Ipsilateral breast tumor recurrence (IBTR) after breast conserving surgery followed by radiation therapy is reported in about 4–7% after 5 years and in about 10–15% in the long term follow up in the literature. Recent data from long term follow up from the MD Anderson Cancer Center report on improving local control rates of 2% in the last years.1 Many factors contribute to this decline in IBTR rates, as modern surgical techniques, the extensive pathologic evaluation of specimen and margins, the increasing use of adjuvant systemic therapies and the extensive use of radiation therapy. On the basis of low IBTR rates, the concept of whole breast irradiation comes up for discussion and partial breast irradiation increasingly is under consideration. Different methods for partial breast irradiation are under investigation, but short and long term results are not available. The standard of care for patients with breast conservation is the postoperative radiotherapy to the whole breast with 50–55 Gy. With this therapy, an excellent local tumor control can be achieved. The additional application of an external boost radiation of 10–16 Gy to the tumor bed can reduce the failure rate by 40%.2, 3, 4 The exact knowledge of the tumor bed is essential to apply the boost precisely to the target volume. Several methods are used to localize the tumor bed exactly. Tumor bed clipping with titanium clips, perioperative brachytherapy or intraoperative radiotherapy (IORT) are used to localize the target volume. A partial topographic miss of the target volume could be a reason for local recurrence after boost radiation.5 Intraoperative radiotherapy is referred to the delivery of a single high dose of irradiation directly to the tumor bed during surgery. This single high dose radiation is delivered to a surgically defined target volume, while uninvolved or dose limiting tissues are displaced, with the goal of enhanced locoregional tumor control. The surgeon and the radiotherapist precisely define the target volume by direct visualization and intraoperative ultrasound, and the high precision boost can be delivered and anticipated into the surgical procedure. The Salzburg concept of IORT in breast cancer patients applies electrons from a linear accelerator in a decicated unit directly to the tumor bed in boost modality during surgery. Postoperative whole breast irradiation is a part of this concept. We reported on this concept earlier6 with excellent short term results. Now we reanalysed our data after a median follow up of 51 months.
Aim of this study is to show that ipsilateral breast tumor recurrence (IBTR) after breast conserving surgery can be reduced by proper surgery and modern radiotherapy techniques. Three hundred and seventy eight women with stage I or II breast cancer had breast conserving surgery and received 51–56.1 Gy of postoperative radiation to the whole breast in 1.7 Gy fractions, but patients received different boost strategies. Group 1 (n = 188) received electron boost radiation of 12 Gy subsequent to the irradiation to the whole breast, group 2 (n = 190) received intraoperative electron boost radiation of 9 Gy directly to the tumor bed, followed by whole breast irradiation. After a median follow up period of 81.0 months in group 1 and a median follow up period of 51.1 months in group 2, 12 IBTRs (6.4%) could be observed in group 1 and no IBTR could be observed in group 2 (0.0%). The 5-year actuarial rates of IBTR were 4.3% (95% CI, 1.9–8.3%) and 0.0% (95% CI, 0.0–1.9%), respectively (p = 0.0018). The 5-year actuarial rates of distant recurrence were 8.6% (95% CI, 4.9–13.5%) and 4.2% (95% CI, 1.8–8.2%), respectively (p = 0.08). The 5 year disease-free survival rates were 90.9% (95% CI, 85.8–94.7%) in group 1 and 95.8% (95% CI, 91.8–98.2%) in group 2 (p = 0.064). Immediate IORT-boost and whole breast irradiation yields excellent local control at 5 years, and was associated with a statistically significant decreased rate of IBTR compared with a similar cohort of patients treated with whole breast irradiation and conventional electron boost. © 2005 Wiley-Liss, Inc.
Patients and methods
Between 1996 and 2001, 378 patients were treated with breast conserving surgery for stage I and II invasive breast cancer at the Breast Center Salzburg. From January 1996 to October 1998, 188 patients were treated with breast conserving surgery and postoperative irradiation to the whole breast followed by postoperative external beam boost irradiation to the tumor bed (group 1). From November 1998 to March 2001, 190 patients were treated with breast conserving surgery, intraoperative electron boost directly to the tumor bed and postoperative irradiation to the whole breast (group 2). The groups are comparable with regard to age, menopausal status, tumor size, histological type, grading and axillary lymph node status (Table I). Included were patients with invasive breast cancers pT1/pT2/pN0/pN1/M0, who were eligible for breast conserving surgery with histologically clear margins of at least 3–5 mm. Exclusion criteria were neoadjuvant chemotherapy, tumors >pT2, multicentricity and ductal carcinoma in situ.
|Group 1||Postop boost (%)||Group 2||IORT boost (%)||p-value|
|Number of patients||n = 188||100.0||n = 190||100.0|
|Invasive ductal carcinoma||163||86.7||167||87.9||0.726|
|Invasive lobular carcinoma||25||13.3||23||12.1||0.769|
|Hormone receptor status|
|ER neg/PR neg||38||20.2||27||14.2||0.123|
|ER neg/PR pos||2||1.1||2||1.1||0.992|
|ER pos/PR neg||23||12.2||16||8.4||0.224|
|ER pos/PR pos||125||66.5||145||76.3||0.035*|
|Combined chemotherapy and hormonal therapy||12||6.4||9||4.7||0.485|
|No adjuvant therapy||22||11.7||6||3.2||0.002*|
In group 1, 128 patients presented with T1 tumors (68.1%) and 60 patients with T2 tumors (31.9%). In all patients, wide excision of the tumor was performed with clear margins of 3–5 mm minimally. The tumor bed was clipped with titanium clips to mark the excision cavity for postoperative radiation treatment planning and for the tumor bed control. Axillary dissection of levels I and II was performed in all patients of group 1. Postoperative radiation treatment consisted of external beam radiation therapy (EBRT) of 51 Gy/6 weeks (30 × 1.7 Gy equivalent to 52.2 – 54 Gy ICRU) for patients with invasive ductal breast cancer and 56 Gy (33 × 1.7 Gy equivalent to 57 – 58.8 Gy ICRU) for patients with invasive lobular breast cancer and subsequent external beam electron boost irradiation to the tumor bed with 12 Gy in 2 Gy fractionation. Adjuvant therapy in group 1 consisted of chemotherapy in 26.6% of patients and of hormonal therapy in 55.3% of patients. 6.4% of patients received combined chemotherapy and hormonal therapy. 11.7% of patients received no adjuvant medical therapy.
In group 2, 190 patients were treated with the same surgical technique of breast conserving surgery in a dedicated IORT unit. One hundred and thirty patients patients had T1 tumors (68.4%) and 60 patients T2 tumors (31.6%). If margins were not clear with a minimum of 3–5 mm in the intraoperative pathologic assessment, reexcison was performed in the same surgical procedure prior to IORT. If margins were not clear in the final pathological report (minimum 3 mm), a secondary reexcison had to be performed. Margins were negative in all patients included in this study. Axillary surgery in group 2 was confined to sentinel lymph node biopsy only, if sentinel lymph nodes were negative. Complete axillary clearance of levels I and II was performed, if sentinel lymph nodes were positive. After complete tumor resection and axillary surgery, the tumor bed surrounding tissue was mobilized and temporarily approximated by sutures, to get the tumor surrounding tissue into the target volume. For depth dose prescription, ultrasound of the approximated tissue was performed and documented. Mean depth dose was 19 mm (6–39 mm). Applicator tubes with diameters of 50–60 mm were inserted with complete skin sparing and fixed by an external fixation frame. Then, patients were “air-docked“ to the linear accelerator in the operating room. IORT was performed with a linear accelerator (Philips Elektra SL 18). A single fractional dose of 9 Gy was applied to the 90% reference isodose with energies ranging from 4 to 18 MeV. The exact technique of IORT was described previously.7, 8 After wound healing, EBRT to the whole breast with opposed tangential fields was performed with single fractional doses of 1.7 Gy (minimum target dose) per day for 30 days, totaling 51 Gy for patients with invasive ductal breast cancer, and with fractional doses of 1.7 Gy per day for 33 days, totaling 56.1 Gy for patients with invasive lobular breast cancers. Postoperative EBRT was equal in both groups. Adjuvant therapy in group 2 consisted of chemotherapy in 18.9% of patients and of hormonal therapy in 73.2% of patients. 4.7% of patients received combined chemotherapy and hormonal therapy and 3.2% of patients received no adjuvant medical treatment. For patients, who received adjuvant chemotherapy, EBRT was sequenced after completion of chemotherapy. The time delay between IORT and EBRT for those patients was up to 20 weeks.
The endpoints for treatment comparisons were IBTR and distant recurrence. IBTR was defined as any recurrence within the ipsilateral breast as first event with or without simultaneous distant recurrence. IBTR-free survival was calculated as the time from the date of surgery to the date of the first documented evidence of IBTR confirmed by biopsy. Distant-recurrence-free survival was calculated as the time from the date of surgery to the date of the first documented evidence of distant recurrence. The events included for analysis of disease-free survival were the first recurrence of disease at local, regional or distant site. The Kaplan–Meier plots were generated to estimate IBTR-free survival, distant-recurrence-free survival and disease-free survival and the corresponding survival functions were compared with the Gehan's Wilcoxon test, Cox-F test, the Cox-Mantel test, the log-rank test and the Breslow's test. All reported p-values are based on 2-sided tests. p-values less than or equal to 0.05 were considered to be statistically significant. All data analyses were performed using STATISTICA 5.5 (StatSoft, Inc.2000, Tulsa, OK, USA) and SPSS 10.0 (SPSS, Inc.1999, Chicago, IL, USA).
Ipsilateral breast tumor recurrence (Fig. 1)
After a median follow up period of 81.0 months in group 1 and a median follow up period of 51.1 months in group 2, 12 IBTRs (6.4%) could be observed in group 1 and no IBTR could be observed in group 2 (0.0%). The 5 year actuarial rates of IBTR were 4.3% (95% CI, 1.9–8.3%) and 0.0% (95% CI, 0.0–1.9%), respectively (p = 0.0018). Mean and median time to IBTR were 45.9 months and 42.5 months, respectively. Seven of 12 patients with IBTR developed distant metastases simultaneously or within 52 months. Five of 12 patients with IBTR are alive without evidence of disease, 2 of 12 patients with IBTR died of disease. The characteristics of patients with local recurrences are shown in Table II.
|Patient number||Age (y)||Menopausal status||Histological type||Grading||Tumour size (pT-stage)||Tumour size (mm)||Number of positive nodes/total nodes||Estrogen receptor||Progesterone receptor||Adjuvant therapy||Time to IBTR (months)||Time to distant recurrence (months)||Status|
|4||36||pre||idc + ilc||2||pT1c||18||17/19||pos||Pos||ec||46||47||awd|
|6||47||pre||idc||1||pT1c||18||3/17||pos||Pos||cmf + tam||45||71||awd|
|8||49||pre||idc||3||pT2||45||17/21||pos||Pos||cmf + tam||40||52||awd|
Distant recurrence (Fig. 2)
Distant recurrences occurred in 24 patients (12.8%) in group 1 and in 8 patients (4.2%) in group 2. The 5-year actuarial rates of distant recurrence were 8.6% (95% CI, 4.9–13.5%) and 4.2% (95% CI, 1.8–8.2%), respectively (p = 0.08). Mean and median time to distant recurrence were 41.3 months and 33.5 months, respectively, in group 1, and 36,1 months and 41.5 months, respectively, in group 2. In group 1, 11/24 patients with metastases died of disease and 13/24 patients are alive with disease. In group 2, 4/8 patients with metastases died of disease and 4/8 patients are alive with disease (Table III).
|Patient number||Age (y)||Meno pausal status||Histological type||Grading||Tumor size (pT-stage)||Tumor size (mm)||Number of positive nodes/total nodes||Estrogen receptor||Progesterone receptor||Adjuvant therapy||Time to distant recurrence (months)||Status|
|Group 1 (postop boost)|
|4||36||pre||idc + ilc||2||pT1c||18||17/19||pos||pos||ec||47||awd|
|6||42||pre||idc||3||pT2||28||21/30||pos||pos||ec + tam||44||awd|
|8||47||pre||idc||1||pT1c||18||3/17||pos||pos||cmf + tam||71||awd|
|10||47||pre||ilc||2||pT1c||18||1/18||pos||pos||cmf + tam||46||awd|
|11||49||pre||idc||3||pT2||45||17/21||pos||pos||cmf + tam||52||awd|
|12||50||post||idc||3||pT2||40||13/21||pos||neg||ec + tam||26||dod|
|18||58||post||idc||2||pT2||50||30/30||pos||pos||ec/cmf + tam||79||awd|
|Group 2 (intraop boost)|
|1||46||pre||idc||2||pT2||30||4/21||pos||pos||ec + tam||53||awd|
Disease-free survival (Fig. 3)
The crude disease-free survival was 84.6 % in group 1 (29/188 first events, 15.4%) and 95.8% in group 2 (8/190 first events, 4.2%), respectively. The 5 year disease-free survival rates were 90.9% (95% CI, 85.8–94.7%) in group 1 and 95.8% (95% CI, 91.8–98.2%) in group 2 (p = 0.064).
Five-year ipsilateral breast tumor recurrence rates after breast conserving surgery and radiotherapy vary between 4% and 7%, and correspond to a local failure rate of 1–2% per year. Data from the recent literature show that IBTR after breast conserving surgery can still be decreased. In our series from the Salzburg concept of IORT, we observed no IBTR after a median follow-up of 51 months. These figures support that IBTR rates can be further decreased with modern and highly sophisticated methods of radiotherapy. One weakness of our study is the fact, that this is not a randomized study. Nevertheless, both groups are comparable according to age, menopausal status, tumor size, lymph node size and grading. There was no difference in patients who received adjuvant chemotherapy, but there was a slight but significant difference in patients who received adjuvant hormone therapy in favour of group 2. How far this difference contributes to the fact that no IBTR developed in group 2 cannot be answered, though there is evidence that hormone treatment can reduce local failure rates. Otherwise, all the patients except 1 patient, who developed IBTR in group 1, received adjuvant chemotherapy and adjuvant hormone therapy. Only 3 out of 24 patients, who developed distant metastases had no further adjuvant treatment in group 1.
Extensive intraductal component is a risk factor for local recurrence. In our series, there were more patients with intraductal component in group 2 (IORT group), what perhaps might indicate a worse IBTR rate, but we observed no IBTR in this group.
The development of breast conserving surgery was accompanied by whole breast irradiation. Exclusively whole breast irradiation allowed for breast conserving surgery with equal results as mastectomy concerning local failure rates.9, 10, 11, 12, 13, 14, 15 The additional application of a boost radiation to the tumor bed further reduced local recurrence rates. In the EORTC boost versus non boost trial, a 5-year IBTR rate of 4.3% was achieved by an additional 16-Gy boost compared to a 5-year IBTR rate of 7.3% in patients without boost irradiation.2, 3 At present, no group of patients can be identified, which does not benefit from postoperative radiation therapy after breast conserving surgery.16 Breast irradiation was shown to reduce recurrence rates, but there was no statistically significant reduction in mortality.17
The risk of recurrence in early breast cancer is highest during the first 5 years following diagnosis, irrespective of baseline prognostic factors, and there is still an ongoing discussion if IBTR has an impact on long term survival. Several large series have reported that patients who develop local recurrences within 2 years have a significantly worse outcome than patients who develop local recurrence after more than 5 years, that means the longer the interval between initial treatment and IBTR, the better the outcome.18, 19 The reason for this is not yet quite clear, but may be the biologic aggressiveness of the tumor or the factor that some IBTRs, especially late IBTRs, are not true recurrences but actually new primaries. The 5-year distant recurrence rates after IBTR range between 25 and 40%.20, 21, 22 The question of adjuvant medical treatment on IBTR is still controversial. In patients after neoadjuvant chemotherapy, the occurrence of IBTR is a significant predictive factor of distant metastases.20, 21, 22, 23, 24, 25 Furthermore, there are a number of reports in the literature proposing the development of distant metastases to be increased after formation of IBTR.26, 27 These studies support the hypothesis, that IBTR is a prognostic and predictive factor for distant metastases or even overall survival. For this reason as well as for psychological reasons and health economy reasons, the reduction of local failure rates should be our ambition.
Local failure rates have to be discussed in the primary treatment of breast cancer. Several authors propose partial breast irradiation with different systems28, 29, 30, 31, 32, 33, 34 with regard to the reduction of radiation time and costs. The promising short term results of those studies must not fail to mention that the risk of local or regional recurrence could be increased with those techniques. Furthermore, IBTR could be jointly responsible for potential distant metastases and a potential reduction in overall survival.
All these unanswered questions have to be considered in the propagation of accelerated partial breast irradiation.
With our study, we could demonstrate that immediate IORT boost yields excellent local control and a further reduction of local failure rates is possible compared to standard radiation schemes. Assuming that local failure is responsible for the decrease in survival for patients with breast cancer treated with conservative surgery and postoperative radiotherapy as proposed by some authors,35, 36, 37, 38, 39 our intent should be to further reduce local recurrence rates whenever possible. At this moment exclusively the Salzburg model of IORT considers this fact, all the other models of (accelerated) partial breast irradiation run the risk of an increase in local failure rates associated with the risk of the decrease in overall survival. In this context, one should consider the data provided by Perera et al.40 In this article, the 5-year actuarial rates of ipsilateral breast recurrence was 16.2% in patients who received HDR brachytherapy to the lumpectomy site as the sole radiation. If one discusses the reasons for this high local failure rate, one should bear in mind the risks of partial breast irradiation if not appropriately performed. The reduction of radiation time is extremely attractive from the patient's point of view and the fact not to be irradiated for a time of approximately 6 weeks looks alluring at the first sight, but is not meaningful, as the above mentioned methods (MammoSite, conventional brachytherapy or IMRT) take at least 1 week of therapy. Solely the IORT technique (Milan Model) and the Intrabeam system are the 2 systems that apply the radiation dose intraoperatively as a single dose and do not perform postoperative radiation. As mentioned in an arcticle of Kuerer et al.,41 3.3% of patients would develop local failure after partial breast irradiation in the non-irradiated breast tissue. With the estimated 216,000 new cases of invasive breast cancer in 2005 in the United States and an estimated breast conserving surgery rate of 70%, approximately 5,000 cases of unnecessary local failures per year could be expected in the United States. The Austrian Health System covers irradiation for all patients with breast cancer, and distances to radiation facilities within Austria are below 2 hr of travel time for nearly all patients. Therefore, the main goal of the Salzburg concept is not to reduce radiation time (which is reduced by 7–10 days), but the main goal is to reduce local failure rates. The non-availability of radiation facilities in the near distance to the patients residence may play a crucial role in other health care systems and, therefore, may play a role in the approach to the problem if a patient decides for mastectomy instead of breast conserving surgery for this reason.42, 43 The longest experience with partial breast irradiation is documented with brachytherapy. Nevertheless, this method is somehow medieval to interlard the breast with multiple catheters, and there is no evidence that the proposed MammoSite system, although FDA approved, can produce the same results. Another problem is that only 20–25% of patients with breast conserving surgery are eligible for the MammoSite procedure.41, 44 The NSABP B-39/RTOG 0413 trial comparing whole breast irradiation followed by optional boost with partial breast irradiation with 3 different techniques (multi-catheter brachytherapy, MammoSite balloon catheter, 3D conformal external beam radiation) started recently to recruit patients. This trial will accrue 3,000 patients over a period of 2 years and 5 months, and it will need years until results will be available. This applies analogously to the Milan model and the Intrabeam system. Until these data are available, it is not yet the time to abandon whole breast irradiation after breast conserving surgery.