Repeat breast-conserving surgery for in-breast local breast carcinoma recurrence

The potential role of partial breast irradiation


  • Henry M. Kuerer M.D., Ph.D.,

    Corresponding author
    1. Department of Surgical Oncology, The University of Texas M. D. Anderson Cancer Center, Houston, Texas
    • Department of Surgical Oncology, Unit 444, The University of Texas M. D. Anderson Cancer Center, Houston, TX 77030
    Search for more papers by this author
    • Fax: (713) 792-4689

  • Douglas W. Arthur M.D.,

    1. Department of Radiation Oncology, Virginia Commonwealth University, Medical College of Virginia Campus, Richmond, Virginia
    Search for more papers by this author
  • Bruce G. Haffty M.D.

    1. Department of Radiation Oncology, Yale University Medical School, New Haven, Connecticut
    Search for more papers by this author


Mastectomy is the current standard of care for in-breast local recurrence of breast carcinoma. The objective of the current study was to critically review the rationale for and the theoretic and actual risks and benefits of repeat breast-conserving surgery followed by partial breast irradiation (PBI) for in-breast local recurrence of breast carcinoma. The main outcomes of interest were local control and survival after in-breast local recurrence and side effects, complications, and cosmesis after reirradiation of the breast. The risk of local recurrence was not found to be eliminated with mastectomy; approximately 2–32% of patients treated with mastectomy develop a chest wall recurrence. The interpretation of local control rates in evaluating repeat breast-conserving surgery studies is difficult because of the lack of information regarding preoperative diagnostic mammography to rule out concurrent multicentric disease and microscopic margin status after surgery. Rates of subsequent local recurrence in these studies appeared to be between 19–50%, similar to reported rates of in-breast local recurrence in patients with a first diagnosis of breast carcinoma who were treated with conservative surgery without irradiation. Early follow-up studies of breast reirradiation suggest that catheter-based interstitial brachytherapy and standard external beam radiation therapy can be delivered to the breast more than once without significant side effects in most patients and with acceptable cosmesis in some patients. Mastectomy may not be necessary in all patients with an in-breast local recurrence of breast carcinoma. Recent advances in conformal radiation delivery and single-center published reports concerning repeat breast-conserving therapy support well designed prospective trials to formally test this hypothesis. Cancer 2004. © 2004 American Cancer Society.

At present, the standard surgical approach for the treatment of in-breast local recurrence of breast carcinoma is mastectomy. The rationale for the general consensus for this recommendation is threefold. First, a third local recurrence after repeat conservative surgery is reported in approximately 35% of patients. Second, a second lumpectomy would yield a cosmetically unacceptable outcome. Third, reirradiation of the whole breast after breast-conserving surgery is considered inappropriate because of the likelihood of serious tissue damage. Recently, novel brachytherapy devices and three-dimensional conformal radiation therapy techniques have brought accelerated partial breast irradiation (APBI) to the forefront as a potential treatment approach in women with in-breast local recurrence of breast carcinoma. To explore the possibility that repeat breast-conserving surgery and PBI may be an appropriate treatment option for women with local in-breast recurrence, we conducted an extensive review of currently available published world medical literature regarding the results of mastectomy for in-breast local recurrence, the results of repeat breast-conserving surgery with and without postoperative reirradiation for in-breast local recurrence, and techniques of PBI. The findings of the current study challenge the accepted dogma that mastectomy after local recurrence is the only acceptable treatment for in-breast local recurrence.

Rationale for Breast-Conserving Surgery and APBI for the Treatment of In-Breast Local Recurrences

Extent of local recurrence on the pathologic examination of mastectomy specimens at the time of mastectomy

One hypothetical way to predict the safety of repeat breast-conserving therapy after in-breast local recurrence is to review mastectomy specimens removed at the time of recurrence for evidence of multicentricity. In what to our knowledge is the largest study of this kind, which included 110 mastectomy specimens from patients initially treated with lumpectomy on National Surgical Adjuvant Breast and Bowel Project (NSABP) protocol B-06, 14% of specimens had evidence of multicentricity.1 The majority of patients in this series did not receive breast irradiation as part of their initial cancer treatment. In 2 smaller series, 22% and 31%, respectively, of mastectomy specimens demonstrated tumor within other quadrants of the breast.2, 3 Theoretically, if patients with multicentric disease at the time of in-breast local recurrence received only localized therapy, then the other regions of the breast might be at risk for additional recurrence. However, these were pathologic studies. Patients in these reports were not clinically evaluated with the aim of selecting appropriate patients for repeat breast-conserving surgery; neither modern diagnostic mammography nor stringent margin-control techniques were used.

For some patients, it is clear that treatment with repeat breast-conserving surgery would not be appropriate. Approximately 6–7% of patients who present with an in-breast local recurrence are considered inoperable because of diffuse local disease or an inflammatory recurrence.4–6

Local recurrence risk after mastectomy for in-breast local recurrence

The risk of chest wall recurrence after mastectomy for in-breast local recurrence appears to depend on whether the in-breast recurrence is primarily invasive or noninvasive.7, 8 In a study from the Joint Center for Radiation Therapy performed in patients with noninvasive or focally invasive recurrent tumors, there were no chest wall recurrences observed at a median follow-up of 39 months.8 In contrast, 7% of patients who presented with an invasive in-breast local recurrence developed a subsequent chest wall recurrence. In a series of patients from Sweden, chest wall recurrences occurred in 19% of 65 patients after mastectomy for local in-breast recurrence at a median follow-up of 13 years.9 A chest wall recurrence was reported in only 3% of 93 patients with an invasive in-breast recurrence who were reported by Doyle et al. from the University of Pennsylvania.10 This extremely low rate of subsequent chest wall recurrence is likely related to the inclusion of only women with initial AJCC Stage I and Stage II breast carcinoma and the median follow-up period of 44 months. A similarly low rate of chest wall recurrence (4%) was reported by Salvadori et al. and also may be the result of the inclusion of patients with initial Stage I and Stage II disease.6

Huang et al.11 found that the rate of local recurrence after mastectomy was significantly higher in patients with a true local recurrence at the site of the initial breast-conserving surgery compared with patients with recurrent tumors detected elsewhere in the breast (18% vs. 2%).

Rates of subsequent chest wall recurrence after mastectomy for in-breast local recurrence from several large series are shown in Table 1.3, 4, 6–17 These studies demonstrated that a second local recurrence in the form of a chest wall recurrence occurs in approximately 2–32% of patients. Thus, mastectomy after in-breast local recurrence does not necessarily eliminate the risk of subsequent secondary local recurrences at the chest wall.

Table 1. Outcome of Patients Treated with Mastectomy after Local Recurrence in Patients Initially Treated with Conservative Surgery
ReferenceNo. of patientsHistology of local recurrenceMedian follow-up after local recurrence (mos)Subsequent chest wall recurrence rate (%)Distant recurrence rate (%)5-Year overall survival rate (%)
  • a


  • b

    Data reported for patients presenting with an invasive recurrence.

  • c

    Includes chest wall and regional lymph node failures.

  • d

    Disease-free survival of entire series of 5 years.

  • e

    10-year distant metastasis rate.

Clarke et al., 1985121212 invasive26a2738Not stated
Recht et al., 198976555 invasive329b25aNot stated
Kurtz et al., 1989443Not stated5312cNot stated53
Forquet et al., 19891339Not stated63Not stated2573
Fowble et al., 199035247 invasive25Not statedNot stated84
Osborne et al., 1992144636 invasive2831bNot stated76
Haffty et al., 19911541Not stated60Not statedNot stated59
Abner et al., 1993810682 invasive397b30b79
Cajucom et al., 1993162524 invasive52324865
Dalberg et al., 1998965Majority invasive156193859d
Voogd et al., 199917      
Salvadori et al., 19996134Not stated7344770
Doyle, et al., 20011011293 invasive443b53e83
Huang et al., 200211118114 invasive84   
  No. of true recurrences 1865b52b
  New primary tumor 219b86b

Survival after mastectomy for in-breast local recurrence

Survival after mastectomy for isolated in-breast local recurrence is dependent on several factors. Recurrences that are primarily invasive are associated with worse survival than recurrences that are primarily noninvasive.8, 18 A purely noninvasive in-breast recurrence is reported to occur in approximately 10% of patients initially treated with breast conservation for invasive disease as opposed to approximately 60% of patients treated with breast conservation for ductal carcinoma in situ (DCIS).1, 3, 18, 19 Death after a noninvasive breast recurrence is extremely rare and generally results from distant metastases associated with the originally diagnosed invasive breast carcinoma. Patients with an isolated invasive in-breast local recurrence after breast-conserving surgery are at increased risk for distant metastases and death compared with patients without an in-breast local recurrence.20 The relative risk of breast carcinoma-associated death is estimated to increase by a factor of 3.4 to 4.6 in patients who develop a local in-breast recurrence compared with patients who do not.20, 21 This fact is likely related to the biology of disease rather than a cause of subsequent distant metastases, although this remains a controversial area of discussion.22 The overall survival of patients with in-breast local recurrences at 5 years reportedly ranges from 52–84% (Table 1); virtually all deaths appear to occur in patients with an invasive recurrence. Distant recurrence reportedly occurs in 19–65% of patients diagnosed with an in-breast local recurrence (Table 1).3, 4, 6–17 Large differences in overall survival among series are likely due to the inclusion of patients with concomitant distant recurrence and may be due in part to differences in the metastatic workup performed at the time of local recurrence and follow-up times after initial recurrence.

Associated simultaneous skin recurrence is reported to occur in approximately 1% of cases of local recurrence and generally is associated with concurrent or rapid distant metastases and a poor overall survival rate, in the range of 20%.18 Several series published to date also have shown that patients with a recurrence elsewhere in the breast have significantly less risk of developing distant metastases and dying of breast carcinoma compared with patients with local recurrence near the site of initial surgery.11, 15, 17 Clarke et al. found that survival after local recurrence is worse in patients presenting with dermal involvement or lymph node involvement.12 Cajucom et al.16 found that the size of the local recurrence and the number of lymph nodes with metastases at the time of mastectomy were statistically significant factors related to overall survival. Similarly, in a large multicenter study from the Netherlands, multivariate analyses of clinical and pathologic factors in patients with local recurrence after breast conservation identified several factors that appeared to have an adverse impact on survival: skin involvement, extent of local recurrence > 1 cm, and positive axillary lymph node status at the time of diagnosis of the primary disease. Haffty et al. at Yale15 also found that a greater extent of local recurrence (> 3 cm) was associated with adverse survival. Investigators also have shown that the detection of in-breast local recurrence by mammography alone is associated with a better prognosis, which may be a reflection of lesser disease extent at the time of local recurrence.3, 10, 22 Similar to the interval between mastectomy and chest wall recurrence, the interval between breast-conserving therapy and in-breast local recurrence has been shown to be a clinically significant prognostic factor.23 With regard to survival, Whelan et al.24 found that a time to local recurrence of < 1 year was associated with a worse survival, whereas Doyle et al.10 found that a time to local recurrence of < 2 years and Haffty et al.25 found that a time to local recurrence of < 4 years was associated with worse survival. Patients with an early local recurrence are reported to subsequently develop distant metastases approximately 50% of the time.21 Veronesi et al.21 found that the risk of developing distant metastases for patients with a local recurrence in the first year after surgery was 6.6 that of patients in whom a local recurrence developed > 3 years after the completion of initial conservative therapy. Similarly, in what to our knowledge is the largest series to date evaluating survival after in-breast local recurrence, which involved 391 patients from Sweden with a median follow-up of 95 months, the 5-year overall survival rate was significantly lower (68%) in patients presenting with a local in-breast recurrence in the same quadrant as the initial tumor within 2 years after the completion of initial therapy compared with patients in whom a recurrence developed > 2 years after the completion of initial therapy (97%).26

Patient outcomes after conservative surgery alone for in-breast local recurrence

We found only five published studies regarding the outcome of patients treated with conservative surgery alone for an in-breast local recurrence (Table)2.6, 8, 9, 17, 27 The evaluation of these studies is difficult because of the lack of information regarding margin status and use of breast imaging at the time of in-breast local recurrence. Many of the series do not indicate whether clear surgical margins were obtained at the time of local excision of the recurrence, and numerous patients were treated before modern-day standard mammography and sonography were available to select appropriate patients for breast-conserving approaches (i.e., no evidence of multicentricity or diffuse suspicious microcalcifications).

Table 2. Outcome of Patients Treated with Conservative Surgery Alone after an In-Breast Local Recurrence
ReferenceNo. of patientsHistology of recurrenceMargin statusMedian follow-up from local recurrence (mos)Local recurrence rate (%)5-year overall survival rate (%)
Kurtz et al., 19912755Invasive and noninvasiveNegative, positive, or unknown5127Not stated
Abner et al., 1993816Invasive and noninvasiveUnknown393181
Dalberg et al., 1998914Not statedUnknown15650Not stated
Voogd et al., 19991716InvasiveUnknown5238Not stated
Salvadori et al., 1999657Not statedUnknown731985

Abner et al.8 reported a local recurrence rate of 31% at a median follow-up of 39 months, and Voogd et al.17 reported a local recurrence rate of 38% at a median follow-up of 52 months. It is interesting to note that in the series by Voogd et al.,17 the local recurrence rate after local excision was not found to be significantly different from that reported after mastectomy (38% vs. 25%; P = 0.27).

Kurtz et al.27 from Marseille reported on the feasibility and outcome of breast-conserving surgery after local recurrence in 50 patients. In this important study, the authors demonstrated that 5-year local control rates were higher in patients who had negative margins of resection compared with patients with positive margins of resection (73% vs. 36%) and were higher in patients with a local recurrence occurring after 5 years compared with patients with shorter intervals to recurrence (92% vs. 49%%).27

In a study from Sweden, Dalberg et al. found that the local recurrence rate after local excision alone in 14 patients at a median follow-up of 13 years was an alarming 50%.9 However, the authors did not explain why these patients were specifically chosen to undergo local excision alone, whether mammography or sonography was utilized to select appropriate patients, and whether negative margins of resection were obtained.

To our knowledge, the largest series reported to date evaluating repeat local resection for in-breast local recurrence was reported by investigators at the European Institute of Oncology in Milan, Italy.6 This was not a randomized study; at a median follow-up of 73 months after the second surgery, the overall survival rate at 5 years was 70% in the 134 patients who had undergone a mastectomy compared with 85% of patients who underwent local excision. No difference was noted with regard to the disease-free survival between patients who underwent a mastectomy and those who underwent local reexcision. Approximately 19% of patients treated with local excision developed a second in-breast local recurrence. In comparison, 4% of patients in the mastectomy group developed a chest wall recurrence.

In the NSABP B-24 trial, a randomized trial of breast conservation therapy with or without tamoxifen for patients with DCIS, 54 of the 150 patients with a local recurrence (36%) underwent a second lumpectomy rather than a mastectomy as their definitive treatment.28 Women were significantly more likely to undergo a second lumpectomy if they had a noninvasive recurrence. The results of this NSABP study indicate that repeat conservative surgery is offered to some women in the U.S. However, to our knowledge, no follow-up information regarding the rate of subsequent in-breast local recurrence after repeat conservative surgery in patients enrolled in the NSABP B-24 trial had been published at the time of our review.

Investigators at The University of Texas M. D. Anderson Cancer Center evaluated the rate of contralateral breast carcinoma occurrence in 116 patients who had developed an ipsilateral recurrence after initial breast-conserving surgery.11 The rate of contralateral breast carcinoma occurrence at 10 years was 8% in patients with a recurrence at the surgical site (as opposed to elsewhere in the breast) and 29% in patients with an in-breast occurrence of a new primary breast tumor outside the initial lumpectomy bed.11 This rate of new primary breast tumor occurrence is similar to reported rates of ipsilateral recurrence in patients treated with wide local excision alone for in-breast local recurrence after breast-conserving therapy. Although a rate of 29% appears high, it would be difficult to recommend a prophylactic contralateral mastectomy in these patients because approximately 71% of patients will not develop contralateral breast carcinoma by 10 years. Taken in this context, recommending mastectomy in all patients with an in-breast local recurrence also appears extreme.

On the basis of these few studies, survival does not appear to be worse with repeat breast-conserving surgery compared with mastectomy for in-breast local recurrence, and local recurrence rates with conservative surgery appear to be on average in the range of 35%. This is similar to rates of local recurrence in patients initially treated with breast-conserving surgery without postoperative radiation therapy. The question then becomes whether these rates of local recurrence can be lowered with repeat partial breast irradiation in appropriately selected patients.

Reported experiences with reirradiation of the surgical bed after in-breast local recurrence

In 1989, Recht et al.7 from the Joint Center for Radiation Therapy reported on one patient who originally was treated with whole breast irradiation who refused a mastectomy at the time of local recurrence and underwent wide local excision followed by iridium implantation. The patient died without evidence of disease 72 months after retreatment.7 There also have been a few reports concerning the use of brachytherapy for locally recurrent soft tissue sarcomas arising in a previously irradiated field to avoid amputation.29–31 In these series, which included approximately 70 patients, postoperative wound complications (wound dehiscence and ulceration) were noted in 12–23% of patients, and local control rates were approximately 62–88%.29–31 The relatively high rates of dehiscence may reflect the fact that most of these recurrences involved the extremities, which are prone to this type of wound complication.

Investigators from France reported their experience with resection followed by brachytherapy in 15 patients with in-breast local recurrences who either refused mastectomy or were not candidates for the procedure.32 Patients were not selected on the basis of mammographic or ultrasound criteria, and resection margin status was not reported. After local resection of invasive tumors (mean dimension of 2.4 cm), patients received interstitial brachytherapy with a total dose of 30 grays (Gy) (dose rate not mentioned). At a mean follow-up of 48 months, 4 patients (26%) had developed a second local recurrence. Cosmetic outcomes were reported for eight patients. Five had minor or no sequelae, and three patients had major sequelae. One patient had skin necrosis that was treated successfully with local wound care, and one patient had extensive erythema and underwent mastectomy (no tumor was evident on pathologic examination).

Deutsch33 recently reported his experience with repeat external beam irradiation for in-breast tumor recurrence after prior lumpectomy and whole breast irradiation in 39 patients treated since 1985 at the University of Pittsburgh Medical Center. Thirty-one patients had invasive disease and 8 patients had DCIS. The author did not report eligibility and selection criteria with respect to lesion size or whether modern diagnostic mammography or sonography was utilized to exclude multicentric disease. Patients underwent repeat segmental resection of the recurrence, and 15% of patients had positive margins of resection. Patients were retreated with external beam radiation therapy to the surgical bed with 50 Gy in 25 fractions at a median of 63 months after their initial radiation treatment. At a median follow-up of 52 months, 21% of patients had developed an in-breast local recurrence. Four of these patients (44%) also developed distant metastases. It is interesting to note that the rate of contralateral breast carcinoma in this cohort study also was 21%. The 5-year overall and disease-free survival rates were 78% and 69%, respectively. These survival rates are similar to survival rates reported in women treated with mastectomy for an in-breast local recurrence. Although this was not a prospective study with strict criteria for the reporting of side effects and cosmesis, there were no reports of radiation-induced necrosis and the cosmesis was reported as excellent or good in 69% of patients. The remaining patients had an obvious deformity, a marked difference in the size of the breast, or excessive skin pigmentation. This study did not specifically report problems with wound healing.

We found only one prospective study regarding reirradiation for in-breast local recurrence.34 In this study from Vienna University, 9 patients were treated with partial breast irradiation with interstitial pulse dose rate brachytherapy with 40.2–50 Gy, and 8 patients were treated with repeat whole breast irradiation with 30 Gy combined with an additional pulse dose rate interstitial brachytherapy boost dose of 12.5 Gy to the surgical bed. Recurrences were diagnosed at a median follow-up of 50 months (range, 11–208 months) after the completion of treatment for the initial breast tumor. All the recurrent tumors were infiltrating ductal carcinomas. Tumors ranged in size from 0.5–2.5 cm. The authors did not indicate whether preoperative mammography or sonography was performed to determine patient eligibility. All patients underwent resection of the recurrence to obtain negative margins. At 5 years of follow-up, 4 patients were reported to have developed a local recurrence, with a median time to recurrence of 8 months. All local recurrences occurred in the patients who received combined whole breast radiation with a brachytherapy boost. Two of the patients with local recurrence developed a concomitant distant recurrence and died of disease. A remarkable finding of this study was that despite the previous administration of whole breast irradiation, side effects were limited to NCI CTC Grade 1 or 2 fibrosis. However, longer-term follow-up might alter these results. Cosmetic outcome as rated by a physician and the patient was excellent or good in approximately one-third of the patients and moderate or acceptable in the remaining patients. The cosmetic outcome was not rated as unacceptable in any patients. This finding indicates that patients most likely were required to have sufficient breast volume with which to sustain an additional therapeutic breast resection.

These reports indicate that approximately 75% of women who underwent breast-conserving surgery and reirradiation of the surgical bed for the treatment of an in-breast local recurrence ultimately avoided mastectomy for their local recurrence and had no need for further local therapy. It is likely that the local control rates would be higher in patients with DCIS only compared with patients with invasive carcinoma. We believe the results of these studies justify a Phase II prospective study with clear entry and exclusion criteria to evaluate the safety and measurable outcome of repeat breast-conserving surgery followed by partial breast irradiation delivered using emerging novel techniques.

Potential Methods of Delivering Partial Breast Irradiation after Repeat Breast-Conserving Surgery

Rapidly emerging technologic advances in radiation delivery have brought APBI to the forefront as a potential alternative to standard whole breast irradiation after initial breast-conserving surgery. APBI refers to radiation therapy delivered over a shorter period than the standard 5–6 weeks (“accelerated”) and delivered to only a portion of the breast (“partial”). APBI can be delivered using brachytherapy; intraoperatively with orthovoltage techniques, a linear accelerator, or electrons; or using conformal external beam radiation therapy. Each of these techniques has its own potential benefits and limitations. Because APBI by definition limits radiation to only a portion of the breast at risk for subsequent failure, the use of APBI in patients previously treated with whole breast radiation has the potential to limit radiation-induced side effects.


In the 1920s, the English surgeon Sir Geoffrey Keynes utilized interstitial radium needles to treat not only primary breast tumors but also the regional lymphatics.35 Brachytherapy—the placement of radioactive sources within the tumor bed (interstitial brachytherapy) or very close to the tumor bed (intracavitary brachytherapy)—has the main potential benefit of limiting damage to healthy tissue while delivering the maximum dose to the tissue at risk for disease. This potential benefit is due to the inverse square law, which states that radiation intensity decreases as the inverse square of the distance from the point source of the radiation.

Generally, the radiation source is added to multiple catheters surrounding the tumor bed by automated afterloading technology that minimizes the potential radiation exposure to health care providers. To our knowledge, the optimal radiation dose and dose rate with which to maximize therapeutic effectiveness and minimize complications have yet to be determined. Brachytherapy can be delivered with either low-dose rate (LDR) or high-dose rate (HDR) radiation sources. With LDR techniques, a dose of 45–50 Gy is delivered to the clinical target volume at a rate of approximately 30–70 centigrays/hour. This technique requires that the patient be admitted to the hospital for approximately 4 days. HDR techniques are now much more common. Various fractionation schemes have been reported for HDR brachytherapy, which is delivered on an outpatient basis. Recently, based on the initial experience with brachytherapy after lumpectomy for patients with early-stage breast carcinoma, a total dose of 34 Gy delivered in twice-daily fractions of 3.4 Gy over 5–7 days has evolved as an attractive and well tolerated schedule.47

The use of brachytherapy has several theoretic potential advantages for patients, one of the most appealing of which is a reduction in treatment time. With brachytherapy, the delivery of radiation therapy to the breast is generally completed over a period of 4–5 days, and all local therapy can be completed before systemic therapy is initiated. There is a growing body of literature regarding the use of brachytherapy as the sole type of radiation therapy after breast-conserving surgery.

Catheter-based interstitial brachytherapy.

The role of standard catheter-based interstitial brachytherapy as the sole modality of radiation therapy for breast carcinoma (using both LDR and HDR techniques) has been investigated in multiple single-institution and multiinstitution studies involving approximately 500 patients.36–47 On the basis of extremely low in-breast local recurrence rates and favorable toxicity profiles, the NSABP together with the Radiation Therapy Oncology Group has designed a 3000-patient randomized study comparing the efficacy of 6-week standard whole breast irradiation and APBI using the accelerated fractionation scheme described earlier.

Balloon intracavitary brachytherapy.

Potential disadvantages of standard catheter-based interstitial brachytherapy include the technical difficulty involved in learning to place the catheters correctly and the limited number of clinicians in the U.S. who routinely utilize this method.50 Another problem with standard brachytherapy is that many patients and health care providers find the appearance of traditional brachytherapy catheters disturbing, describing them as appearing “barbaric” and “very painful looking.” This impression of brachytherapy catheters might have limited their more widespread use in the U.S.

Recently, there has been extreme interest on the part of physicians and patients alike regarding the technique of balloon intracavitary brachytherapy, which offers several potential advantages over traditional interstitial brachytherapy for patients with breast carcinoma. Balloon intracavitary brachytherapy after lumpectomy can be administered with a new breast brachytherapy applicator called the MammoSite© (MammoSite Radiation Therapy System; Proxima Therapeutics, Alpharetta, GA).48, 49 The MammoSite device is comprised of an HDR source at the center of an inflatable balloon that can be placed into the lumpectomy cavity at the time of surgery or after surgery when the definitive margin status is known. The results of the initial clinical experience with the MammoSite breast brachytherapy device in women with invasive ductal carcinomas < 2 cm recently was reported. The results of this study led the U.S. Food and Drug Administration to approve the catheter for clinical use and led the authors to conclude that the applicator provides a simpler method for performing reproducible breast brachytherapy.

Intraoperative radiation therapy

Delivery of a single dose of radiation at the time of surgery for definitive local treatment would be an extremely attractive alternative for patients undergoing lumpectomy for breast carcinoma. Intraoperative radiation therapy (IORT) has been utilized in patients with breast carcinoma to provide an intraoperative boost dose of 9 Gy to the local tumor bed, with IORT followed by additional external beam radiation therapy to the whole breast over the course of 6 weeks.51–53 Investigators utilizing IORT for the boost dose state that directed application of the radiation dose can avoid the problem of a geographic miss and can ensure that the skin is not irradiated and therefore that cosmesis is improved. However, complications (primarily at the skin and the chest wall and ribs) are a major concern and limit the intraoperative dose. IORT can be delivered using a mobile linear accelerator or using a minielectron–beam, low-energy X-ray source.

To investigate whether external radiation therapy can be safely replaced by IORT in the treatment of breast tumors measuring ≤ 2.5 cm in dimension, an ongoing randomized trial at the European Institute of Oncology compares IORT at a total dose of 21 Gy (prescribed at the 90% isodose) with the standard external radiation therapy regimen of 50 Gy to the whole breast and 10 Gy as a boost to the tumor bed. The recruitment phase for this study is expected to be completed in 2 years.

Three-dimensional conformal external beam radiation therapy

Formenti et al.54 have studied three-dimensional (3D) conformal partial breast irradiation using external beam radiation therapy with the patient imaged and treated on a dedicated computed tomography scanner. The planning target volume used by these investigators was the tumor bed plus a 1–2-cm margin defined on postlumpectomy computed tomography. Nine patients in this study received a total dose of 25–30 Gy in 5 fractions. At a median follow-up of 3 years, all patients were found to have good to excellent cosmesis. The authors concluded that hypofractionated conformal breast irradiation is feasible and that further studies are warranted. Recently, the technologic advances of intensity-modulated radiation therapy (IMRT) also have been explored as a potential method for delivering a more uniform and standardized radiation dose in patients treated with breast-conserving surgery.55–57 IMRT makes unnecessary the usual reliance on flat radiation fields and replaces that simple paradigm with a variable-intensity pattern that is determined with the aid of a computerized optimization algorithm.58 The main goal of the majority of IMRT and optimization work is the delivery of more conformal therapy.58

One limitation of 3D conformal external beam radiation therapy as a method for APBI is the difficulty in controlling movement associated with deep breathing.59–60 The advantages of 3D conformal external beam radiation therapy are that it is less invasive than brachytherapy and that most radiation facilities in the U.S. already have the tools required for this method of radiation delivery. Furthermore, because the overwhelming majority of practicing radiation oncologists do not have experience with brachytherapy, external beam radiation therapy might be more feasible than brachytherapy for APBI. However, more large-scale studies are needed to ensure that the 3D method will deliver a radiation dose that is adequate to ensure low breast recurrence rates.

A multiinstitutional Phase II study designed to address some of these factors is currently being conducted by the Radiation Therapy Oncology Group.

Proposal for a Multiinstitutional Phase II Study to Test the Safety of Repeat Breast-Conserving Surgery and APBI after In-Breast Local Recurrence: Patient Selection and Other Considerations

It is clear that most patients with primary breast carcinoma who are treated with breast-conserving surgery without whole breast irradiation will not develop a local recurrence. However, most patients and their physicians believe that the long-term local recurrence rate of 30–40% without breast irradiation is unacceptably high. Whole breast irradiation can decrease in-breast local recurrence rates by approximately 67%. We hypothesize that secondary in-breast local recurrence rates also can be reduced by 30–40% with irradiation and that the toxicity of repeat irradiation will be acceptable when APBI is utilized. To test this hypothesis, we propose a Phase II multicenter study to evaluate the potential of repeat breast-conserving surgery followed by PBI.

With regard to patients with invasive breast carcinoma, there is broad consensus that it is difficult to select any particular group of patients in whom the risk of local recurrence is so low with conservative surgery alone that the addition of breast irradiation would be unnecessary.61 However, many experts contend that breast irradiation for small, low-grade, or intermediate-grade DCIS with clearly negative margins may not be necessary; this hypothesis currently is being tested in a randomized study by the Radiation Therapy Oncology Group.62 This hypothesis also may apply to women who develop a noninvasive in-breast local recurrence, although to our knowledge this has yet to be tested formally.

There are some obvious necessary inclusion criteria for patients selected to participate in a trial of repeat breast-conserving surgery. Patients will need to desire repeat breast–conserving surgery, have no evidence of multicentricity by diagnostic mammography, and have negative microscopic margins of resection. The addition of magnetic resonance imaging (MRI) at the time of local recurrence to exclude MRI evidence of multicentricity would further help to identify the most suitable candidates for this approach. Tumor size and lymph node status are other important considerations; patients with invasive recurrences > 2 cm with simultaneous lymph node recurrence have been shown to have a worse prognosis than patients with smaller invasive recurrences without lymph node recurrence. Finally, time to recurrence should be taken into account. early recurrences after initial conservative therapy have been shown to have a worse prognosis than later recurrences.

For our Phase II trial, we propose the following eligibility criteria: recurrent invasive ductal carcinoma or DCIS measuring < 2 cm, no evidence of multicentric disease on mammography and MRI, no clinical evidence of simultaneous lymph node recurrence, and an interval of at least 3 years between the completion of initial radiation therapy and recurrence. Patients who receive neoadjuvant systemic therapy prior to repeat conservative surgery also would be eligible for participation.

PBI would potentially be comprised of multicatheter brachytherapy, MammoSite brachytherapy, or 3D conformal radiation therapy (Fig. 1).

Figure 1.

Proposed Phase II multicenter trial to evaluate repeat breast-conserving surgery and accelerated partial breast irradiation for the local recurrence of breast carcinoma.

Endpoints to be evaluated in this Phase II study would include: 1) wound healing and potentially wound-related side effects; 2) reproducibility of partial breast irradiation techniques (evaluated by measurement of dose homogeneity and coverage of target volume); 3) patient satisfaction with the treatment as measured by a questionnaire; 4) cosmesis as judged by the patient and by an independent evaluator based on the Harvard Scale and documented by serial digital photographs; 5) ipsilateral breast carcinoma recurrence and contralateral breast carcinoma occurrence; and 6) mastectomy-free survival.

On the basis of the results of this Phase II study, further consideration will be given to Phase III studies in the management of in-breast local recurrences.

Impact of Systemic Therapy Given at the Time of In-Breast Local Recurrence

Many studies have shown that the use of systemic therapy in the form of tamoxifen or cytotoxic chemotherapy reduces the absolute risk of local breast carcinoma recurrence in patients treated with breast conservation therapy. The absolute risk reduction is in the range of 10% at 8 years compared with the risk in patients not receiving systemic therapy. Although we are aware of no studies that address the effect of systemic therapy on subsequent local control after repeat breast-conserving surgery, we hypothesize that systemic therapy also will reduce the risk of local recurrence in this setting.

It is widely accepted that early invasive local recurrence after breast-conserving surgery is associated with an adverse prognosis. Given this, it is logical to hypothesize that the administration of systemic therapy at the time of first isolated local recurrence may have a positive impact on survival after local recurrence. To our knowledge, no specific studies to date have addressed this hypothesis in patients with a local recurrence after breast conservation therapy. However, this hypothesis was addressed in 167 patients with a first isolated local and regional recurrence after mastectomy for breast carcinoma in a Swiss Phase III multicenter study comparing tamoxifen with observation after local therapy.63, 64 Although a local recurrence of breast carcinoma after mastectomy appears to confer a worse prognosis in general than local recurrence after breast conservation, the results of this multicenter study have implications for the latter clinical scenario.23, 65, 66 At a median follow-up of 11.6 years in surviving patients, tamoxifen was associated with improved disease-free survival (61% vs. 33%; P = 0.006).66 However, it is important to note that the difference in disease-free survival was due mainly to a reduction in the rate of further local recurrences (P = 0.011). For the entire study population, no detectable difference in overall survival was found between treatment groups. Thus, these results suggest that systemic therapy with tamoxifen and, potentially, the newer aromatase inhibitors may have a favorable influence on locoregional control in women who choose to undergo complete reexcision of a local estrogen receptor-positive, in-breast local recurrence.

In addition to this study evaluating tamoxifen, we found only two additional, very small trials of systemic therapy for isolated locoregional recurrence. One study randomized 32 patients to treatment with dactinomycin versus observation67; the other study randomized 32 patients to treatment with interferon-α versus observation.68 In the dactinomycin study, the investigators found an absolute increase in local control of 19% in the chemotherapy arm compared with those patients treated with observation alone, without any demonstrable effect on survival.67 As would be expected, no effect on local control or survival was noted in the small study evaluating interferon-α.68

Motivated by the paucity of randomized trial data evaluating the role of systemic therapy after isolated locoregional breast carcinoma recurrence, investigators at the International Breast Cancer Study Group have initiated a randomized multicenter international trial of chemotherapy versus observation for the treatment of patients with resected locoregional recurrence of breast carcinoma.69 Those patients with in-breast or chest wall recurrence are eligible. The trial was designed to allow the treating clinician to choose the chemotherapy drugs to ensure adequate participation and rapid patient accrual. For patients randomized to receive chemotherapy, the duration of chemotherapy is at least three cycles. All patients in both arms are required to receive hormonal therapy if the locoregional recurrence is estrogen receptor positive. The investigators assume on the basis of multiple previously cited studies that the overall 5-year survival for the group receiving no chemotherapy will be in the 50% range. Based on this assumption, an overall total of 347 events will be required to detect an improvement in 5-year survival to 60% with an 80% power. Because approximately 5% of the cases are expected to be nonvaluable, 245 patients will need to be accrued each year over the course of 4 years for a total of 977 total patients.

Summary and Future Directions

Rapidly emerging technologic advances in radiation delivery in the form of balloon-based intracavitary brachytherapy and 3D conformal external beam radiation therapy have quickly brought the concept of APBI to the forefront as a potential alternative to standard whole breast irradiation in patients treated with initial breast-conserving surgery. Although there is a growing amount literature regarding the effectiveness of APBI after initial breast-conserving surgery, to our knowledge extremely few patients have been treated with repeat breast-conserving surgery followed by modern techniques of APBI. Because the radiation therapy is delivered directly to the area at potential risk for recurrence, APBI holds the promise of potentially limiting the toxicity of reirradiation of breast tissue.

Based on limited studies regarding repeat breast-conserving surgery alone for the treatment of in-breast local recurrences and emerging conformal techniques of radiation delivery, we hypothesize that mastectomy is not necessary in all patients after local recurrence of breast carcinoma and that repeat breast-conserving surgery followed by PBI in appropriately selected patients (no multicentric disease as indicated by mammography and MRI and negative microscopic resection margins) will result in local recurrence rates similar to those obtained with mastectomy with acceptable toxicity and acceptable cosmetic results. Optimal patient selection for a Phase II study addressing this hypothesis would include women with tumor recurrences measuring < 2 cm in greatest dimension without a concurrent lymph node recurrence diagnosed at least 3 years after the initial treatment of breast carcinoma. There currently are ongoing prospective trials evaluating the role of systemic therapy for breast carcinoma after locoregional recurrence. Well controlled prospective local therapy trials also are needed at this time to challenge the accepted dogma that mastectomy is required in all patients after the local recurrence of breast carcinoma.