A new three-dimensional conformal radiotherapy (3DCRT) technique for large breast and/or high body mass index patients: evaluation of a novel fields assessment aimed to reduce extra–target-tissue irradiation

Abstract

Objective:

To develop an alternative three-dimensional treatment plan with standardized fields class solution for whole-breast radiotherapy in patients with large/pendulous breast and/or high body mass index (BMI).

Methods:

Two treatment plans [tangential fields and standardized five-fields technique (S5F)] for a total dose of 50 Gy/25 fractions were generated for patients with large breasts [planning target volume (PTV) >1000 cm³ and/or BMI >25 kg m⁻²], supine positioned. S5F plans consist of two wedged tangential beams, anteroposterior: 20° for the right breast and 340° for the left breast, and posteroanterior: 181° for the right breast and 179° for the left breast. A field in field in medial–lateral beam and additional fields were added to reduce hot spot areas and extra–target-tissue irradiation and to improve dose distribution. The percentage of PTV receiving 95% of the prescribed dose (PTV V_95%), percentage of PTV receiving 105% of the prescribed dose (PTV V_105%), maximal dose to PTV (PTV D_max), homogeneity index (HI) and conformity index were recorded. V_10%, V_20%, V_105% and V_107% of a “proper” normal tissue structure (body-PTV healthy tissue) were recorded. Statistical analyses were performed using SYSTAT v.12.0 (SPSS, Chicago, IL).

Results:

In 38 patients included, S5F improved HI (8.4 vs 10.1; p ≤ 0.001) and significantly reduced PTV D_max and PTV V_105%. The extra–target-tissue irradiation was significantly reduced using S5F for V_105% (cm³) and V_107% (cm³) with a very high difference in tissue irradiation (46.6 vs 3.0 cm³, p ≤ 0.001 for V_105% and 12.2 vs 0.0 cm³, p ≤ 0.001 for V_107% for tangential field and S5F plans, respectively). Only a slight increase in low-dose extra–target-tissue irradiation (V_10%) was observed (2.2719 vs 1.8261 cm³, p = 0.002).

Conclusion:

The S5F technique in patients with large breast or high BMI increases HI and decreases hot spots in extra-target-tissues and can therefore be easily implemented in breast cancer radiotherapy.

Advances in knowledge:

The treatment planning strategy proposed in this study has several advantages: (a) it is extremely reliable as the standard supine positioning is used; (b) the standardized class solution allows for widespread use; (c) time and cost of treatment are not increased; and (d) it can be used for both large breasted and obese patients not compliant to different treatment positioning.

INTRODUCTION

The breast-conserving approach with the adjuvant use of radiotherapy (RT) has gained an important role in the treatment for early-stage breast cancer with excellent long-term local control and survival.¹ During or shortly after the course of breast cancer RT, a large portion of patients will experience acute radiation dermatitis and breast oedema to some degree.² There is accumulating clinical evidence that acute reactions are associated with the development of late toxicity which includes telangiectasia, breast induration and pain.^3–5

The severity of RT-induced breast damage relates to the biological damage to the epidermis, dermis and connective tissue during treatment.⁶ Although all patients undergoing RT are at risk of these reactions, there are a number of intrinsic and extrinsic risk factors that can affect the severity.⁷ Intrinsic factors include age, current or previous history of smoking, malnourishment, concomitant disease/medication, obesity, the presence of skin folds and previously irradiated areas.^8–10 Extrinsic factors relate to the type of treatment with regard to the energy of the radiation beam, the dose distribution with hot spots and/or the field size of the treated area.

Although the skin and subcutaneous tissues are not a dose-limiting tissue, skin toxicity, breast induration and pain are associated with impairment of the patients' quality of life, causing pain and discomfort and limiting activities.^2,11

Furthermore, surgery, chemotherapy and RT can have a significant impact on health-related quality of life and can be worse in females with higher body mass index (BMI).¹² The majority of existing studies on obesity and health-related quality of life in females with breast cancer have not specially focused on patients undergoing RT.^12–14 Obesity has been associated with greater acute toxicity with adjuvant RT for breast cancer.^12,13 Higher BMI has also previously been associated with breast cancer treatment-related lymphoedema and inferior disease-specific survival in both pre- and post-menopausal females.^13–15 Overweight and obese females are at higher risk of developing treatment-related symptoms, with higher BMI found to be associated with development of long-term pain after breast cancer treatment.^15,16

Moreover, the treatment of large or pendulous breasts has been associated with impaired cosmetic outcomes due to inhomogeneous dose distribution.¹⁷ This anatomic feature, characterized by the tendency of the breast to fall laterally and/or superiorly, leads to the inclusion of a larger portion of the lung and the heart in the treatment fields and increases inframammary folds yielding a bolus effect.¹⁸

In these patients, the challenge is to minimize these side effects without losing efficacy of the treatment.^19,20

Prone position had been used in order to reduce the dose received by normal tissues and the size of hot spots.^21,22 However, coexisting large or pendulous breasts and high BMI make this kind of approach difficult because of some positioning complications.

Introducing improved radiation techniques, such as intensity-modulated RT (IMRT), has led to a reduction in acute skin toxicity and late fibrosis and improved cosmetic outcome.^19,20 This type of treatment is, however, not available in all RT centres, whereas breast cancer RT is one of the most diffused treatments in radiation oncology.

The aim of our study was to develop an alternative simple and easily available three-dimensional (3D) treatment plan with a standardized fields class solution for patients with large/pendulous breast and/or high BMI in order to improve homogeneity and to reduce hot spots and extra–target-tissue dose irradiation.

METHODS AND MATERIALS

A retrospective cohort of patients who underwent conservative surgery and adjuvant breast irradiation with 3D conformal RT technique were selected for this study. Patients were selected based on their anatomy (large or pendulous breasts, CT simulation breast volume >1000 cm³ or when breast tissue tended to fall towards the mid axillary line, and/or a BMI of >25 kg m⁻²). Simulations were performed with patients positioned supine on a breast board with the ipsilateral arm raised above the head. The scan was extended from the jugular notch to 5 cm below the lower edge of the breast with a scan interval of 5 mm. The target volume, heart and lungs were manually contoured on each CT slice by a single radiation oncologist following the Radiation Therapy Oncology Group guidelines²³ and heart atlas published by Feng et al.²⁴ The clinical target volume was defined as the entire breast excluding the outer 4 mm from the skin surface. The planning target volume (PTV) was defined as clinical target volume + 5 mm in the direction of the chest wall. An apposite normal tissue structure (body-PTV healthy tissue) was generated subtracting the PTV from the body volume.

Treatment planning techniques

Eclipse™ v. 7.3.10 (Varian Medical Systems, Palo Alto, CA) treatment planning was used to generate two different treatment plans for each patient [a = “tangential fields” (TF); b = “standardized five-fields technique” (S5F)] described below. The pencil beam convolution 7.5.18.0 algorithm was used.

The total prescribed dose was 50 Gy at the isocentre in accordance with the International Commission on Radiation Units Measurement. All plans were optimized according to the following constraint for the PTV: V_95% ≥ 95% (the volume receiving 95% of the prescription dose or more must be ≥95% of the PTV). This was considered as the primary constraint. As a secondary constraint, we considered the following: PTV V_105% ≤ 5% of the prescribed dose.

Tangential fields treatment plans

Treatment plans consisted of a simple wedged tangential plan (with gantry angles optimized to match divergence of the posterior edges of the beam) to avoid contralateral breast irradiation and to minimize the ipsilateral lung and heart area in the beam's eye view or field-in-field technique, in which the dose on each of the two tangential beams was split into two different segments. The first segment was designed to encompass the entire breast. A second segment was then directed to this area of underdosing, in order to compensate for the drop in dose and to reduce hot spot areas. The weights of the two segments were determined through an iterative process repeated until optimal results were achieved. The weight of these segments is typically in the range of 10–15% of the total. The energy of the photon beams was 6 MV for TFs and 15 MV for field-in-field beams; in some cases, to increase dose coverage in depth, the energy of the tangential beams was also 15 MV.

Standardized five-fields technique treatment plans

Five-fields treatment plans were made with two wedged tangential beams such as above (fields 1–2), with an anteroposterior (AP) field with 20° rotation starting from the zero-position (linear accelerator's gantry 20° for right breast and 340° for left breast treatment) (field 3), and a posteroanterior (PA) field with 179° rotation starting from the zero position (linear accelerator's gantry 181° for right-breast treatment and 179° for left-breast treatment) in order to avoid collision between the linear accelerator and table (field 4). Moreover, a field in field with a weight of 10–15% and the same geometry of the medial–lateral beam (field 5) was added to reduce hot spot areas, to homogenize dose to the target and to improve dose distribution in the chest wall region. The fields AP and PA, with a weight of 10–15%, were optimized to compensate the dose fall localized at the centre of the breast, avoiding the extra-target-tissues at the level of the scapula and including not >1 cm of the lung in the field's beam's eye view. The energy of photon beams was 6 MV for TFs and 15 MV for field-in-field, AP and PA beams.

Statistical analysis

Dose–volume histograms were generated for PTV and organs at risk for all plans. For PTV, the percentage of PTV receiving 95% of the prescribed dose (V_95%), percentage of PTV receiving 105% of the prescribed dose (V_105%), maximal dose to PTV (D_max) and a homogeneity index (HI) defined as HI = 100 × (D_2% − D_98%)/D_p, where D_p is the prescribed dose, were chosen as parameters for comparison. Lower HI values indicate a more homogeneous target dose. V_105% was chosen to specify the target volume receiving high doses. In addition, a conformity index (CI) was defined as CI = V_ri/V_PTV, where V_ri is the volume encompassed by the reference isodose for the PTV (95% of the prescribed dose) and V_PTV is the PTV volume expressed in cm³. Body-PTV healthy tissue V_10%, V_20%, V_105% and V_107% were defined as surrogates for extra–target-tissue irradiation. For each dosimetric variable, normality was tested using the Shapiro–Wilk test. For the ipsilateral lung, the two plans were compared in terms of mean lung dose, V_2.5Gy, V_5Gy, V_20Gy, V_40Gy and V_50Gy and for the heart, mean heart dose, V_2.5GyV_5Gy, V_15Gy and V_25Gy. Comparisons were made by means of Mann–Whitney test.

The statistical analysis was performed using SYSTAT v. 12.0 (SPSS, Chicago, IL).

RESULTS

Patients' characteristics

38 patients were included in the analysis. Patients' characteristics are listed in Table 1. 17 (44.7%) patients were right sided, 21 left sided (55.3%). The mean BMI was 33.6 (standard deviation 7.1). 22 (57.9%) patients had a breast CT simulation volume >1000 cm³.

Table 1.

Patients' characteristics

Patients' characteristics	n
Patients	38
Age, years
Median	65
Range	40–82
Tumour side
Right	17
Left	21
Breast separation (cm)
Mean	18.7
SD	2.6
BMI
Mean	33.6
SD	7.1
PTV (cm3)
Mean	1077.0
SD	362.4

BMI, body mass index; PTV, planning target volume; SD, standard deviation.

Target coverage

The analysis results for target coverage are reported in Table 2. The CI, as well as V_95%, was not significantly different between the two groups (p = not significant). However, the constraint PTV V_95% > 95% was achieved for all plans realized with S5F, but not for TF treatment plans. S5F improved dose homogeneity (8.4 vs 10.1; p ≤ 0.001). In particular, D_max and PTV V_105% were significantly reduced with this approach (106.9% vs 108.6%, 1.9% vs 4.46% respectively; p ≤ 0.001) (Figure 1).

Table 2.

Comparison of target coverage data (median values, range)

Dosimetric data	Standard tangential	5-fields technique	p-value
HI
Median	10.1	8.4	<0.001
Range	7.1–38.6	6.8–16.5
CI
Median	1.0	1.0	0.795
Range	0.7–1.0	0.9–1.0
PTV V95% (%)
Median	98.9	98.9	0.752
Range	71.9–99.9	95.1–99.9
PTV V105% (%)
Median	4.5	1.9	<0.001
Range	0.1–9.3	0.0–6.4
PTV Dmax% (%)
Median	108.9	106.9	<0.001
Range	105.1–117.0	105.2–111.0
PTV Dmax (Gy)
Median	54.1	53.4	<0.001
Range	52.7–57.5	52.5–55.3

CI, conformity index; HI, homogeneity index; PTV, planning target volume.

HI = 100 × (D_2% − D_98%)/prescribed dose.

CI = volume of PTV encompassed by 95% of the prescribed dose/PTV volume.

Figure 1.

(a, b) 95% Isodose distribution: (a) tangential fields (TF) technique; (b) standardized five-fields technique (S5F). (c, d) Body-planning target volume (PTV) healthy tissue 105% isodose distribution (extra–target-tissue irradiation): (c) TF technique; (d) S5F. (e, f) Body-PTV healthy tissue 10% isodose distribution (extra–target-tissue irradiation): (e) TF technique; (f) S5F. (g) Dose–volume histogram (DVH): PTV, ipsilateral lung, heart and body-PTV healthy tissue DVH for TF (dashed line) and Standardized five-fields technique (solid line).

Normal tissue irradiation

Normal tissue dosimetric data are summarized in Table 3 (median value and range).

Table 3.

Comparison of organ at risk (median values, ranges)

Organs at risk	Standard tangential	Five-fields technique	p-value
Lung (ipsilateral)
MLD (Gy)
Median	3.8	4.1	0.585
Range	1.8–12.1	2.2–12.2
V2.5Gy (%)
Median	25.7	28.8	0.010
Range	13.4–45.8	20.8–45.7
V5Gy (%)
Median	13.0	14.8	0.112
Range	4.9–37.3	7.2–38.9
V20Gy (%)
Median	5.0	5.2	0.986
Range	1.0–23.3	0.6–23.3
V40Gy (%)
Median	2.8	2.8	0.901
Range	0.1–16.9	0.0–14.9
V50Gy (%)
Median	0.0	0.0	0.124
Range	0.0–1.8	0.0–0.9
Heart (left sided)
MHD (Gy)
Median	2.0	1.9	0.562
Range	1.1–2.6	1.3–2.5
V2.5Gy (%)
Median	17.1	16.5	0.656
Range	3.5–28.8	6.5–27.9
V5Gy (%)
Median	4.3	4.1	0.519
Range	0.0–8.2	0.0–7.5
V15Gy (%)
Median	0.5	0.4	0.769
Range	0.0–1.7	0.0–1.6
V25Gy (%)
Median	0.2	0.1	0.934
Range	0.0–0.8	0.0–0.8
Heart (right sided)
MHD (Gy)
Median	0.8	0.9	0.534
Range	0.4–1.2	0.5–1.2
V2.5Gy (%)
Median	0.0	0.0	0.820
Range	0–8.5	0.0–4.2
V5Gy (%)
Median	0.0	0.0	0.264
Range	0.0–0.0	0.0–0.0
V15Gy (%)
Median	0.0	0.0	1000
Range	0.0–0.0	0.0–0.0
V25Gy (%)
Median	0.0	0.0	1000
Range	0.0–0.0	0.0–0.0
Body-PTV healthy tissue
V10% (cm3)
Median	1826.1	2271.9	0.002
Range	942.6–3472.8	1117.2–3601.0
V20% (cm3)
Median	1561.0	1587.0	0.720
Range	810.2–2974.4	824.2–2577.5
V105% (cm3)
Median	46.6	3.0	<0.001
Range	0.0–394.6	0.0–41.0
V107% (cm3)
Median	12.2	0.0	<0.001
Range	0.0–298.8	0.0–5.4

MLD, mean lung dose; MHD, mean heart dose; PTV, planning target volume.

Lung and heart dosimetric parameters analysed showed no differences between the two techniques (Table 3), with the exception of lung V_2.5Gy which was increased in S5F treatment plans (28.8% vs 25.7%, p = 0.010). Particularly, heart dosimetry (mean heart dose, V_2.5Gy, V_5Gy, V_15Gy, V_25Gy) was not different regardless of the breast cancer side. On the contrary, high-dose extra–target-tissue irradiation (body-PTV healthy tissue) was significantly reduced using S5F; in particular, absolute V₁₀₅ values are reduced by 15 times (46.6 vs 3.0 cm³, p ≤ 0.001) and V_107% from 12.2 vs 0 cm³ (p ≤ 0.001) (Figure 1). However, a slight increase in low-dose extra–target-tissue irradiation (body-PTV healthy tissue V_10%) was observed using S5F (2.2719 vs 1.8261 cm³, p = 0.002) (Figure 1).

DISCUSSION

The aim of this study was to develop a simple and easily available alternative to the tangential 3D treatment plan with a standardized fields class solution for patients with large/pendulous breast and/or high BMI in order to reduce hot spots particularly at the inframammary fold and extra–target-tissues irradiation. In our series, the S5F showed improved dose homogeneity (8.4 vs 10.1; p ≤ 0.001), lower PTV V_105% (4.5 vs 1.9%; p ≤ 0.001) and extra–target-tissue irradiation (body-PTV healthy tissue) (46.6 vs 3.0 cm³, p ≤ 0.001).

Several attempts had been carried out in order to achieve a better dose distribution and thus reduce radiation-induced toxicity, including the use of novel techniques (inverse/forward IMRT), different patient positioning (lateral/prone decubitus) and using immobilization devices that displace the breast from the chest wall.

Pignol et al¹¹ in a multicentre randomized trial with a total of 358 patients showed how IMRT technique improved radiation treatment quality by reducing the clinical significant maximum dose gradient (105% vs 110%; p ≤ 0.001), PTV V_105% (7.5% vs 16.9%), and reported how these dosimetric benefits translated into a lower proportion of patients experiencing acute moist desquamation (31.2% vs 47%; p = 0.002). A British study with median follow-up of 5 years, which included 1145 treated patients, showed that the use of forward planned IMRT reduces the rates of telangiectasia (odds ratio 0.58, 95% confidence interval: 0.36–0.92, p = 0.021).²⁵

Similarly, a study carried out after 4.7 years of median follow-up showed a reduction in chronic breast oedema (3% vs 30%; p = 0.007) and hyperpigmentation (3 vs 41%; p = 0.001), especially in patients with larger breasts (volume >1600 cm³).²⁶

Even though breast cancer RT is one of the most diffused applications of radiation oncology, this technique is not, however, always available in all RT centres, as the use of IMRT is not always justified due to the cost of technology, particularly when less expensive alternatives that yield similar target dose coverage and normal organ sparing are readily available.

Lateral decubitus position decreases separation and improves dose homogeneity in the target by changing breast shape. Recently, Kirova et al²⁷ reported in a study carried out on 56 patients that a maximum dose to the breast of 53.48 Gy on average, a low mean lung and heart dose of 0.96 and 1.35 Gy, respectively, and an incidence of acute grade 3 dermatitis of 1.8% was acceptable.

The greatest challenge of this positioning lies in the complexity of daily set-up. Prone set-up yields the same advantages of lateral decubitus RT with a simpler set-up and better accuracy. Moreover, several studies have investigated prone RT where dosimetric studies^22,28 showed increased homogeneity and reduced lung doses with prone positioning in comparison with supine positioning, whereas in a large single-centre study by Stegman et al,²⁹ that reviewed the data of 245 patients treated over a 12-year period, planning was with opposed coplanar beams and the median hot spot percentage was 106% (interquartile range 104–108%) in the majority of cases.

Grade 3 acute skin toxicity reported with prone positioning ranges from 2% to 14.5%.^29,30 Even though prone set-up is considered a reasonable option for large breasted patients, it, however, requires patients' compliance and some older patients may not be able to maintain the position. Additionally, in our series 63.0% of patients had a BMI >30% and 94.3% had a BMI >25. In these overweight/obese patients, if a large bore CT scan is not available, prone set-up may not be feasible.

Finally, approaches were proposed which included the construction of a thermoplastic mould or reinforced polyvinyl chloride ring and Styrofoam™ (The Dow Chemical Company, Midland, MI)^17,31,32 to pull the lateral breast tissue anteriorly and upright and to decrease breast separation.

Recently Arenas et al¹⁸ reported on the use of breast cups on patients with large or pendulous breasts.

The use of breast cups resulted in a significant reduction of the PTV volume (from 1640 to 1283 cm³) of the irradiated volume (from 2154 to 1477 cm³) and of the CI (from 1383 to 1213). Furthermore, the use of breast cups also led to significant dose reductions in V₂₀ for the lung (from 13.7% to 1.7%) and V₅ for the heart (from 9.8% to 2.7%).

Despite these encouraging results, however, these devices may be limited in standard application due to decreased reproducibility, patient discomfort and costs.

On the contrary, the treatment planning strategy proposed in this study has several advantages: (a) it is extremely reliable as a standard supine positioning is used; (b) the standardized class solution allows for widespread use; (c) time and cost of treatment are not increased; and (d) it can be used for both large breasted and obese patients not compliant to different treatment positioning.

There are some drawbacks in our study. First, treatment plans were performed by an “expert” operator. Second, low doses (lung V_2.5Gy and body-PTV healthy tissue V_10%) were slightly increased using S5F due to the contribution derived both from the AP and PA field. However, we believe this is limited as the two fields had low weight (10–15%). Furthermore, we have no clinical data on the efficacy of this treatment in reducing skin toxicity. For this reason, we are currently evaluating this 3D treatment strategy in a clinical setting.

In conclusion, S5F in patients with large breast or high BMI increases HI and decreases D_max hot spots in extra-target-tissues by a factor of 15 and can therefore be easily implemented in post-operative breast cancer RT.