- Email: [email protected]
Robustness to source displacement in dual air kerma strength planning for focal low-dose-rate brachytherapy of prostate cancer
Brachytherapy
-
(2016)
-
Robustness to source displacement in dual air kerma strength planning for focal low-dose-rate brachytherapy of prostate cancer S. Sara Mahdavi1,*, Ingrid T. Spadinger2, Nicholas T. Chng3, W. James Morris1 1
Department of Radiation Oncology, Vancouver Centre, British Columbia Cancer Agency, Vancouver, BC, Canada 2 Department of Medical Physics, Vancouver Centre, British Columbia Cancer Agency, Vancouver, BC, Canada 3 Department of Medical Physics, Centre for the North, British Columbia Cancer Agency, Prince George, BC, Canada
ABSTRACT
PURPOSE: To describe the use of dual source strength implants for focal low-dose-rate brachytherapy. METHODS AND MATERIALS: An interneedle dual source strength planning strategy is described for focal low-dose-rate brachytherapy of the prostate. The implanted treatment plans were designed using peripheral (except near the rectum) needles loaded with high strength (0.9 U) sources and central needles loaded with low strength (0.4 U) sources (‘‘interneedle’’ dual strength planning). This approach has been applied for focally treating 3 patients. In this article, we compare the characteristics and robustness to source motion of interneedle dual strength planning with four alternative planning strategies (single strength high, low, and intermediate, and intraneedle dual strength) on 50 simulated cases. RESULTS: Interneedle dual source strength planning results in greater robustness to source motion and overall lower seed and needle density compared to the standard low source strength planning currently used in our centre. This planning approach is also significantly superior to single strength high, single strength intermediate and intraneedle dual strength planning strategies in terms of high dose to the urethral avoidance structure. CONCLUSIONS: The use of interneedle dual source strength treatment plans for focal low-doserate brachytherapy is possibly the practical solution for limiting the density of sources required to deliver the prescribed dose while limiting proximity of high strength sources to organs at risk. Ó 2016 American Brachytherapy Society. Published by Elsevier Inc. All rights reserved.
Keywords:
Low-dose-rate brachytherapy; Prostate cancer; Dual source strength; Robustness; Source displacement
Introduction Focal treatment for prostate cancer, in which only diseased regions of the gland are treated, is an emerging topic of interest. By sparing healthy prostate tissue, fewer adverse effects on quality of life, compared to conventional wholegland treatment methods, are expected. However, in order for this approach to become widely acceptable as a standard Received 7 March 2016; received in revised form 8 April 2016; accepted 16 April 2016. Financial disclosure: This study was funded by the BC Cancer Foundation (DRG01829). Conflict of interest: The authors report no proprietary or commercial interest in any product mentioned or concept discussed in this article. * Corresponding author. Department of Radiation Oncology, Vancouver Centre, British Columbia Cancer Agency, 600 W 10th Ave., V5Z 4E6, Vancouver, BC, Canada. Tel.: þ1 604 877 6000; fax: þ1 604 877 0505. E-mail address: [email protected] (S.S. Mahdavi).
treatment option, it needs accurate methods for detecting and localizing cancer within the prostate, clear definitions of eligibility criteria and treatment success, safe and reliable ablation methods, and sufficient long-term followup results. Low-dose-rate prostate brachytherapy (LDR-PB) is a favorable ablation method for focal therapy as it can be easily modified to treat portions of the gland, as done in standard practice for salvage treatments and reimplantations. However, when planning for the small target volumes usually seen in focal treatment, obtaining the desired coverage with a reasonable source density while avoiding overdose to the rectum and urethra poses new challenges that are less easily met with a uniform source strength plan. Whole-gland LDR-PB treatment planning strategies vary among centers in terms of needle placement and loading, source (seed) strength, and the isotope used, with the common goals of improving dose coverage and plan robustness, while simplifying treatment delivery. General
1538-4721/$ - see front matter Ó 2016 American Brachytherapy Society. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.brachy.2016.04.388
2
S.S. Mahdavi et al. / Brachytherapy
guidelines have been developed to aid practitioners in managing treatment. The American Brachytherapy Society recommends a prescription dose of 140e160 Gy for 125I sources as a monotherapy option for National Comprehensive Cancer Networkedefined low- and intermediate-risk prostate cancer (1, 2). The original planning approach for delivering this dose was the use of a purely uniform seed distribution by the Seattle group (3). This eventually evolved into a ‘‘peripheral,’’ ‘‘modified peripheral,’’ or ‘‘modified uniform loading’’ distribution to avoid the high central doses (O400 Gy) caused by uniform loading, which have been associated with urethral toxicity (4e6). There is currently no consensus on the optimal source strength (or, alternatively, optimal source density). When treating the whole gland, acceptable dose coverage using the modified peripheral loading approach is achievable with single source strength plans of both low and high source density (7e10). In the RTOG clinical trials, 0.3e0.55 U is the reported range of air kerma strengths for 125I seeds (1). Low source strength (LSS) (i.e., high source density) plans lead to better dose homogeneity, smaller hot spots, and less influence on dosimetry in case of seed migration or implantation errors (11), whereas high source strength (HSS) plans provide better dose coverage with fewer seeds, hence, less prostate trauma, and less operating time and procedure costs (12). In whole-gland treatment, the cumulative dose from peripheral sources provides interior coverage without the need for source placement in the prostate center, close to the urethra. However, when the target volume is small, as in the case of focal treatment, the standard air kerma strength used in our center (~0.4 U) leads to a very dense plan. Higher air kerma strength may be used to reduce source density, but there are concerns that the consequences of source placement errors or source motion, which are commonly seen postimplant, might be more severe in a high strength plan. In addition, the modified peripheral approach is not feasible because the urethra is likely to lie adjacent to the planning target volume (PTV) periphery. To overcome these issues, we propose a dual source strength (DSS) planning strategy for focal low-dose-rate prostate brachytherapy (focal LDR-PB) in which needles are loaded with either high strength seeds or low strength seeds. Three patients have been treated using this strategy in the British Columbia Cancer Agency focal LDR-PB pilot study (13). The concept of uneven radiation distribution has been discussed since the 1930s (14) and was investigated for prostate implants, in addition to the effect of source placement errors, by Narayana et al. (15) in the mid1990s. However, to the best of our knowledge, this is the first report of a practical application of dual strength plans for focal treatment of prostate cancer. In this article, we provide a brief overview of the focal LDR-PB pilot study and focus on the robustness of dual source planning to source motion compared to four alternative planning strategies.
-
(2016)
-
Methods The focal LDR-PB pilot study The focal LDR-PB pilot study at the British Columbia Cancer Agency Vancouver Centre aims to determine the feasibility of applying focal LDR-PB as a treatment option for men with low risk prostate cancer. Institutional ethics approval was acquired before recruitment, and 3 patients have received focal LDR-PB. Consented patients are screened with multiparametric magnetic resonance imaging (mpMRI) multiparametric transrectal ultrasound (mpTRUS) imaging and transperineal template mapping biopsy (TTMB) (16). TTMB is used for cancer localization and is performed in the operating suite using a standard LDR-PB setup with a 5 5 mm brachytherapy grid. Based on the volume of the prostate, 20e50 cores are extracted, approximately one every 1 cc of the prostate volume, and their locations in the axial plane and craniocaudal depths (with a precision of 0.5 cmethe TRUS image spacing) are recorded. The biopsy samples are reviewed by a single pathologist to identify the presence of cancer, Gleason score, tumor location, and length of involvement within the core. Eligibility to receive focal LDR-PB includes TTMB results of #4 positive cores (Gleason score #3 þ 4) within a single lobe and spanning no more than two adjacent sectors (base, midgland, apex). Based on the cancerous core locations and tumor extension within the cores, the gross tumor volume is mapped onto the TRUS images collected immediately before TTMB. The PTV is defined as the gross tumor volume plus a 5e8 mm margin in all three dimensions. Extraprostatic extension of the PTV is allowed up to 4e5 mm from the prostate boundary except for the posterior. The rectum is delineated from 1 cm superior to the base to 1 cm inferior to the apex. An intraprostatic avoidance structure is outlined as a circle of 5-mm radius based on the ‘‘deviated’’ urethra model recommendations in Bucci et al. (17), which assumes a 30 anterior deviation in the superior half of the prostate. This structure is used as a surrogate for the urethra. Contouring and PTV mapping are performed with the MIM Symphony (MIM Software, Cleveland, OH), and planning is performed with the VariSeed 8.0 planning system (Varian Medical Systems, Palo Alto, CA). Oncoseed 6711 sources (Oncura Inc., Arlington Heights, IL) were used to generate the treatment plans. In our ‘‘interneedle dual source strength’’ plans, high strength seeds (0.8e0.9 U) were used in needles located in the PTV periphery (except near the rectum and urethral avoidance region) and low strength seeds (0.4 U) in needles located elsewhere, with the goal of meeting standard dosimetric constraints for V100 and D90. Source strength was kept uniform within needles largely for logistical reasons, as we wanted to use stranded, preloaded seed trains in all the needles. Implantation was performed with a standard low-dose-rate brachytherapy setup.
S.S. Mahdavi et al. / Brachytherapy
Further details of imaging procedures and clinical implementation and outcomes are the subject of a separate article. Dosimetric robustness to source displacement Seed displacement during and in the weeks after prostate implantation has been widely reported. In extreme cases, seeds can be completely lost through ejaculation, transurethral routes, or through the venous plexus to the lung, but the incidence of such events is reduced if stranded seeds are used (18e21) among others. Smaller displacements from the planned positions are more common and are caused by implantation error due to factors such as needle deflection, gland motion, and prostate edema. Regardless of the cause, unpredictable changes in the seed locations from the original plan may significantly affect dose coverage, leading to overdosing adjacent critical structures and/or underdosing the target region. We have analyzed the potential effect of local source displacements on dosimetric parameters on 50 simulated cases. We compared the robustness of interneedle dual source strength (DSS) planning to source motion with four alternative planning strategies: high source strength (HSS), intermediate source strength (ISS), low source strength (LSS), and ‘‘intraneedle’’ dual source strength (D2SS) planning. To generate the 50 simulated cases, sample PTVs were manually created from the whole-gland contours of 50 patients randomly chosen, to cover a range of PTV volumes, from our large pool of standard whole prostate implant patients. Hemigland PTVs were assumed, as an approximate
Fig. 1. Illustration of a sample PTV (cyan) defined for robustness analysis on simulated cases. A margin of 1e2 mm is drawn inside the urethral avoidance structure (green). The midportion of the gland is excluded by drawing the PTV 15e20 off from the midline (dashed). PTV [ planning target volume. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)
-
(2016)
-
3
simulation of the PTVs seen in the three focally treated patients, and to reflect our criteria of focally treating cases with only unilaterally identified cancer (Fig. 1). These were defined as half of the prostate gland (the clinical target volume) minus a portion of the gland midline (approximately 15e20 off from the ‘‘D’’ grid column located on the prostate midline) and a portion of the urethral avoidance structure, with no extraprostatic margin. A margin of 1e2 mm inside the urethral avoidance structure was included, and the entire length of the prostate from the base to the apex was covered. Air kerma strengths used were: 0.941 U for HSS plans, 0.417 U for LSS plans, 0.679 U (the average of 0.941 and 0.417) for ISS plans, and a combination of 0.941 U and 0.417 U for DSS and D2SS plans. D2SS plans differed from DSS plans in having two air kerma strengths within the needles if necessary. The D2SS planning strategy was similar to the DSS strategy, in that high strength sources were not placed close to the urethral avoidance region or rectum, but they could be placed in predominately low source strength trains in slices where the critical structure was not close to the needle path. Similarly, some seeds in predominately high strength needles were replaced by low strength seeds to achieve better conformity or reduce the V150% inside the PTV. Plans were created by manual optimization with the goal of meeting the following constraints: minimum peripheral dose 5 144 Gy, PTV V100% $95% of PTV volume, PTV V150% between 50% and 70% of PTV volume, PTV D90% between 95% and 125% of prescription dose, urethral avoidance structure V100% #30% of the structure’s volume, and rectal V50% #5% of rectal volume. Rectal and urethral dose constraints are lower than typical values used for whole-gland monotherapy and were established during the course of planning for smaller and unilateral PTVs. Attempts were made to minimize the overall number of needles (i.e., avoiding needles with just one seed, and the average number of seeds per needle was ~3 or greater) and maintain uniform 1 cm source spacing as much as possible. Source spacing within a needle was always at least 1 cm. Changes in the planned dosimetric parameters due to source displacements were studied using a Monte Carlo approach. In this method, for each of the 50 simulated cases, random displacements were applied to each individual seed and changes in the dosimetric constraints were measured. This was repeated over 500 iterations, and statistics were derived. MATLAB (The MathWorks, Inc. MA) was used to apply source displacements in plans exported from VariSeed, to compute dosimetric parameters, and to obtain robustness results. MATLAB dose calculation was in accordance with the AAPM TG-43 (22), assuming a point source, and the results were validated against VariSeed. The random displacement statistics used were 0 2.5 mm (mean and one standard deviation) lateral and posterior motions and 4 3 mm inferior motion. A normal distribution of displacements was assumed in the modeling.
4
S.S. Mahdavi et al. / Brachytherapy
These values were chosen on the basis of posttreatment seed displacements measured in the literature (21,23e26). The dosimetric parameters reported in the Results for the rectum and urethral avoidance region are different from the standard metrics recommended by AAPM (27) to better describe our focal plans created on smaller PTVs. CI100 is the conformity index, ideally being 1, and is defined as (28): CI100 5
Vr ðVu þ Vh Þ Vr
where Vr is the PTV volume, Vu the portion of the PTV receiving less than 100% of the prescribed dose, and Vh is the volume of the region outside the PTV that is receiving $ 100% of the prescribed dose. D1cc is the dose received by 1 cc of the structure, and Vxx is the volume of the structure receiving xx% of the prescribed dose, reported in either cc or percent of the structure volume. D90 is the minimum dose delivered to 90% of the structure.
-
(2016)
-
Results Figure 2 shows an example of the five plan types for a simulated patient. Table 1 provides the description of the five plan types including the average number of needles and sources used, and PTV, urethral avoidance, and rectal volumes and dosimetric parameters. As summarized in this table, the average number of needles and sources used in the LSS plans is approximately 2 times that of HSS and 1.5 times the number used in the other three plan types. Per cc of PTV, lower air kerma strength was required in LSS plans, compared to HSS plans, to achieve the desired dosimetric constraints, due to the higher degree of conformity that could be achieved with the low strength sources. Results of robustness to source displacements in simulated plans are summarized in Fig. 3. The mean and 95% confidence interval of the change in each dosimetric parameter, over the 50 cases, are displayed. A negative value indicates a decrease in the dose parameter after displacing the sources.
Fig. 2. Example of the five plan types created for one of the simulated patients. In the needle distribution (Row 1), the icons triangle, square, diamond, and inverted triangle represent needles planned to retraction planes 0.5 cm, 1 cm, 1.5 cm, and 2 cm inferior to the base. Source distributions are shown for the base, midgland, and apex (Rows 2e4). The number of sources and needles used in this case are as follows: DSS (18 high strength and 17 low strength sources, 5 high and 4 low strength needles), HSS (26 sources and 8 needles), LSS (51 sources and 14 needles), ISS (33 sources and 10 needles), and D2SS (17 high strength and 15 low strength sources in 9 needles).
S.S. Mahdavi et al. / Brachytherapy
-
(2016)
-
5
Table 1 Description of the interneedle dual source strength (DSS), high source strength (HSS), low source strength (LSS), intermediate source strength (ISS), and intraneedle dual source strength (D2SS) plans and the dosimetric parameters in the 50 simulated cases Plan type
DSS
HSS
LSS
ISS
D2SS
Source strength (U) Total source strength/PTV volume # Needles (H, L)b # Sources (H, L) PTV vol. (cc) V100 (%) V150 (%) D90 (Gy) CI100 Rectum vol. (cc) D1cc (Gy) Urethral avoid vol. (cc) V125 (cc) V30 (%)
0.941, 0.417 1.42 0.34a 4.7 1.0; 4.7 1.1 15.3 4.4; 16.6 5.2
0.941 1.47 0.36 7.4 1.1 23.2 6.0
0.417 1.29 0.31 13.7 2.3 45.9 12.2
0.679 1.38 0.33 9.5 1.6 30.4 8.2
0.941, 0.417 1.40 0.31 9.0 1.4 16.3 4.7; 13.5 4.5
16.3 93.57 63.41 155.74 0.37
7.3, range (4.2e43.8) 1.34 94.64 4.09 62.89 4.55 159.05 0.21 0.35
1.10 5.15 4.11 0.23
93.71 56.89 155.11 0.48
1.36 4.99 4.18 0.16
94.17 61.45 156.93 0.42
1.43 3.87 4.66 0.20
94.15 61.05 156.59 0.42
1.19 4.43 4.08 0.21
19.2 8.5 64.24 12.2
69.06 15.74
64.11 12.98
66.38 13.64
63.97 11.86
4.2 0.6 0.03 0.02 84.37 4.57
0.04 0.02 87.19 4.36
0.04 0.02 81.99 4.00
0.04 0.02 85.17 4.38
0.03 0.02 85.74 5.08
PTV 5 planning target volume. a Mean and standard deviation. b (H, L): high strength and low strength, if applicable.
The plots illustrate less robustness of the LSS plans to source displacement in terms of PTV V100%, V150%, and D90. CI100 robustness in LSS plans is also less than
that of DSS and HSS plans. There is no significant difference in PTV dose metrics among the other four plan types. DSS plans show significantly less increase in
Fig. 3. Robustness of dual, high, low, intermediate, and interneedle dual source strength plans to source displacement. Plots display the average and 95% confidence interval of change in PTV V100%, V150%, D90 (Gy), and CI100, urethral avoidance V125 (cc) and V30%, and rectum D1cc (Gy) over 50 cases (aeg). For each case, 500 iterations of source displacement were applied. PTV [ planning target volume; DSS [ dual source strength; HSS [ high source strength; LSS [ low source strength; ISS[ intermediate source strength; D2SS [ intraneedle dual source strength.
6
S.S. Mahdavi et al. / Brachytherapy
-
(2016)
-
Fig. 3. (continued).
the urethral V125% compared to the HSS, ISS, and D2SS plans. We attribute the larger 95% confidence interval in the D2SS plans, especially in the rectum D1cc, to greater plan-to-plan variability in the anatomical location of high strength vs. low strength seeds, compared to the other four plan types. By comparing the original urethral V30% values (Table 1) to the changes in Fig. 3f, we see that the planned original V30% seems to have a negative correlation with the change in urethral V30% due to source displacement. However, the changes to V30% are generally very small, whereas V30% is quite large, so this observation may simply be an artifact of the particular planning strategy used. In summary, robustness of DSS plans is significantly superior to ISS, HSS, and D2SS plans in terms of high dose (as measured by V125%) to the urethral avoidance structure, which overlaps the PTV. This, combined with the overall lower seed and needle density required by DSS plans in comparison to LSS plans, as well as greater robustness with respect to PTV coverage, leads us to conclude that they are superior, for focal planning, to adopting an ISS or HSS strategy, or to retaining the LSS strategy that is our current standard for whole prostate implantation.
Discussion We would like to emphasize that postimplant dosimetry of dual- or mixed-air kerma strength implants is not possible unless the strength of each source on postimplant imaging (e.g., CT) can be clearly identified. Our unique plan reconstruction software (29) can match seeds from postimplant images to the corresponding strands in the plans, with a high accuracy of 97.7%. This software is currently being routinely used for whole-gland plan reconstruction of stranded implants and can be easily applied to focal LDR-PB plans. Seed displacement in LDR-PB has been studied by various groups. A common method of measuring seed motion is postimplant seed localization (e.g., from CT imaging or intraoperative fluoroscopy), followed by some form of seed cloud registration between the planned seed cloud and postimplant seed cloud (26, 30). To measure targeting errors (i.e., errors resulting from intentional or unintentional misplacement of the seeds by the radiation oncologist), the postimplant seed cloud is derived immediately after the implant. At approximately 1 month postimplant (21), seed migration occurs. However, by registering the seed clouds alone, it is not possible to measure the displacement of seeds with respect to the boundary. This is commonly seen, especially in the craniocaudal direction.
S.S. Mahdavi et al. / Brachytherapy
Measuring this type of displacement is challenging, and very few groups have reported it (23e25). The displacement distribution used in this article is chosen to cover various displacement values reported in the literature and to include an approximation of a systematic shift, which has the greatest potential to affect PTV coverage. Inferior misplacement of the sources due to targeting errors is generally considered to be the most likely type of systematic displacement for focal LDR-PB, so this was chosen as the systematic shift. Derivation of a more detailed region-based seed displacement model and understanding edema in focal LDR-PB (similar to that reported in Sloboda et al. (31) for whole-gland treatment) and its effect on postimplant dosimetry could further refine robustness analysis of focal LDR-PB planning. Furthermore, we have applied random displacements to each individual seed as opposed to the whole strands. Comparison of strand vs. loose seed displacement, especially in the craniocaudal direction, is also worth investigating. Our results show that the use of higher strength sources, alone or combined with lower strength sources, can lead to fewer needles; however, an increase in the source strength can have negative affects on the dose delivered to critical structures, particularly those that overlap the PTV. The potential benefits of high strength plans have been shown in the literature, specifically the reduced costs and fewer needle puncture wounds, although there are inconsistent results with respect to adverse effects associated with needle trauma. Martin et al. (8) produced excellent implants with fewer high strength seeds (0.8 U). By using an average number of seeds and needles of 54.6 14.4 and 17.6 3.1 and inverse planning for optimal seed positioning, they report a 60-month biochemical failure-free survival of 94.6% according to the nadir þ 2 ng/mL definition, on 396 patients (70% of which were low risk). In a study on 105 patients implanted with low strength seeds (0.44 mCi) and 94 implanted with higher strength seeds (0.61 mCi), Massucci et al. (9) report no statistical difference in patients’ International Prostate Symptom Score or dosimetric parameters calculated for the rectum based on Day 30 CT; however, the quality of implants measured at Day 30 CT (V100 and D90) was significantly better in the higher strength cohort, although this group also showed higher V150 and V200. They observed no significant difference in short-term urinary toxicity (up to 8 months), although increased dose inhomogeneity using high strength sources may lead to increased long-term urinary toxicity. Beaulieu et al. (7) studied the effect of a wide range of air kerma strengths on the dosimetric properties of plans created by inverse planning and their robustness to seed displacement and migration. They report that with a properly tuned inverse planning algorithm, the choice of preferred seed strength, in a range of up to about 0.7 mCi (0.89 U), is open in terms of adequate dose coverage and protection for the
-
(2016)
-
7
organs at risk in the presence of seed placement errors. However, the placement of high air kerma strength seeds (0.9 U and above) becomes more critical to ensuring robustness to displacement and migration of seeds. Usmani et al. (32) report improved postoperative dosimetry when transitioning from low (median 0.398 U) to intermediate (median 0.494 U) strength sources, with fewer needles and sources used, and decreased need for catheterization. However, more rectal complications were observed in the intermediate source strength cohort. On the other hand, Thomas et al. (33) report no significant difference between low-density source implants (equivalent to high strength implants) and high-density source implants in any of the measured postimplant dosimetric parameters for the whole gland or quadrants, urethra, or rectum. The above studies all report on whole-gland treatment. We show that DSS plans have the benefit of requiring fewer needles and show better robustness of the dose delivered to critical structures, with respect to simulated seed displacements, compared to all other single source strength plan types. DSS planning is possibly the only way to plan small PTV sizes to achieve desirable results. As a final word, it is important to recognize the limitations of this study. Although we have attempted to include a fairly wide range of PTV sizes with varying degrees of proximity to the rectum, other factors, such as overlap with the urethral avoidance structure, were kept relatively constant. The full range of possible clinical scenarios with respect to PTVs location and organs’ at risk locations relative to the PTV was therefore not sampled. However, we can perhaps generalize our findings by suggesting that the results for the urethral avoidance structure represent the general trend for an overlapping organ at risk, whereas the results for the rectum represent the trend for a proximal, but nonoverlapping, organ at risk. Furthermore, although all planning was done by one individual subject to strict dosimetric constraints and seed placement and spacing criteria, it may be argued that a manual planning approach is too easily subject to bias. An inverse-planning approach, as has been used by other groups that have undertaken similar studies (7, 10), may be deemed preferable. Although we considered this approach, we concluded that the tuning, constraints, and acceptance criteria used with an inverse planning algorithm would also not be free of a subjective component. Conclusions Using DSS planning is a feasible approach for focal LDR-PB and seems to result in improved plans in terms of number of sources and dose coverage, as well as increased robustness to source displacements. Unambiguous identification of the seeds and association to the correct source strength is necessary for practical implementation of this approach. This strategy could be extended to wholegland treatment with potentially similar benefits.
8
S.S. Mahdavi et al. / Brachytherapy
References [1] Davis BJ, Horwitz EM, Lee WR, et al. American Brachytherapy Society consensus guidelines for transrectal ultrasound-guided permanent prostate brachytherapy. Brachytherapy 2012;11:6e19. [2] Rivard MJ, Butler WM, Devlin PM, et al. American Brachytherapy Society recommends no change for prostate permanent implant dose prescriptions using iodine-125 or palladium-103. Brachytherapy 2007;6:34e37. [3] Ragde H, Blasko JC, Grimm PD, et al. Interstitial iodine-125 radiation without adjuvant therapy in the treatment of clinically localized prostate carcinoma. Cancer 1997;80:442e453. [4] Nath R, Anderson LL, Meli JA, et al. Code of practice for brachytherapy physics: report of the AAPM Radiation Therapy Committee Task Group No. 56. Med Phys 1997;24:1557e1598. [5] Wallner K, Roy J, Harrison L. Dosimetry guidelines to minimize urethral and rectal morbidity following transperineal I-125 prostate brachytherapy. Int J Radiat Oncol Biol Phys 1995;32:465e471. [6] Butler WM, Merrick GS, Lief JH, Dorsey AT. Comparison of seed loading approaches in prostate brachytherapy. Med Phys 2000;27: 381e392. [7] Beaulieu L, Archambault L, Aubin S, et al. The robustness of dose distributions to displacement and migration of 125 I permanent seed implants over a wide range of seed number, activity, and designs. Int J Radiat Oncol Biol Phys 2004;58:1298e1308. [8] Martin AG, Roy J, Beaulieu L, et al. Permanent prostate implant using high activity seeds and inverse planning with fast simulated annealing algorithm: a 12-year Canadian experience. Int J Radiat Oncol Biol Phys 2007;67:334e341. [9] Masucci GL, Donath D, Tetreault-Laflamme A, et al. Comparison between high and low source activity seeds for I-125 permanent seed prostate brachytherapy. Int J Radiat Oncol Biol Phys 2010;78:781e786. [10] Sloboda RS, Pedersen JE, Hanson J, Halperin RM. Dosimetric consequences of increased seed strength for I-125 prostate implants. Radiother Oncol 2003;68:295e297. [11] Su Y, Davis BJ, Furutani KM, et al. Dosimetry accuracy as a function of seed localization uncertainty in permanent prostate brachytherapy: increased seed number correlates with less variability in prostate dosimetry. Phys Med Biol 2007;52:3105. [12] Narayana V, Troyer S, Evans V, et al. Randomized trial of high-and low-source strength 125 I prostate seed implants. Int J Radiat Oncol Biol Phys 2005;61:44e51. [13] Mahdavi SS, Morris WJ, Salcudean SE, et al. Combining multimodality imaging and transperineal mapping biopsy to guide the focal application of low-dose-rate brachytherapy for prostate cancer: an ethics approved pilot study. Brachytherapy 2014;13:S122eS123. [14] Paterson R, Parker HM. A dosage system for interstitial radium therapy. Br J Radiol 1938;11:252e266. [15] Narayana V, Roberson PL, Winfield RJ, et al. Optimal placement of radioisotopes for permanent prostate implants. Radiology 1996;199:457e460. [16] Bott SR, Henderson A, Halls JE, et al. Extensive transperineal template biopsies of prostate: modified technique and results. Urology 2006;68:1037e1041.
-
(2016)
-
[17] Bucci J, Spadinger I, Hilts M, et al. Urethral and periurethral dosimetry in prostate brachytherapy: is there a convenient surrogate? Int J Radiat Oncol Biol Phys 2002;54:1235e1242. [18] Spadinger I, Hilts M, Keyes M, et al. Prostate brachytherapy postimplant dosimetry: a comparison of suture-embedded and loose seed implants. Brachytherapy 2006;5:165e173. [19] Sugawara A, Nakashima J, Shigematsu N, et al. Prediction of seed migration after transperineal interstitial prostate brachytherapy with I-125 free seeds. Brachytherapy 2009;8:52e56. [20] Saibishkumar EP, Borg J, Yeung I, et al. Sequential comparison of seed loss and prostate dosimetry of stranded seeds with loose seeds in 125 I permanent implant for low-risk prostate cancer. Int J Radiat Oncol Biol Phys 2009;73:61e68. [21] Usmani N, Chng N, Spadinger I, Morris WJ. Lack of significant intraprostatic migration of stranded iodine-125 sources in prostate brachytherapy implants. Brachytherapy 2011;10:275e285. [22] Rivard MJ, Coursey BM, DeWerd LA, et al. Update of AAPM Task Group No. 43 Report: a revised AAPM protocol for brachytherapy dose calculations. Med Phys 2004;31:633e674. [23] Roberson PL, Narayana V, McShan DL, et al. Source placement error for permanent implant of the prostate. Med Phys 1997;24: 251e257. [24] Badwan HO, Shanahan AE, Adams MA, et al. AnchorSeed for the reduction of source movement in prostate brachytherapy with the Mick applicator implant technique. Brachytherapy 2010;9:23e26. [25] Gao M, Wang JZ, Nag S, Gupta N. Effects of seed migration on postimplant dosimetry of prostate brachytherapy. Med Phys 2007;34: 471e480. [26] Lobo JR, Moradi M, Chng N, et al. Use of needle track detection to quantify the displacement of stranded seeds following prostate brachytherapy. IEEE Trans Med Imaging 2012;31:738e748. [27] Nath R, Bice WS, Butler WM, et al. AAPM recommendations on dose prescription and reporting methods for permanent interstitial brachytherapy for prostate cancer: Report of Task Group 137. Med Phys 2009;36:5310e5322. [28] Otto K, Clark BG, Huntzinger C. Exploring the limits of spatial resolution in radiation dose delivery. Med Phys 2002;29:1823e1831. [29] Chng N, Spadinger I, Morris WJ, et al. Prostate brachytherapy postimplant dosimetry: automatic plan reconstruction of stranded implants. Med Phys 2011;38:327e342. [30] Taschereau R, Roy J, Pouliot J. Monte Carlo simulations of prostate implants to improve dosimetry and compare planning methods. Med Phys 1999;26:1952e1959. [31] Sloboda R, Usmani N, Pedersen J, et al. Time course of prostatic edema post permanent seed implant determined by magnetic resonance imaging. Brachytherapy 2009;8:126. [32] Usmani N, Martell K, Ghosh S, et al. Comparison of low and intermediate source strengths for 125 I prostate brachytherapy implants. Brachytherapy 2013;12:442e448. [33] Thomas CW, Kruk A, McGahan CE, et al. Prostate brachytherapy post-implant dosimetry: a comparison between higher and lower source density. Radiother Oncol 2007;83:18e24.
-
(2016)
-
Robustness to source displacement in dual air kerma strength planning for focal low-dose-rate brachytherapy of prostate cancer S. Sara Mahdavi1,*, Ingrid T. Spadinger2, Nicholas T. Chng3, W. James Morris1 1
Department of Radiation Oncology, Vancouver Centre, British Columbia Cancer Agency, Vancouver, BC, Canada 2 Department of Medical Physics, Vancouver Centre, British Columbia Cancer Agency, Vancouver, BC, Canada 3 Department of Medical Physics, Centre for the North, British Columbia Cancer Agency, Prince George, BC, Canada
ABSTRACT
PURPOSE: To describe the use of dual source strength implants for focal low-dose-rate brachytherapy. METHODS AND MATERIALS: An interneedle dual source strength planning strategy is described for focal low-dose-rate brachytherapy of the prostate. The implanted treatment plans were designed using peripheral (except near the rectum) needles loaded with high strength (0.9 U) sources and central needles loaded with low strength (0.4 U) sources (‘‘interneedle’’ dual strength planning). This approach has been applied for focally treating 3 patients. In this article, we compare the characteristics and robustness to source motion of interneedle dual strength planning with four alternative planning strategies (single strength high, low, and intermediate, and intraneedle dual strength) on 50 simulated cases. RESULTS: Interneedle dual source strength planning results in greater robustness to source motion and overall lower seed and needle density compared to the standard low source strength planning currently used in our centre. This planning approach is also significantly superior to single strength high, single strength intermediate and intraneedle dual strength planning strategies in terms of high dose to the urethral avoidance structure. CONCLUSIONS: The use of interneedle dual source strength treatment plans for focal low-doserate brachytherapy is possibly the practical solution for limiting the density of sources required to deliver the prescribed dose while limiting proximity of high strength sources to organs at risk. Ó 2016 American Brachytherapy Society. Published by Elsevier Inc. All rights reserved.
Keywords:
Low-dose-rate brachytherapy; Prostate cancer; Dual source strength; Robustness; Source displacement
Introduction Focal treatment for prostate cancer, in which only diseased regions of the gland are treated, is an emerging topic of interest. By sparing healthy prostate tissue, fewer adverse effects on quality of life, compared to conventional wholegland treatment methods, are expected. However, in order for this approach to become widely acceptable as a standard Received 7 March 2016; received in revised form 8 April 2016; accepted 16 April 2016. Financial disclosure: This study was funded by the BC Cancer Foundation (DRG01829). Conflict of interest: The authors report no proprietary or commercial interest in any product mentioned or concept discussed in this article. * Corresponding author. Department of Radiation Oncology, Vancouver Centre, British Columbia Cancer Agency, 600 W 10th Ave., V5Z 4E6, Vancouver, BC, Canada. Tel.: þ1 604 877 6000; fax: þ1 604 877 0505. E-mail address: [email protected] (S.S. Mahdavi).
treatment option, it needs accurate methods for detecting and localizing cancer within the prostate, clear definitions of eligibility criteria and treatment success, safe and reliable ablation methods, and sufficient long-term followup results. Low-dose-rate prostate brachytherapy (LDR-PB) is a favorable ablation method for focal therapy as it can be easily modified to treat portions of the gland, as done in standard practice for salvage treatments and reimplantations. However, when planning for the small target volumes usually seen in focal treatment, obtaining the desired coverage with a reasonable source density while avoiding overdose to the rectum and urethra poses new challenges that are less easily met with a uniform source strength plan. Whole-gland LDR-PB treatment planning strategies vary among centers in terms of needle placement and loading, source (seed) strength, and the isotope used, with the common goals of improving dose coverage and plan robustness, while simplifying treatment delivery. General
1538-4721/$ - see front matter Ó 2016 American Brachytherapy Society. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.brachy.2016.04.388
2
S.S. Mahdavi et al. / Brachytherapy
guidelines have been developed to aid practitioners in managing treatment. The American Brachytherapy Society recommends a prescription dose of 140e160 Gy for 125I sources as a monotherapy option for National Comprehensive Cancer Networkedefined low- and intermediate-risk prostate cancer (1, 2). The original planning approach for delivering this dose was the use of a purely uniform seed distribution by the Seattle group (3). This eventually evolved into a ‘‘peripheral,’’ ‘‘modified peripheral,’’ or ‘‘modified uniform loading’’ distribution to avoid the high central doses (O400 Gy) caused by uniform loading, which have been associated with urethral toxicity (4e6). There is currently no consensus on the optimal source strength (or, alternatively, optimal source density). When treating the whole gland, acceptable dose coverage using the modified peripheral loading approach is achievable with single source strength plans of both low and high source density (7e10). In the RTOG clinical trials, 0.3e0.55 U is the reported range of air kerma strengths for 125I seeds (1). Low source strength (LSS) (i.e., high source density) plans lead to better dose homogeneity, smaller hot spots, and less influence on dosimetry in case of seed migration or implantation errors (11), whereas high source strength (HSS) plans provide better dose coverage with fewer seeds, hence, less prostate trauma, and less operating time and procedure costs (12). In whole-gland treatment, the cumulative dose from peripheral sources provides interior coverage without the need for source placement in the prostate center, close to the urethra. However, when the target volume is small, as in the case of focal treatment, the standard air kerma strength used in our center (~0.4 U) leads to a very dense plan. Higher air kerma strength may be used to reduce source density, but there are concerns that the consequences of source placement errors or source motion, which are commonly seen postimplant, might be more severe in a high strength plan. In addition, the modified peripheral approach is not feasible because the urethra is likely to lie adjacent to the planning target volume (PTV) periphery. To overcome these issues, we propose a dual source strength (DSS) planning strategy for focal low-dose-rate prostate brachytherapy (focal LDR-PB) in which needles are loaded with either high strength seeds or low strength seeds. Three patients have been treated using this strategy in the British Columbia Cancer Agency focal LDR-PB pilot study (13). The concept of uneven radiation distribution has been discussed since the 1930s (14) and was investigated for prostate implants, in addition to the effect of source placement errors, by Narayana et al. (15) in the mid1990s. However, to the best of our knowledge, this is the first report of a practical application of dual strength plans for focal treatment of prostate cancer. In this article, we provide a brief overview of the focal LDR-PB pilot study and focus on the robustness of dual source planning to source motion compared to four alternative planning strategies.
-
(2016)
-
Methods The focal LDR-PB pilot study The focal LDR-PB pilot study at the British Columbia Cancer Agency Vancouver Centre aims to determine the feasibility of applying focal LDR-PB as a treatment option for men with low risk prostate cancer. Institutional ethics approval was acquired before recruitment, and 3 patients have received focal LDR-PB. Consented patients are screened with multiparametric magnetic resonance imaging (mpMRI) multiparametric transrectal ultrasound (mpTRUS) imaging and transperineal template mapping biopsy (TTMB) (16). TTMB is used for cancer localization and is performed in the operating suite using a standard LDR-PB setup with a 5 5 mm brachytherapy grid. Based on the volume of the prostate, 20e50 cores are extracted, approximately one every 1 cc of the prostate volume, and their locations in the axial plane and craniocaudal depths (with a precision of 0.5 cmethe TRUS image spacing) are recorded. The biopsy samples are reviewed by a single pathologist to identify the presence of cancer, Gleason score, tumor location, and length of involvement within the core. Eligibility to receive focal LDR-PB includes TTMB results of #4 positive cores (Gleason score #3 þ 4) within a single lobe and spanning no more than two adjacent sectors (base, midgland, apex). Based on the cancerous core locations and tumor extension within the cores, the gross tumor volume is mapped onto the TRUS images collected immediately before TTMB. The PTV is defined as the gross tumor volume plus a 5e8 mm margin in all three dimensions. Extraprostatic extension of the PTV is allowed up to 4e5 mm from the prostate boundary except for the posterior. The rectum is delineated from 1 cm superior to the base to 1 cm inferior to the apex. An intraprostatic avoidance structure is outlined as a circle of 5-mm radius based on the ‘‘deviated’’ urethra model recommendations in Bucci et al. (17), which assumes a 30 anterior deviation in the superior half of the prostate. This structure is used as a surrogate for the urethra. Contouring and PTV mapping are performed with the MIM Symphony (MIM Software, Cleveland, OH), and planning is performed with the VariSeed 8.0 planning system (Varian Medical Systems, Palo Alto, CA). Oncoseed 6711 sources (Oncura Inc., Arlington Heights, IL) were used to generate the treatment plans. In our ‘‘interneedle dual source strength’’ plans, high strength seeds (0.8e0.9 U) were used in needles located in the PTV periphery (except near the rectum and urethral avoidance region) and low strength seeds (0.4 U) in needles located elsewhere, with the goal of meeting standard dosimetric constraints for V100 and D90. Source strength was kept uniform within needles largely for logistical reasons, as we wanted to use stranded, preloaded seed trains in all the needles. Implantation was performed with a standard low-dose-rate brachytherapy setup.
S.S. Mahdavi et al. / Brachytherapy
Further details of imaging procedures and clinical implementation and outcomes are the subject of a separate article. Dosimetric robustness to source displacement Seed displacement during and in the weeks after prostate implantation has been widely reported. In extreme cases, seeds can be completely lost through ejaculation, transurethral routes, or through the venous plexus to the lung, but the incidence of such events is reduced if stranded seeds are used (18e21) among others. Smaller displacements from the planned positions are more common and are caused by implantation error due to factors such as needle deflection, gland motion, and prostate edema. Regardless of the cause, unpredictable changes in the seed locations from the original plan may significantly affect dose coverage, leading to overdosing adjacent critical structures and/or underdosing the target region. We have analyzed the potential effect of local source displacements on dosimetric parameters on 50 simulated cases. We compared the robustness of interneedle dual source strength (DSS) planning to source motion with four alternative planning strategies: high source strength (HSS), intermediate source strength (ISS), low source strength (LSS), and ‘‘intraneedle’’ dual source strength (D2SS) planning. To generate the 50 simulated cases, sample PTVs were manually created from the whole-gland contours of 50 patients randomly chosen, to cover a range of PTV volumes, from our large pool of standard whole prostate implant patients. Hemigland PTVs were assumed, as an approximate
Fig. 1. Illustration of a sample PTV (cyan) defined for robustness analysis on simulated cases. A margin of 1e2 mm is drawn inside the urethral avoidance structure (green). The midportion of the gland is excluded by drawing the PTV 15e20 off from the midline (dashed). PTV [ planning target volume. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)
-
(2016)
-
3
simulation of the PTVs seen in the three focally treated patients, and to reflect our criteria of focally treating cases with only unilaterally identified cancer (Fig. 1). These were defined as half of the prostate gland (the clinical target volume) minus a portion of the gland midline (approximately 15e20 off from the ‘‘D’’ grid column located on the prostate midline) and a portion of the urethral avoidance structure, with no extraprostatic margin. A margin of 1e2 mm inside the urethral avoidance structure was included, and the entire length of the prostate from the base to the apex was covered. Air kerma strengths used were: 0.941 U for HSS plans, 0.417 U for LSS plans, 0.679 U (the average of 0.941 and 0.417) for ISS plans, and a combination of 0.941 U and 0.417 U for DSS and D2SS plans. D2SS plans differed from DSS plans in having two air kerma strengths within the needles if necessary. The D2SS planning strategy was similar to the DSS strategy, in that high strength sources were not placed close to the urethral avoidance region or rectum, but they could be placed in predominately low source strength trains in slices where the critical structure was not close to the needle path. Similarly, some seeds in predominately high strength needles were replaced by low strength seeds to achieve better conformity or reduce the V150% inside the PTV. Plans were created by manual optimization with the goal of meeting the following constraints: minimum peripheral dose 5 144 Gy, PTV V100% $95% of PTV volume, PTV V150% between 50% and 70% of PTV volume, PTV D90% between 95% and 125% of prescription dose, urethral avoidance structure V100% #30% of the structure’s volume, and rectal V50% #5% of rectal volume. Rectal and urethral dose constraints are lower than typical values used for whole-gland monotherapy and were established during the course of planning for smaller and unilateral PTVs. Attempts were made to minimize the overall number of needles (i.e., avoiding needles with just one seed, and the average number of seeds per needle was ~3 or greater) and maintain uniform 1 cm source spacing as much as possible. Source spacing within a needle was always at least 1 cm. Changes in the planned dosimetric parameters due to source displacements were studied using a Monte Carlo approach. In this method, for each of the 50 simulated cases, random displacements were applied to each individual seed and changes in the dosimetric constraints were measured. This was repeated over 500 iterations, and statistics were derived. MATLAB (The MathWorks, Inc. MA) was used to apply source displacements in plans exported from VariSeed, to compute dosimetric parameters, and to obtain robustness results. MATLAB dose calculation was in accordance with the AAPM TG-43 (22), assuming a point source, and the results were validated against VariSeed. The random displacement statistics used were 0 2.5 mm (mean and one standard deviation) lateral and posterior motions and 4 3 mm inferior motion. A normal distribution of displacements was assumed in the modeling.
4
S.S. Mahdavi et al. / Brachytherapy
These values were chosen on the basis of posttreatment seed displacements measured in the literature (21,23e26). The dosimetric parameters reported in the Results for the rectum and urethral avoidance region are different from the standard metrics recommended by AAPM (27) to better describe our focal plans created on smaller PTVs. CI100 is the conformity index, ideally being 1, and is defined as (28): CI100 5
Vr ðVu þ Vh Þ Vr
where Vr is the PTV volume, Vu the portion of the PTV receiving less than 100% of the prescribed dose, and Vh is the volume of the region outside the PTV that is receiving $ 100% of the prescribed dose. D1cc is the dose received by 1 cc of the structure, and Vxx is the volume of the structure receiving xx% of the prescribed dose, reported in either cc or percent of the structure volume. D90 is the minimum dose delivered to 90% of the structure.
-
(2016)
-
Results Figure 2 shows an example of the five plan types for a simulated patient. Table 1 provides the description of the five plan types including the average number of needles and sources used, and PTV, urethral avoidance, and rectal volumes and dosimetric parameters. As summarized in this table, the average number of needles and sources used in the LSS plans is approximately 2 times that of HSS and 1.5 times the number used in the other three plan types. Per cc of PTV, lower air kerma strength was required in LSS plans, compared to HSS plans, to achieve the desired dosimetric constraints, due to the higher degree of conformity that could be achieved with the low strength sources. Results of robustness to source displacements in simulated plans are summarized in Fig. 3. The mean and 95% confidence interval of the change in each dosimetric parameter, over the 50 cases, are displayed. A negative value indicates a decrease in the dose parameter after displacing the sources.
Fig. 2. Example of the five plan types created for one of the simulated patients. In the needle distribution (Row 1), the icons triangle, square, diamond, and inverted triangle represent needles planned to retraction planes 0.5 cm, 1 cm, 1.5 cm, and 2 cm inferior to the base. Source distributions are shown for the base, midgland, and apex (Rows 2e4). The number of sources and needles used in this case are as follows: DSS (18 high strength and 17 low strength sources, 5 high and 4 low strength needles), HSS (26 sources and 8 needles), LSS (51 sources and 14 needles), ISS (33 sources and 10 needles), and D2SS (17 high strength and 15 low strength sources in 9 needles).
S.S. Mahdavi et al. / Brachytherapy
-
(2016)
-
5
Table 1 Description of the interneedle dual source strength (DSS), high source strength (HSS), low source strength (LSS), intermediate source strength (ISS), and intraneedle dual source strength (D2SS) plans and the dosimetric parameters in the 50 simulated cases Plan type
DSS
HSS
LSS
ISS
D2SS
Source strength (U) Total source strength/PTV volume # Needles (H, L)b # Sources (H, L) PTV vol. (cc) V100 (%) V150 (%) D90 (Gy) CI100 Rectum vol. (cc) D1cc (Gy) Urethral avoid vol. (cc) V125 (cc) V30 (%)
0.941, 0.417 1.42 0.34a 4.7 1.0; 4.7 1.1 15.3 4.4; 16.6 5.2
0.941 1.47 0.36 7.4 1.1 23.2 6.0
0.417 1.29 0.31 13.7 2.3 45.9 12.2
0.679 1.38 0.33 9.5 1.6 30.4 8.2
0.941, 0.417 1.40 0.31 9.0 1.4 16.3 4.7; 13.5 4.5
16.3 93.57 63.41 155.74 0.37
7.3, range (4.2e43.8) 1.34 94.64 4.09 62.89 4.55 159.05 0.21 0.35
1.10 5.15 4.11 0.23
93.71 56.89 155.11 0.48
1.36 4.99 4.18 0.16
94.17 61.45 156.93 0.42
1.43 3.87 4.66 0.20
94.15 61.05 156.59 0.42
1.19 4.43 4.08 0.21
19.2 8.5 64.24 12.2
69.06 15.74
64.11 12.98
66.38 13.64
63.97 11.86
4.2 0.6 0.03 0.02 84.37 4.57
0.04 0.02 87.19 4.36
0.04 0.02 81.99 4.00
0.04 0.02 85.17 4.38
0.03 0.02 85.74 5.08
PTV 5 planning target volume. a Mean and standard deviation. b (H, L): high strength and low strength, if applicable.
The plots illustrate less robustness of the LSS plans to source displacement in terms of PTV V100%, V150%, and D90. CI100 robustness in LSS plans is also less than
that of DSS and HSS plans. There is no significant difference in PTV dose metrics among the other four plan types. DSS plans show significantly less increase in
Fig. 3. Robustness of dual, high, low, intermediate, and interneedle dual source strength plans to source displacement. Plots display the average and 95% confidence interval of change in PTV V100%, V150%, D90 (Gy), and CI100, urethral avoidance V125 (cc) and V30%, and rectum D1cc (Gy) over 50 cases (aeg). For each case, 500 iterations of source displacement were applied. PTV [ planning target volume; DSS [ dual source strength; HSS [ high source strength; LSS [ low source strength; ISS[ intermediate source strength; D2SS [ intraneedle dual source strength.
6
S.S. Mahdavi et al. / Brachytherapy
-
(2016)
-
Fig. 3. (continued).
the urethral V125% compared to the HSS, ISS, and D2SS plans. We attribute the larger 95% confidence interval in the D2SS plans, especially in the rectum D1cc, to greater plan-to-plan variability in the anatomical location of high strength vs. low strength seeds, compared to the other four plan types. By comparing the original urethral V30% values (Table 1) to the changes in Fig. 3f, we see that the planned original V30% seems to have a negative correlation with the change in urethral V30% due to source displacement. However, the changes to V30% are generally very small, whereas V30% is quite large, so this observation may simply be an artifact of the particular planning strategy used. In summary, robustness of DSS plans is significantly superior to ISS, HSS, and D2SS plans in terms of high dose (as measured by V125%) to the urethral avoidance structure, which overlaps the PTV. This, combined with the overall lower seed and needle density required by DSS plans in comparison to LSS plans, as well as greater robustness with respect to PTV coverage, leads us to conclude that they are superior, for focal planning, to adopting an ISS or HSS strategy, or to retaining the LSS strategy that is our current standard for whole prostate implantation.
Discussion We would like to emphasize that postimplant dosimetry of dual- or mixed-air kerma strength implants is not possible unless the strength of each source on postimplant imaging (e.g., CT) can be clearly identified. Our unique plan reconstruction software (29) can match seeds from postimplant images to the corresponding strands in the plans, with a high accuracy of 97.7%. This software is currently being routinely used for whole-gland plan reconstruction of stranded implants and can be easily applied to focal LDR-PB plans. Seed displacement in LDR-PB has been studied by various groups. A common method of measuring seed motion is postimplant seed localization (e.g., from CT imaging or intraoperative fluoroscopy), followed by some form of seed cloud registration between the planned seed cloud and postimplant seed cloud (26, 30). To measure targeting errors (i.e., errors resulting from intentional or unintentional misplacement of the seeds by the radiation oncologist), the postimplant seed cloud is derived immediately after the implant. At approximately 1 month postimplant (21), seed migration occurs. However, by registering the seed clouds alone, it is not possible to measure the displacement of seeds with respect to the boundary. This is commonly seen, especially in the craniocaudal direction.
S.S. Mahdavi et al. / Brachytherapy
Measuring this type of displacement is challenging, and very few groups have reported it (23e25). The displacement distribution used in this article is chosen to cover various displacement values reported in the literature and to include an approximation of a systematic shift, which has the greatest potential to affect PTV coverage. Inferior misplacement of the sources due to targeting errors is generally considered to be the most likely type of systematic displacement for focal LDR-PB, so this was chosen as the systematic shift. Derivation of a more detailed region-based seed displacement model and understanding edema in focal LDR-PB (similar to that reported in Sloboda et al. (31) for whole-gland treatment) and its effect on postimplant dosimetry could further refine robustness analysis of focal LDR-PB planning. Furthermore, we have applied random displacements to each individual seed as opposed to the whole strands. Comparison of strand vs. loose seed displacement, especially in the craniocaudal direction, is also worth investigating. Our results show that the use of higher strength sources, alone or combined with lower strength sources, can lead to fewer needles; however, an increase in the source strength can have negative affects on the dose delivered to critical structures, particularly those that overlap the PTV. The potential benefits of high strength plans have been shown in the literature, specifically the reduced costs and fewer needle puncture wounds, although there are inconsistent results with respect to adverse effects associated with needle trauma. Martin et al. (8) produced excellent implants with fewer high strength seeds (0.8 U). By using an average number of seeds and needles of 54.6 14.4 and 17.6 3.1 and inverse planning for optimal seed positioning, they report a 60-month biochemical failure-free survival of 94.6% according to the nadir þ 2 ng/mL definition, on 396 patients (70% of which were low risk). In a study on 105 patients implanted with low strength seeds (0.44 mCi) and 94 implanted with higher strength seeds (0.61 mCi), Massucci et al. (9) report no statistical difference in patients’ International Prostate Symptom Score or dosimetric parameters calculated for the rectum based on Day 30 CT; however, the quality of implants measured at Day 30 CT (V100 and D90) was significantly better in the higher strength cohort, although this group also showed higher V150 and V200. They observed no significant difference in short-term urinary toxicity (up to 8 months), although increased dose inhomogeneity using high strength sources may lead to increased long-term urinary toxicity. Beaulieu et al. (7) studied the effect of a wide range of air kerma strengths on the dosimetric properties of plans created by inverse planning and their robustness to seed displacement and migration. They report that with a properly tuned inverse planning algorithm, the choice of preferred seed strength, in a range of up to about 0.7 mCi (0.89 U), is open in terms of adequate dose coverage and protection for the
-
(2016)
-
7
organs at risk in the presence of seed placement errors. However, the placement of high air kerma strength seeds (0.9 U and above) becomes more critical to ensuring robustness to displacement and migration of seeds. Usmani et al. (32) report improved postoperative dosimetry when transitioning from low (median 0.398 U) to intermediate (median 0.494 U) strength sources, with fewer needles and sources used, and decreased need for catheterization. However, more rectal complications were observed in the intermediate source strength cohort. On the other hand, Thomas et al. (33) report no significant difference between low-density source implants (equivalent to high strength implants) and high-density source implants in any of the measured postimplant dosimetric parameters for the whole gland or quadrants, urethra, or rectum. The above studies all report on whole-gland treatment. We show that DSS plans have the benefit of requiring fewer needles and show better robustness of the dose delivered to critical structures, with respect to simulated seed displacements, compared to all other single source strength plan types. DSS planning is possibly the only way to plan small PTV sizes to achieve desirable results. As a final word, it is important to recognize the limitations of this study. Although we have attempted to include a fairly wide range of PTV sizes with varying degrees of proximity to the rectum, other factors, such as overlap with the urethral avoidance structure, were kept relatively constant. The full range of possible clinical scenarios with respect to PTVs location and organs’ at risk locations relative to the PTV was therefore not sampled. However, we can perhaps generalize our findings by suggesting that the results for the urethral avoidance structure represent the general trend for an overlapping organ at risk, whereas the results for the rectum represent the trend for a proximal, but nonoverlapping, organ at risk. Furthermore, although all planning was done by one individual subject to strict dosimetric constraints and seed placement and spacing criteria, it may be argued that a manual planning approach is too easily subject to bias. An inverse-planning approach, as has been used by other groups that have undertaken similar studies (7, 10), may be deemed preferable. Although we considered this approach, we concluded that the tuning, constraints, and acceptance criteria used with an inverse planning algorithm would also not be free of a subjective component. Conclusions Using DSS planning is a feasible approach for focal LDR-PB and seems to result in improved plans in terms of number of sources and dose coverage, as well as increased robustness to source displacements. Unambiguous identification of the seeds and association to the correct source strength is necessary for practical implementation of this approach. This strategy could be extended to wholegland treatment with potentially similar benefits.
8
S.S. Mahdavi et al. / Brachytherapy
References [1] Davis BJ, Horwitz EM, Lee WR, et al. American Brachytherapy Society consensus guidelines for transrectal ultrasound-guided permanent prostate brachytherapy. Brachytherapy 2012;11:6e19. [2] Rivard MJ, Butler WM, Devlin PM, et al. American Brachytherapy Society recommends no change for prostate permanent implant dose prescriptions using iodine-125 or palladium-103. Brachytherapy 2007;6:34e37. [3] Ragde H, Blasko JC, Grimm PD, et al. Interstitial iodine-125 radiation without adjuvant therapy in the treatment of clinically localized prostate carcinoma. Cancer 1997;80:442e453. [4] Nath R, Anderson LL, Meli JA, et al. Code of practice for brachytherapy physics: report of the AAPM Radiation Therapy Committee Task Group No. 56. Med Phys 1997;24:1557e1598. [5] Wallner K, Roy J, Harrison L. Dosimetry guidelines to minimize urethral and rectal morbidity following transperineal I-125 prostate brachytherapy. Int J Radiat Oncol Biol Phys 1995;32:465e471. [6] Butler WM, Merrick GS, Lief JH, Dorsey AT. Comparison of seed loading approaches in prostate brachytherapy. Med Phys 2000;27: 381e392. [7] Beaulieu L, Archambault L, Aubin S, et al. The robustness of dose distributions to displacement and migration of 125 I permanent seed implants over a wide range of seed number, activity, and designs. Int J Radiat Oncol Biol Phys 2004;58:1298e1308. [8] Martin AG, Roy J, Beaulieu L, et al. Permanent prostate implant using high activity seeds and inverse planning with fast simulated annealing algorithm: a 12-year Canadian experience. Int J Radiat Oncol Biol Phys 2007;67:334e341. [9] Masucci GL, Donath D, Tetreault-Laflamme A, et al. Comparison between high and low source activity seeds for I-125 permanent seed prostate brachytherapy. Int J Radiat Oncol Biol Phys 2010;78:781e786. [10] Sloboda RS, Pedersen JE, Hanson J, Halperin RM. Dosimetric consequences of increased seed strength for I-125 prostate implants. Radiother Oncol 2003;68:295e297. [11] Su Y, Davis BJ, Furutani KM, et al. Dosimetry accuracy as a function of seed localization uncertainty in permanent prostate brachytherapy: increased seed number correlates with less variability in prostate dosimetry. Phys Med Biol 2007;52:3105. [12] Narayana V, Troyer S, Evans V, et al. Randomized trial of high-and low-source strength 125 I prostate seed implants. Int J Radiat Oncol Biol Phys 2005;61:44e51. [13] Mahdavi SS, Morris WJ, Salcudean SE, et al. Combining multimodality imaging and transperineal mapping biopsy to guide the focal application of low-dose-rate brachytherapy for prostate cancer: an ethics approved pilot study. Brachytherapy 2014;13:S122eS123. [14] Paterson R, Parker HM. A dosage system for interstitial radium therapy. Br J Radiol 1938;11:252e266. [15] Narayana V, Roberson PL, Winfield RJ, et al. Optimal placement of radioisotopes for permanent prostate implants. Radiology 1996;199:457e460. [16] Bott SR, Henderson A, Halls JE, et al. Extensive transperineal template biopsies of prostate: modified technique and results. Urology 2006;68:1037e1041.
-
(2016)
-
[17] Bucci J, Spadinger I, Hilts M, et al. Urethral and periurethral dosimetry in prostate brachytherapy: is there a convenient surrogate? Int J Radiat Oncol Biol Phys 2002;54:1235e1242. [18] Spadinger I, Hilts M, Keyes M, et al. Prostate brachytherapy postimplant dosimetry: a comparison of suture-embedded and loose seed implants. Brachytherapy 2006;5:165e173. [19] Sugawara A, Nakashima J, Shigematsu N, et al. Prediction of seed migration after transperineal interstitial prostate brachytherapy with I-125 free seeds. Brachytherapy 2009;8:52e56. [20] Saibishkumar EP, Borg J, Yeung I, et al. Sequential comparison of seed loss and prostate dosimetry of stranded seeds with loose seeds in 125 I permanent implant for low-risk prostate cancer. Int J Radiat Oncol Biol Phys 2009;73:61e68. [21] Usmani N, Chng N, Spadinger I, Morris WJ. Lack of significant intraprostatic migration of stranded iodine-125 sources in prostate brachytherapy implants. Brachytherapy 2011;10:275e285. [22] Rivard MJ, Coursey BM, DeWerd LA, et al. Update of AAPM Task Group No. 43 Report: a revised AAPM protocol for brachytherapy dose calculations. Med Phys 2004;31:633e674. [23] Roberson PL, Narayana V, McShan DL, et al. Source placement error for permanent implant of the prostate. Med Phys 1997;24: 251e257. [24] Badwan HO, Shanahan AE, Adams MA, et al. AnchorSeed for the reduction of source movement in prostate brachytherapy with the Mick applicator implant technique. Brachytherapy 2010;9:23e26. [25] Gao M, Wang JZ, Nag S, Gupta N. Effects of seed migration on postimplant dosimetry of prostate brachytherapy. Med Phys 2007;34: 471e480. [26] Lobo JR, Moradi M, Chng N, et al. Use of needle track detection to quantify the displacement of stranded seeds following prostate brachytherapy. IEEE Trans Med Imaging 2012;31:738e748. [27] Nath R, Bice WS, Butler WM, et al. AAPM recommendations on dose prescription and reporting methods for permanent interstitial brachytherapy for prostate cancer: Report of Task Group 137. Med Phys 2009;36:5310e5322. [28] Otto K, Clark BG, Huntzinger C. Exploring the limits of spatial resolution in radiation dose delivery. Med Phys 2002;29:1823e1831. [29] Chng N, Spadinger I, Morris WJ, et al. Prostate brachytherapy postimplant dosimetry: automatic plan reconstruction of stranded implants. Med Phys 2011;38:327e342. [30] Taschereau R, Roy J, Pouliot J. Monte Carlo simulations of prostate implants to improve dosimetry and compare planning methods. Med Phys 1999;26:1952e1959. [31] Sloboda R, Usmani N, Pedersen J, et al. Time course of prostatic edema post permanent seed implant determined by magnetic resonance imaging. Brachytherapy 2009;8:126. [32] Usmani N, Martell K, Ghosh S, et al. Comparison of low and intermediate source strengths for 125 I prostate brachytherapy implants. Brachytherapy 2013;12:442e448. [33] Thomas CW, Kruk A, McGahan CE, et al. Prostate brachytherapy post-implant dosimetry: a comparison between higher and lower source density. Radiother Oncol 2007;83:18e24.