Accelerated partial breast irradiation
Consensus statement of 3 German Oncology societies
Breast-conserving surgery followed by whole-breast radiotherapy (WBRT) has become the standard treatment for the majority of patients with early breast cancer. Whereas the indications for systemic adjuvant treatment have continuously expanded, there is a tendency to restrict postoperative radiotherapy to accelerated partial breast irradiation (APBI) instead of WBRT.
The different techniques of APBI are described and their respective advantages or potential drawbacks outlined. Moreover, the results described in the literature are briefly reviewed as a basis for the consensus statements and recommendations of the German Society of Radiation Oncology, the German Society of Senology, and the Working Group for Gynecological Oncology of the German Cancer Society.
The methods mainly used for APBI are: interstitial radiotherapy with multicatheter technique, intraoperative radiotherapy (IORT) using either electrons produced by linear accelerators or 50 kV x-rays (Intrabeam), the balloon-catheter technique (MammoSite), or 3D conformal external beam radiotherapy. These techniques have marked differences in dose distribution and homogeneity. The published range of local recurrence rates varies between 0% to 37%, the median follow-up from 8 to 72 months.
To date, follow-up times mostly do not yet permit a definite judgment concerning the long-term effectiveness and side effects of APBI. The relevant societies in Germany support randomized clinical studies comparing APBI with WBRT in a well-defined subset of low-risk patients. However, the authors expressly discourage the routine use of APBI outside clinical trials. Until definite results show that APBI neither impairs therapeutic outcome nor cosmetic results, WBRT remains the gold standard in the treatment of early breast cancer. Cancer 2007. © 2007 American Cancer Society.
Breast-conserving surgery and adjuvant whole-breast radiotherapy (WBRT) have become a standard treatment for the majority of patients with early breast cancer. Usually, the whole breast is irradiated with a dose of about 50 Gy. Depending on risk factors, an additional local boost dose of 10–20 Gy is applied to the tumor bed.1, 2 According to the tumor stage and nodal status, the lymphatic pathways may be included in the treatment volume.3, 4 Whereas the 10-year local recurrence rate is 29.2% in node-negative patients (N−) and 46.5% in node-positive patients (N+) after breast-conserving surgery alone, WBRT provides an absolute reduction of in-breast-recurrences of 19.2% (N−) and 33.4% (N+) and an absolute gain in survival of 5.4% after 15 years.5
As most in-breast recurrences are observed in the neighborhood of the tumor bed and as carcinomas occurring in distant parts of the same breast are often classified as secondary primaries,6 it has been hypothesized that some of the patients might be adequately treated by partial breast irradiation alone. Several studies have been initiated using different concepts and techniques.7, 8–19 All these methods are based on the concept of accelerated partial breast irradiation (APBI) such as brachytherapy with interstitial implants or balloon catheters, intraoperative radiotherapy with electrons or 50 kV x-rays, or 3D conformal external beam radiotherapy. The rationale for using these methods of APBI instead of WBRT is mainly derived from the extensive experience in their use for boost irradiation.
The aim of the present study was to analyze the specific properties of the aforementioned techniques and to outline a general consensus statement of the German Society of Radiation Oncology, the German Society of Senology, and the Working Group for Gynecological Oncology of the German Cancer Society regarding APBI as a sole method of adjuvant radiotherapy for early breast cancer.
OVERVIEW OF THE DIFFERENT MODALITIES
Interstitial Multicatheter Technique
Between 6 to 12 weeks after breast-conserving surgery (BCS) several plastic catheters were inserted into the tumor bed using a parallel geometry and a safety margin of about 20 mm. Usually the implants are inserted in 3 planes with a distance of 10–20 mm from each other. Irradiation is performed with the fractionated afterloading technique using high dose rate (HDR) iridium-192 (eg, 32 Gy in 8 fractions or 30.1 Gy in 7 fractions in 2 daily sessions) or as pulsed dose rate brachytherapy (24 hours per day, total dose 50 Gy).
Partial breast irradiation with the multicatheter technique is an invasive procedure that requires extensive clinical experience in target volume definition, dosage, and fractionation. The method has the following advantages:
In contrast to intraoperative procedures, the histopathologic report concerning the removed breast tumor and the axillary lymph node status are already present at the time of irradiation.
The flexibility of the method permits individual conformation of the irradiated volume precisely adapted to the anatomical conditions and the radiobiologically necessary dose but sparing surrounding tissue and skin.
The geometry of the dose distribution is reproducible.
Acute and late toxicities are acceptable and the long-term cosmetic results are mostly good or excellent,15, 20 although in 2 studies grade ≥3 late toxicity occurred in 33 and 59%.21, 22
The interstitial multicatheter technique is the only method of APBI with reported experience and data covering > 5 years (range, 11–144 months).23 Several phase 1 and 2 protocols suggest that in a selected patient collective the tumor control rates in early breast cancer are comparable to those with conventional WBRT.11, 15, 17, 24 The in-breast tumor control rate in these studies was between 92.3% and 100% after an observation time of up to 7 years. Whereas local recurrence rates in the early phase were up to 37%,25 they range between 0% to 16% in more recent series.24, 26–28
For studies concerning APBI with the multicatheter technique, the following restrictive inclusion criteria are recommended: tumor size 3 cm (no bilateral or secondary carcinoma); unifocal tumor, no multifocality or multicentricity; free resection margins with a minimal safety margin of more than 2 mm; no vascular invasion; no lymphatic (small) vessel invasion; no Paget-Carcinoma, no skin infiltration; no extensive intraductal component; limited ductal carcinoma in situ (DCIS)-Van Nuys Prognostic Index29 less than 8 of 12; negative axillary lymph nodes (pN0 or pNmi); and age >40 years. However, these strict criteria are not generally applied. For example, the RTOG 95-17 study permitted accrual of patients with up to 3 positive axillary nodes and also patients of age 18 to 40 years. However, only 3% of the patients in this study were younger than 40 years and 20% had involved axillary nodes.30
Currently, the European Brachytherapy Breast Cancer Working Group of the GEC-ESTRO is performing an international phase 3 study investigating the effectiveness and side effects of brachytherapy alone in comparison with a conventional WBRT in patients older than 40 years of age presenting with unifocal DCIS or invasive cancer of 3 cm or smaller and no more than 1 micrometastasis in the axillary lymph nodes.31 The primary endpoint is locoregional tumor control; secondary endpoints are incidence and severity of side effects and the cosmetic outcome. We planned to randomize 1170 patients.
Intraoperative Radiotherapy (IORT)
For IORT either electrons produced by linear accelerators or 50 kV x-rays (Intrabeam) are available. Both modalities are used for either applying a local boost before WBRT or as partial breast irradiation during BCS. The rationale for this modality is based on the idea that radiotherapy of the tumor bed should be performed without delay immediately after removal of the tumor. Moreover, with an intraoperative modality the application of a high dose precisely targeted to a limited volume can be performed while sparing the surrounding tissue, thus providing good cosmetic results. One disadvantage is that at the time of IORT definite information about the resection margin, pathohistologic features, and lymph node status is not yet available.
Intraoperative radiotherapy with electrons (IOERT)
Historically, APBI with electrons is based on boost treatment. Dobelbower et al.32 published the first results of an intraoperative electron boost before WBRT in 1989. The first long-term results were presented in 1997 together with a French group.33, 34 This modality was also implemented in 1998 in Salzburg with a conventional linear accelerator dedicated to intraoperative radiotherapy. After tumor excision, margins were temporarily adapted with a suture and encompassed into the radiation field with a safety margin.35 The results of a study with 541 patients who received IOERT as an anticipated boost were published after a median follow-up time of 30.5 months, with no local recurrence being detected.36 The Milan group started IOERT in 1999 with a dedicated mobile linear accelerator (Novac 7) providing 3–12 MeV.9 Patients received an anticipated boost of 10–15 Gy before homogeneous WBRT. After the above-mentioned pilot phase using IOERT as boost, the Milan group switched to higher intraoperative single fractions (22.3 Gy), using IOERT as a sole modality and omitting radiotherapy of the whole breast. A single fraction of 20–22 Gy is considered biologically equivalent to a conventionally fractionated dose of about 50–60 Gy. The procedure in Milan differs from surgery in other institutions as breast-conserving therapy is mostly performed as a quadrantectomy, which is not regarded as optimal in terms of cosmetic outcome. It is required to achieve free resection margins and an additional safety margin of 10 mm. After mobilization of the mammary gland, a lead shield is put between the gland and muscle to avoid necroses of the ribs and chest wall. The authors state that this modality is not associated with significant side effects and is well accepted by patients.9
Several issues appear to be critical with this method: Preparation of all those areas at risk for potential subclinical tumor invasion and centering these areas into the radiation field require extensive surgical and radio-oncologic expertise, especially with the aspect that no further treatment of the breast is planned. The method is demanding with regard to technique and radiation protection, and therefore not available everywhere.
Nonetheless, the experiences with IOERT in the European Institute of Oncology in Milan as a sole APBI14, 37 are encouraging. After a median follow-up of 27 months, no local recurrence was observed. For further validation of the technique and confirmation of the concept as the sole modality, prospective randomized clinical studies are required and ongoing.
Intraoperative radiotherapy with 50 kV x-rays (Intrabeam)
For the Intrabeam technique a 50-kV x-ray machine (Orthovolt) is used whose radiation is scattered by a spherical applicator for homogeneous delivery of x-rays into the resection cavity. The applicator size is variable to be adapted to the size of the resection cavity. The dose prescription is 20 Gy on the applicator surface12; this corresponds physically to a dose of 6 Gy in 10 mm tissue depth. Low-energy x-rays have a higher relative biological efficiency.38
Several questions arise when this modality is used as a sole treatment for breast cancer: 1) The physical characteristics of the 50 kV-machine provide a restricted spherical distribution of the dose without any individual adaptation, for instance, to individual shape of the resection cavity or the resection margins and the consecutive risk situation. 2) The dose decrease from the applicator surface (20 Gy) to the surrounding tissue is extremely steep and substantially depending on the size of the applicator. The assumption that a dose of 6 Gy in 10 mm is sufficient to sterilize microscopic residual disease is based on radiobiologic calculations that are still experimental and not yet clinically validated.
The technique had first been studied for intraoperative boost irradiation12, 39, 40 and was then introduced for APBI alone in the TARGIT (targeted intraoperative radiotherapy) protocol. Treatment was feasible and well tolerated. Vaidya et al.41 started a pilot study in 2001 including 25 patients and initiated an international randomized trial (Great Britain, Australia, USA, Germany) to investigate whether the outcome with intraoperative radiotherapy using Intrabeam is equivalent to conventional WBRT.
Inclusion criteria are restricted to patients with a supposed low risk for failure: age > 50 years; <2 cm unifocal carcinoma; ductal-invasive subtype; free margins of 10 mm; no lymphatic vessel invasion. If the pathohistologic workup shows that the above criteria are not met or reveals additional risk factors (tumor size over 2 cm, free margin of less than 1 cm, extensive intraductal component [EIC] over 25% of the tumor volume, or another type of histology), a subsequent external WBRT with 46 Gy is added. In the case of positive resection margins the surgery is repeated. First results are expected in the near future.
Balloon Catheter Technique (MammoSite)
Recently, the balloon catheter has been introduced, which is a rather simple method for intracavitary APBI after BCS. A catheter whose balloon is filled with saline solution is inserted into the resection cavity to form a spatial geometry. Irradiation is performed by a central iridium-192 source. Radiotherapy is delivered to a spherical volume with a 10-mm safety margin on 5 days with 8 to 10 fractions to a total dose of 32–35 Gy with the high-dose-rate technique. The advantage of this method is its simple handling, which requires only a short learning curve.42 However, there are some substantial drawbacks:
The target volume and dosimetry are standardized, thus individual requirements, eg, resulting from the size of the resection margins, cannot be met. The spherical symmetry of the dose distribution cannot be adapted to individual target volumes.
The therapeutic range is only 10 mm, which may be too short for residual disease, as even with margins of 10 to 20 mm a substantial risk for minimal residual disease is still present and assumed to be up to 50% according to tumor characteristics.1, 43–45
The spherical dose distribution can also result in an unfavorable cosmetic result, as overdosage to the skin may not be avoided. Therefore, it is mandatory to have a skin source distance of at least 10 to 15 mm between balloon and skin. In the early phase after introduction of this system, acute side effects were reported and appeared higher than those of WBRT. In a phase 1 study assessing the feasibility and acute toxicity in 43 patients, erythema was observed in 57%, breast pain in 42%, and edema in 15% of patients.10 So far, there exist only limited clinical data from randomized studies on APBI alone with the balloon catheter technique. Preliminary data from a registry trial of the American Society of Breast Surgeons46 were published after a median follow-up of 5 months. In 8% an infection occurred, but the cosmetic result after 12 months was good or excellent in 92%. In the latest publication of the William Beaumont hospital, the local recurrence rate was 2.5% after a median follow-up of 22 months.47 In 2004 the NSABP/RTOG initiated a prospective randomized multicenter study investigating APBI alone, using either the multicatheter technique, the balloon catheter technique, or 3D conformal external APBI in comparison to conventional WBRT.48 The results from this study with sufficient follow-up are expected to provide more meaningful conclusions concerning the MammoSite technique and will be available in a few years.
3D Conformal External Beam Radiotherapy
As an alternative to invasive procedures, 3D conformal external beam radiation (3D-CRT) has been introduced for APBI.8, 19, 49 With this method, additional surgical procedures are avoided and a homogenous coverage of the target volume is possible. Moreover, the goal is to avoid fat necrosis and achieve an improved overall cosmetic result by restricting radiotherapy to the affected part of the breast. However, the breast is a mobile organ; therefore, the required volume may be larger to avoid a geographical miss of the target volume. 3D planning is performed on the basis of computed tomography (CT) or magnetic resonance imaging (MRI) scans and the planning target volume is assessed with an additional safety margin of 1–2 cm.
Formenti et al.49 applied 30 Gy in 5 fractions with single doses of 6 Gy. After a follow-up time of 18 months toxicity was minimal and no recurrence was observed. In the RTOG Study19 treatment was performed twice daily in 10 fractions of 3.85 Gy up to a total dose of 38.5. A minimum of a 10-mm safety margin around the clinical target volume (CTV) was required. Inclusion criteria were: AJCC stage I or II (T1N0-T2N1); tumor size ≤3 cm; unifocal tumor; negative surgical margins. Dose volume constraints in this study are strictly defined for lungs, heart, and the uninvolved breast. Ideally, <50% of the whole breast should receive ≥50% of the dose. Forty-two patients were eligible, feasibility and reproducibility were excellent, and adverse effects were minimal. Follow-up and patient numbers do not yet permit pertinent conclusions about efficacy and the long-term side effects of this method.
The Ontario Clinical Oncology Group is currently performing a randomized phase 3 study comparing treatment results of WBRT and 3D conformal APBI. Patients receive 38.5 GY in 10 fractions delivered twice daily over a time period of 5–8 days.50 In contrast to the above-mentioned study, negative axillary nodes are required as an inclusion criterion.
DOSIMETRIC COMPARISON OF THE DIFFERENT MODALITIES
The 4 methods described above for APBI have different characteristics in dose rate, dose distribution, and consecutively in their radiobiologic implications.
A comparative analysis of treatment plans and dose volume histograms for HDR brachytherapy, balloon-catheter technique, and 3D-CRT by Weed et al.51 showed the best coverage of the target volume using 3D-CRT. However, the integral dose to the normal breast was higher compared with the balloon catheter technique or interstitial HDR brachytherapy. The authors concluded that the choice of the best APBI technique depends on the individual clinical situation.
Dose homogeneity to the target volume is an important parameter for the effectiveness of radiotherapy. A recent study used planning target volumes (PTV) and dose volume histograms (DVH) to compare the respective properties of the different APBI methods (except 3D-CRT) on the basis of a mathematical model. The results suggest that the smallest dose inhomogeneities are achieved by IOERT, followed by balloon catheters, orthovolt technique, and interstitial implants. Another important parameter for avoiding late normal tissue complications is the dose to the surrounding tissues. These were most effectively spared by IOERT followed by orthovolt treatment. However, target volumes frequently show an asymmetric shape that seem to be matched best by IOERT and interstitial brachytherapy.52
From a scientific point of view, innovative treatment approaches are always welcome if they provide the option for definite improvement with regard to cure rates and decreased side effects or the prospect of a better quality of life. Such a change of paradigms is best illustrated by the history of breast-conserving therapy replacing mastectomy. BCS became generally accepted as a routine tool but the former gold standard of mastectomy was not given up before numerous randomized clinical trials had confirmed equivalent local control and overall survival of this conservative strategy.1, 53–57 The latest meta-analysis of the Early Breast Cancer Trialist's Collaborative Group5 explicitly states that WBRT does not only substantially reduce the risk for local recurrence, but that for every 4 recurrences prevented, 1 breast cancer death is avoided over the next 15 years. This is in accordance with the results of several studies investigating whether in defined subgroups of patients WBRT could be omitted after BCS without compromising outcome.58–60 All of these studies failed to identify a subgroup in which omitting radiotherapy was not associated with a substantial increase of local recurrences.
Whereas indications for systemic adjuvant chemotherapy are hardly ever questioned, although its impact on survival is not necessarily greater than that of radiotherapy,61 the justification of WBRT has recently become an issue of a very controversial discussion that sometimes seems to reach the dimension of an ideological debate. Even in the American societies there are differences in the recommendations: the American Society of Breast Surgeons suggests in their consensus that “outside of multi-institutional studies patients should be carefully selected for APBI and properly informed of the risks and benefits.”62 In contrast, the American Brachytherapy Society explicitly states that “brachytherapy as a sole modality is considered investigational and should be performed in controlled clinical trials.”63
One rationale for APBI is to avoid side effects such as impeded cosmetic results caused by irradiation of the whole breast. Although there are promising data, the question whether APBI may achieve a better cosmetic result than WBRT cannot yet be answered. From a radiobiological point of view it is possible that long-term cosmetic results may be even worse than those of conventional radiotherapy. The latency period for manifestation of radiation-induced late fibroses is 4.7 years in 90% of the cases.64 This time is much longer than the follow-up in most studies investigating APBI. Moreover, recent radiobiological investigations gave rise to doubts if biologically effective doses of an APBI correspond to a conventional fractionated homogeneous irradiation at all.65 Regarding the excellent cosmetic results achievable with modern 3D-conformal WBRT the aspects of reducing side effects do not appear to be of dominant clinical relevance.
An interesting argument for APBI comes from 1 of the protagonists of APBI and an initiator of the TARGIT-Study, J.S. Vaidya. His group66 systematically investigated pathologic mastectomy specimens and found that 80% of occult tumor foci were outside of the affected quadrant. However, later they argued that 90% of manifest local recurrences occur within the originally affected quadrant and therefore such distant microfoci were not clinically relevant. They conclude that a restriction of irradiation to the affected tumor area was adequate.41 These considerations do not seem logical, as an alleged lack of clinical relevance of those distant tumor foci is most probably due to the finding that a manifest recurrence in the affected quadrant leads to mastectomy before a tumor focus becomes morphologically apparent. This hypothesis would rather favor a local boost of the tumor area after WBRT rather than omitting radiotherapy of the remaining breast.
Protagonists of APBI argue furthermore that APBI could “save breasts” for women who would otherwise undergo mastectomy as they could not afford fractionated radiotherapy.18 This may be an argument in countries with limited resources but should not play a role when health insurance systems cover the costs for WBRT.
An interesting cost comparison of APBI versus WBRT from the University of Michigan including direct medical costs and nonmedical costs like time and travel was performed by Suh et al.67 The authors came to the conclusion that (in the US) reduction of overall treatment time did not necessarily save costs. The least expensive approach of APBI was 3D-CRT, whereas brachytherapy provided higher total social costs. The least expensive method was accelerated WBRT using 42.5 Gy in 16 fractions.
Saving time is another advantage of APBI and may be tempting for women who are working or have family obligations. In countries with limited radiotherapy institutions acceleration of radiotherapy delivery may be an advantage in terms of capacity. However, in light of the implications of local tumor control on the further life of the affected women, our scientific societies agree with a statement in a commentary publishing the results of a workshop on partial breast irradiation in Bethesda 2002: Convenience, availability of therapies, and financial considerations should not jeopardize the substantial and good results of BCS plus WBRT.68
APBI alone after BCS and omitting homogeneous breast irradiation is still an experimental treatment even in patients with a low risk, such as older women with favorable prognostic factors. For all the techniques discussed in this article the relative biological effectiveness, side effects, and long-term complications are not yet reliably predictable.
Because of systemic treatment, recurrences may occur with a substantial time delay; therefore, a final judgment is only possible after many years of follow-up. Until then BCS with WBRT—in defined cases with an additional boost—continues to be the ‘gold standard’ in the treatment of early breast cancer. Worldwide, more than 1,000,000 women per year are affected by breast cancer.69 In an editorial addressing this issue, Whelan and Levine70 stated in 2005 that changes of treatment standards, even if only applied in subgroups of patients, are likely to have an impact on thousands of women and stressed “We owe it to our patients that our management strategies use the highest possible level of evidence.”
Therefore, the German Society for Senology, the Working Group for Gynecological Oncology of the German Cancer Society, and the German Society of Radiooncology71 agreed to support randomized clinical studies using the above-defined inclusion criteria but to explicitly discourage the use of APBI in any other setting or as a routine tool.