For early breast cancer, breast conserving surgery followed by postoperative radiotherapy is accepted as equally effective as mastectomy [
In the last decades, there is a growing interest in the delivery of a simultaneously integrated boost (SIB). With this approach a daily boost dose is given to the lumpectomy area during the WBI. This results in less radiotherapy fractions. Advanced imaging techniques for accurate pretreatment staging and positioning have made it possible to more accurately delineate the boost area in the postoperative setting. With the developments in dosimetry software, an integrated boost dose within another dose volume can be calculated. Availability of daily image guidance before every radiotherapy session allows accurate repositioning of the breast and the boost volume within it.
SIB is explored earlier for other cancers [
We present a dosimetric comparison of WBI with a sequential boost, compared to 3 different techniques of WBI with SIB.
From a pool of available computed tomography (CT) scans of early breast cancer patients treated in an earlier trial [
Contouring of the target volumes and OAR was done at the time of and according to the protocol of the phase III trial [ Ipsilateral and contralateral lung: autosegmentation with manual verification. Heart: beginning from the level in which the pulmonary trunk branches into the left and right pulmonary arteries to its most extent near the diaphragm, excluding pericardial fat. Contralateral breast: all visible breast tissue from 5 mm under the skin to the anterior surface of the pectoralis, serratus anterior muscles.
In the earlier dosimetric comparison [
For the current comparison, we performed a summation of the dosimetry of a WBI with 2 tangential compensated fields on a conventional linear accelerator with the dosimetry of boost irradiation on Vero. This was the reference for the sequential boost technique and was compared to 3 SIB techniques, one with CMS XIO planning software (Elekta AB, Stockholm, Sweden) and 2 with the TomoTherapy system using the helical TomoTherapy as well as the static application (TomoDirect) for tangential IMRT (Accuray Inc., Madison, USA) (Table
Planning software and corresponding type of calculation algorithm.
Technique | Planning software | Type of calculation algorithm [ |
---|---|---|
CMS XIO | CMS XIO Release V4.62.00.13 | b |
Vero | iPlan RT Dose 4.1.2 for Vero | b |
Helical TomoTherapy | TomoTherapy Planning Station H-Art Version 4.0.5 | b |
TomoDirect | TomoTherapy Planning Station H-Art Version 4.0.5 | b |
The dose prescription to the whole breast was 50 Gy, 2 Gy per fraction. For the sequential boost a dose of 16 Gy, 2 Gy per fraction, was delivered to the initial tumor bed. For the SIB a daily integrated dose of 0.4 Gy was delivered, giving a total dose of 60 Gy to the initial tumor bed. The fraction dose of 2.4 Gy was chosen to be biologically equivalent to 66 Gy in 2 Gy per fraction according to the following formula [
The planning aims were to cover 95% of the volume of the PTV with at least 95% of the prescribed dose, but not more than 107%. For the OAR constraints and priorities were prescribed (Table
Constraints to organs at risk and priority list.
Priority | Organ | Constraint |
---|---|---|
1 | Heart | V30 ≤ 10% |
2 | Ipsilateral lung | V20 ≤ 20% |
3 | Contralateral lung | V5 ≤ 5% |
4 | Contralateral breast | V10 ≤ 5% |
To analyse conformity of the boost dose distribution, 2 parameters were evaluated. The mean volume of breast tissue that received 107% of the prescribed 50 Gy (Vol107) was compared as a measure of spilling of the dose by adding the boost. A conformity index (CI) was calculated, using the following formula [
In this setting the TV is the EVboost and the PIV is the 95% isodose of the total dose (66 Gy for sequential boost, 60 Gy for SIB).
The differences of means between the plans were compared and analyzed with ANOVA with a post hoc analysis test using a Bonferroni correction for multiple testing with IBM SPSS version 19.
The pathological T-stage ranged from T1b to T2. The mean maximal diameter of the tumor was 1.6 cm (range: 0.6–2.7 cm). In 3 patients the deepest border of the tumor was located at a minimal distance of 3.5 cm from the skin surface. In 9 patients the PTVboost reached the skin surface. The mean PTVboost volume was 71.73 cc (range: 24.91–137.88 cc). The mean volume of the PTVbreast was 704.26 cc (range: 252.56–1234.25 cc). The mean PTVboost to PTVbreast ratio was 11.26% (range: 5.49%–19.71%). Six patients had seroma with a mean volume of 9.26 cc (range: 0.00–45.84 cc). No correlation was found between the tumor size, the PTVboost volume, or the volume of the ipsilateral breast with the CI or with the dose to the OAR.
Figure
Summary of dosimetric evaluation.
Sequential boost | SIB | Significant (ANOVA) | |||
---|---|---|---|---|---|
CMS | Helical TomoTherapy | TomoDirect | |||
# violation EVboost coverage | 1 | 2 | 1 | 1 | No |
# violation EVbreast coverage | 2 | 1 | 0 | 0 | No |
# CI ≥ 50% | 0 | 9 | 3 | 5 | Yes ( |
Vol107 ipsilateral breast (cc) | 463.32 (139.50–724.95) | 230.47 (90.92–442.41) | 346.61 (118.29–605.28) | 365.15 (179.19–767.29) | Yes ( |
Dmean heart left sided (Gy) | 3.04 (1.30–7.35) | 3.12 (1.61–7.66) | 2.97 (1.59–6.89) | 4.49 (1.84–8.95) | No |
V5 heart left sided (%) | 9.61 (3.66–23.72) | 10.10 (4.29–24.70) | 9.91 (2.54–25.65) | 16.50 (4.82–27.27) | No |
V20 heart left sided (%) | 3.65 (0.83–13.20) | 3.65 (0.81–13.28) | 2.85 (0.00–11.10) | 5.98 (0.65–15.32) | No |
V30 heart left sided (%) | 2.55 (0.15–10.49) | 2.55 (0.04–10.60) | 1.44 (0.00–6.73) | 3.74 (0.32–11.79) | No |
D2 heart right sided (Gy) | 2.04 (1.25–3.72) | 3.34 (1.17–9.52) | 1.85 (1.29–2.38) | 4.30 (1.11–11.52) | No |
V20 ipsilateral lung (%) | 10.95 (6.63–16.48) | 11.15 (6.63–16.71) | 10.32 (0.00–13.84) | 11.33 (6.02–17.26) | No |
V5 ipsilateral lung (%) | 20.89 (12.59–28.87) | 24.96 (12.66–47.32) | 21.89 (4.36–28.49) | 30.45 (16.42–61.51) | No |
Dmean ipsilateral lung (Gy) | 6.26 (3.69–8.64) | 6.72 (3.73–9.20) | 6.13 (2.15–7.96) | 7.24 (4.85–11.14) | No |
Dmean contralateral breast (Gy) | 0.36 (0.03–1.25) | 0.44 (0.14–1.45) | 1.17 (0.27–3.60) | 0.46 (0.18–0.56) | Yes ( |
D2 contralateral breast (Gy) | 2.26 (0.16–12.93) | 2.39 (0.40–12.96) | 4.61 (0.86–19.46) | 1.19 (0.52–1.50) | No |
In case more than 1 value is given, the first number is the mean value; the numbers between brackets are the minimum and maximum values.
D2: near-maximum dose, as a surrogate for the point maximum dose as suggested in the ICRU Report 83 [
Example of dose distribution. Legend: red: 107% of total dose ((a) 70.62 Gy, (b) 64.2 Gy), orange: 100% of total dose ((a) 66 Gy, (b) 60 Gy), yellow: 95% of total dose ((a) 62.7 Gy, (b) 57 Gy), dark green: 55 Gy, middle green: 50 Gy, light green: 47.5 Gy, blue: 40 Gy, pink: 30 Gy, old pink: 20 Gy, purple: 5 Gy.
Sequential boost: 25 × 2 Gy + 8 × 2 Gy (total dose: 66 Gy)
SIB with CMS XIO: 25 × 2 à 2.4 Gy (total dose: 60 Gy)
None of the techniques achieved a CI of 70% or more. The CI was between 60 and 70% in none of the patients with the sequential technique, in 1 patient with CMS XIO, in 1 patient with helical TomoTherapy, and in 2 patients with TomoDirect. The CI was between 50 and 60% in none of the patients with the sequential technique, in 8 patients with CMS XIO, in 2 patients with helical TomoTherapy, and in 3 patients with TomoDirect. CI was less than 50% in all of the patients with the sequential technique, in 1 patient with CMS XIO, in 7 patients with helical TomoTherapy, and in 5 patients with TomoDirect. When using a cut off of 50%, there were more patients with a higher CI for the SIB techniques compared to the sequential boost. This difference was statistically significant (
The difference in dose to the heart and ipsilateral lung was not statistically significant between the 4 treatment techniques. The V5 of the contralateral breast was <5% for all patients with the sequential technique, with CMS XIO and with TomoDirect. With helical TomoTherapy the V5 was >5% for 3 patients. The mean dose to the contralateral breast was low with all techniques but did show a statistically significant difference with the ANOVA test (
The schedules and techniques for postoperative radiotherapy are developed in an era when there was no CT-scan to adjust the fields to the anatomy of the individual patient and when there were less advanced imaging techniques available to perform an accurate pretreatment staging and positioning. Electron boost dosimetry was not possible and positioning of the boost area was done clinically based upon not much more than the localization of the surgical scar and palpation of postoperative changes within the breast, guided by preoperative mammography, which deforms the breast to achieve a good image quality.
The last decades have shown a tremendous evolution in techniques for imaging and radiotherapy delivery. Yet, very often, the breast radiotherapy technique has not evolved with it. Developments in imaging techniques, dosimetry software, and treatment machines have led to the possibility of giving postoperative WBI with SIB for breast cancer patients.
The cosmetic outcome after breast conserving therapy for early breast cancer is of high importance. Less dose spilling within the treated breast theoretically diminishes the risk of developing fibrosis or late skin reactions. But a higher fraction dose could increase the risk of late skin or breast toxicity [
The dosimetric advantages of SIB for breast treatment were examined earlier [
Hijal et al. compared a SIB technique with helical TomoTherapy to a 3D conformal SIB technique, showing that the helical TomoTherapy delivers better homogeneity and less dose spilling than the 3D technique [
In our comparison we find that an arc technique in most situations delivers a better homogeneity than static fields for SIB delivery. This is different in case of a sequential boost. In an earlier dosimetric comparison of different noninvasive sequential boost techniques [
Franco et al. delivered SIB with TomoDirect [
Compared to the earlier comparison of sequential boost techniques [
For our analyses, we use the formula suggested by Paddick [
SIB does not cause higher dose to the surrounding OAR compared to sequential boost. We do not expect an increased cardiac or pulmonary toxicity because of a SIB, since it delivers equal dose to the heart and lungs. Thus, SIB should be equally feasible as a sequential boost for patients who benefit of a systemic treatment.
Though only low mean dose to the contralateral breast is seen, it was highest with helical TomoTherapy. This seems inherent to the rotational technique and not caused by the SIB [
Based upon the results of this comparison, we changed the planning target values in our daily routine. We apply 2 levels of target values because we want to find the best possible dosimetry, not only a dosimetry that fits the constraints, but one that with some extra effort could be better. The first level contains very strict target values that we always strive for, derived from the mean results of the CMS XIO dosimetry in this comparison. At this level, we aim for a V30 of the heart of less than 2% for a right breast and less than 5% for a left breast irradiation. We try to keep the V10 at 0% for a right breast and less than 20% for a left breast irradiation. The mean dose should not exceed 2 Gy for a right breast and 4 Gy for a left breast. The heart has first priority, since a recent population-based case-control study showed that the rates of major coronary events increased linearly with the mean dose to the heart by 7.4% per Gy, with no apparent threshold [
We aim to keep the V20 of the ipsilateral lung beneath 15% when there is no lymph node irradiation and beneath 20% when the lymph node areas are part of the target. We aim for a V5 of the ipsilateral lung less than 30%. The contralateral breast is outside the field.
A minor violation exists if these very strict target values cannot be met. Minor violations are no reason to reject the dosimetry. There is a second level with absolute constraints, which are more commonly used and are based primarily upon the presumed risk of developing toxicity. The V30 of the heart should not exceed 5% for a right breast, 15% for a left breast. The maximal tolerable V10 is 10% for a right breast, 30% for a left breast, and the mean dose has to be less than 5 Gy for both sides.
The V20 of the ipsilateral lung should be less than 30%, for both lungs less than 20% with a mean dose not exceeding 15 Gy. The V5 of the ipsilateral lung should be as low as possible, but not more than 70%.
If one of these constraints cannot be met, a major violation exists. In case of one or more major violations, the treating radiation therapist has to decide if the dosimetry is clinically acceptable or should be rejected.
When a SIB is given, the lumpectomy area is irradiated with a higher dose from the start of the radiotherapy course. It is important that the delay between surgery and start of radiotherapy is long enough for postoperative changes to heal. If there is a postoperative seroma or hematoma of more than 30 cc, the boost area becomes relatively large, which has a negative impact on cosmetic outcome [
Next to the dosimetric advantages, SIB has practical advantages. There are less fractions, which has an economic benefit, not only for the radiotherapy department [
In conclusion, we confirm the dosimetric advantages of SIB for breast irradiation, even when compared to an advanced and highly conformal sequential technique. SIB can be performed with acceptable dosimetry on a conventional linear accelerator or on TomoTherapy. With helical TomoTherapy, there is a risk of higher dose to the contralateral breast. On a conventional linear accelerator, a technique with tangential compensated fields for WBI and arc technique for SIB is preferable in most situations. SIB seems a safe alternative and can be easily implemented in clinical routine.
The authors declare that there is no conflict of interests regarding the publication of this paper.
Hilde Van Parijs and Truus Reynders contributed equally to the paper.