporary PPI, and evaluated it against 3D-CRT and CPBRT. Patients and Methods: A total of 249 patients were treated with PPI at the University of California, San Francisco, and the outcomes were compared with those from a 3D-CRT cohort and the published results of a high-dose CPBRT boost (CPBRTB)more » trial. For each comparison, subsets of the PPI cohort were selected with patient and disease criteria similar to those of the reference group. Results: With a median follow-up of 5.3 years, the bNED rate at 5 and 7 years achieved with PPI was 92% and 86%, respectively, using the American Society for Therapeutic Radiology and Oncology (ASTRO) definition, and 93% using the PSA nadir plus 2 ng/mL definition. Using the ASTRO definition, a 5-year bNED rate of 78% was achieved for the 3D-CRT patients compared with 94% for a comparable PPI subset and 93% vs. 92%, respectively, using the PSA nadir plus 2 ng/mL definition. The median PSA nadir for patients treated with PPI and 3D-CRT was 0.10 and 0.40 ng/mL, respectively (p < .0001). For the CPBRT comparison, the 5-year bNED rate after a CPBRTB was 91% using the ASTRO definition vs. 93% for a similar group of PPI patients. A greater proportion of PPI patients achieved a lower PSA nadir compared with those achieved in the CPBRTB trial (PSA nadir <=0.5 ng/mL, 91% vs. 59%, respectively). Conclusion: We have demonstrated excellent outcomes in low- to intermediate-risk patients treated with PPI, suggesting at least equivalent 5-year bNED rates and a greater proportion of men achieving lower PSA nadirs compared with 3D-CRT or CPBRTB.« less