Dosimetric comparison of stereotactic body radiotherapy in different respiration conditions: A modeling study

Radiotherapy and Oncology 81 (2006) 97–104 www.thegreenjournal.com

Dosimetry

Dosimetric comparison of stereotactic body radiotherapy in different respiration conditions: A modeling study Kristina Kontrisova1, Markus Stock1, Karin Dieckmann, Joachim Bogner, Richard Po ¨tter, Dietmar Georg* Department of Radiotherapy and Radiobiology, AKH Vienna, Medical University Vienna, Vienna, Austria

Abstract Purpose: To evaluate the dosimetric consequences for irradiated lung tissue for different respiration conditions for hypofractionated stereotactic body radiotherapy (SBRT). Methods and materials: Thirteen patients with lung lesion undergoing SBRT treatment in shallow breathing with abdominal compression (SB + AP) underwent additional multislice CT studies in free breathing (FB), deep inspiration and expiration breath hold (DIBH, DEBH). For each patient 6 different treatment plans were designed for the various respiration conditions applying standard (7/7/10 mm), reduced (5/5/5 mm) and individual margins. The FB plan with standard margins was used as a reference. The percentage of volume of the ipsilateral lung receiving total doses P12, P15 and P18 Gy, mean lung dose (Dmean), NTCP corrected for fractionation effects and the total monitor units (MU) were evaluated. Results: With DIBH it was possible to reduce all lung dose parameters by about 20%. Applying reduced margins in DIBH, this reduction was even increased to about 40%. The standard technique (SB + AP) with individual margins showed similar results as DIBH with standard margins. DEBH showed some improvement over FB only when reduced margins were applied. Only for 5/13 patients NTCP values >1% were obtained. For these patients a significant NTCP reduction was achieved with DIBH techniques. Conclusions: In SBRT shallow breathing with abdominal compression produces acceptable results concerning lung DVHs. DIBH, especially with reduced margins, showed the best lung sparing. For the clinical implementation of such a technique some form of gating is advisable. However, there are some practical limitations due to high fractional doses. c 2006 Elsevier Ireland Ltd. All rights reserved. Radiotherapy and Oncology 81 (2006) 97–104.



Keywords: Stereotactic irradiation; Body frame; Hypofractionation; Lung cancer; Breath hold

The strategies behind the concept of ‘‘stereotactic body radiotherapy’’ (SBRT), namely to deliver a high dose to the target and low doses elsewhere, have always been the aim of radiotherapy. From a radiobiological point of view, inhomogeneous doses are delivered to small targets which are located in large ‘‘parallel’’ organs [5]. Hypofractionated treatments for targets in the lung, abdomen, liver, and pelvis have gained popularity and various groups have described their treatment techniques and/or clinical results [1,8–11,19,32,36,37]. Based on the poor results obtained for non-small cell lung cancer (NSCLC) using total doses around 74 Gy in daily fractions of 2 Gy, new treatment concepts were investigated and SBRT was identified as a promising option. On the other hand, because of the excellent local control, the short overall treatment time and the basically non-invasive approach, hypofractionated treatments became the treatment of 1

These authors contributed equally to this paper.



choice for inoperable lung metastases. Consequently, primary and metastatic lung lesions are the main focus for current SBRT applications. A summary of published treatment concepts and results of stereotactic radiotherapy in lung can be found in a recent publication [38]. For successful small volume radiotherapy applied to targets in the lung accurate target definition, target positioning and the compensation for uncertainties are essential. While the reproducibility and setup accuracy of immobilization devices designed for SBRT have been investigated and described in several publications [4,7,17,20,35], respiration induced target motion remains a challenge in SBRT. For advanced conformal radiotherapy several approaches have been proposed to overcome respiration movements in conformal radiotherapy, ranging from active breathing control or deep inspiration breath hold techniques to automated respiratory gated radiotherapy based on external or internal marker tracking [18,21,23,25,30,34]. Using such approaches the internal margin could be successfully reduced [12].

0167-8140/$ - see front matter c 2006 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.radonc.2006.08.006

98

Respiration control for SBRT: Dosimetric considerations

The delivery of high fractional doses, e.g. 20 Gy at the isocenter, with beams of small sizes is associated with a large number of monitor units. Even with modern linear accelerators working at high dose rates treatment times are of the order of several minutes. If SBRT treatments are performed under some form of automated respiration gating monitor unit efficiency will be compromised. Thus the benefit of reduced margins in terms of reduced normal tissue irradiation is counterbalanced by extended treatment times. The aim of the present study was to investigate the potential dosimetric benefit of reduced margins under respiration control and various respiration conditions. In a treatment planning comparison, a typical SBRT technique performed under free breathing with the patient immobilized with a body frame was compared with treatments performed in deep inspiration and expiration using standard and reduced margins and in shallow breathing with abdominal compression and individual margins.

Materials and methods Patients Thirteen patients undergoing hypofractionated stereotactic body radiotherapy at the Department of Radiotherapy and Radiobiology, Medical University Vienna, were included in this study. All patients were treated either for primary lung lesions (2/13) or lung metastases (11/13). The tumors were located in upper lobe (5/13), middle lobe (2/13) and lower lobe (6/13). The mean patient age was 64 years (range 42–81). The eligibility criteria were: (i) being able to hold breath in deep inspiration (DIBH) and deep expiration (DEBH) for P20 s; (ii) patient with good bounded lesions; and (iii) patient’s agreement to participate in this study.

SBRT technique The SBRT technique applied clinically at the Medical University Vienna was used for comparison and is briefly described in the following. Patient immobilization was performed using a stereotactic body frame (SBF, Elekta, Crawley, UK) and an individually adapted vacuum pillow attached to the SBF [1,16]. The SBF was modified in the department’s mechanical workshop to integrate an armrest, which was attached to a modified base plate of the SBF (Fig. 1). The patient position in the body frame was verified with 6 small tattoos on the patient’s breast and one on the leg. The relative position of these points with respect to the external reference system was defined using laser pointers. During treatment planning (see Target definition and treatment planning) the CTV was delineated. For the actual treatment, the PTV margin was determined individually according to the respiratory movement. The margin in cranio-caudal direction was derived from 2D fluoroscopy by measuring the difference between the tumor position in inspiration and expiration. If the tumor was not visible, we evaluated the diaphragm motion as a surrogate for tumor motion. The margins in lateral and anterior–posterior direction were determined with respect to the tumor location according to our clinical experiences. A diaphragm con-

Fig. 1. Stereotactic body frame with an integrated armrest attached to the base plate.

trol, which can be attached to the SBF, was generally used to minimize respiration movements and consequently tumor movements. In a recent study based on the same SBRT technique the following interfractional set-up reproducibility has been reported: 3.5 mm in cranio-caudal (CC), 2.2 mm in anterior–posterior (AP) and 3.9 mm in lateral (lat) direction, respectively [35]. All SBRT patients were treated with 3 · 12.5 Gy prescribed to the 65% isodose level, delivered within 5 days. Prior each fraction the patient position was verified with a verification CT. The planning and control CT series were co-registered and predefined contours (CTV, PTV, patient outline, lung and spinal cord) were transferred from the planning CT to the actual control CT scan. If there was a discrepancy larger than 3 mm in any direction between these two scans an isocenter correction was performed.

CT data acquisition in different respiration conditions For routine treatment planning all patients underwent a planning CT scan (Siemens, Somatom Plus S, Erlangen, Germany) in treatment position in the SBF in shallow breathing with abdominal compression. Images were acquired with 4 mm slice thickness. Additionally, rapid multislice CT scans (Siemens, Somatom Sensation 16, Erlangen, Germany) were performed in spiral mode with a pitch factor of 1.5, 3 mm slice width and 2 mm reconstruction. These multislice CT studies were performed under (a) free breathing without abdominal pressure (FB) in a random point of the breathing cycle, (b) deep inspiration breath hold (DIBH) (1 scan) and (c) deep expiration breath hold (DEBH) (1 scan). During

K. Kontrisova et al. / Radiotherapy and Oncology 81 (2006) 97–104

the acquisition the patients received acoustic breathing instructions from the CT operator.

Target definition and treatment planning FB, DIBH and DEBH CT scans were transferred to a 3 D treatment planning system (XiO V4.2, CMS, USA). For each CT set the lung was contoured using an auto-contour tool and the gross tumor volume (GTV) was manually excluded from lung volume. For each patient, clinical target volumes (CTVs) were delineated by the same physician (KD) in each scan applying the same window level setting and using a margin of about 2–3 mm around the GTV. Different PTVs were generated from CTVs using a 3D auto-margin tool of the treatment planning system. For FB, DIBH and DEBH scans standard margins of 7 mm in the axial slice and 10 mm in CC direction were used [1,16]. Additionally, an isotropic margin of 5 mm was applied in both DIBH and DEBH data sets. The 5 mm margin implies an internal margin (IM) of 2 mm using some form of respiration control [26,27], a set-up margin (SM) of 3 mm [35] and a simple summation of both margins for hypofractionated stereotactic applications. Including the actual treatment plan, for each patient 6 different 3D conformal treatment plans were designed for the various respiration conditions. Table 1 summarizes the different plans based on the various CT scans and PTV margins. As mentioned earlier, for the actual treatment plan #6 (SB + abdominal compression), an individual respiration assessment was made during fluoroscopy to define margins. The following margins were applied clinically: 5 mm in AP direction in 9 cases, 6 mm in 1 case and 7 mm in 3 cases; in lateral direction 4 mm in 1, 5 mm in 7, 6 mm in 1, 7 mm in 2 and 10 mm in 2 cases, respectively. In CC direction we added a 5 mm margin in 4 cases, 6 mm in 1 case, 7 mm in 7 cases and 10 mm in 1 case. For each patient all plan categories were based on the same beam setup and photon beam quality. Usually, 6–7 coplanar or non-coplanar 6 or 15 MV photon beams were combined. The shapes of beams were individually optimized using a MLC (leaf width of 1 cm) and a beams-eye-view technique in such a way that the prescription isodose level (65%) was encompassing at least 99% of the PTV. Dose distribu-

Table 1 Summary of treatment plan categories including the margins applied Plan

CT scan

Total margin (AP/lat/CC)

# # # # # #

FB DIBH DIBH DEBH DEBH SB + abdominal control

7–7–10 7–7–10 5–5–5 7–7–10 5–5–5 Individual margins (4–10 mm)a

1 2 3 4 5 6a a

. . .treated plan with individual margins (4–10 mm) in AP as well as in lateral direction (depending on tumor location) were added to the CTV. The margin in longitudinal direction was based on fluoroscopic measurement of tumor movement under free breathing in AP view. The margin in CC direction was at least 5 mm.

99

tions were calculated with a point dose kernel (superposition) algorithm to provide a high dose calculation accuracy [3,29]. The voxel matrix size was set to 2.5 mm. The normalization point (100%) was defined in the center of PTV. The conformity criterion of Paddick was applied during treatment planning for simple (manual) optimization [22]: CI ¼

TVPI2 ; PI  TV

where TVPI represents the part of the planning target volume encompassed by the 65% isodose, TV represents the total planning target volume and PI the total volume of the 65% prescription isodose. For all treatment plans a CI P 0.75 was achieved with intra-patient differences not larger than 0.07 for plans #1 to #6. The maximal dose per fraction to spinal cord was restricted to 64 Gy. Applying the linear quadratic model and assuming an a/b-ratio of 2, 3 · 4 Gy corresponds to a total dose of about 18 Gy given in 2 Gy daily fractions.

Evaluation criteria The ratio PTV to ipsilateral lung volume was determined for each treatment plan. For the ipsilateral lung the percentage of volume receiving total doses P12, 15 and 18 Gy (V12 Gy, V15 Gy, V18 Gy), respectively, delivered in 3 fractions and the mean lung dose (Dmean) were evaluated. Dosimetric data of different plans were compared with a paired two tailed Student’s t-test and for p < 0.05 statistical significance was assumed. Applying the linear quadratic model (LQ model) and assuming an a/b = 3 Gy for the ipsilateral lung cells, we calculated biologically weighted DVHs similar to a method described by Van den Heuvel [31]. The dose axis of the DVH was rescaled to an EQD2 (dose equivalent to a 2 Gy fractionation scheme). A Makro in Microsoft Excel was developed to calculate the normal tissue complication probability (NTCP) using the Lyman–Kutcher–Burman (LKB) model for each plan. The following parameter set for NTCP calculations were used: TD50 = 24.5 Gy, n = 0.87, m = 0.18 [2,14]. Finally, the total number of monitor units (MU) necessary to deliver each plan was compared and the treatment time for each technique was estimated. The estimation of the treatment time was based on the following parameters: dose rate 400 MU/min, duration of breath hold 20 s, break between 2 breath holds 20 s, machine setup (table and gantry position) between 2 fields on average 2 min.

Results Target characteristics Mean CTV volumes and mean differences of the ipsilateral lung volume are shown in Table 2, where the lung volume in FB was used as reference. CTV volumes, defined on each CT study for the different respiration conditions, varied on average from 1% to +3%. There was no significant increase or decrease in mean CTVs for different respiration conditions (p  0.05). For individual patients variations were observed, which can be explained by tumor compression or

100

Respiration control for SBRT: Dosimetric considerations

Table 2 Summary of average CTV and PTV as well as average increase of ipsilateral lung volume, including ranges for the different respiratory conditions Mean CTV (±SD) [cm3]

Range [cm3]

Mean increase lung vol. (±SD)

Range [cm3]

Margins

Mean PTV (±SD) [cm3]

PTV/Lung volume (%)

FB DIBH

17.4 (±19.8) 17.1 (±18.5)

1.3–72.0 1.4–67.0

– 47% (±21%)

– +14 to +75%

DEBH

17.4 (±19.2)

1.3–70.0

14% (±9%)

36 to 2%

SB with abdominal compression

16.5 (±18.7)

1.5–67.0

7% (±9%)

26 to +7 %

Standard Standard Reduced Standard Reduced Individual

64.1 63.6 42.3 64.5 42.9 50.2

3.9 2.8 1.9 4.6 3.0 3.4

Plan # # # # # #

1 2 3 4 5 6

(±50.6) (±48.4) (±35.7) (±49.8) (±36.9) (±49.9)

The last column shows the ratio PTV/ipsilateral lung volume for the different breathing scans and margins.

stretching in inspiration and expiration, depending on target location and size. A maximum CTV increase of 31% was measured in deep inspiration in case of small volumes (1.5 cm3). Comparing FB and SB with abdominal compression, a difference larger than 20% was observed for 3 patients. These CTV differences are mainly attributed to respiration motion artifacts in the slow scan mode. For DIBH, the ipsilateral lung volume showed a mean increase of 47% while for DEBH and at shallow breathing with abdominal compression a mean decrease of 14% and 7% were obtained. In both DIBH and DEBH the effect of the margin reduction resulted in an average PTV reduction of about 20 cm3 compared to other respiration conditions (p < 0.001). The actual gain was depending on CTV volume. The average PTVs, when applying individual margins in SB with abdominal compression, were also significantly different from FB conditions (p < 0.001).

Treatment plan evaluation As an example, a cumulative dose–volume histogram (DVH) of the ipsilateral lung of a representative patient is displayed in Fig. 2 for FB, SB with abdominal compression and DIBH with different margins. For this particular patient

DVH of plans #4 and #5 in DEBH were almost identical with the DVHs of plans #1 and #2 (agreement within 2%). A similar trend was observed for all patients. For this patient, the maximum differences in CTV volumes were 15%. For plan #6 isotropic margins of 7 mm were used. The lung volume in DIBH increased by 19%, in DEBH and SB with abdominal compression it decreased by 16 and 5%, respectively. With margin reduction in DIBH this patient achieved a 33% decrease in PTV volume. Table 3 summarizes the mean percentage of the ipsilateral lung receiving doses P12, P15 and P18 Gy, and the mean lung dose. For the 13 patients, the average values under FB (plan #1) for V12 Gy, V15 Gy and V18 Gy were 14.3%, 12.0% and 9.9%, respectively. In DIBH with standard margins (plan category #2) these values decreased by about 20% and by about 40% or more when applying reduced margins in DIBH (plan #3). The DEBH technique with standard margins (plan category #4) did not show any improvement over the FB technique (p  0.05). DEBH with reduced margins and the SB technique with abdominal compression and individual margins were comparable with the DIBH technique with standard margins, for all dosimetric parameters shown in Table 3. All three dose–volume parameters (V12 Gy, V15 Gy

40 #1: FB standard margins #2: DIBH standard margins

35

#3: DIBH reduced margins #6: SB with abd. pressure indiv. margins

Volume [%]

30 25 20 15 10 5 0 6

12

18

24

30

36

42

48

54

Dose [Gy] Fig. 2. Comparison of a typical ipsilateral lung DVH for treatment plans #1–#3, #6 (for explanation of treatment plan categories see Table 1), for a full treatment consisting of 3 fractions (3 · 12.5 Gy prescribed to 65% isodose).

5.19 4.23 3.24 5.51 4.22 4.44 – 24.8 46.5 5.0 25.2 19.8

V18 Gy (%) V15 Gy (%)

– 23.2 43.1 4.2 21.9 19.6 – 19.1 39.5 5.4 19.9 16.9 (±7.6) (±5.3) (±4.7) (±7.5) (±6.2) (±7.2) 9.9 7.5 5.3 10.4 7.4 8.0 (±8.2) (±6.0) (±4.9) (±8.1) (±6.8) (±7.7) 12.0 9.2 6.8 12.5 9.3 9.6 (±8.7) (±6.7) (±5.7) (±8.5) (±7.4) (±8.0) 14.3 11.6 8.7 15.1 11.5 11.9 SB with abdominal compression

DEBH

Standard Standard Reduced Standard Reduced Individual 1 2 3 4 5 6

FB DIBH

# # # # # #

V12 Gy (%)

(±2.60) (±2.01) (±1.72) (±2.59) (±2.15) (±2.57)

– 18.4 37.5 +6.3 18.7 14.5

Reduction – Dmean rel. to #1 (%) Dmean (±SD) [Gy] Reduction rel. to #1 V18 Gy (±SD) V15 Gy (±SD) V12 Gy (±SD) Margins Plan category

Table 3 Mean percent of ipsilateral lung volume receiving doses P12, P15, P18 Gy (in 3 fractions) and average mean lung dose Dmean for the different treatment plan categories

K. Kontrisova et al. / Radiotherapy and Oncology 81 (2006) 97–104

101

and V18 Gy) reached statistical significance (p < 0.001) when comparing free breathing with DIBH with both reduced and standard margins, DEBH with reduced margins and SB + abdominal compression. Average Dmean values for the ipsilateral lung in FB and DEBH with standard margins were between 5 and 5.5 Gy. These values were about 1 Gy lower for plan #2 (DIBH with standard margins), plan #5 (DEBH with reduced margins), and plan #6 (SB + abdominal compression and individual margins) and the lowest Dmean values were observed for plan #3 (average difference around 2 Gy). Similar to DVH parameters, Dmean differences reached statistical significance (p < 0.001) when comparing free breathing (plan #1) with DIBH with both reduced and standard margins (plan #2 and 3), DEBH with reduced margins (plan #4) and SB with abdominal compression (plan #6). For each patient and plan category, the mean dose to the contra-lateral lung did not exceed 2 Gy at all. Consequently, for 8/13 patients the NTCP values calculated were less than 1% for all plan categories. For the other 5 patients, the resulting NTCP values are displayed in Fig. 3. In DIBH, the increase in lung volume reduced the NTCP dramatically. The additional margin reduction helped to decrease the NTCP further. DEBH, on the other hand, showed an improvement over FB technique only if reduced margins were applied. When comparing the total number of monitor units (MU) for all plan categories, no large differences were observed, although there was a slight increase for plans with margin reduction (+3% for DIBH, +5% for DEBH). Total monitor units, duration of the beam delivery and the estimated treatment time (in minutes) per fraction are shown in Table 4. Compared to free breathing or shallow breathing, both breath hold techniques increased the overall treatment times by at least 7 min.

Discussion DIBH and gating techniques are well-known and applied concepts used in the treatment of NSCLC or liver tumors using standard fractionation schemes. However, there is little information on gating or breath hold techniques applied clinically in SBRT. Whyte et al. placed two to four small metal fiducials into the tumor to track its position for radiosurgery [33]. Kimura et al. reported about the good reproducibility of voluntary breath hold at the end-inspiration and end-expiration using a spirometer for SBRT [13]. Based on standard linear accelerator technology and the calculations summarized in Table 4, the total beam-on time in SBRT can be divided into approximately 19 breath holds. This could result in a relative exhausting treatment and a strict patient selection is required, which can be based on criteria such as overall patient condition, age, lung function (FEV1, VC, FEV1%), etc. With every patient, we performed at least 4 or more training sessions, each included 5 or more breath-holds. This was done for the purpose of another study [27]. The breath hold was tolerated well by all patients and they did not find DIBH more fatiguing than DEBH. Tests with healthy volunteers simulating the duration of a full treatment were also well tolerated.

102

Respiration control for SBRT: Dosimetric considerations

Fig. 3. NTCP for ipsilateral lung for plan categories #1–6 for the five patients with NTCP values >1% in one of the plans. PTV and lung volume in FB were 66 cm3 and 1375 cm3 for patient 1, 126 cm3 and 945 cm3 for patient 2, 82 cm3 and 1933 cm3 for patient 3, 113 cm3 and 2531 cm3 for patient 4, 181 cm3 and 1525 cm3 for patient 5.

By applying the same margins the impact of different lung volumes could be assessed separately from that of target immobilization. The applied margin reduction (from 7/ 7/10 to 5/5/5 mm) demonstrated the additional gain due to breath hold and consequently target immobilization. However, current clinical practice in SBRT based on a stereotactic body frame aims to reduce tumor mobility with an abdominal compression [1,16,35]. This standard procedure was used to benchmark DIBH and DEBH techniques. The typical standard SBRT technique in shallow breathing with abdominal compression reduced the V12 Gy to V18 Gy volume by about 19% compared to free breathing. For this technique, all DVH values were comparable with the one of the DIBH plan with standard margins and the DEBH plan with reduced margins. In other words, the standard SBRT technique with abdominal pressure is relatively optimal. Any additional improvement could be only achieved with the DIBH technique and reduced margins. Similar conclusions resulted when comparing the Dmean values. In order to be able to achieve a uniform margin of 5 mm a gating method, which is based on the correlation between the external markers and the relative tumor position, was developed at our institution. With this method both breath

hold stability and reproducibility could be significantly improved [26,27]. The application of generic standard margins can lead to under- or overestimation of tumor mobility and thus to irradiation of more healthy lung tissue than necessary, or a geographic miss. A simple and fast method to determine internal margins are fluoroscopic measurements. The main disadvantages of this method are poor image quality in lateral views and in addition, for approximately one-third of our patients the tumor was not detectable and the diaphragm motion was used as a surrogate. For that reason CT studies are considered to be more appropriate to determine individual margins. In several reports different parameters have been suggested to predict pulmonary toxicity, e.g. percentage of the ipsilateral and total lung volume receiving dose greater than 20, 25 or 30 Gy, mean dose or NTCP calculations. For example, Graham et al. reported on a strong correlation between the frequency of pneumonitis and V20 Gy [6]. On the other hand, Kwa et al. described a correlation between the rate of radiation induced pneumonitis and the mean total lung dose [15]. Usually, these patients received a total dose up to 70 Gy delivered in 1.8–2 Gy daily fractions.

Table 4 Total monitor units, duration of the beam delivery and the treatment time in minutes per fraction for the different treatment plan categories Plan category

Margins

Total MU/fx

Tbeam

# # # # # #

Standard Standard Reduced Standard Reduced individual

2633 2673 2743 2628 2735 2661

6.6 6.7 6.9 6.6 6.8 6.7

1 2 3 4 5 6

FB DIBH DEBH SB with abdominal compression

on

[min]

Ttreatment [min] 18.6 25.0 25.4 24.8 25.3 18.7

K. Kontrisova et al. / Radiotherapy and Oncology 81 (2006) 97–104

Although the application of the standard biological models and parameters in SBRT is still unclear, the linear quadratic model has been applied to convert SBRT fractionation schemes into equivalent doses based on 2 Gy fractions [5]. In our study, the LQ model and the Lyman–Kutcher–Burman model were applied to estimate NTCP. The relative low NTCP values for some patients can be explained by the small PTVs and the steep dose gradient of the stereotactic treatment approach using inhomogeneous target doses. Only for 5 patients NTCP was higher than 1%. For these patients NTCP could be reduced dramatically when applying DIBH. Based on NTCP considerations, it can be concluded that DIBH facilitates dose escalation for a sub-group of patients. For the majority of patients fractional doses of 3 · 12.5 Gy (prescribed to the 65%) seem to be safe for SBRT applications with respect to lung toxicity. On the other hand, there is still space for dose escalation to improve local control. Currently, phase one studies are carried out to explore the different fractionation schemes (e.g. [24,28]).

Conclusion The main benefit of a DIBH will result from a relative decrease of the irradiated lung volume and the reduced lung density. However, self-breath holding can be improved by using a respiratory monitoring and ideally a feedback device. Only in such conditions the same (deep) inspiration or expiration level can be achieved during image acquisition for treatment planning and treatment delivery. A margin reduction might be enabled in such conditions. For SBRT, a stereotactic body frame and shallow breathing with abdominal compression compares favorably with DIBH and DEBH techniques, especially if treatment delivery efficiency is considered.

Acknowledgements The authors would like to thank Stefan Lang and the Reviewers for many useful suggestions during the development of this paper. K. Kontrisova was supported by the EC project HCMPT-2001-0318. * Corresponding author. Dietmar Georg, Division Medical Radiation Physics, Department of Radiotherapy and Radiobiology, AKH Vienna, Medical University Vienna, Waehringer Guertel 18-20, A-1090 Vienna, Austria. E-mail address: [email protected] Received 23 January 2006; received in revised form 1 August 2006; accepted 15 August 2006; Available online 8 September 2006

References [1] Blomgren H, Lax I, Na ¨slund I, et al. Stereotactic high dose fraction radiation therapy of extracranial tumors using an accelerator. Clinical experience of the first thirty-one patients. Acta Oncol 1995;34:861–70. [2] Burman C, Kutcher GJ, Emami B, et al. Fitting of normal tissue tolerance data to an analytic function. Int J Radiat Oncol Biol Phys 1991;21:123–35. [3] Dvorak P, Georg D, Bogner J, et al. Impact of IMRT and leaf width on stereotactic body radiotherapy of liver and lung lesions. Int J Radiat Oncol Biol Phys 2005;61:1572–81.

103

[4] Engelsman M, Damen EM, De Jaeger K, et al. The effect of breathing and set-up errors on the cumulative dose to a lung tumor. Radiother Oncol 2001;60:95–105. [5] Fowler JF, Tome WA, Fenwick JD, et al. A challenge to traditional radiation oncology. Int J Radiat Oncol Biol Phys 2004;60:1241–56. [6] Graham M, Purdy J, Emami B, et al. Clinical dose–volume histogram analysis for pneumonitis after 3D treatment for nonsmall cell lung cancer (NSCLC). Int J Radiat Oncol Biol Phys 1999;45:323–9. [7] Hara R, Itami J, Kondo T, et al. Stereotactic single high dose irradiation of lung tumors under respiratory gating. Radiother Oncol 2002;63:159–63. [8] Herfarth KK, Debus J, Lohr F, et al. Extracranial stereotactic radiation therapy: set-up accuracy of patients treated for liver metastases. Int J Radiat Oncol Biol Phys 2000;46:329–35. [9] Hof H, Herfarth KK, Munter M, et al. Stereotactic single-dose radiotherapy of stage I non-small-cell lung cancer (NSCLC). Int J Radiat Oncol Biol Phys 2003;56:335–41. [10] Hof H, Herfarth K, Mu ¨nter M, et al. The use of the multislice CT for the determination of respiratory lung tumor movement in stereotactic single-dose irradiation. Strahlenther Onkol 2003;179:542–7. [11] Hoyer M, Roed H, Sengelov L, et al. Phase-II study on stereotactic radiotherapy of locally advanced pancreatic carcinoma. Radiother Oncol 2005;76:48–53. [12] International Commission on Radiation Units and Measurements. ICRU report 62: Prescribing, recording and reporting photon beam therapy (supplement to ICRU Report 50). Bethesda; 1999. [13] Kimura T, Hirokawa Y, Murakami Y, et al. Reproducibility of organ position using voluntary breath-hold method with spirometer for extracranial stereotactic radiotherapy. Int J Radiat Oncol Biol Phys 2004;60:1307–13. [14] Kutcher GJ, Burman C. Calculation of complication probability for non-uniform tissue irradiation: the effective volume method. Int J Radiat Oncol Biol Phys 1989;16:1623–30. [15] Kwa SL, Theuws JC, Wagenaar A, et al. Evaluation of two dose–volume histogram reduction models for the prediction of radiation pneumonitis. Radiother Oncol 1998;48:61–9. [16] Lax I, Blomgren H, Na ¨slund, et al. Stereotactic radiotherapy of malignancies in the abdomen. Methodological aspects. Acta Oncologica 1994;6:677–83. [17] Lohr F, Debus J, Frank C, et al. Noninvasive patient fixation for extracranial stereotactic radiotherapy. Int J Radiat Oncol Biol Phys 1999;45:521–7. [18] Mah D, Hanley J, Rosenzweig KE, et al. Technical aspects of the deep inspiration breath-hold technique in the treatment of thoracic cancer. Int J Radiat Oncol Biol Phys 2000;48:1175–85. [19] Nagata Y, Negoro Y, Aoki T, et al. Clinical outcomes of 3D conformal hypofractionated single high-dose radiotherapy for one or two lung tumors using a stereotactic body frame. Int J Radiat Oncol Biol Phys 2002;52:1041–6. [20] Negoro Y, Nagata Y, Aoki T, et al. The effectiveness of an immobilization device in conformal radiotherapy for lung tumor: reduction of respiratory tumor movement and evaluation of the daily setup accuracy. Int J Radiat Oncol Biol Phys 2001;50:889–98. [21] Onishi H, Kuriyama K, Komiyama T, et al. A new irradiation system for lung cancer combining linear accelerator, computed tomography, patient self-breath-holding, and patientdirected beam-control without respiratory monitoring devices. Int J Radiat Oncol Biol Phys 2003;56:14–20. [22] Paddick I. A simple scoring ratio to index the conformity of radiosurgical treatment plans. Technical note. J Neurosurg 2000;93:219–22.

104

Respiration control for SBRT: Dosimetric considerations

[23] Pedersen AN, Korreman S, Nystrom H, et al. Breathing adapted radiotherapy of breast cancer: reduction of cardiac and pulmonary doses using voluntary inspiration breath-hold. Radiother Oncol 2004;72:53–60. [24] Radiation Therapy Oncology Group RTOG 0236. A Phase II Trial of Stereotactic Body Radiation Therapy (SBRT) in the Treatment of Patients with Medically Inoperable Stage I/II Non-Small Cell Lung Cancer. Version Date: November 21, 2005. [25] Rosenzweig K, Hanley J, Mah D, et al. The deep inspiration breath-hold technique in the treatment of inoperable nonsmall-cell lung cancer. Int J Radiat Oncol Biol Phys 2000;48:81–7. [26] Stock M, Nagyova K, Dieckmann K, et al. Impact of a feedback system on a precise gated breath hold technique during SBRT. Radiother Oncol 2005;76:S155–6. [27] Stock M, Kontrisova K, Dieckmann K, et al. Development and application of a real-time monitoring and feedback system for deep inspiration breath hold based on external marker tracking. Med Phys 2006;33:2868–77. [28] Timmerman R, Papiez L, McGarry R, et al. Extracranial stereotactic radioablation: results of a phase I study in medically inoperable stage I non-small cell lung cancer. Chest 2003;124:1946–55. [29] Traberg Hansen A, Petersen JB, Hoyer M, Christensen JJ. Comparison of two dose calculation methods applied to sextracranial stereotactic radiotherapy treatment planning. Radiother Oncol 2005;77:96–8.

[30] Tsunashima Y, Sakae T, Shioyama Y, et al. Correlation between the respiratory waveform measured using a respiratory sensor and 3D tumor motion in gated radiotherapy. Int J Radiat Oncol Biol Phys 2004;60:951–8. [31] Van den Heuvel F. Decomposition analysis of differential dose–volume histograms. Med Phys 2006;33:297–307. [32] Wersall PJ, Blomgren H, Lax I, et al. Extracranial stereotactic radiotherapy for primary and metastatic renal cell carcinoma. Radiother Oncol 2005;77:88–95. [33] Whyte RI, Crownover R, Murphy MJ, et al. Stereotactic radiosurgery for lung tumors: preliminary report of a phase I trial. Ann Thorac Surg 2003;75:1097–101. [34] Wong JW, Sharpe MB, Jaffray DA, et al. The use of active breathing control (ABC) to reduce margin for breathing motion. Int J Radiat Oncol Biol Phys 1999;44:911–9. [35] Wulf J, Haedinger U, Oppitz U, et al. Stereotactic radiotherapy of extracranial targets: CT-simulation and accuracy of treatment in the stereotactic body frame. Radiother Oncol 2000;57:225–36. [36] Wulf J, Hadinger U, Oppitz U, et al. Impact of target reproducibility on tumor dose in stereotactic radiotherapy of targets in the lung and liver. Radiother Oncol 2003;66:141–50. [37] Wulf J, Hadinger U, Oppitz U, et al. Stereotactic boost irradiation for targets in the abdomen and pelvis. Radiother Oncol 2004;70:31–6. [38] Wulf J, Baier K, Mueller G, Flentje MP. Dose–response in stereotactic irradiation of lung tumors. Radiother Oncol 2005;77:83–7.