Effect of imaging frequency on PTV margins and geographical miss during image guided radiation therapy for prostate cancer

    Effect of Imaging Frequency on PTV Margins and Geographical Miss during Image Guided Radiotherapy for Prostate Cancer Meetakshi Gupta, Poonam Gamre, Sadhana Kannan, Ganesh Rokde, Rahul Krishnatry, Vedang Murthy PII: DOI: Reference:

S1879-8500(17)30267-9 doi: 10.1016/j.prro.2017.09.010 PRRO 823

To appear in:

Practical Radiation Oncology

Received date: Revised date: Accepted date:

26 August 2017 13 September 2017 20 September 2017

Please cite this article as: Gupta Meetakshi, Gamre Poonam, Kannan Sadhana, Rokde Ganesh, Krishnatry Rahul, Murthy Vedang, Effect of Imaging Frequency on PTV Margins and Geographical Miss during Image Guided Radiotherapy for Prostate Cancer, Practical Radiation Oncology (2017), doi: 10.1016/j.prro.2017.09.010

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Title. Effect of Imaging Frequency on PTV Margins and Geographical Miss during Image Guided Radiotherapy for Prostate Cancer Short running Title: Imaging frequency and PTV margins in prostate cancer treated with IGRT

T

Author names and affiliations. 1,

MD Radiation Oncology, Department of Radiation Oncology 2

M.Sc. Physics, Department of Radiation Oncology 2,

MA NU

Sadhana Kannan [email protected]

SC

Poonam Gamre , [email protected]

RI P

Meetakshi Gupta [email protected]

M. SC. Biostatistics, Department of Biostatistics 2,

Ganesh Rokde [email protected]

PG Diploma Radiotherapy, Department of Radiation Oncology 1,

ED

Rahul Krishnatry [email protected]

MD Radiation Oncology, Department of Radiation Oncology 2,

PT

Vedang Murthy [email protected]

CE

MD Radiation Oncology, Department of Radiation Oncology

AC

1. Tata Memorial Centre, Mumbai, Maharashtra, India- 400012 2. Tata Memorial Centre- Advanced Centre for Treatment, Research and Education in Cancer, Kharghar, Navi Mumbai, Maharashtra, India- 410210 Corresponding author: Dr. Vedang Murthy Professor Department of Radiation Oncology Tata Memorial Centre- Advanced Centre for Treatment, Research and Education in Cancer (ACTREC) Sector 22, Utsav Chowk - CISF Road, Kharghar, Navi Mumbai, Maharashtra 410210 India E-mail: [email protected]

ACCEPTED MANUSCRIPT Disclosure: There was no funding source involved. We do not have any conflict of interest to disclose.

T

Background: The relationship between frequency of imaging during Image Guided Radiotherapy (IGRT) and

RI P

Planning Target Volume (PTV) margin remains unclear. This issue is of practical significance given resource and time intensive nature of image guided radiation therapy (RT). The purpose of this study was to evaluate PTV

SC

margins with predefined and commonly used less-than-daily IGRT schedules using data obtained from patients treated with daily IGRT for prostate cancer.

MA NU

Methods and materials: Daily set up error and 3 dimensional daily alignment data for a total of 108 consecutive patients with prostate cancer treated with 2700 fractions of daily image guidance on Tomotherapy was retrospectively analysed. Five IGRT scenarios were simulated, namely alternate day, twice weekly, once weekly, first 3 days only and no image guidance. The daily alignment data was modelled to simulate the 5

ED

predefined scenarios by applying appropriate corrections to determine the PTV margin for each image guidance scenario. The data was also analysed to predict possible geographical miss in any direction using two

PT

frequently used PTV margins of 7 mm and 5 mm, for all the scenarios.

CE

Results: Decreasing frequency of image guidance increased the mean systematic error and the standard deviation of the systematic error. With decrease in image guidance frequency, an increase in PTV margins was

AC

required to achieve adequate coverage of the clinical target volume. With reduction in image guidance from 50 to 12%, a gradual increase in percentage of fractions with predicted geographical miss using an isotropic PTV margin of 7 or 5 mm was seen. With every 15% decrease in imaging, a 5% increased risk of geographical miss was estimated. Conclusion: The use of less than daily image-guided RT requires larger PTV margins for patients treated with IMRT for prostate cancer. With every 15% reduction a 5% increased risk of geographical miss was estimated.

ACCEPTED MANUSCRIPT Introduction: Inter- and intra-fraction organ motion and deformation pose a challenge in planning and implementation of

T

radiotherapy for prostate cancer. The prostate is a non-rigid and deformable structure and its position varies in

RI P

three dimensions depending on bladder and rectal filling(1). With increasing dose and dose conformity with Intensity Modulated Radiotherapy (IMRT), the uncertainties in dose delivery to the target and organs at risk

SC

(OAR) increase. This uncertainty can be reduced with the application of Image Guided Radiotherapy (IGRT). With increasing use of hypofractionation in radiotherapy of the prostate, IGRT has become essential in

MA NU

improving the therapeutic ratio as it improves the precision of treatment delivery and helps in reducing the errors.

Significant uncertainty exists regarding the frequency of IGRT used in clinical practice as technical precision needs to be balanced with an unacceptable workload on the system. Several authors have advocated the use

ED

of daily IGRT for prostate cancer patients due to large variations in intra- and inter-fraction prostate motion(2, 3). Less than daily IG schedules such as once or twice weekly and alternate day imaging are commonly used in

PT

clinical practice (4, 5)for logistic, commercial and technical reasons. This is however done without altering the PTV margin, assuming constant systematic errors on the days imaging is not performed (5). Some centres alter

CE

IGRT frequency after treatment starts by assessing geometric shifts in the beginning of treatment and either

AC

increasing or decreasing the frequency of imaging accordingly(6). The relationship between frequency of imaging and PTV margin size remains unclear although it is likely that a larger PTV is needed with a less than daily IGRT approach. Image guided treatment has the potential for false reassurance if used inappropriately, leading to inappropriate margin reduction. This issue is of practical significance given the resource-intensive and time-intensive nature of image guided RT(7). The purpose of this study was to evaluate the setup uncertainty with a number of predefined and commonly used temporal variations of less-than-daily image guided RT using data obtained from patients treated with daily imageguided RT for prostate cancer.

Materials and methods: Daily set up error and 3 dimensional daily alignment data for a total of 108 consecutive patients with nonmetastatic, high risk, localised prostate cancer treated with definitive intensity modulated radiotherapy using

ACCEPTED MANUSCRIPT Tomotherapy with daily image guidance, at a single institute was selected for the study. This provided baseline data from 2700 fractions, collected prospectively and used for analysis retrospectively in this study. The process of simulation, planning and treatment execution to obtain the data was uniform in all patients, and is

T

described below.

RI P

Simulation: Patients were simulated in the supine position, on a helical kilovoltage computed tomography (kV CT) scanner with a comfortably full bladder. To achieve this, patients were asked to void and then drink 500 ml

SC

water. Imaging was done after 30-45 minutes. Patients were asked to take a light diet for 1 week and laxative for 3 days prior to simulation to ensure an empty rectum. A knee rest was used for immobilisation without any

MA NU

other immobilisation devices. Non-contrast CT images of 2.5mm slice thickness were taken from first lumbar vertebra down to the mid-thigh. Three fiducials were placed over the pelvis just above the pubic symphysis for patient positioning. The marks were tattooed for future reference. The images were transferred to treatment planning station for contouring.

ED

Target delineation: Radiation Therapy Oncology Group (RTOG) guidelines were followed for prostate and nodal contouring. Gross Tumour Volume (GTV) was delineated on the CT images. Magnetic Resonance Imaging (MRI)

PT

scan images were registered with the CT images for target delineation. Clinical Target Volume (CTV) consisted of the GTV, whole prostate and seminal vesicles (whole organ if involved or else only base). If pelvic nodes

CE

were to be treated, bilateral external and internal iliac vessels were delineated and 7 mm margin was given

AC

around the vessels to generate nodal CTV. Bladder and rectum were delineated as OARs. PTV was generated by adding 7mm to the CTV in all directions. Planning: Each treatment was planned on Helical TomoTherapy Hi-Art® system (TomoTherapy Inc., Madison, WI), TomoPlan (TomoTherapy planning system) v. 4.2.0 for HT-IMRT. For each plan, the treatment field width (FW), pitch (defined as the ratio of the distance travelled per rotation to the axial FW used for treatment) and modulation factor (MF) were selected. The dose distribution for each beamlet was calculated by a convolution/superposition algorithm and iterative least-squares minimization was used to optimize the objective function. A fine calculation grid (256 × 256 pixels) was used both in the optimization and calculation processes. Organs at risk (OAR) objectives were described by a maximum dose, a DVH-based constraint and their respective penalties. Planned dose to primary PTV was 68 Gray in 25 fractions over 5 weeks and nodal PTV was 50 Gray in 25 fractions as a simultaneous integrated boost.

ACCEPTED MANUSCRIPT Treatment set up and delivery: Final plans were transferred to the treatment console for implementation. Bladder and rectum protocol was followed strictly daily during treatment as described for simulation. Patients were positioned on the Tomotherapy couch and the simulation position was reproduced. Initial alignment was

T

performed by aligning patients straight along the vertical laser and matching horizontal and sagittal room

RI P

lasers with the tattoo marks. A Megavoltage Cone Beam CT scan (MVCT) was acquired using a 3.5 MV photon beam and 3mm pitch. The MVCT scan width was adjusted to encompass the entire PTV. The MVCT image was

SC

fused with the planning CT image using automatic registration. Image superimposition was verified first by matching pelvic bones followed by soft tissue matching to ensure that the prostate was within the planned

MA NU

PTV. In cases where both the prostate and pelvic nodes were treated, position covering the whole of prostate was given precedence over nodal region coverage. Couch alignment data was acquired in 6 dimensions (vertical, lateral, longitudinal, roll, pitch and yaw). In presence of more than 2° pitch or yaw, realignment was warranted. Roll was accounted for and any value below 2° was accepted. The couch alignment parameters in 4

ED

dimensions (vertical, lateral, longitudinal and roll) were hence implemented for the day and treatment delivered. This was repeated for every fraction for the planned 25 fractions. Patient alignment and matching

PT

was done by the clinician for the first 3-5 fractions depending on reproducibility and by a trained therapist thereafter. Any untoward observations in alignment parameters were reported to the clinician and

AC

CE

appropriate measures were taken to correct the underlying issue.

IGRT scenarios: For the purpose of this work, five IGRT scenarios were simulated assuming varying imaging frequency. The daily alignment data was modelled to simulate the 5 assumed scenarios by applying appropriate corrections to determine the magnitude of the residual errors if IG was used less frequently than daily(8).

1) Alternate day IGRT-In this scenario, the patient would be imaged every other day. Each time imaging is done, the patient position was corrected according to the MVCT–kVCT registration. A running average shift was determined from the image guided (IG) fractions, and was subtracted from the shifts of the non IG days to get the residual shifts of the non IG fractions. The imaging frequency was 50%.

ACCEPTED MANUSCRIPT 2) Twice weekly IGRT- In this scenario, the patient would be imaged twice a week, e.g. one, three, six, eight fractions and so forth. After each imaging, a mean shift was calculated from all the previous corrections and applied to subsequent fractions. The imaging frequency was 40%.

T

3) Weekly IGRT- In this scenario, the patient would be imaged once a week. After each imaging, a mean

RI P

shift was calculated from all the previous corrections and applied to subsequent fractions. The imaging frequency was 20%.

SC

4) First 3 fractions only IGRT- In this scenario, the patient would be imaged only for the first 3 fractions. A mean shift was calculated from imaging during the first three fractions and then applied on all

MA NU

subsequent treatment days. The imaging frequency was 12%.

5) No IGRT- In this scenario each patient’s position was only corrected for a machine-specific systematic positioning error. Because no MVCT images were acquired, this scenario yielded the least workload. The imaging frequency was 0%.

ED

The residual shifts in each scenario were used to calculate the random and systematic errors.

PT

We also analysed the residual shifts for a possible geographical miss in any direction using two commonly used PTV margins of 7 mm and 5 mm, with daily IGRT. For each scenario, any of the 2700 fractions with a residual

CE

shift value of more than 7 mm was considered to have a potential geographical miss if image guidance was not

AC

used on that day. The same was done for an assumed PTV margins of 5 mm. Geographical miss in lateral, superoinferior and anteroposterior directions and also the overall effect in any direction were plotted against the imaging frequency in a line diagram to observe the trends.

Statistical analysis: Systematic (Σ) and random (σ) errors were calculated as per conventionally defined norms (ref). The systematic error was assessed by mean values of all the displacements for the whole population. For each patient, dispersion around the systematic displacement was calculated to assess the random displacement. For the whole population, the distribution of random displacements was expressed by the root mean square of standard deviation of all patients.

ACCEPTED MANUSCRIPT Given the systematic and random errors observed with each scenario, the treatment margins (as determined by van Herk’s formula and Stroom’s formula)(9, 10) were calculated for each scenario in each dimension for each less than daily IGRT scenario.

T

Probability of geographical miss was estimated by subtracting the absolute value of residual errors in for all

RI P

non-image guided fractions in all assumed scenarios, from 7 mm and similarly from 5 mm. Any positive value

SC

in was considered a potential risk for missing the target.

Results:

MA NU

The mean systematic error and random error for each scenario and the PTV margins generated using Stroom’s and van Herk’s formulae are summarised in Table 1. As observed, decreasing frequency of image guidance increases the mean systematic error and the standard deviation of the systematic error. However, the random error was not affected with decreasing imaging frequency and was in the range of 3-4 mm in all scenarios.

ED

With a decrease in image guidance frequency, an increase in PTV margins was required to achieve adequate coverage of the clinical target volume (CTV).

PT

The difference in PTV margins with decreasing imaging frequency was the least in lateral direction and maximum in anteroposterior directions. Figure 1 represents the progressive increase in PTV margins in all

CE

three directions with decreasing frequency of imaging.

AC

The influence of reducing imaging frequency on the potential geographical miss of the target was assessed. Fractions where the residual errors in any of the 3 directions were more than 7 mm were considered to be a potential geographical miss. Table 2 shows the number of fractions with residual errors more than 7 mm in lateral, superoinferior and anteroposterior directions. Also shown is the overall number of fractions with residual shifts of more than 7 mm in any direction for decreasing frequency of image guidance. Figure 2 shows that the largest number of fractions with residual shifts more than 7 mm was seen in superoinferior directions. With reduction in image guidance from alternate day to first 3 days only, i.e. from 50 to 12%, an increase in the percentage of fractions with residual shifts more than 7 mm was seen. With every 10-20% decrease in imaging frequency, a 5% increased risk of missing the target was seen. The same analysis was repeated assuming an isotropic PTV margin of 5 mm. Table 2 and Figure 3 depict the number of fractions with residual errors more than 5 mm in any direction and the overall predicted geographical miss in any direction.

ACCEPTED MANUSCRIPT Discussion: Prostate motion and deformation during radiotherapy are well known phenomenon. Careful interplay of image guidance frequency with PTV margins is required to ensure adequate target coverage as daily variation

T

in anatomy may affect both tumor control and toxicity rates(8, 11). In the present study, five different

RI P

potential commonly used non-daily imaging strategies were tested in a cohort of patients undergoing 2700 fractions with daily image guidance (100% frequency), to study the influence of reduced imaging frequency on

SC

PTV margin generation. The imaging frequency ranged from 0% to 50% for the 5 different scenarios. Systematic errors increased with decreasing imaging frequency. However, the random errors were unaffected

MA NU

and contributed to significant residual error on the days when imaging was not performed. Decreasing imaging frequency resulted in substantial increase in the size of PTV margins needed to counter the residual random error and adequately cover the target.

Margin reduction is an important benefit of online image guidance. Without image guidance, PTV margins may

ED

need to be as large as 13 mm anteroposteriorly, 12 mm craniocaudally and 7 mm laterally(12, 13). Uncertainties could be reduced with on-line IGRT(14, 15). Several authors have shown that CTV to PTV margins

PT

could be reduced to 3-8mm with daily IGRT with CBCT which represents the residual error after correction of inter- and intrafraction motion(15-17).However, reducing treatment margins may lead to geographical miss

CE

and underdosage of the CTV if daily imaging is not used, especially in hypofractionated schedules(18). We used

AC

a PTV margin of 7 mm in all directions, for all patients chosen for the study. Hence, assuming an isotropic PTV margin of 7mm for all these patients depending on the frequency of imaging, about 11-30% of the fractions would have suffered a potential geographical miss if daily image guidance was not carried out. This proportion increases substantially if an isotropic PTV margin of 5 mm is considered. As high as 39% of the fractions could have a potential geographical miss if the imaging frequency was reduced to 12%. The geographical miss was greatest in the anteroposterior direction probably due to the influence of rectal filling on prostate position and deformation. The lowest risk of a geographical miss was observed in the lateral direction as prostate motion is smallest side-to-side. Although daily image guidance is hypothesized to improve local tumour control and reduce treatment related side effects leading to improvement in quality of life, this comes at the expense of increasing the treatment time per fraction considerably(7).A cost-outcome analysis reported by Ploquin et al showed that prostate IGRT used solely for translational patient repositioning increased cost with relatively little improvement in

ACCEPTED MANUSCRIPT dosimetric quality unless a margin reduction was incorporated with resultant improvement in the therapeutic ratio (19).IGRT also increases the complexity of radiation planning and delivery process, mandating stringent quality assurance at every level for effective and safe treatment. Time and/ or equipment may not be available

T

in all centres to accomplish online correction of setup errors daily. Feasibility and availability of daily image

RI P

guidance has not kept pace with the technological advancement in many parts of the world. According to a survey conducted in the United Kingdom in 2014 to assess the practice of IGRT for prostate cancer, more than

SC

half of the centres were not using daily image guidance and almost two-thirds were not using daily volumetric imaging(4).First 3 days or once weekly schedules were commonly used to compensate for limited availability

MA NU

of IGRT equipment. Another survey in the United States found that the most common method of verification used was online for the first few fractions only, followed by off-line verification for all subsequent fractions (6).

Although some authors have suggested that less-than-daily image-guided RT may potentially compromise

ED

target coverage, the relationship between frequency of imaging and patient outcomes remains unclear(8).In a smaller study conducted by Kupelian et al(8) different imaging frequencies were replayed in a cohort of 74

PT

patients treated with daily IGRT. They showed that significant residual errors > 3 mm were seen in ~40% of fractions and > 5 mm in ~25% of fractions even with alternate day imaging.PTV margins needed to increase

CE

with decreasing frequency of image guidance. There are no randomized clinical trials evaluating the benefit of

AC

daily image guidance for patient outcomes. Two retrospective reviews have reported a decrease in biochemical control (Prostate Specific Antigen levels) following prostate radiotherapy when failing to account for the systematic error of a distended rectum which caused a geographical miss of the prostate gland (20, 21). Another retrospective series demonstrated a significant reduction in late urinary toxicity with IGRT patients compared with the non-IGRT patients with localized prostate cancer(22).

In busy radiotherapy centres where daily image guidance may not be feasible, we propose the use of two separate plans for days when image guidance is used and days when it is not, allowing margin reduction at least on the days that imaging is done, while 3-7 mm PTV margin is used for IGRT. Margins for the non-image guidance days need to be tailored according the accuracy of patient set-up and frequency of imaging protocol. However, the effectiveness of this approach of adaptive radiotherapy needs to be tested in a controlled prospective setting. We are planning a prospective study using an adaptive approach wherein a PTV margin of

ACCEPTED MANUSCRIPT 8-10 mm will be used in all non-image guided fractions as per the data generated in this study. A post treatment scan will be taken to assess if smaller margins were used without image guidance, would the target be covered adequately? This will be correlated with the dosimetric data and subsequently clinical outcomes.

T

We recognise that the present study has some limitations. Although it is retrospective in nature, the large

RI P

number of patients and treatment fractions used for the analysis in this study would have diluted any uncertainty in daily alignment data. Also, errors because of intra-fraction motion, rotation, and soft-tissue

SC

deformation were not accounted for in the PTV generation. Nodal coverage and OAR doses have not been studied. The less-than-daily image guidance scenarios were hypothetical and no patients were treated with

MA NU

these schedules. The dosimetric and clinical benefits of different imaging frequencies have also not been addressed in this study and need prospective validation.

Conclusion: The use of less than daily image-guided RT requires larger PTV margins for patients treated with

ED

IGRT for prostate cancer. With reduction in image guidance to less-than-daily schedules, an increase in risk of geographical miss was seen. With every 15% decrease in imaging frequency there is a 5% increased risk of

AC

CE

and non-IGRT fractions.

PT

missing the target. We propose an adaptive strategy of using separate plans with varying PTV margins for IGRT

ACCEPTED MANUSCRIPT References:

AC

CE

PT

ED

MA NU

SC

RI P

T

1. Drabik DM, MacKenzie MA, Fallone GB. Quantifying appropriate PTV setup margins: analysis of patient setup fidelity and intrafraction motion using post-treatment megavoltage computed tomography scans. International journal of radiation oncology, biology, physics. 2007;68(4):1222-8. 2. Kupelian PA, Lee C, Langen KM, Zeidan OA, Mañon RR, Willoughby TR, et al. Evaluation of ImageGuidance Strategies in the Treatment of Localized Prostate Cancer. International Journal of Radiation Oncology • Biology • Physics.70(4):1151-7. 3. Schulze D, Liang J, Yan D, Zhang T. Comparison of various online IGRT strategies: The benefits of online treatment plan re-optimization. Radiotherapy and oncology : journal of the European Society for Therapeutic Radiology and Oncology. 2009;90(3):367-76. 4. Ariyaratne H, Chesham H, Alonzi R. Image-guided radiotherapy for prostate cancer in the United Kingdom: a national survey. The British journal of radiology. 2017;90(1070):20160059. 5. Nabavizadeh N, Elliott DA, Chen Y, Kusano AS, Mitin T, Thomas CR, Jr., et al. Image Guided Radiation Therapy (IGRT) Practice Patterns and IGRT's Impact on Workflow and Treatment Planning: Results From a National Survey of American Society for Radiation Oncology Members. International Journal of Radiation Oncology • Biology • Physics.94(4):850-7. 6. Nabavizadeh N, Elliott DA, Chen Y, Kusano AS, Mitin T, Thomas CR, Jr., et al. Image Guided Radiation Therapy (IGRT) Practice Patterns and IGRT's Impact on Workflow and Treatment Planning: Results From a National Survey of American Society for Radiation Oncology Members. International journal of radiation oncology, biology, physics. 2016;94(4):850-7. 7. Deodato F, Cilla S, Massaccesi M, Macchia G, Ippolito E, Caravatta L, et al. Daily on-line set-up correction in 3D-conformal radiotherapy: is it feasible? Tumori. 2012;98(4):441-4. 8. Kupelian PA, Lee C, Langen KM, Zeidan OA, Manon RR, Willoughby TR, et al. Evaluation of imageguidance strategies in the treatment of localized prostate cancer. International journal of radiation oncology, biology, physics. 2008;70(4):1151-7. 9. van Herk M, Remeijer P, Rasch C, Lebesque JV. The probability of correct target dosage: dosepopulation histograms for deriving treatment margins in radiotherapy. International journal of radiation oncology, biology, physics. 2000;47(4):1121-35. 10. Stroom J, Gilhuijs K, Vieira S, Chen W, Salguero J, Moser E, et al. Combined recipe for clinical target volume and planning target volume margins. International journal of radiation oncology, biology, physics. 2014;88(3):708-14. 11. Ghilezan M, Jaffray D, Siewerdsen J, Van Herk M, Shetty A, Sharpe M, et al. Prostate gland motion assessed with cine-magnetic resonance imaging (cine-MRI). International journal of radiation oncology, biology, physics. 2005;62. 12. Meijer GJ, de Klerk J, Bzdusek K, van den Berg HA, Janssen R, Kaus MR, et al. What CTV-to-PTV Margins Should Be Applied for Prostate Irradiation? Four-Dimensional Quantitative Assessment Using ModelBased Deformable Image Registration Techniques. International Journal of Radiation Oncology • Biology • Physics.72(5):1416-25. 13. Poli ME, Parker W, Patrocinio H, Souhami L, Shenouda G, Campos LL, et al. An assessment of PTV margin definitions for patients undergoing conformal 3D external beam radiation therapy for prostate cancer based on an analysis of 10,327 pretreatment daily ultrasound localizations. International journal of radiation oncology, biology, physics. 2007;67(5):1430-7. 14. C. C. Target and organ motion considerations. In: Valicenti RK, Dicker AP, Jaffray DA, editors. Image-guided radiation therapy of prostate cancer. New York: Informa Healthcare. 2008:p. 51–64. 15. Letourneau D, Martinez AA, Lockman D, Yan D, Vargas C, Ivaldi G, et al. Assessment of residual error for online cone-beam CT-guided treatment of prostate cancer patients. International journal of radiation oncology, biology, physics. 2005;62(4):1239-46. 16. Wu Q, Ivaldi G, Liang J, Lockman D, Yan D, Martinez A. Geometric and dosimetric evaluations of an online image-guidance strategy for 3D-CRT of prostate cancer. International journal of radiation oncology, biology, physics. 2006;64(5):1596-609. 17. Gill SK, Reddy K, Campbell N, Chen C, Pearson D. Determination of optimal PTV margin for patients receiving CBCT-guided prostate IMRT: comparative analysis based on CBCT dose calculation with four different margins. Journal of applied clinical medical physics. 2015;16(6):5691.

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA NU

SC

RI P

T

18. Deutschmann H, Kametriser G, Steininger P, Scherer P, Scholler H, Gaisberger C, et al. First clinical release of an online, adaptive, aperture-based image-guided radiotherapy strategy in intensity-modulated radiotherapy to correct for inter- and intrafractional rotations of the prostate. International journal of radiation oncology, biology, physics. 2012;83(5):1624-32. 19. Ploquin N, Dunscombe P. A cost-outcome analysis of Image-Guided Patient Repositioning in the radiation treatment of cancer of the prostate. Radiotherapy and oncology : journal of the European Society for Therapeutic Radiology and Oncology. 2009;93(1):25-31. 20. de Crevoisier R, Tucker SL, Dong L, Mohan R, Cheung R, Cox JD, et al. Increased risk of biochemical and local failure in patients with distended rectum on the planning CT for prostate cancer radiotherapy. International journal of radiation oncology, biology, physics. 2005;62(4):965-73. 21. Heemsbergen WD, Hoogeman MS, Witte MG, Peeters ST, Incrocci L, Lebesque JV. Increased risk of biochemical and clinical failure for prostate patients with a large rectum at radiotherapy planning: results from the Dutch trial of 68 GY versus 78 Gy. International journal of radiation oncology, biology, physics. 2007;67(5):1418-24. 22. Zelefsky MJ, Kollmeier M, Cox B, Fidaleo A, Sperling D, Pei X, et al. Improved clinical outcomes with high-dose image guided radiotherapy compared with non-IGRT for the treatment of clinically localized prostate cancer. International journal of radiation oncology, biology, physics. 2012;84(1):125-9.

ACCEPTED MANUSCRIPT Table 1. Errors and PTV margins generated using Stroom’s and van Herk’s formulae for each IGRT scenario Standard deviation of systematic error (mm)

Random error (mm)

Lateral shift

0.1

1.8

3.7

Supero-inferior shift Antero-posterior shift Twice weekly imaging

-0.2

1.5

0.1

1.5

Lateral shift

-0.04

2

Supero-inferior shift

-0.02

1.8

Antero-posterior shift

0.2

Margin using van Herk’s formula (mm)

day

7.1

3.2

5.3

6.1

3.7

5.6

6.4

3.8

6.6

7.6

4.4

6.8

7.7

2

4.7

7.2

8.2

2.1

3.9

6.9

7.9

1.7

3.4

5.7

6.5

0.6

2.1

3.5

6.6

7.6

Lateral shift

-0.2

2.8

3.6

8.2

9.6

Supero-inferior shift

-0.2

2.3

3.4

7

8.1

Antero-posterior shift

-0.3

3.1

4.0

9.1

10.7

No image guidance Lateral shift

0.5

3.1

3.7

8.8

10.4

Supero-inferior

-1.8

6.6

3.2

15.5

18.8

Weekly imaging

-0.3

AC

Supero-inferior shift Antero-posterior shift First 3 days only imaging

0.1

MA NU

ED

CE

Lateral shift

SC

6.2

PT

Alternate imaging

Margin using Stroom’s formula (mm)

T

Systematic error (mm)

RI P

Scenario

ACCEPTED MANUSCRIPT shift 0.6

8.9

3.3

20.1

AC

CE

PT

ED

MA NU

SC

RI P

T

Antero-posterior shift

24.5

ACCEPTED MANUSCRIPT

Table 2. Number of fractions estimated to have geographical miss with 7 mm PTV margin for different imaging frequency Superoinferior

Direction

Direction

Anteroposterior Direction

T

Lateral

RI P

Scenario

Any Direction

5 mm

7 mm

5 mm

7 mm

5mm

7 mm

5 mm

Alternate Day

99

236

60

276

193

329

304

656

%

3%

9%

2%

10%

7%

12%

11%

24%

Twice Weekly

121

311

158

298

288

388

466

708

%

4%

11%

6%

11%

10%

14%

17%

26%

Once Weekly

192

448

120

307

324

525

563

889

%

7%

16%

4%

11%

12%

19%

22%

33%

First 3 Days

284

545

149

400

495

675

792

1058

%

10%

20%

15%

18%

25%

29%

39%

AC

CE

MA NU

ED

PT

5%

SC

7 mm

AC

CE

PT

ED

MA NU

SC

RI P

T

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA NU

SC

RI P

T

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA NU

SC

RI P

T

ACCEPTED MANUSCRIPT