K. Verhoeven

ESTRO consensus guideline on target volume delineation for elective radiation therapy of early stage breast cancer

BACKGROUND AND PURPOSE Delineation of clinical target volumes (CTVs) is a weak link in radiation therapy (RT), and large inter-observer variation is seen in breast cancer patients. Several guidelines have been proposed, but most result in larger CTVs than based on conventional simulator-based RT. The aim was to develop a delineation guideline obtained by consensus between a broad European group of radiation oncologists. MATERIAL AND METHODS During ESTRO teaching courses on breast cancer, teachers sought consensus on delineation of CTV through dialogue based on cases. One teacher delineated CTV on CT scans of 2 patients, followed by discussion and adaptation of the delineation. The consensus established between teachers was sent to other teams working in the same field, both locally and on a national level, for their input. This was followed by developing a broad consensus based on discussions. RESULTS Borders of the CTV encompassing a 5mm margin around the large veins, running through the regional lymph node levels were agreed, and for the breast/thoracic wall other vessels were pointed out to guide delineation, with comments on margins for patients with advanced breast cancer. CONCLUSION The ESTRO consensus on CTV for elective RT of breast cancer, endorsed by a broad base of the radiation oncology community, is presented to improve consistency.

Breathing adapted radiation therapy in comparison with prone position to reduce the doses to the heart, left anterior descending coronary artery, and contralateral breast in whole breast radiation therapy.

PURPOSE To compare 3 different treatment positions in whole breast radiation therapy in terms of target volume coverage and doses to the organs at risk (OAR). METHODS AND MATERIALS Thirty-four breast cancer (BC) patients (17 right-sided and 17 left-sided) were included in this dosimetric planning study. They all underwent a computed tomography (CT) scan in standard supine position in free-breathing (FB), supine position with gating in deep inspiratory breath hold (DIBH)(G), and prone position (P). Three-dimensional treatment plans were made for all 3 CTs. Target coverage and OAR sparing were evaluated. RESULTS Breast volumes varied between 209 and 2814 cm(3). The target coverage, expressed as the mean volume of the breast receiving at least 95% of the prescription dose, was similar for the 3 treatment positions. The mean lung dose and the volume of the lungs receiving >20 Gy were significantly lower in P (1.7 Gy; 2.3%) compared with G (3.4 Gy; 5.6%; P < .0001) and FB (4 Gy; 7.3%; P < .0001). The volume of the contralateral breast receiving >5 Gy was significantly lower in G (P = .001) or FB (P = .004) versus prone. The supine position with gating in DIBH significantly reduced the volume of the heart receiving >30 Gy (V30(heart)), the mean heart (D(heart)), and mean left anterior descending coronary artery (LAD) dose (D(LAD)) for left-sided BC patients (V30(heart) 0.9%, D(heart) 1.6 Gy, DLAD 22.4 Gy) with respect to FB (V30(heart) 4.3%, D(heart) 3.5 Gy, DLAD 30.9 Gy)(V30(heart) and mean D(heart): P ≤ .0001; mean D(LAD): P = .008) and P (V30(heart) 7.9%, D(heart) 5.4 Gy, D(LAD) 36.4 Gy)(V30(heart) and mean D(heart): P = .0004; mean D(LAD): P = .01). CONCLUSIONS The coverage of the planning target volume breast was equal for the 3 treatment positions. The lowest doses to the lungs were achieved in prone. The heart, LAD, and contralateral breast were best spared in the supine position with gating in DIBH.

Vessel based delineation guidelines for the elective lymph node regions in breast cancer radiation therapy - PROCAB guidelines

OBJECTIVE A national project to improve the quality of breast radiation therapy was started, named PROCAB (PROject on CAncer of the Breast). One of the objectives was to reach a national consensus guideline for the delineation of the regional lymph node areas in breast radiation therapy. METHODS The realization of the new guidelines was a step by step process that started with multiple expert meetings where the existing guidelines were analyzed and the delineations of the lymph node regions were performed together with a surgeon, specialized in the anatomy of the drainage of the breast. RESULTS The delineation guidelines are vessel-based. Since the occurrence of pathological lymph nodes is typically around the veins, the cranial and caudal borders of all different nodal regions are based on a 5mm margin around the veins, except for the parasternal lymph node area. Compared to the existing guidelines there are some major changes. CONCLUSION With this project a national as well as a European (ESTRO) consensus guideline for the delineation of the regional lymph node areas in breast RT is reached. The new delineation atlas is vessel-based and no longer field-based.

A comparison of three different radiotherapy boost techniques after breast conserving therapy for breast cancer

PURPOSE Compare different boost techniques after breast conserving therapy (BCT) in terms of local and loco-regional recurrences. MATERIALS AND METHODS From 2000 to 2005, patients treated with BCT for invasive breast cancer (BC) were included. An electron boost (EB) was performed for a superficial boost-volume (less than 29 mm under the epidermis), in all other cases a brachytherapy boost (BTB) was proposed. When patients refused a BTB or it was not possible for technical reasons, a photon boost (PB) was given. The primary endpoints were local and loco-regional recurrences. Secondary endpoints were metastasis-free and overall survival. RESULTS 1379 patients were eligible for analysis. Most patients (1052) received an EB, 225 a BTB and 76 a PB. At a median follow-up of 8.8 years, 35 (2.5%) patients developed a local or loco-regional recurrence. Ten years local relapse-free rate was 97.9%. No differences between boost techniques were observed in relapse risk, metastasis-free and overall survival after multivariate analyses. CONCLUSION In women treated with BCT followed by a boost irradiation to the tumor bed, no difference in local and loco-regional recurrence, metastasis-free and overall survival was observed comparing three different boost techniques. Outcome was excellent regardless of the boost technique.

The photon dose calculation algorithm used in breast radiotherapy has significant impact on the parameters of radiobiological models.

The comparison of the pencil beam dose calculation algorithm with modified Batho heterogeneity correction (PBC‐MB) and the analytical anisotropic algorithm (AAA) and the mutual comparison of advanced dose calculation algorithms used in breast radiotherapy have focused on the differences between the physical dose distributions. Studies on the radiobiological impact of the algorithm (both on the tumor control and the moderate breast fibrosis prediction) are lacking. We, therefore, investigated the radiobiological impact of the dose calculation algorithm in whole breast radiotherapy. The clinical dose distributions of 30 breast cancer patients, calculated with PBC‐MB, were recalculated with fixed monitor units using more advanced algorithms: AAA and Acuros XB. For the latter, both dose reporting modes were used (i.e., dose‐to‐medium and dose‐to‐water). Next, the tumor control probability (TCP) and the normal tissue complication probability (NTCP) of each dose distribution were calculated with the Poisson model and with the relative seriality model, respectively. The endpoint for the NTCP calculation was moderate breast fibrosis five years post treatment. The differences were checked for significance with the paired t‐test. The more advanced algorithms predicted a significantly lower TCP and NTCP of moderate breast fibrosis then found during the corresponding clinical follow‐up study based on PBC calculations. The differences varied between 1% and 2.1% for the TCP and between 2.9% and 5.5% for the NTCP of moderate breast fibrosis. The significant differences were eliminated by determination of algorithm‐specific model parameters using least square fitting. Application of the new parameters on a second group of 30 breast cancer patients proved their appropriateness. In this study, we assessed the impact of the dose calculation algorithms used in whole breast radiotherapy on the parameters of the radiobiological models. The radiobiological impact was eliminated by determination of algorithm specific model parameters. PACS numbers: 87.55.dh, 87.55.dkThe comparison of the pencil beam dose calculation algorithm with modified Batho heterogeneity correction (PBC-MB) and the analytical anisotropic algorithm (AAA) and the mutual comparison of advanced dose calculation algorithms used in breast radiotherapy have focused on the differences between the physical dose distributions. Studies on the radiobiological impact of the algorithm (both on the tumor control and the moderate breast fibrosis prediction) are lacking. We, therefore, investigated the radiobiological impact of the dose calculation algorithm in whole breast radiotherapy. The clinical dose distributions of 30 breast cancer patients, calculated with PBC-MB, were recalculated with fixed monitor units using more advanced algorithms: AAA and Acuros XB. For the latter, both dose reporting modes were used (i.e., dose-to-medium and dose-to-water). Next, the tumor control probability (TCP) and the normal tissue complication probability (NTCP) of each dose distribution were calculated with the Poisson model and with the relative seriality model, respectively. The endpoint for the NTCP calculation was moderate breast fibrosis five years post treatment. The differences were checked for significance with the paired t-test. The more advanced algorithms predicted a significantly lower TCP and NTCP of moderate breast fibrosis then found during the corresponding clinical follow-up study based on PBC calculations. The differences varied between 1% and 2.1% for the TCP and between 2.9% and 5.5% for the NTCP of moderate breast fibrosis. The significant differences were eliminated by determination of algorithm-specific model parameters using least square fitting. Application of the new parameters on a second group of 30 breast cancer patients proved their appropriateness. In this study, we assessed the impact of the dose calculation algorithms used in whole breast radiotherapy on the parameters of the radiobiological models. The radiobiological impact was eliminated by determination of algorithm specific model parameters. PACS numbers: 87.55.dh, 87.55.dk.

Is the use of a preoperative computed tomography beneficial to reduce the interobserver variability of the CTVboost delineation for breast radiation therapy

PURPOSE To determine whether the use of a preoperative (preop) computed tomography (CT) reduces (1) the clinical target volume boost (CTVboost) and (2) the interobserver variability (IOV) of the delineated CTVboost in breast radiation therapy. METHODS AND MATERIALS In patients treated with breast-conserving therapy, 3 CT scans in treatment position were performed: (1) preop; (2) after surgery, prechemotherapy (postop); and (3) postchemotherapy (postchemo). Six radiation-oncologists delineated the tumor bed and CTVboost before and after fusion of the preop CT. To assess the IOV, the Jaccard index was used. Linear mixed models were performedfor all analyses. RESULTS Eighty-two lumpectomy cavities were evaluated in 22 patients. No difference in CTVboost using the fusion of the preop CT (50.0 cm3; 95% confidence interval [CI], 35.6-64.4) compared with no fusion (49.0 cm3; 95% CI, 34.6-63.4) (P = .6) was observed. A significant increase in IOV was shown with the fusion of the preop CT; the mean Jaccard index of the CTVboost delineation of postop and postchemo CT together without the fusion of the preop CT was 0.53 (95% CI, 0.49-0.57) versus 0.50 (95% CI, 0.46-0.53) with fusion (P < .0001). CONCLUSIONS There is no benefit of using a preop CT to reduce the volume or the interobserver variability of the delineated CTVboost for breast radiation therapy.

Efficacy and workload analysis of a fixed vertical couch position technique and a fixed-action–level protocol in whole-breast radiotherapy

Quantification of the setup errors is vital to define appropriate setup margins preventing geographical misses. The no‐action–level (NAL) correction protocol reduces the systematic setup errors and, hence, the setup margins. The manual entry of the setup corrections in the record‐and‐verify software, however, increases the susceptibility of the NAL protocol to human errors. Moreover, the impact of the skin mobility on the anteroposterior patient setup reproducibility in whole‐breast radiotherapy (WBRT) is unknown. In this study, we therefore investigated the potential of fixed vertical couch position‐based patient setup in WBRT. The possibility to introduce a threshold for correction of the systematic setup errors was also explored. We measured the anteroposterior, mediolateral, and superior–inferior setup errors during fractions 1–12 and weekly thereafter with tangential angled single modality paired imaging. These setup data were used to simulate the residual setup errors of the NAL protocol, the fixed vertical couch position protocol, and the fixed‐action–level protocol with different correction thresholds. Population statistics of the setup errors of 20 breast cancer patients and 20 breast cancer patients with additional regional lymph node (LN) irradiation were calculated to determine the setup margins of each off‐line correction protocol. Our data showed the potential of the fixed vertical couch position protocol to restrict the systematic and random anteroposterior residual setup errors to 1.8 mm and 2.2 mm, respectively. Compared to the NAL protocol, a correction threshold of 2.5 mm reduced the frequency of mediolateral and superior–inferior setup corrections with 40% and 63%, respectively. The implementation of the correction threshold did not deteriorate the accuracy of the off‐line setup correction compared to the NAL protocol. The combination of the fixed vertical couch position protocol, for correction of the anteroposterior setup error, and the fixed‐action–level protocol with 2.5 mm correction threshold, for correction of the mediolateral and the superior–inferior setup errors, was proved to provide adequate and comparable patient setup accuracy in WBRT and WBRT with additional LN irradiation. PACS numbers: 87.53.Kn, 87.57.‐s

Accuracy of a new paired imaging technique for position correction in whole breast radiotherapy.

Image‐guided position verification in breast radiotherapy is accurately performed with kilovoltage cone beam CT (kV‐CBCT). The technique is, however, time‐consuming and there is a risk for patient collision. Online position verification performed with orthogonal‐angled mixed modality paired imaging is less time‐consuming at the expense of inferior accuracy compared to kV‐CBCT. We therefore investigated whether a new tangential‐angled single modality paired imaging technique can reduce the residual error (RE) of orthogonal‐angled mixed modality paired imaging. The latter was applied to 20 breast cancer patients. Tangential‐angled single modality paired imaging was investigated in 20 breast and 20 breast cancer patients with locoregional lymph node irradiation. The central lung distance (CLD) residual error and the longitudinal residual error were determined during the first 5 treatment fractions. Off‐line matching of the tangential breast field images, acquired after online position correction, was used. The mean, systematic, and random REs of each patient group were calculated. The systematic REs were checked for significant differences using the F‐test. Tangential‐angled single modality paired imaging significantly reduced the systematic CLD residual error of orthogonal‐angled mixed modality paired imaging for the breast cancer patients, from 2.3 mm to 1.0 mm, and also significantly decreased the systematic longitudinal RE from 2.4 mm to 1.3 mm. PTV margins, which account for the residual error (PTVRE), were also calculated. The PTVRE margin needed to account for the RE of orthogonal‐angled mixed modality paired imaging (i.e., 8 mm) was halved by tangential‐angled single modality paired imaging. The differences between the systematic REs of tangential‐angled single modality paired imaging of the breast cancer patients and the breast cancer patients with locoregional lymph node irradiation were not significant, yielding comparable PTVRE margins. In this study, we showed that tangential‐angled single modality paired imaging is superior to orthogonal‐angled mixed modality paired imaging to correct the position errors in whole breast radiotherapy. PACS numbers: 87.57N‐, 87.56Da, 87.53Kn

Boost delineation in breast radiation therapy: Isotropic versus anisotropic margin expansion

PURPOSE The purpose of this article is to compare isotropic and anisotropic margin expansion with regard to the size of the clinical target volume boost (CTVboost) and the interobserver variability (IOV). METHODS AND MATERIALS Lumpectomy cavities marked with 3 or more surgical clips were delineated by 6 radiation oncologists who specialized in breast radiation therapy. CTVboost anisotropic was created by manually expanding the tumor bed with an anisotropic margin of 15 mm (20 mm in case of extensive intraductal component) minus the surgical free margins in 6 directions (anteroposterior, craniocaudal, and superoinferior). For the CTVboost isotropic, the tumor bed was enlarged with an isotropic margin of 15 mm (20 mm in case of extensive intraductal component) minus the minimal surgical free margin. The volumes of the delineated CTVboost (cm3) were measured. To assess the IOV, the Jaccard index (JI), defined as the intersection divided by the size of the union of the sample sets, was used (ideal value = 1). The JI was calculated for each case and each observer pair. Linear mixed models were used for all analyses. RESULTS A total of 444 delineated tumor beds were evaluated. The mean volume of the CTVboost almost doubled by expanding the tumor bed with an isotropic margin compared with anisotropic margins (CTVboost isotropic 94 mL [12.5-331.0] vs CTVboost anisotropic 50 mL [3.2-332.7]; P = .0006). The IOV, assessed by the JI, significantly decreased by using isotropic versus anisotropic margin expansion (JICTV boost isotropic 0.73 [0.02-0.92] vs JICTV boost anisotropic 0.51 [0.0-0.8]; P< .0001). Because of the known positive correlation of the IOV and larger volumes, we corrected for CTVboost volumes. With this correction, the difference in IOV remains highly significant (P < .0001) in favor of isotropic margin expansion. CONCLUSIONS The use of anisotropic margin expansion from tumorbed to CTVboost isotropic significantly reduced the volume of the delineated CTVboost with a factor of 1.9 compared with isotropic margin expansion, but it substantially increased the interobserver variability.

PO-0778: Delineation of the regional nodal areas in breast radiotherapy: What are the most problematic regions?

volume delineation. Reasons for non-compliance are mostly problems with target coverage. Thirty-eight percent of the centers reported issues with target coverage; in 1 center they interpret the Dose Volume Histogram differently and are now looking at the 85% coverage of the planning target volume instead of the 95%. Conclusions: The introduction of new delineation guidelines for the RNA in breast radiotherapy has a major impact on the treatment planning and dosimetry with especially introducing newer treatment techniques to achieve better target coverage. Surprisingly, not all centers use the centrally reviewed and corrected target delineation to guide their radiotherapy to the RNA.