doi: /j.ijrobp

Transcription

1 doi: /j.ijrobp Int. J. Radiation Oncology Biol. Phys., Vol. 79, No. 4, pp , 2011 Copyright Ó 2011 Elsevier Inc. Printed in the USA. All rights reserved /$ see front matter ASTRO GUIDELINE LIMITATIONS OF THE AMERICAN SOCIETY OF THERAPEUTIC RADIOLOGY AND ONCOLOGY CONSENSUS PANEL GUIDELINES ON THE USE OF ACCELERATED PARTIAL BREAST IRRADIATION FRANK VICINI, M.D.,* DOUGLAS ARTHUR, M.D., y DAVID WAZER, M.D., z PETER CHEN, M.D.,* CHRISTINA MITCHELL, R.N.,* MICHELLE WALLACE, R.N.,* LARRY KESTIN, M.D.,* AND HONG YE, M.S.* *Department of Radiation Oncology, William Beaumont Hospital, Royal Oak, MI; y Department of Radiation Oncology, Virginia Commonwealth University, Richmond, VA; and z Department of Radiation Oncology, Tufts University School of Medicine, Boston, MA Purpose: We applied the American Society of Therapeutic Radiology and Oncology (ASTRO) Consensus Panel (CP) guidelines for the use of accelerated partial breast irradiation (APBI) to treated with this technique to determine the ability of the guidelines to differentiate with significantly different clinical outcomes. Methods and Materials: A total of 199 treated with APBI and 199 with whole-breast irradiation (WBI) (matched for tumor size, nodal status, age, margins, receptor status, and tamoxifen use) were stratified into the three ASTRO CP levels of suitability ( suitable, cautionary, and unsuitable ) to assess rates of ipsilateral breast tumor recurrence (IBTR), regional nodal failure, distant metastases, disease-free survival, cause-specific survival, and overall survival based on CP category. Median follow-up was 11.1 years. Results: Analysis of the APBI and WBI patient groups, either separately or together (n = 398), did not demonstrate statistically significant differences in 10-year actuarial rates of IBTR when stratified by the three ASTRO groups. Regional nodal failure and distant metastasis were generally progressively worse when comparing the suitable to cautionary to unsuitable CP groups. However, when analyzing multiple clinical, pathologic, or treatment-related variables, only patient age was associated with IBTR using WBI (p = 0.002). Conclusions: The ASTRO CP suitable group predicted for a low risk of IBTR; however, the cautionary and unsuitable groups had an equally low risk of IBTR, supporting the need for continued refinement of patient selection criteria as additional outcome data become available and for the continued accrual of to Phase III trials. Ó 2011 Elsevier Inc. Breast-conserving therapy, Brachytherapy, Radiation, Partial breast irradiation, Breast cancer. INTRODUCTION Over the past decade, there has been increased interest in the use of accelerated partial breast irradiation (APBI) as an alternative to whole breast irradiation (WBI) for adjuvant radiation treatment after breast-conserving surgery. Accelerated partial breast irradiation offers decreased overall treatment times and a potential decrease in the radiation dose delivered to nontarget portions of the breast and adjacent tissues. However, there are several possible disadvantages to APBI. One primary concern is that occult foci of cancer in remote areas of the breast (away from the tumor bed) may remain untreated. Given the tremendous interest in APBI, numerous multicenter, prospective randomized clinical trials (including the NSABP B39/RTOG 0413 Phase III trial) have been initiated or completed to compare the effectiveness and safety of APBI vs. WBI (1 3). Despite limited long-term clinical data, the use of APBI outside of clinical trials has increased sharply. In the past, few data were available to help guide physicians as to which could be safely treated with APBI and which should more appropriately receive WBI. In light of increasing evidence that postoperative adjuvant radiation therapy significantly improves long-term overall survival after breastconserving surgery (4), it seems prudent that conservative patient selection criteria for APBI should be followed until additional long-term data are available to indicate otherwise. Unfortunately, the optimal set of selection criteria to use when applying APBI remains poorly defined. Given these concerns, the American Society of Therapeutic Radiology and Oncology (ASTRO) Health Services Research Committee convened a Task Force of breast cancer experts from various oncologic specialties to develop Reprint requests to: Frank A. Vicini, M.D., FACR, Department of Radiation Oncology, Beaumont Cancer Institute, William Beaumont Hospital, 3601 W. 13 Mile Road, Royal Oak, MI Tel: (248) ; Fax: (248) ; fvicini@ beaumont.edu 977 Conflict of interest: none. Received Nov 3, 2009, and in revised form Dec 15, Accepted for publication Dec 15, 2009.

2 978 I. J. Radiation Oncology d Biology d Physics Volume 79, Number 4, 2011 a consensus statement addressing patient selection criteria and best practices for the application of APBI outside of clinical trials (2). These recently published recommendations were based on the results of a systematic literature review of the limited APBI data and were also supplemented by the expert opinions of the Task Force members using data applying WBI to treat. Because knowledge regarding APBI is rapidly accumulating, the Task Force acknowledged that the consensus statement would require frequent updates and modifications to account for ongoing research findings. It was hoped that the recommendations contained within the consensus statement would help to provide guidance not only for the use of APBI outside of a clinical trial but also to serve as a framework to promote additional clinical investigations into the optimal role of APBI in the treatment of breast cancer. To that end, we applied the ASTRO Consensus Panel (CP) guidelines for the application of APBI to a large group of treated at our institution with this technique to help clarify their ability to differentiate with significantly different clinical outcomes. Whole-breast irradiation To compare the rate of IBTR in a comparable group of treated with WBI, each APBI patient was matched with one WBI patient at WBH. The patient database at WBH extends from 1980 through 1997 and consists of 1,503 consecutively treated with WBI (all treatment data were collected retrospectively). Each WBI patient underwent lumpectomy and axillary lymph node dissection followed by standard WBI at a median total dose to the tumor bed of 60 Gy. Of these WBI therapy, 709 (47%) were initially selected according to the following APBI protocol selection criteria: infiltrating ductal histology, absence of an extensive intraductal component, negative margins, tumor size <3.0 cm, performance of an axillary lymph-node dissection, and three or fewer involved lymph nodes. Each APBI patient was randomly matched to a unique patient from among the 709 WBI therapy who met the following additional matching criteria. Patients were matched one-to-one according to age (10 years), tumor size (5 mm), nodal status (negative vs. one to three positive nodes), estrogen receptor status (positive vs. negative vs. unknown), and the use of adjuvant tamoxifen therapy (yes vs. no). Outcome (i.e., any type of tumor recurrence) was blinded for both the WBI and APBI when matching was performed. METHODS AND MATERIALS Study participants The study population consists of 199 treated with APBI and 199 treated with WBI. All (see below) were categorized according to the ASTRO CP guidelines for APBI and analyzed for the ability of these groupings to predict differences in clinical outcome compared with traditional clinical, pathologic, and treatment-related variables. APBI A total of 199 consecutive with invasive early-stage breast cancer were prospectively treated at William Beaumont Hospital (WBH) with APBI using interstitial brachytherapy directed only at the region of the tumor bed as part of their BCT from 1993 to All eligible had Stage I/II breast cancer. Eligibility criteria have been published in previous experiences and have included infiltrating ductal carcinomas less than 3.0 cm in diameter, negative surgical margins ($2 mm), more than 40 years of age at diagnosis, and negative lymph node status (5). Patients with an extensive intraductal component, infiltrating lobular histology, ductal carcinoma in situ, and clinically significant areas of lobular carcinoma in situ were generally excluded (6). Patients who did not meet all of these eligibility criteria for APBI were also included in this analysis. Several of these had more than one criterion that did not meet all eligibility criteria (i.e., positive lymph nodes). Follow-up was complete through September 2008 for the APBI. All were treated with radiation techniques according to the guidelines of one of three different protocols for APBI brachytherapy (5). A total of 120 (60%) were treated as in with a low-dose-rate (LDR) implant, and 79 (40%) were treated as out with a high-dose-rate (HDR) implant. A total of 139 (70%) received adjuvant therapy after completion of APBI radiation therapy: 25 (13%) received adjuvant systemic chemotherapy, and 114 (57%) received adjuvant tamoxifen therapy. ASTRO CP groups All (APBI and WBI) were categorized into one of three ASTRO CP groups (outlined in two recent publications) as described below. Suitable. Patients were required to have all the following characteristics: age $60 years, BRCA1/2 mutation absent, tumor size #2 cm, stage T1, margins negative by at least 2 mm, any grade, no lymphatic vessel invasion (LVI), ER status was positive, tumors must be unicentric only and clinically unifocal with total size #2.0 cm, only invasive ductal histology or other favorable subtypes, pure DCIS was not allowed, EIC not allowed, associated LCIS allowed, stage pn0 (i-, i+), sentinel biopsy or axillary lymph node dissection required, and neoadjuvant therapy not allowed. Cautionary. Patients were classified as of cautionary status if they had any one or more of the following characteristics: age 50 to 59 years, tumor size 2.1 to 3.0 cm, stage T0 or T2, margins close (<2 mm), LVSI limited/focal, ER status negative, clinically unifocal with total size 2.1 to 3.0 cm, invasive lobular, pure DCIS #3 cm, EIC #3 cm. Unsuitable. Patients in this group had any one or more of the following: age <50 years, BRCA1/2 mutation present, tumor size >3 cm, Stage T3 to T4, margins positive, LVSI extensive, multicentricity present, microscopically multifocal >3 cm in total size or if clinically multifocal, pure DCIS >3 cm in size, EIC >3 cm in size, Stage pn1, pn2, pn3 nodal surgery not performed, or neoadjuvant therapy used. Outcome measures An IBTR was defined as the recurrence of cancer in the treated breast before or at the time of regional failure or metastases. Contralateral breast failure was defined as the development of breast cancer in the opposite, untreated breast. Overall survival (OS) was defined as all deaths from any cause. Cause-specific survival (CSS) was defined as all deaths attributed only to breast cancer. Disease-free survival (DFS) was defined as all events attributed to breast cancer, i.e., local, regional, or distant recurrence.

3 Limitations of ASTRO Consensus Panel guidelines on APBI d F. VICINI et al. 979 ASTRO grouping Table 1. Study population vs. ASTRO guideline group All (n = 398) APBI (n = 199) Statistical analysis The estimated likelihood of events for IBTR, contralateral breast failure, DFS, OS, and CSS were calculated by the Kaplan-Meier method, and the statistical significance of differences between treatment and CP groups were calculated using the log-rank test. The association of categorical variables (e.g., margin status) with treatment group was analyzed using Fisher s two-tailed exact test. The statistically significant differences between two sample means for continuous variables (e.g., patient age in years) were analyzed using the Student s unpaired t-test. Regression analysis was performed using the Cox proportional hazards model. All time intervals were calculated from the date of completion of RT. A p value of #0.05 was considered statistically significant. Statistical analyses were performed using SPSS version 17.0 (SPSS Inc., Chicago, IL), and all statistical tests were two-sided. RESULTS EBRT matched (n = 199) N (%) N (%) N (%) p Value Suitable 176 (44.2%) 95 (47.7%) 81 (40.7%) Cautionary 143 (35.9%) 63 (31.7%) 80 (40.2%) Unsuitable 79 (19.8%) 41 (20.6%) 38 (19.1%) Abbreviations: APBI = accelerated partial breast irradiation; AS- TRO = American Society of Therapeutic Radiology and Oncology; EBRT = external beam radiation therapy. The median follow-up was 11.1 years for all (9.33 years for APBI and for WBI, p < 0.001). The median age at diagnosis was years for all (65.0 years for APBI and for WBI, p = 0.10) Patients were categorized according to the three ASTRO CP groups in Tables 1 to 7. There were no statistically significant differences in the number of in each ASTRO grouping based on treatment technique (Table 1). Suitable For APBI, 47.7% of had all of the suitable ( S ) characteristics and for WBI, 40.7% had all of these findings (p = 0.197). Cautionary For APBI, 31.7% of had one or more of the cautionary ( C ) characteristics and for WBI, 40.2% had these findings. For APBI, the distribution of characteristics that placed them into this CP group was as follows: age 50 to 59 years (n = 36), tumor size 2.1 to 3 cm (n = 10), negative estrogen receptor status (n = 19), margins negative but <2 mm (n = 4), presence of an EIC but <3 cm (n = 8), and pure DCIS <3 cm or lobular histology (n = 1). Some had more than one of these findings. For WBI, the distribution of characteristics that placed them into this CP group was as follows: age 50 to Table 2. Ten-year outcome for all (n = 398) vs. ASTRO group Suitable Cautionary Unsuitable p Value Local recurrence 3.3% 3.9% 5.5% Regional nodal 0% 0.7% 4.0% 0.017* Failure Distant metastases 3.7% 7.6% 12.1% Disease-free 93.1% 89.3% 85.2% Survival Overall survival 75.9% 81.7% 83.0% 0.011* Cause-specific Survival 96.5% 93.6% 90.9% Abbreviation: ASTRO = American Society of Therapeutic Radiology and Oncology. * p < years (n = 38), tumor size 2.1 to 3 cm (n = 12), negative estrogen receptor status (n = 21), margins negative but <2 mm (n = 0), presence of an EIC but < 3 cm (n = 22), and pure DCIS <3 cm or lobular histology (n = 6). Some had more than one of these findings. Unsuitable For APBI, 20.6% of had any of the unsuitable ( U ) characteristics, and for WBI, 19.1% had any of these findings. For APBI, the distribution of characteristics that placed them into this CP group was as follows: age <50 years (n = 19), tumor size >3 cm (n = 0), positive margins (n = 1), positive lymph nodes (n = 23), and DCIS >3 cm (n = 0). Some had more than one of these findings. For WBI, the distribution of characteristics that placed them into this CP group was as follows: age <50 years (n = 15), tumor size >3 cm (n = 1), positive margins (n = 1), positive lymph nodes (n = 23), and DCIS >3 cm (n = 0). Some had more than one of these findings. Tables 2 to 4 present the 10-year actuarial rates of various clinical endpoints for all (n = 398), APBI only (n = 199), and WBI only (n = 199) vs. the three ASTRO groupings. In APBI (Table 3), use of Table 3. Ten-year outcome for APBI (n = 199) vs. ASTRO group Suitable (n = 95) Cautionary (n = 63) Unsuitable (n = 41) p Value Local recurrence 2.6% 7.8% 2.5% Regional nodal 0% 1.7% 5.9% failure Distant metastases 0% 7.1% 11.2% 0.018* Disease-free 97.4% 85.3% 88.9% survival Overall survival 73.6% 74.3% 82.1% Cause-specific Survival 98.3% 91.4% 92.2% Abbreviations: APBI = accelerated partial breast irradiation, AS- TRO = American Society of Therapeutic Radiology and Oncology. * p < 0.05.

4 980 I. J. Radiation Oncology d Biology d Physics Volume 79, Number 4, 2011 Table 4. Ten-year outcome for WBI (n = 199) vs. ASTRO group Suitable (n = 81) Cautionary (n = 79) Unsuitable (n = 39) p Value Local recurrence 4.1% 1.2% 8.4% Regional nodal failure 0% 0% 2.7% Distant metastases 7.7% 8.0% 13.2% Disease-free survival 88.4% 92.1% 81.5% Overall survival 78.8% 86.2% 84.2% Cause-specific survival 94.6% 94.6% 89.5% Abbreviations: ASTRO = American Society of Therapeutic Radiology and Oncology; WBI = whole breast irradiation. Table 6. Risk of 10-year regional nodal failure vs. ASTRO group and treatment ASTRO grouping All (n = 398) APBI (n = 199) EBRT matched (n = 199) p Value Suitable (n = 176) 0% 0% 0% N/A Cautionary (n = 132) 0.7% 1.7% 0% Unsuitable (n = 90) 4.0% 5.9% 2.7% p Value 0.017* Abbbreviations: APBI = Accelerated partial breast irradiation; ASTRO = American Society of Therapeutic Radiology and Oncology, EBRT = External beam radiation therapy; N/A = not available. * p < the CP groupings resulted in statistically significant differences in the rate of DM (p = 0.018). Table 5 presents the 10-year actuarial rates of IBTR vs. ASTRO groups and treatment technique. There were no statistically significant differences in the IBTR rate regardless of CP group or treatment technique. Table 6 presents the 10-year actuarial rates of RNF vs. AS- TRO groups and treatment technique. There were no statistically significant differences in RNF rates regardless of CP group or treatment technique Table 7 presents the 10-year actuarial rates of DMs vs. AS- TRO groups and treatment technique. As noted above, application of the CP groupings resulted in statistically significant differences in the rate of DM (p = 0.018) only in APBI. Table 8 presents a univariate analysis of variables associated with IBTR in all combined and in APBI or WBI separately. In WBI, only age at diagnosis (both as a categorical variable, p = 0.007, and as a continuous variable, p = 0.002) was associated with IBTR. The ASTRO CP groupings were not associated with IBTR in treated with APBI. Table 9 presents a univariate analysis of variables associated with RNF or DM for all (n = 398) and RNF for APBI alone (n = 199). On multivariate analysis (data not shown), only the presence of an EIC was associated with RNF in all (p = 0.018) and in APBI (p = 0.028). The ASTRO CP groupings were not associated with any endpoint (RNF or DM) for either group of (WBI or APBI). Table 5. Risk of 10-year local recurrence vs. ASTRO group and treatment type ASTRO grouping All (n = 398) APBI (n = 199) EBRT matched (n = 199) p Value Suitable (n = 176) 3.3% 2.6% 4.1% Cautionary (n = 132) 3.9% 7.8% 1.2% Unsuitable (n = 90) 5.5% 2.5% 8.4% p Value Abbreviations: APBI = accelerated partial breast irradiation; AS- TRO = American Society of Therapeutic Radiology and Oncology; EBRT = external beam radiation therapy. DISCUSSION We applied the ASTRO CP guidelines for the use of APBI to treated with this technique at William Beaumont Hospital to evaluate the predictive ability of the described levels of suitability in regard to significant differences in long-term clinical outcome and appropriateness for treatment with APBI. With a median follow-up of 11.1 years, analysis of both APBI and a matched group of WBI (either together or by method of treatment alone) when stratified by the three ASTRO CP groups did not demonstrate statistically significant differences in the 10-year actuarial rates of IBTR. Regional nodal failure and DM were generally progressively worse when comparing the S to C to U groups. However, when analyzing multiple variables potentially associated with IBTR, only age at diagnosis was associated with IBTR when using WBI (p = 0.002). The ASTRO CP groupings did not independently differentiate who may or may not be more appropriately treated with this technique. These results await validation from other large APBI databases and suggest that further refinement of these groupings may be needed. They also support the continued enrollment of high-risk considered for APBI in Phase III clinical trials. Overview Despite limited long-term data available on the use of APBI, treatment outside of clinical trials has increased Table 7. Risk of 10-year distant metastases vs. ASTRO group and treatment type ASTRO grouping All APBI EBRT matched subjects (n = 398) (n = 199) (n = 199) p Value Suitable (n = 176) 3.7% 0% 7.7% 0.006* Cautionary (n = 132) 7.6% 7.1% 8.0% Unsuitable (n = 90) 12.1% 11.2% 13.2% p Value * Abbreviations: APBI = accelerated partial breast irradiation; AS- TRO = American Society of Therapeutic Radiology and Oncology; EBRT = external beam radiation therapy. * p < 0.05.

5 Limitations of ASTRO Consensus Panel guidelines on APBI d F. VICINI et al. 981 Table 8. Univariate analysis of variables associated with local recurrence All (n = 398) WBI (n = 199) APBI (n = 199 Variable* OR (95% CI) p Value OR (95% CI) p Value OR (95% CI) p Value Age <50 vs. $ ( ) ( ) ( ) continuous ( ) ( ) ( ) Tumor stage T1 vs. T ( ) ( ) Tumor size (mm) #20 vs. > ( ) ( ) ( ) Histology ductal vs. lobular EIC absent vs. present ( ) ( ) ( ) Grade I/II vs. III ( ) ( ) ( ) ER status ( ) vs. (+) ( ) ( ) ( ) Margins #2 vs. >2 mm ( ) Nodal status (+) vs. ( ) ( ) ( ) ( ) Treatment WBI vs. APBI ( ) NA NA NA NA ASTRO grouping I/II vs. III ( ) ( ) ( ) I vs. II/III ( ) ( ) ( ) Abbreviations: APBI = accelerated partial breast irradiation; ASTRO = American Society of Therapeutic Radiology and Oncology; ASTRO grouping: I = suitable; II = cautionary; III = unsuitable; CI = confidence interval; ER = estrogen receptor; EIC = extensive intraductal component; LVI = lymphatic vessel invasion; OR = odds ratio (comparing the reference group). * First group is reference group for the following variables. markedly (1). For example, it is estimated that more than 45,000 women in the United States have been treated with the MammoSite (Hologic, Bedford, MA) breast brachytherapy catheter alone to deliver APBI. This does not include the thousands of other women treated with external beam APBI techniques, intraoperative, single fraction radiation (both in the United States and Europe) or other newer forms (devices) of APBI currently being used in clinic practice (7 9). Combined with increasing evidence that postlumpectomy RT improves long-term overall survival, most breast cancer experts agree that conservative patient selection criteria for APBI should be developed and followed until additional data from Phase III trials are available. Given these concerns, the ASTRO Health Services Research Committee convened a Task Force of breast cancer experts to generate a consensus statement regarding optimal patient selection criteria and best practices for the application of APBI. These recommendations were based on the results of a systematic literature review of APBI data and were supplemented by the expert opinions of the Task Force members. A remarkable achievement of the Task Force was that after a comprehensive literature review, a broad-based consensus was achieved to definitively identify that can be appropriately treated with APBI outside the context of a clinical trial. However, the Task Force also readily acknowledged that there were insufficient data to determine which were clearly inappropriate for APBI. This critical absence of data drove the Task Force to rely heavily upon expert opinion in formulation of the guideline categories of cautionary and unsuitable as applied to APBI that is offered outside the confines of a clinical trial. In retrospect, the specific word choice in labeling these categories may have been misleading, as it did not fully reflect the scope of contemporary data for APBI. As the current study clearly indicates, characterized as cautionary and unsuitable are, in fact, appropriate for APBI provided that it is in the context of ongoing structured clinical investigation. The ASTRO Task Force explicitly recognized the need for continued data accumulation for in the cautionary and unsuitable categories and fully anticipated that the current guidelines would require periodic update and modification. Development of the ASTRO Consensus Panel Guideline criteria Developing selection criteria for the optimal use of APBI requires the identification of subsets of with a very low risk of clinically occult disease remote from the lumpectomy cavity. Unfortunately, few data have been conclusively identified from pathologic studies or prospective clinical

6 982 I. J. Radiation Oncology d Biology d Physics Volume 79, Number 4, 2011 Table 9. Univariate analysis of variables associated with regional nodal failure or distant metastases in all (n = 398) and APBI (n = 199) Regional nodal failure (n = 398) Distant metastases (n = 398) APBI only, regional nodal failure Variable* OR (95% CI) p Value OR (95% CI) p Value OR (95% CI) p Value Age <50 vs. $ ( ) ( ) continuous ( ) ( ) ( ) Tumor stage T1 vs. T ( ) ( ) ( ) Tumor size (mm) #20 vs. > ( ) ( ) ( ) Histology ductal vs. lobular EIC Absent vs. present ( ) ( ) ( ) Grade I/II vs. III ( ) ER status ( ) vs. (+) ( ) ( ) ( ) Margins #2 vs. >2 mm ( ) Nodal status (+) vs. ( ) ( ) ( ) ( ) Treatment WBI vs. APBI ( ) ( ) NA NA ASTRO grouping I/II vs. III ( ) ( ) I vs. II/III ( ) Abbreviations: APBI = accelerated partial breast irradiation; ASTRO grouping: I = suitable, II = cautionary, III = unsuitable; CI = confidence interval; ER = estrogen receptor; EIC = extensive intraductal component; LVI = lymphatic vessel invasion; NA = not applicable; OR = odds ratio (comparing the reference group). * First group is reference group for the following variables. studies of APBI to clearly define this subgroup of breast cancer (10). As a result, the proposed clinical pathologic selection criteria for the ASTRO suitable group were developed (according to the CP authors) based entirely upon the common inclusion criteria and characteristics of enrolled in prospective single-arm studies of APBI that had a minimum of 4 years follow-up and that reported a low risk of IBTR (11). These data were then supplemented by the Task Force s general knowledge of risk factors for IBTR with WBI. It was acknowledged by the authors that there may be other groups of for whom APBI will prove to be suitable, but the Task Force believed that there was insufficient evidence to identify them at the time of CP publication. The ASTRO cautionary group included those for whom the Task Force expressed uncertainty regarding the appropriateness of APBI, although they acknowledged at least some in this group had been included in single-arm, prospective trials. Again, there were no APBI data available that demonstrated (or even suggested) in this cautionary group had significantly different (worse) rates of IBTR, RNF or DM when applying APBI. Finally, the unsuitable group included those for whom there was very limited evidence (too few treated with these characteristics) from clinical trials to support the use of APBI. Furthermore, the Task Force strongly suggested that those who remain eligible for the ongoing RTOG 0413/ NSABP B-39 randomized clinical trial ( aged <50 years or any age with estrogen receptor [ER] negative tumors or any age with pn1 tumors) should be strongly encouraged to participate in this important trial. It should be noted again that this category was not developed because of negative data on the application of APBI in this patient group. In fact, there have been several small, recently published studies in these higher-risk that have demonstrated acceptable rates of IBTR, further supporting the ethical and appropriate randomization of these on the NSABP B39/RTOG 0413 phase III trial so that the breadth of APBI application can be correctly defined (12, 13). Several other societies and organizations have also provided guidelines for the application of APBI both on and off-protocol. These include statements by the American Brachytherapy Society (ABS) (11, 14), the American Society of Breast Surgeons (ASBrS) (15), the American College of Radiation Oncology (ACRO), and a consensus statement from three German oncology societies (the German Society of Radiation Oncology, the German Society of Senology, and the Working Group for Gynecological Oncology of the German Cancer Society) (16). Many of the selection criteria from these other societies overlap with those of the ASTRO CP guidelines; but, generally, the methodology used to formulate

7 Limitations of ASTRO Consensus Panel guidelines on APBI d F. VICINI et al. 983 these recommendations is similar. Typically, all of these groups (1) support the participation of in randomized clinical studies comparing APBI with WBRT, (2) discourage the routine use of APBI outside clinical trials for higher risk, and (3) suggest that until definite results show that APBI neither impairs therapeutic outcome nor cosmetic results, WBI should remain the gold standard in the treatment of early breast cancer. This analysis did demonstrate that the CP suitable group could be used as a tool to help select appropriate for the application of APBI on or off protocol, yet failed to verify that the cautionary and unsuitable groups identify experiencing a higher rate of IBTR or are inappropriate for APBI. These findings need to be viewed cautiously, however. This was a relatively small group of treated with only one form of APBI (at a single institution). In addition, the number of events (IBTRs) was low, thus limiting the statistical power to validate any associations. Clearly, additional studies using other forms of APBI are needed to determine what refinements in these groupings may be needed. It may also be possible that other variables need to be explored as potential risk factors for the development of an IBTR when applying APBI. These as-yet-unidentified risk factors may or may not be similar to those observed with conventional WBI. Current analysis Only a few other groups have published their experience on potential risk factors for the development of an IBTR with the application of APBI. Most APBI studies have too few treated, a relatively short follow-up period, and a limited number of IBTR events. However, in a series of 80 treated with the MammoSite breast brachytherapy catheter to deliver APBI, Chao et al. identified young age as a potential risk for the development of an IBTR (17). More recently, Vicini et al. analyzed 1,449 cases also treated with the MammoSite to deliver APBI on the ASBrS MammoSite registry trial. With a median follow-up of 53 months, they found that estrogen receptor negative tumors (p = ) were associated with IBTR at 5 years in with invasive breast cancer and young age (p = ) and margin staus (p = ) in with DCIS. It is hoped that additional data such as these will become available from larger series of treated with APBI to help clarify these important associations. Until then, conservative selection criteria should be applied in identifying to be treated off-protocol while higher-risk (classified by the CP guidleines as unsuitable) should continue to be directed toward enrollment in phase III trials such as the NSABP B39/RTOG 0413 study. CONCLUSION The ASTRO CP guidelines for the use of APBI, when applied to a large database of treated with APBI, identified a select group of who are appropriately treated with APBI experiencing low rates of IBTR, but the guidelines were unable to independently differentiate cases that may be more appropriately treated with WBI. These results await validation from other large APBI databases and suggest that further refinements of these ASTRO CP groupings may be needed. These findings also support the continued enrollment of high-risk treated with APBI in Phase III clinical trials and emphasize the importance of the completion of these trials so that, in the future, treatment decisions can be made on existing outcome data. REFERENCES 1. Swanson TA, Vicini FA. Overview of accelerated partial breast irradiation. Curr Oncol Rep 2008;10: Smith BD, Arthur DW, Buchholz TA, et al. Accelerated partial breast irradiation consensus statement from the American Society for Radiation Oncology (ASTRO). J Am Coll Surg 2009; 209: Smith BD, Arthur DW, Buchholz TA, et al. Accelerated partial breast irradiation consensus statement from the American Society for Radiation Oncology (ASTRO). Int J Radiat Oncol Biol Phys 2009;74: Clarke M, Collins R, Darby S, et al. Effects of radiotherapy and of differences in the extent of surgery for early breast cancer on local recurrence and 15-year survival: An overview of the randomised trials. Lancet 2005;366: Antonucci JV, Wallace M, Goldstein NS, et al. Differences in patterns of failure in treated with accelerated partial breast irradiation vs. whole-breast irradiation: A matched-pair analysis with 10-year follow-up. Int J Radiat Oncol Biol Phys 2009;74: Vicini FA, Kestin L, Chen P, et al. Limited-field radiation therapy in the management of early-stage breast cancer. J Natl Cancer Inst 2003;95: Vicini FA, Chen P, Wallace M, et al. Interim cosmetic results and toxicity using 3D conformal external beam radiotherapy to deliver accelerated partial breast irradiation in with early-stage breast cancer treated with breast-conserving therapy. Int J Radiat Oncol Biol Phys 2007;69: Brown S, McLaughlin M, Pope K, et al. Initial radiation experience evaluating early tolerance and toxicities in undergoing accelerated partial breast irradiation using the Contura Multi-Lumen Balloon breast brachytherapy catheter. Brachytherapy 2009;8: Veronesi U, Orecchia R, Luini A, et al. Full-dose intraoperative radiotherapy with electrons during breast-conserving surgery: Experience with 590 cases. Ann Surg 2005;242: Vicini FA, Kestin LL, Goldstein NS. Defining the clinical target volume for with early-stage breast cancer treated with lumpectomy and accelerated partial breast irradiation: A pathologic analysis. Int J Radiat Oncol Biol Phys 2004;60: Arthur DW, Vicini FA, Kuske RR, et al. Accelerated partial breast irradiation: An updated report from the American Brachytherapy Society. Brachytherapy 2003;2: Patel RR, Christensen ME, Hodge CW, et al. Clinical outcome analysis in high-risk vs. low-risk eligible for national surgical adjuvant breast and bowel B-39/radiation therapy oncology group 0413 trial: Five-year results. Int J Radiat Oncol Biol Phys 2008;70:

8 984 I. J. Radiation Oncology d Biology d Physics Volume 79, Number 4, Khan AJ, Vicini F, Beitsch P, et al. Local control, toxicity, andcosmesisinwomenyoungerthan50enrolledontothe American Society of Breast Surgeons MammoSite Radiation Therapy System registry trial. Ann Surg Oncol 2009;16: Arthur D, Vicini F, Kuske RR, et al. Accelerated partial breast irradiation: An updated report from the American Brachytherapy Society. Brachytherapy 2002;1: Jeruss JS, Vicini FA, Beitsch PD, et al. Initial outcomes for treated on the American Society of Breast Surgeons MammoSite clinical trial for ductal carcinoma-in-situ of the breast. Ann Surg Oncol 2006;13: Sauer R, Sautter-Bihl ML, Budach W, et al. Accelerated partial breast irradiation: Consensus statement of 3 German oncology societies. Cancer 2007;110: Chao KK, Vicini FA, Wallace M, et al. Analysis of treatment efficacy, cosmesis, and toxicity Using the MammoSite Breast Brachytherapy Catheter to deliver accelerated partial-breast irradiation: The William Beaumont Hospital Experience. Int J Radiat Oncol Biol Phys 2007;69:32 40.