Dosimetric comparison of three‐dimensional conformal radiotherapy and intensity‐modulated radiotherapy for left‐sided chest wall and lymphatic irradiation

Abstract Introduction The aim of this study was to compare five different techniques for chest wall (CW) and lymphatic irradiation in patients with left‐sided breast carcinoma. Methods Three‐dimensional conformal radiotherapy (3DCRT), forward‐planned intensity‐modulated radiotherapy (FP‐IMRT), inverse‐planned IMRT (IP‐IMRT; 7‐ or 9‐field), and hybrid IP‐/FP‐IMRT were compared in 10 patients. Clinical target volume (CTV) included CW and internal mammary (IM), supraclavicular (SC), and axillary nodes. Planning target volumes (PTVs), CTVs, and organs at risks (OARs) doses were analyzed with dose–volume histograms (DVHs). Results No differences could be observed among the techniques for doses received by 95% of the volume (D95%) of lymphatics. However, the FP‐IMRT resulted in a significantly lower D95% dose to the CW‐PTV compared to other techniques (P = 0.002). The 9‐field IP‐IMRT achieved the lowest volumes receiving higher doses (hotspots). Both IP‐IMRT techniques provided similar mean doses (Dmean) for the left lung which were smaller than the other techniques. There was no difference between the techniques for maximum dose (Dmax) of right breast. However, FP‐IMRT resulted in lower Dmean and volume of right breast receiving at least 5 Gy doses compared to other techniques. Conclusion The dose homogeneity in CW‐CTV was better using IMRT techniques compared to 3DCRT. Especially 9‐field IP‐IMRT provided a more homogeneous dose distribution in IM and axillary CTVs. Moreover, the OARs volumes receiving low radiation doses were larger with IP‐IMRT technique, while volumes receiving high radiation doses were larger with FP‐IMRT technique. Hybrid IMRT plans were found to have the advantages of both FP‐ and IP‐IMRT techniques.


| INTRODUCTION
Radiotherapy has a major role in the management of breast cancer that reduces the risk of loco-regional recurrences and improves overall survival both for early stage breast cancer after breast-conserving surgery and locally advanced disease after mastectomy. 1,2 However, irradiation of the chest wall (CW) and regional lymph nodes is one of the most difficult and challenging techniques in radiation oncology. Coverage of CW and regional lymphatics including supraclavicular (SC), axillary, and internal mammary (IM) lymph nodes needs special attention to the doses of lungs, heart, and the opposite breast tissue.
Treatment planning is most commonly implemented using a three-dimensional (3D) conformal technique. Arthur et al. showed improved target coverage and reduction in normal tissue doses with partially wide tangent fields when compared with widened tangents or a 5-field technique using a photon/electron mix. 3 However, when treating left-sided breast cancer, the concave shape of the CW results in suboptimal dose homogeneity, target coverage, and unavoidable irradiation to portions of the underlying lung and heart with 3D conformal radiotherapy (3DCRT).
Intensity-modulated radiotherapy (IMRT) is an advanced form of 3DCRT that has been shown to improve target coverage and dose homogeneity compared to 3DCRT. 4 In addition, IMRT can decrease the dose to heart and lung. Forward-planned IMRT (FP-IMRT) with field-in-field (FINF) technique allows more homogenous dose distribution with reduced lung and heart doses. In recent years, inverseplanned IMRT (IP-IMRT) has been shown to improve breast and regional node coverage while decreasing dose to lung, heart, and contralateral breast tissue. 5 Patients with larger breasts who frequently have large-dose inhomogeneities are most likely to benefit from IMRT.
Recent trials have demonstrated that the use of the simplified IMRT technique resulted in improved dose homogeneity in the breast while resulting in lower skin and soft tissue toxicity than two-dimensional treatment planning. [6][7][8] IMRT technique can also be an alternative for left-sided breast cancers to decrease cardiac dose. 4,[9][10][11][12] Although there are several IMRT techniques and many dosimetric and planning studies in the literature showed better dose homogeneity with decreased organs at risk (OAR) doses, the role and consequence of IP-IMRT in breast carcinoma need to be defined. The primary aim of this study was to compare five different techniques including 3DCRT, FP-IMRT, IP-IMRT, and hybrid IP-IMRT/FP-IMRT for CW and full lymphatic irradiation with respect to target volumes and doses to critical structures in patients with left-sided breast carcinoma.
This study also aimed to define the ideal treatment plan according to the treatment planning system (TPS) and the dosimetric analysis.

2.A | Patient selection, positioning, and computed tomography scans
The computed tomography (CT) scans of 10 patients with left-sided breast carcinoma were used for this dosimetric study. All patients were positioned supine with left arm raised above the head on a breast board. Patients were scanned from the level of the mandibula through to the upper abdomen, including left and right lungs, with a 2.5-mm slice thickness on a GE™ HiSpeed NX/i (GE™ HiSpeed NX/i; GE Medical Systems, Little Chalfont, UK) CT system.

2.B | Target volumes and critical structures
All clinical target volumes (CTVs) were contoured according to the RTOG contouring atlas by a single radiation oncologist using Varian Eclipse Operation version 8.9 treatment planning software (Eclipse treatment planning system; Varian Medical Systems, Palo Alto, CA). 13 A planning target volume (PTV) margin of 5 mm was added to all CTVs in order to overcome the set-up uncertainty and internal organ motion. A PTVeval was also produced to pull back the PTV 1 mm below the skin surface for CW and 5 mm for lymphatics and exclude left lung from posterior rib surface. Normal tissues including the heart, lung, brachial plexus, spinal cord, thyroid, right breast, esophagus, and head of humerus were contoured as OARs. In addition, a structure "external" encompassed all normal tissues not otherwise specified (such as the arm, posterior thorax, and soft tissues) by substracting the volumes of the targets and specified normal tissues from the external surface. A fractionation schedule of 50 Gy in 2 Gy per fractions was used. The details of dose prescription are given in Table 1.

2.D | Dosimetric measurements
An individual simulation was done using an Alderson rando R phantom (Alderson rando phantom; Phantom Laboratory, Salem, New York, USA) for each technique. Field borders were defined, and CT markers were placed to delineate margins. The CT scans of Alderson rando R phantom were transferred to the TPS, and treatments were planned with five different techniques. After determination of treatment fields in rando phantom, thermoluminescent dosimeters (TLDs) were placed at certain depths which were thought to represent the SC, axillary, and IM lymphatics. Additional TLDs were put on several points that were thought to represent the right breast, CW, lung, and heart. In order to determine surface doses, Gafchromic TM EBT (Gafcromic EBT films; International Specialty Products, New Jersey, USA) dosimetry films were placed on the CW and right breast with 0.5 mm tissue equivalent bolus material or without bolus. In this way, two different measurements were taken for each plan. TLD and Gafchromic TM EBT dosimetry films were calibrated before treatment.     (Fig. 2-2). However, the differences were not statistically significant.   The percent difference between TPS and TLD was between 0.6-11.1% for 3DCRT and 0.7-10.9% for FP-IMRT. The biggest difference was obtained in heart for both techniques. For IP-IMRT techniques, the percent difference between TPS and TLD was between 1.4 and 7.8% for 7-field IP-IMRT and between 1.2 and 9.6% for 9-field IP-IMRT techniques. Abbreviations: 3DCRT = three-dimensional conformal radiotherapy; CTV = clinical target volume; D 95% = lowest dose received by at least 95% of the volume, for example, D 98% is the lowest dose received by at least 98% of the volume, etc.; D max = Maximum dose; FP-IMRT = forward-planned intensity-modulated radiotherapy; IP-IMRT = inverse-planned IMRT; PTV = planning target volume; SD = standard deviation; V 5Gy = percentage of the volume receiving 5 Gy or more, for example, V 10Gy is the percentage of the volume receiving 10 Gy or more, etc. *statistically significant values < 0.005. For IP-IMRT, these figures were in the range of 125 ± 4 cGy to 136 ± 2 cGy when there was no bolus. Five mm bolus caused a more homogeneous and effective surface doses ranging between 206 ± 3 cGy and 216 ± 6 cGy.

| DISCUSSION
In this study, we compared five different techniques 3DCRT, FP-IMRT, 7-and 9-field IP-IMRT, and hybrid IMRT in order to find the optimal technique in ten patients with left-sided breast carcinoma with intent to treat the CW and entire lymphatics. We found similar PTV coverage both in chest wall and lymphatics in all techniques except forward planning IMRT in which the D95% of chest wall PTV was slightly but significantly lower than the other techniques. However, when evaluating the Dmax and D2% of chest wall CTV, the dose distribution of 3DCRT technique revealed significantly higher doses compared to the other techniques. The dose homogeneity in chest wall CTV was better in IMRT techniques compared to the 3DCRT. Again IMRT especially 9-field IP-IMRT provided more homogeneous dose distribution in internal mammary and axillary CTVs. In literature, the partially wide tangential field technique is regarded as the best 3DCRT technique for breast/CW and full lymphatic irradiation. [14][15][16] However, in recent years with the introduction of modern treatment machines, FP-IMRT using FINF technique is accepted as standard technique in breast carcinoma radiatıon therapy. IP-IMRT has been shown to increase the dose homogeneity in CTVs, decrease the high-dose regions in OARs while increasing the lowdose areas in normal tissues. 17 The most widely used technique in IP-IMRT is with either seven or nine fields.
In our planning and dosimetric study, we found similar PTV coverage both in CW and lymphatics in all techniques except FP-IMRT in which the D95% of CW-PTV was slightly but significantly lower than the other techniques. However, when evaluating the Dmax and D2% of CW-CTV, the dose distribution of 3DCRT technique revealed significantly higher doses compared to the other techniques. The dose homogeneity in CW-CTV was better in IMRT techniques compared to the 3DCRT. Again IMRT, especially 9-field IP-IMRT provided more homogeneous dose distribution in IM and axillary CTVs.
The improvement of target coverage of chest wall and local lymphatics with IP-IMRT has been shown by several studies. 18 In a study by Dogan et al., 6-and 9-field IP-IMRT were shown to provide much better coverage compared to the 2-and 4-field IP-IMRT techniques. 5 In this particular study, the largest difference in average D95% of the internal mammary PTV between 3DCRT and IMRT was reported to be with 9-field IP-IMRT which is similar to ours.
Both FP-IMRT and IP-IMRT techniques in our study resulted in similar CW and regional nodal coverage. A strong relationship between radiation exposure and cancer has been shown especially by epidemiological studies conducted for survivors of atomic bombings in Japan as well as studies for medically exposed patients or radiation workers. 27,28 The overall risk of fatal cancers is estimated to be 8%/1 Gy. 29 In a study by Zhang et al.
among 5248 patients treated for breast cancer between 1965 and 1994, there was an increased risk of all second cancers combined following radiotherapy with a relative risk of 1.22. 30 The increased risk in that particular study was apparent 5 or more years after radiotherapy. In addition, it has been estimated that the conversion of treatment plan from 3D conformal radiotherapy to IMRT, due to the increase of volumes receiving lower doses, will cause additional 0.5% of surviving patients with second malignancy. 29 Moreover, an additional 0.25% of surviving patients will develop a radiation-induced malignancy because of the increase in monitor units in IMRT.
As expected, 3DCRT and forward-planned IMRT had significantly lower MUs than inverse-planned IMRT in our study. The mean MU was 420 in 3DCRT and forward-planned IMRT, whereas it was calculated as 1330 MU ± 205 for 7-field inverse-planned IMRT, and 1541 ± 254 for 9-field inverse-planned IMRT which may lead to increase in radiation-induced secondary malignancy in long-term follow-up. Moreover, V5Gy in the contralateral breast was significantly less in techniques using either 3D-CRT or field-in-field technique.
There was decrease of V5Gy with hybrid technique, but this decrease was small. Since it is a dosimetric study that did not aim to focus on secondary malignancies, we cannot estimate the risk of radiation-induced malignancies.

| CONCLUSION
In left chest wall and lymphatic irradiation, IP-IMRT techniques provide more homogeneous dose coverage when compared to 3DCRT and produce significant sparing in the heart tissue and left lung. Forward-planned IMRT has an advantage to achieve significantly lower dose to the contralateral breast tissue when compared to IP-IMRT.
The hybrid technique seems to combine the advantages of both techniques and be promising. The secondary malignancy risk and heart complications should be kept in mind when treatment plans are made.

CONFLI CT OF INTEREST
No conflict of interest.