Accelerated Radiotherapy after Breast-Conserving Surgery for Early Stage Breast Cancer

Transcription

1 Technology Evaluation Center Accelerated Radiotherapy after Breast-Conserving Surgery for Early Stage Breast Cancer Assessment Program Volume 27, No. 6 February 2013 Executive Summary Background Radiation therapy is the standard care for patients undergoing breast-conserving surgery, as an individual-level meta-analysis based on multiple randomized, controlled trials shows that such therapy reduces recurrences and lengthens survival. The conventional radiation therapy regimen consists of about 25 treatments of 2 Gray (Gy; a measure of absorbed radiation dose), each delivered over 5 to 6 weeks. Not all patients undergo radiation therapy following breast-conserving surgery. The duration and logistics of treatment may be barriers to some women. Therefore, several accelerated radiotherapy approaches have been proposed to make the regimen less burdensome. n Accelerated (also called hypofractionated ) whole-breast irradiation reduces the number of fractions and the duration of treatment to about 3 weeks. This approach has been commonly used in Canada and Europe. n Accelerated partial-breast irradiation (APBI) irradiates a limited part of the breast in and close to the tumor cavity. By reducing the area irradiated, fewer treatments are needed and the total treatment takes about 1 week. Several approaches can be used to deliver APBI, including interstitial brachytherapy, balloon brachytherapy, external beam, or intraoperative (which occurs on only 1 day) radiation. Objective To evaluate the outcomes, particularly local recurrence, of 2 accelerated alternatives to conventional whole-breast radiotherapy after breast-conserving surgery (BCS): 1) accelerated, hypofractionated, whole-breast irradiation therapy; and 2) accelerated, partial-breast irradiation (APBI) therapy. Search Strategy To update the literature, a search was performed on MEDLINE and Embase in May 2012 using the following strategy: accelerated whole breast irradiation OR (hypofractionated AND (radiotherapy OR radiation therapy OR irradiation) AND breast). For accelerated partial-breast irradiation, searches were performed on MEDLINE and Embase in late April 2012, using the same terms listed in the literature search cited in the consensus statement of the American Society of Radiation Oncology (ASTRO). BlueCross BlueShield Association An Association of Independent Blue Cross and Blue Shield Plans Selection Criteria All comparative English-language articles in peer-reviewed journals on accelerated radiotherapy following breast-conserving surgery were included. Uncontrolled (i.e., single-arm) studies were excluded, because indirect comparisons in this setting are not reliable. Initial patient differences may account for perceived differences in treatment outcomes; they cannot be controlled for adequately to ensure the validity of the conclusions. Abstracts were excluded. NOTICE OF PURPOSE: TEC Assessments are scientific opinions, provided solely for informational purposes. TEC Assessments should not be construed to suggest that the Blue Cross Blue Shield Association, Kaiser Permanente Medical Care Program or the TEC Program recommends, advocates, requires, encourages, or discourages any particular treatment, procedure, or service; any particular course of treatment, procedure, or service; or the payment or non-payment of the technology or technologies evaluated Blue Cross and Blue Shield Association. Reproduction without prior authorization is prohibited. 1

2 Technology Evaluation Center Main Results Accelerated Whole-breast Irradiation. Four randomized, controlled trials compared accelerated whole-breast radiotherapy to 5-week whole-breast radiotherapy and 2 nonrandomized studies. Three of the 4 randomized trials are rated good quality using the modified U.S. Preventive Services Task Force criteria; the fourth trial was rated fair, and the nonrandomized studies, poor. Two of the studies are particularly useful, as they directly compared a 5-week to a 3-week regimen. They are both prospectively designed, noninferiority trials. Both trials accepted a maximum loss of efficacy of 5 percentage points in the accelerated group (one-sided α=0.025 or 0.05). Although the studies differ in the specific fractionation schedules and patient characteristics, they report no difference in local recurrence rates (i.e., recurrence of the cancer in the same breast) across treatment arms. One study from the U.K. includes women with grade 1 3 tumors. About 75% of the women had negative lymph nodes, and about 42% had a radiation boost to the tumor bed. Randomization was stratified for hospital, type of surgery (about 15% had a mastectomy), and plans for tumor bed boost. Systemic therapy, primarily tamoxifen, was used by some patients and appears to be fairly evenly distributed across treatment groups. The treatment arms compared a total dose of 40 Gy in 15 fractions over 3 weeks to 50 Gy in 25 fractions over 5 weeks. The hazard ratios for 40 Gy accelerated whole-breast radiotherapy versus conventional whole-breast radiotherapy were not statistically significant (using the log-rank test) for local or locoregional relapse. The absolute difference in locoregional relapse rates was 0.7% (95% CI: 1.7%, 0.9%). There were statistically significant differences in the 2 treatment regimens for distant relapse and overall survival, with relapse more frequent and survival longer for the 40 Gy accelerated whole-breast irradiation. This unexpected difference between treatment arms began to appear at about 1 year. The authors speculate that it could be due to chance and might change with longer follow-up. The 6-year follow-up period on this trial is too short to reach firm conclusions; follow-up continues. The second randomized, controlled trial from Canada compared accelerated whole-breast irradiation versus whole-breast irradiation. Of 2,429 eligible patients, 51% agreed to participate in the trial. Intention-to-treat analysis was used. The 10-year local recurrence was 6.2% for the 42.5 Gy arm accelerated whole-breast irradiation arm and 6.7% for the conventional 50 Gy whole-breast irradiation arm (absolute difference: 0.5%; 95% CI: 2.5%, 3.5%). Local recurrence rates with accelerated whole-breast radiotherapy were not worse than conventional whole-breast irradiation, when applying a noninferiority margin of 5%. In exploratory subgroup analyses, treatment effects were similar by age, tumor size, estrogen-receptor status, and chemotherapy use (48% had no systemic therapy). However, local recurrence at 10 years for patients with high-grade tumors was 4.7% for the conventional whole-breast irradiation arm and 15.6% for the 42.5 Gy accelerated whole-breast irradiation arm. The absolute difference equals 10.9 percentage points (95% CI: 19.1, 2.8, p=0.01). The patients in this trial all had invasive carcinoma of the breast with negative lymph nodes and surgical margins, and they did not have a radiotherapy boost to the tumor site. Exclusion criteria for the trial were invasive disease or ductal carcinoma in situ involving the margins of excision, tumors that were larger than 5 cm in diameter, and a breast width of more than 25 cm at the posterior border of the medial and lateral tangential beams, which could increase the heterogeneity of the radiation dose to the breast. In the trial, lymph node status was determined by axillary dissection, but recent reports suggest that sentinel lymph node biopsy is likely to be as effective. The overall body of evidence on accelerated whole-breast irradiation compared to conventional whole-breast irradiation suggests local recurrence rates with accelerated whole-breast radiotherapy were not worse than conventional whole-breast irradiation in patients meeting the criteria of the Canadian trial, when applying a noninferiority margin of 5%. Longer follow-up is needed for the U.K. trial. Accelerated Partial-breast Irradiation. There are 2 randomized trials on interstitial, accelerated, partial-breast irradiation compared to conventional whole-breast irradiation, a pragmatic randomized trial on intraoperative radiotherapy compared to whole-breast external radiotherapy, and Blue Cross and Blue Shield Association. Reproduction without prior authorization is prohibited.

3 Accelerated Radiotherapy after Breast-Conserving Surgery for Early Stage Breast Cancer 8 nonrandomized, comparative studies. Ten of the studies focused on interstitial brachytherapy; and one, on intraoperative radiotherapy. Quality of the 2 randomized trials on interstitial brachytherapy was rated as poor. For the first, accrual was stopped before reaching the goal specified to evaluate differences in local recurrence, to allow patients to participate in another trial. The randomization process was unclear; patients deemed technically unsuitable for interstitial brachytherapy were given external-beam APBI, and the patient characteristics and outcomes for each type of APBI were not reported separately. Finally, the sample size of 126 was small, and the longest follow-up reported was 66 months. Similar local and regional failure rates were found in the treatment arms. The other randomized, controlled trial on APBI was reported in 1990 and 1993, and many changes in the care of breast cancer have occurred since then. The study was rated as poor because nodal status was based on clinical exam, among other factors. Recurrence was higher for the limited field treatment arm (analogous to partial-breast irradiation) than for the wide field arm (analogous to whole-breast irradiation), but some of the excess recurrences in the limited field arm were axillary. This may be accounted for by the fact that the axillary area was included in the wide field radiotherapy, but not in the limited field, and the initial work-up for nodal involvement was limited. The follow-up was 65 months, and the sample size 708. The fair quality randomized trial compared intraoperative to external-beam accelerated partialbreast irradiation. It is a noninferiority trial with 28 centers in 9 countries and a sample size of 2,232. An intention-to-treat (ITT) approach was used; 89% of the intraoperative group and 92% of the external radiotherapy group completed treatment. Patients were not blinded to treatment choice. As anticipated in advance, 14% of those in the intraoperative arm received external-beam radiotherapy as well, because of unfavorable pathologic features determined after surgery, e.g., lobular carcinoma. The predefined noninferiority margin was an absolute difference of 2.5% between groups for pathologically confirmed, ipsilateral local recurrence. After 4 years, wound seroma needing more than 3 aspirations was significantly more common in the intraoperative group than in the external radiotherapy group (2.1% vs. 0.8%, respectively; p=0.012). Conversely, Radiation Therapy Oncology Group (RTOG) toxicity grade of 3 or 4 was more common in the external radiotherapy group than in the intraoperative group (2.1% vs. 0.5%, respectively; p=0.002). The 4-year local recurrence rates in the ipsilateral breast were 1.20% (95% CI: 0.53%, 2.71%) in the intraoperative radiotherapy arm vs. 0.95% (95% CI: 0.39%, 2.31%) in the external radiotherapy arm (difference between groups=0.25%, 95% CI: -1.04%, 1.54%; log-rank test, p=0.41). Local recurrence rates after 4 years with intraoperative radiotherapy were not worse than with external irradiation, when applying a noninferiority margin of 2.5%. Fortunately for the patients, the recurrence rates are low: 6 in the intraoperative group versus 5 in the external radiotherapy group. But these small numbers make it more difficult to detect real differences between arms, if they exist. A number of reviews and editorials discussed the preliminary results of the TARGIT-A trial. While recognizing the potential benefits of intraoperative radiotherapy (IORT), including convenience, excellent delineation of the tumour bed under visual control, very good dose homogeneity, and high sparing of normal tissue, a number of concerns have been expressed. They include the following: n If IORT is performed during the surgery to excise the tumor, the definitive pathology is not available when the radiotherapy is performed. Therefore, a subset of patients must undergo whole-breast external beam radiotherapy as well following surgery. The article reports that 14% of patients in the IORT arm also received whole-breast external-beam radiotherapy. When only those who received IORT during initial surgery are considered, 21% received whole-breast external-beam radiotherapy. There are limited data suggesting that breast symptoms and pain following treatment may be greater for patients receiving both IORT and whole-breast external radiotherapy compared with IORT alone and that patients satisfaction is greater for wholebreast external radiotherapy or IORT alone compared to the combined treatment. n Is the radiation dose and type actually equivalent to the standard radiation therapy regimen? Of particular concern is the rapid drop in dose with distance from the applicator and whether any residual disease will eradiated. Some argue that the TARGIT-A trial alleviates this concern, while others do not Blue Cross and Blue Shield Association. Reproduction without prior authorization is prohibited. 3

4 Technology Evaluation Center n The length of follow-up is insufficient to determine long-term toxicity and efficacy, particularly since only 19% of the participants in the TARGIT-A trial completed 4 years of follow-up. The median follow-up is not reported but appears to be around 2 years. While the INTRABEAM device used in this trial is subject to U.S. Food and Drug Administration (FDA) regulation, it does not fall under the regulatory purview of the U.S. Nuclear Regulatory Commission. In some states, the participation of radiation oncologists in delivering radiation is not required. The other 8 nonrandomized, comparative studies were all rated as poor, due to potential baseline differences in treatment groups, lack of multivariable analyses to account for them, inclusion of patients who did not meet eligibility criteria, variations in treatment within arms, and generally small sample sizes and insufficient follow-up. Overall, the body of evidence on interstitial and intraoperative APBI compared to conventional whole-breast irradiation is insufficient, and there are no comparative studies for external-beam or balloon APBI. Author s Conclusions and Comments Accelerated Whole-breast Irradiation. The evidence suggests that accelerated whole-breast irradiation provides a reasonable, shorter alternative to conventional whole-breast irradiation among women who had breast-conserving surgery and are at low risk of recurrence. In a Canadian trial, the 10-year local recurrence was 6.2% for the 42.5 Gy accelerated whole-breast irradiation arm and 6.7% for the conventional whole-breast irradiation arm (absolute difference: 0.5%, 95% CI: 2.5%, 3.5%). Accelerated whole-breast irradiation cuts the radiotherapy treatment time by about 40% without any apparent negative consequences after 10 years of follow-up. The results should be interpreted in the context of a noninferiority design with a margin of 5%. Patient selection is key, and at this point, only patients similar to those in the Canadian trial should be considered for this therapy. Outcomes could vary in women with other disease characteristics. The patients in this trial all had invasive carcinoma of the breast with negative lymph nodes and surgical margins. Exclusion criteria for the trial included invasive disease or ductal carcinoma in situ involving the margins of excision, tumors that were larger than 5 cm in diameter, and a breast width of more than 25 cm at the posterior border of the medial and lateral tangential beams, which could increase the heterogeneity of the radiation dose to the breast. In the trial, lymph node status was determined by axillary dissection, but recent reports suggest that sentinel lymph node biopsy is likely to be as effective. Forty-one percent of the women took tamoxifen, despite the fact that 71% had estrogen-receptor positive tumors. While the current evidence on this radiotherapy regimen is convincing, it is not as compelling as that for conventional whole-breast irradiation. Additional randomized, controlled trials or longer follow-up of the existing trials could uncover additional concerns. Some potential adverse events, such as cardiac ischemia, may take longer to become evident. This regimen has been widely used outside the U.S. without substantial reports of major adverse events. Potential patients should be carefully selected and should make a decision with their physicians based on full information, while the results of longer follow-up for the START B trial are awaited. Accelerated Partial-breast Irradiation. The data on APBI compared to whole-breast irradiation are insufficient to draw any conclusions about the relative effectiveness of these modalities. Furthermore, it is becoming increasingly clear that each type of APBI should be judged on its own merits, and studies comparing different APBI techniques to each other as well as to whole-breast irradiation are needed. Fortunately, a number of large randomized, controlled trials are underway. Given the current level of evidence, it is important for patients to be well-informed about the uncertainty regarding the outcomes of this approach. This information should include failure rates for the specific devices (e.g., explantation for MammoSite, incomplete expansion of the catheters for some of the hybrid devices), as well as the uncertainty regarding their comparative effectiveness. Accelerated whole-breast irradiation may present an intermediate alternative for women who Blue Cross and Blue Shield Association. Reproduction without prior authorization is prohibited.

5 Accelerated Radiotherapy after Breast-Conserving Surgery for Early Stage Breast Cancer meet the criteria for the Canadian trial, and the critical importance of completing radiotherapy for the majority of patients undergoing breast-conserving surgery deserves emphasis. In a 2009 review of the APBI trials currently underway, Mannino and Yarnold (note Yarnold is a lead author on the START A and B trials) raise several concerns regarding variations across the trials. The extent of the initial breast-conserving surgery can vary substantially across studies, as well as the definition of the targeted tumor cavity. A larger margin is usually drawn around the tumor cavity for 3D conformal radiotherapy, for example, because of the need to allow for variations in set-up and respiration motion. Studies of APBI usually distinguish between same site relapse, i.e., close to the irradiated area and elsewhere relapse, yet it is unclear whether what constitutes the same site varies across studies. The percentage of relapses occurring elsewhere in the ipsilateral breast in studies of whole-breast radiotherapy following breast-conserving surgery range from 18% to 42% (these studies may include some patients at higher risk of recurrence). Proponents of APBI have sometimes asserted that elsewhere tumors are rare, that they are mostly new primary tumors (rather than a recurrence), or that earlier studies have shown that radiotherapy is not effective on these tumors in any case. Mannino and Yarnold challenge each of these points in turn, although they also conclude that the results of the trials currently underway will provide level 1 evidence for or against APBI. Based on the available evidence, the Blue Cross and Blue Shield Association Medical Advisory Panel made the following judgments about whether accelerated radiotherapy meets the Blue Cross and Blue Shield Association Technology Evaluation Center (TEC) criteria to decrease recurrence and reduce treatment sequelae after breast-conserving surgery for early stage breast cancer. 1. The technology must have final approval from the appropriate governmental regulatory bodies. The various radiotherapy modalities presented in this Assessment have been approved or cleared for marketing by the U.S. Food and Drug Administration (FDA). The brachytherapy and intraoperative devices, however, contain a black box warning required by the FDA indicating that The safety and effectiveness of the [device] as a replacement for whole-breast irradiation in the treatment of breast cancer has not been established. 2. The scientific evidence must permit conclusions concerning the effect of the technology on health outcomes. Accelerated Whole-breast Irradiation. Two randomized, controlled trials rated as good quality with over 3,000 participants compared accelerated and conventional whole-breast irradiation. Local recurrence rates with accelerated whole-breast radiotherapy were not worse than conventional whole breast irradiation, when applying a noninferiority margin of 5%. Follow-up for one study was 6 years, which is insufficient; results of longer follow-up are being collected. The follow-up period for the other study, however, was 12 years, with 10-year recurrence rates reported. This time period is sufficient to gauge the relative outcomes of these two treatment approaches. However, these results only apply to women similar to those in the Canadian trial, who had invasive carcinoma of the breast with negative lymph nodes and surgical margins. Exclusion criteria for the trial included invasive disease or ductal carcinoma in situ involving the margins of excision, tumors that were larger than 5 cm in diameter, and a breast width of more than 25 cm at the posterior border of the medial and lateral tangential beams, which could increase the heterogeneity of the radiation dose to the breast. In the trial, lymph node status was determined by axillary dissection, but recent reports suggest that sentinel lymph node biopsy is likely to be as effective. Accelerated Partial-breast Irradiation. Two poor quality randomized trials; one fair quality randomized trial; and 8 poor quality nonrandomized, comparative studies on APBI compared to conventional whole-breast radiotherapy were found. Ten of the studies focused on interstitial brachytherapy; and one, on intraoperative radiotherapy. They are insufficient to permit conclusions concerning the effect of interstitial or intraoperative brachytherapy on health outcomes Blue Cross and Blue Shield Association. Reproduction without prior authorization is prohibited. 5

6 Technology Evaluation Center No comparative studies were found on external-beam or balloon APBI, and thus the evidence on these modalities is insufficient as well. 3. The technology must improve the net health outcome. Accelerated Whole-breast Irradiation. Indirect evidence is available on the impact of accelerated whole-breast irradiation on the net health outcome. A number of trials and individual-level meta-analyses have demonstrated that the use of radiation therapy after breast-conserving therapy reduces recurrence and improves survival. In the Canadian trial, local recurrence rates with accelerated whole-breast radiotherapy were not worse than conventional whole-breast irradiation, when applying a noninferiority margin of 5%. Therefore, the available evidence suggests that accelerated whole-breast irradiation also improves the net health outcome compared to no radiotherapy for women who meet the criteria for Canadian clinical trial (i.e., invasive carcinoma of the breast with negative lymph nodes and surgical margins). Accelerated Partial-breast Irradiation. Since available evidence is insufficient to permit conclusions regarding the equivalence of APBI and conventional whole-breast irradiation and since APBI has not and cannot ethically be compared to using no radiation therapy after breastconserving surgery, it cannot be determined whether accelerated radiotherapy improves the net health outcome. 4. The technology must be as beneficial as any established alternatives. Accelerated Whole-breast Irradiation. Evidence is available that accelerated whole-breast irradiation is not worse than the longer whole-breast regimen, when applying a noninferiority margin of 5%. With treatment duration of 3 weeks rather than 5 weeks, it is more convenient and less taxing for patients and may improve adherence with the standard of care that women undergo radiation therapy following breast-conserving surgery. Accelerated Partial-breast Irradiation. Since available evidence is insufficient to permit conclusions, it cannot be determined whether accelerated partial-breast irradiation is as beneficial as whole-breast, standard radiotherapy after breast-conserving surgery for early stage breast cancer. Longer follow-up ideally in randomized controlled trials, several of which are underway for APBI is needed to ascertain whether accelerated radiotherapy is as beneficial as the current standard of care. 5. The improvement must be attainable outside the investigational settings. Accelerated Whole-breast Irradiation. The 2 major studies on accelerated whole-breast irradiation were large, multicenter trials conducted at 10 sites in Canada and 23 sites in the U.K. These results should be attainable outside investigational settings. Accelerated Partial-breast Irradiation. Whether accelerated partial-breast radiotherapy improves health outcomes after breast-conserving surgery for early stage breast cancer has not been demonstrated in the investigational setting. Based on the above, accelerated whole-breast radiotherapy after breast-conserving surgery for early stage breast cancer meets the TEC criteria for women who meet the entry criteria for the Canadian trial (i.e., invasive carcinoma of the breast with negative lymph nodes and technically clear surgical margins; exclusion criteria for the trial included invasive disease or ductal carcinoma in situ involving the margins of excision, tumors that were larger than 5 cm in diameter, and a breast width of more than 25 cm at the posterior border of the medial and lateral tangential beams, which could increase the heterogeneity of the radiation dose to the breast ). Accelerated partial-breast radiotherapy after breast-conserving surgery for early stage breast cancer does not meet the TEC criteria Blue Cross and Blue Shield Association. Reproduction without prior authorization is prohibited.

7 Accelerated Radiotherapy after Breast-Conserving Surgery for Early Stage Breast Cancer Contents Assessment Objective 8 Background 8 Methods 13 Part I: Accelerated Whole-breast Irradiation Formulation of the Assessment 15 Review of Evidence 15 Discussion and Conclusions 49 Summary of Application of the 50 Technology Evaluation Criteria References 53 Appendix 57 Part II: Accelerated Partial-breast Irradiation Formulation of the Assessment 30 Review of Evidence 31 Published in cooperation with Kaiser Foundation Health Plan and Southern California Permanente Medical Group. TEC Staff Contributors Author Barbara Mauger, Ph.D.; TEC Executive Director Naomi Aronson, Ph.D.; TEC Director, Technology Assessments Mark D. Grant, M.D., M.P.H.; Director, Clinical Science Services Kathleen M. Ziegler, Pharm.D.; Research/Editorial Staff Claudia J. Bonnell, B.S.N., M.L.S.; Kimberly L. Hines, M.S. Blue Cross and Blue Shield Association Medical Advisory Panel Allan M. Korn, M.D., F.A.C.P. Chairman, Senior Vice President, Clinical Affairs/Medical Director, Blue Cross and Blue Shield Association; Steven N. Goodman, M.D., M.H.S., Ph.D. Scientific Advisor, Dean for Clinical and Translational Research, Stanford University School of Medicine, Professor, Departments of Medicine, Health Research and Policy; Mark A. Hlatky, M.D. Scientific Advisor, Professor of Health Research and Policy and of Medicine (Cardiovascular Medicine), Stanford University School of Medicine. Panel Members Peter C. Albertsen, M.D., Professor, Chief of Urology, and Residency Program Director, University of Connecticut Health Center; Sarah T. Corley, M.D., F.A.C.P., Chief Medical Officer, NexGen Healthcare Information Systems, Inc. American College of Physicians Appointee; Helen Darling, M.A., President, National Business Group on Health; Josef E. Fischer, M.D., F.A.C.S., William V. McDermott Professor of Surgery, Harvard Medical School American College of Surgeons Appointee; I. Craig Henderson, M.D., Adjunct Professor of Medicine, University of California, San Francisco; Jo Carol Hiatt, M.D., M.B.A., F.A.C.S., Chair, Inter-Regional New Technology Committee, Kaiser Permanente; Saira A. Jan, M.S., Pharm.D., Associate Clinical Professor, Ernest Mario School of Pharmacy, Rutgers, The State University of New Jersey, Residency Director and Director of Clinical Programs Pharmacy Management, Horizon Blue Cross and Blue Shield of New Jersey; Thomas Kowalski, R.Ph., Clinical Pharmacy Director, Blue Cross Blue Shield of Massachusetts; Bernard Lo, M.D., Professor of Medicine and Director, Program in Medical Ethics, University of California, San Francisco; Randall E. Marcus, M.D., Charles H. Herndon Professor and Chairman, Department of Orthopaedic Surgery, Case Western Reserve University School of Medicine; Barbara J. McNeil, M.D., Ph.D., Ridley Watts Professor and Head of Health Care Policy, Harvard Medical School, Professor of Radiology, Brigham and Women s Hospital; William R. Phillips, M.D., M.P.H., Clinical Professor of Family Medicine, University of Washington American Academy of Family Physicians Appointee; Richard Rainey, M.D., Medical Director, Regence BlueShield of Idaho; Rita F. Redberg, M.D., M.Sc., F.A.C.C., Professor of Medicine and Director, Women s Cardiovascular Services, University of California San Francisco; Alan B. Rosenberg, M.D., Vice President, Medical Policy, Technology Assessment and Credentialing Programs, WellPoint, Inc.; Maren T. Scheuner, M.D., M.P.H., F.A.C.M.G., Clinical Genetics and Principal Investigator, Health Services Genomics Program, VA Greater Los Angeles Healthcare System; Health Sciences Associate Clinical Professor, Department of Medicine, David Geffen School of Medicine at UCLA; Natural Scientist, RAND Corporation; J. Sanford Schwartz, M.D., F.A.C.P., Leon Hess Professor of Medicine and Health Management & Economics, School of Medicine and The Wharton School, University of Pennsylvania; Earl P. Steinberg, M.D., M.P.P., Executive Vice President, Innovation and Dissemination & Chief, Geisinger Healthcare Solutions Enterprise. CONFIDENTIAL: This document contains proprietary information that is intended solely for Blue Cross and Blue Shield Plans and other subscribers to the TEC Program. The contents of this document are not to be provided in any manner to any other parties without the express written consent of the Blue Cross and Blue Shield Association Blue Cross and Blue Shield Association. Reproduction without prior authorization is prohibited. 7

8 Technology Evaluation Center Assessment Objective This Assessment evaluates the outcomes of two accelerated alternatives to conventional whole-breast radiotherapy after breast-conserving surgery: 1) accelerated, hypofractionated whole-breast irradiation therapy and 2) accelerated partial-breast irradiation (APBI) therapy. Background Breast Conservation Therapy Survival after breast-conservation therapy is equivalent to survival after mastectomy for patients diagnosed with tumors categorized as stage I or II. Breast-conservation therapy is a multimodality treatment that consists of breastconserving surgery to excise the tumor with adequate margins, followed by whole-breast external-beam radiation therapy administered as 5 daily fractions per week over 5 to 6 weeks. Local boost irradiation to the tumor bed often is added to whole-breast irradiation to provide a higher dose of radiation at the site where recurrence most frequently occurs. For some patients, breast-conservation therapy also includes axillary lymph node dissection, sentinel lymph node biopsy, or irradiation of the axilla. A number of randomized, controlled trials have demonstrated that the addition of radiotherapy after breast-conserving surgery reduces recurrences and mortality. In an individual-level meta-analysis, the Early Breast Cancer Trialists Collaborative Group (EBCTCG 2005) reported 5-year risk of local recurrence of 7.3% among patients who received breast-conserving surgery and who were allocated to radiotherapy versus 25.9% among those not (p< ; n=7,311), for an absolute difference of 18.6 percentage points. Among women with node-negative disease, the 5-year risk for those not assigned to radiotherapy was 22.9% versus 6.7% for those receiving radiotherapy (n=6,097). The corresponding percentages for node-positive disease are 41.1% versus 11.0% (n=1,214), respectively, for those without or with radiotherapy. While initial results did not find any effect of whole-breast irradiation on survival, longer follow-up has shown that postsurgical radiotherapy increases survival, as well as reducing the risk of local recurrence. The Early Breast Cancer Trialists Collaborative Group (2005) found that whole-breast irradiation reduced the 15-year breast cancer mortality risk from 35.9% to 30.5% (RR breast cancer death: 0.83; 95% CI: 0.75, 0.91; p=0.0002; n=7,311); there was a similar reduction in mortality from all causes (5.3%, SE: 1.8, p=0.005). For women with nodenegative disease, the 15-year mortality rate for those assigned to radiotherapy was 26.1% versus 31.2% for those not assigned to radiotherapy (p=0.006); the corresponding numbers for patients with node-positive disease was 47.9% versus 55.0%, respectively (p=0.01). Consequently, radiation therapy is generally recommended following breast-conserving surgery. A potential exception is for older women at low risk of recurrence. For example, the National Comprehensive Cancer Network (NCCN) guidelines (2012) state that women aged 70 years or older may omit radiotherapy if they have estrogen-receptor (ER) positive, T1 tumor; clinically negative lymph nodes; and plan to take adjuvant endocrine therapy. An article on use of lumpectomy and tamoxifen with or without whole-breast radiotherapy in this same population found that the addition of radiation had no impact on rates of mastectomy for local recurrence, distant metastases, or 5-year rates of overall survival (Hughes et al. 2004). However, the rate of local or regional recurrence at 5 years was 1% in the group receiving tamoxifen and radiation and 4% in the group taking tamoxifen alone (p<0.001). The authors concluded that Lumpectomy plus adjuvant therapy with tamoxifen alone is a realistic choice for the treatment of women 70 years of age or older who have early, estrogenreceptor-positive breast cancer. A recent study of Surveillance, Epidemiology, and End Results (SEER) and Medicare data examined outcomes among 7,403 women aged years with ER-positive breast cancer, invasive tumors smaller than 2 cm, and negative lymph nodes who had conservative breast surgery between 1992 and 2002 (Albert et al. 2012). At a median follow-up of 7.3 years, the risk of subsequent mastectomy within 10 years was 3.2% (95% CI: 2.6%, 3.9%) for patients who received radiotherapy versus 6.3% (95% CI: 4.6%, 8.6%) for those who did not. After adjustment (apparently for type of lymph node assessment; use of chemotherapy; age at diagnosis; year of diagnosis; race; SEER registry; Charlson comorbidity score; and tumor size, grade, and histology), the hazard ratio was 0.33 (95% CI: 0.22, 0.48; p<0.001) for those receiving radiotherapy. The authors suggest that these findings are useful as a reflection of patterns in routine practice Blue Cross and Blue Shield Association. Reproduction without prior authorization is prohibited.

9 Accelerated Radiotherapy after Breast-Conserving Surgery for Early Stage Breast Cancer The primary outcome considered in this Assessment, and in many of the studies of post-breast-conserving-surgery radiotherapy, is ipsilateral breast tumor recurrence (IBTR). Although other outcomes, such as survival and adverse events, are also important, IBTR is an informative and pragmatic measure for the effectiveness of therapy. This outcome is also important to the patient, because recurrence leads to further treatment, often mastectomy, and predicts an increased risk of mortality. In addition, the duration of follow-up needed to measure ipsilateral breast recurrence is substantially shorter than for mortality. The EBCTCG individual-level meta-analysis provides data on the pattern of recurrences among patients undergoing breast-conserving surgery with and without radiotherapy for over 10 years. In web figure 6a ( uk/~ebctcg/local2000/annex.pdf), data on the percentage of patients per year with isolated local recurrence is 2.7% and 9.0% for those with and without whole-breast irradiation, respectively, in year 0; 1.5% and 4.8% in years 1 2; 1.1% and 3.7% in years 3 4; 0.8% and 1.9% in years 5 9; and 0.2% for both groups in years 10 and higher (trend χ 2 =4.3, p=0.04). 1 While the recurrence rate falls over time, it persists for more than 10 years. In the Owen et al. (2006) study described below, local recurrence was highest between years 3 to 5 of follow-up, and 67% of events took place in the first 5 years. However, that means that 33% of events took place after 5 years. For APBI, recurrence away from the tumor site is of greatest interest, because that is the area that does not receive radiation therapy. Unfortunately, few articles report ipsilateral breast recurrence by year of follow-up and location near or away from the tumor site. One study (Recht et al. 1988) reported frequencies separately for ipsilateral breast recurrences classified as true relapses or marginal misses (TR/MM), and those that occurred elsewhere in the same breast that permit comparison for recurrences up to 5 years with those after 5 years. While elsewhere recurrences were relatively infrequent (total of 4%), three-fourths of these occurred after 5 years. Additional data on patterns of recurrences, by year of followup and location in the breast, can be found in the prior TEC Assessment on APBI (2007; Vol. 22, No. 4). Another study (Gujral et al. 2011) retrospectively reviewed data on 1,410 women who had been enrolled in a prospective randomized, controlled trial of hypofractionated radiotherapy (Yarnold et al. 2005, described below). At a median follow-up of 10.1 years, there were 150 recurrences: 109 were in the same quadrant as the primary tumor, 20 of which were histologically different; and 31 were in a different quadrant, 7 of which had the same histology as the primary tumor (the quadrant was unknown in 10 cases). The cumulative incidence rates of local recurrence of the primary tumor at 5, 10, and 15 years were 6.7, 8.9, and 11.8; the corresponding data for new primary tumors was 0.8, 2.0, and 3.5, all respectively. It has also been noted that the use of tamoxifen may reduce or delay recurrences (Sautter-Bihl et al. 2010). The Early Breast Cancer Trialists Collaborative Group published another individual-level metaanalysis in 2011, which focuses only on women with breast-conserving therapy (the 2005 study reported on women with mastectomies as well, although those data are not reported above). For women with node-negative disease (n=7,287), radiotherapy halved the annual recurrence rate after 10 years, from 31.0% for those not receiving radiotherapy to 15.6%. It also reduced the 15-year risk of breast cancer death from 20.5% to 17.2% (p=0.005). The rate ratio for any first recurrence among women treated with radiotherapy versus those who were not was 0.45 (SE 0.04) during years 0-4 after treatment; 0.52 (SE 0.08) for years 5-9; and, it is 1.25 (0.23) for years 10+. (Webfigure 1b) For women with node-positive disease (n=1,050), radiotherapy reduced the 1-year recurrence risk from 26.0% to 5.1%. Radiotherapy also reduced the 15-year risk of breast cancer death from 42.8% to 51.3% (p=0.01). The rate ratio for any first recurrence for those treated with radiotherapy versus those who were not was 0.44 (SE 0.07) for years 0-4 after treatment; 0.76 (SE 0.20) for years 5-9; and 1.18 (0.35) for years 10+. (reference Webfigure 1c) The authors state in the text of the article, There was a moderate additional effect over the next few years but little further effect after year 5, for women with nodepositive disease. Although no similar comment is made regarding women with node-negative disease, the rate ratios reported in the appendix and above suggest that the gap in first recurrences between APBI and whole-breast irradiation does not continue to widen after 5 years. Although these are the strongest evidence available on the impact of radiotherapy versus no radiotherapy on recurrences among women with early breast cancer, they include 2013 Blue Cross and Blue Shield Association. Reproduction without prior authorization is prohibited. 9

10 Technology Evaluation Center both recurrences close to the tumor site (which both APBI and whole breast irradiation would irradiate) and recurrences in other parts of the breast (which only whole breast irradiation would include). In a Cancer Management Controversy Article published in the American Journal of Clinical Oncology, Ciervide and Formenti (2012) support the need for longer follow-up and continued investigation of APBI in the context of prospective clinical trials. They note earlier examples of changes in breast cancer therapy that were adopted on the basis of shorter follow-up, whereas longer follow-up indicated no difference between the new and older approach. They also refer to data from a retrospective study of 1,990 cases between 1970 and 1998 using breast-conserving surgery and radiation at a single institution (Freedman et al. 2005). Freedman et al. compared ipsilateral breast tumor recurrences to cancers in the contralateral breast, hypothesizing that if there is a protective effect of whole-breast radiation away from the vicinity of the primary tumor, then differences should become apparent with longterm follow-up between the rates of elsewhere recurrence and contralateral breast cancer. They found that recurrences elsewhere in the irradiated breast were rare at 5 (1%) and 10 (2%) years, but increased to 6% (95% CI: 3 9%) at 15 years. This 15-year rate of elsewhere recurrence was half the rate of contralateral breast cancers of 13% (95% CI: 10 16%). This analysis is based on the assumption that the contralateral breast serves as a proxy for the nonirradiated portion of the ipsilateral breast and that radiotherapy in the ipsilateral breasts has an effect on recurrences years later. It is based on treatment from 1970 to 1998 in a group of women that includes some who would not be recommended to undergo APBI (e.g., age younger than 40; Smith et al. 2009). There is also evidence to suggest that current treatment regimens have lowered recurrence rates (see, for example, Vaidya et al. 2010). In addition to ipsilateral breast tumor recurrences, there are other potential long-term effects from the radiation treatment itself. For example, there is evidence of cardiac toxicity among breast cancer patients treated from the early 1970s through the early 1990s (Darby et al. 2005, 2010). However, those effects usually take 10 to 20 years to become manifest. It has been stated that various forms of APBI reduce the radiation dose to the heart and lungs, but only long-term follow-up will demonstrate whether this is in fact the case in clinical practice. Other approaches to reducing heart exposure, such as prone versus supine positioning, are also being studied (Formenti et al. 2012). The controversy continues on the length of follow-up needed to determine whether APBI is equivalent to whole-breast irradiation; some 10-year data are already available on accelerated whole-breast irradiation. However, the issue may be resolved by statistical issues rather than biological ones. Because recurrences are relatively rare among low-risk early breast cancer patients, it may take considerable time for there to be enough recurrences to achieve sufficient power to compare rates for each radiotherapy approach. For example, in the large NSABP-39/RTOG 0413 trial comparing whole-breast irradiation versus APBI, the enrollment goal is 4,300, which may be achieved around the end of The length of the trial (presumably barring early termination) is determined by the occurrence of a prespecified number (175) of in-breast recurrences. The researchers expect that reaching that number of recurrences will take about 10 years. Most patients diagnosed with stage I or II breast cancer now are offered a choice of breast-conserving therapy or modified radical mastectomy, but breast-conserving therapy is selected less frequently than expected. Studies have shown that those living furthest from treatment facilities are least likely to select breast-conserving therapy instead of mastectomy and most likely to forgo radiation therapy after breast-conserving surgery (Farrow et al. 1992; Athas et al. 2000; Nattinger et al. 2001). A study using data from the National Cancer Institute s Surveillance, Epidemiology, and End Results (SEER) tumor registries from 1992 to 2002 examined how many women with early stage (I or II) breast cancer received radiotherapy within 4 months following breast-conserving surgery (Du and Gor 2007). After adjusting for age, they found that in 2002, 30.8% of Caucasian women and 44.7% of African- American women had not received radiotherapy. Furthermore, these rates had increased from 24.7% for Caucasians and 34.0% for African Americans in A more recent study (Parise et al. 2012) examined patterns of breast cancer cases using data from the California Cancer Registry for cases diagnosed between 2000 and They found that women aged 46 to 69 (versus those younger or older) were Blue Cross and Blue Shield Association. Reproduction without prior authorization is prohibited.

11 Accelerated Radiotherapy after Breast-Conserving Surgery for Early Stage Breast Cancer most likely to receive radiotherapy after breast conserving surgery. African Americans and Hispanics were less likely to receive radiotherapy than white women only in the Los Angeles region. Lower socioeconomic status was also associated with less radiotherapy in Los Angeles, while the trend was the opposite in the North region (north of San Francisco and Sacramento). A study (Dragun et al. 2011) using data from the Kentucky Cancer Registry on women undergoing breast-conserving surgery between 1998 and 2007 found that less use of subsequent radiotherapy was associated with age older than 70 years, being from rural Appalachia, being African American, and being uninsured. Given that duration and logistics appear to be barriers to completion of treatment, there has been interest in developing shorter radiotherapy regimens. Two approaches have been explored. The first method is to provide the same dose to the whole breast in a shorter time by increasing the dose provided per treatment (hypofractionation). This approach was initially avoided out of concern that increasing doses to target the tumor more effectively might induce more severe adverse events from radiation exposure, thus, tipping the balance between benefits and harms. More recent research, some of which is highlighted below, has allayed some of these concerns. Accelerated wholebreast irradiation has been used especially in Canada and Europe. The second approach to reducing radiotherapy treatment time is accelerated partial-breast irradiation. It differs from conventional wholebreast irradiation in several ways. First, the radiation targets only the segment of the breast surrounding the area where the tumor was removed, rather than the entire breast. Second, the duration of treatment is 4 to 5 days rather than 5 to 6 weeks, because the radiation is delivered in fewer fractions at larger doses per fraction to the tumor bed. Third, the radiation dose is intrinsically less uniform within the target volume when APBI uses brachytherapy (i.e., the implantation of radioactive material directly in the breast tissue). The major types of radiotherapy used after breast-conserving surgery are outlined in Table 1. They differ in their techniques, instrumentation, dose delivery, and possibly in their outcomes. The Table 1. Major Types of Radiation Therapy Following Breast-conserving Surgery Radiation Therapy Type Accelerated? Whole (W) or Partial (P) Breast External-beam (E) or Brachytherapy (B)? Approximate Duration of Treatment Published RCTs (length of followup in years) Conventional whole-breast irradiation Accelerated whole-breast irradiation N W E 5 6 weeks Multiple; >15 years Y W E 3 weeks 4; 10 years Interstitial APBI* Y P B 1 week 2; 5.4 years Balloon APBI** Y P B 1 week 0 External-beam APBI*** Intraoperative APBI**** Y P E 1 week 0 Y P Not applicable 1 day 1; 4 years * Interstitial brachytherapy entails placement of multiple hollow needles and catheters to guide placement of the radioactive material by a remote afterloading device. It is more difficult to perform than other types of brachytherapy and has a steep learning curve. ** Balloon brachytherapy, e.g., MammoSite, entails inserting a balloon into the tumor bed, inflating the balloon, confirming its position radiographically, and then using a remote afterloader to irradiate the targeted area. Some brachytherapy systems combine aspects of interstitial and balloon brachytherapy. *** External-beam APBI is delivered in the same way as conventional or accelerated whole-breast radiotherapy but to a smaller area. All 3 external-beam regimens can use 3-dimensional, conformal radiation therapy (3D-CRT) or intensity-modulated radiation therapy (IMRT). ****Intraoperative APBI is performed during breast-conserving surgery, when a single dose of radiation is delivered to the exposed tumor bed. The authors reported 4-year outcomes, but fewer than 20% of participants had longer follow-up; other authors estimate that the median followup was about 2 years Blue Cross and Blue Shield Association. Reproduction without prior authorization is prohibited. 11

12 Technology Evaluation Center use of bead or seed implants is also being researched (e.g., Pignol et al. 2009). To appreciate the differences among radiotherapy techniques, it is useful to understand attributes of radiation delivery. The goals of cancer radiotherapy are usually to provide the tumor or tumor bed with a high dose of homogeneous radiation (e.g., all parts of the tumor cavity receive close to the targeted dose). Areas adjacent to the tumor may be treated with a lower dose of radiation (e.g., with whole-breast irradiation, to treat any unobserved cancerous lesions). Radiation outside the treatment area should be minimal or nonexistent. The goal is to target the tumor or adjacent areas at risk of harboring unseen cancer with an optimum dose, while avoiding healthy tissues. Relevant considerations in selecting a radiotherapy regimen for breast cancer following breastconserving surgery are as follows. 1. Is an optimal dose delivered to all of the tumor bed? Doses that are too low may elevate the risk of recurrence, while doses that are too high can increase toxic effects. 2. Should the rest of the breast, which may harbor unseen cancers, be treated? Not treating these cancers with radiotherapy may increase the number of recurrences, while not irradiating healthy tissue reduces toxic effects, improves cosmesis, and may even reduce the risk of subsequent cancers from the radiation itself. 3. How can the healthy tissues adjacent to the cancer best be protected, especially critical organs such as the lungs and heart (particularly for left-breast tumors)? Exposing these areas to higher levels of radiation may increase the risk of long-term complications. Patient preferences may also play a role in treatment decisions. In a qualitative study of breast cancer patient s experience in making radiation therapy treatment decisions in Australia, Halkett et al. (2005) reported that some women found it difficult to maintain the routine of daily treatment, in terms of having the treatment and experiencing the adverse effects. Hoopes et al. (2012) surveyed 5,000 randomly selected women undergoing mammography (with and without cancer) and 2,150 physicians who were members of the American Society for Radiation Oncology. Thirty-six percent of the women undergoing mammography and 17% of the physicians returned usable responses. The majority of patients (62%) preferred hypofractionated whole-breast irradiation, 28% would have chosen partial-breast irradiation, and 10%, conventional whole-breast irradiation. In contrast, 82% of physicians reported using conventional whole-breast irradiation for more than two-thirds of their patients and 56% never use hypofractionated whole-breast irradiation. Within APBI choices, 62% of women preferred APBI using 3D-CRT, while 38% favored brachytherapy. Thirty-six percent of physicians offer 3D-CRT, while 66% provide brachytherapy. These results deserve further investigation, especially among patients facing the treatment decision. Guidelines and Recommendations According to the NCCN guidelines (2012), patients should be encouraged to undergo partial-breast irradiation as part of a clinical trial. For patients not eligible for inclusion in a trial, partial-breast irradiation should only be performed in accordance with the suitable group in the American Society for Radiation Oncology (ASTRO) consensus statement. For whole-breast radiotherapy, NCCN recommends either a conventional whole-breast irradiation regimen or a total dose of 42.5 Gy with 2.66 Gy per fraction, which equals 16 fractions. Although the NCCN guidelines do not specify the duration of treatment, the latter is presumably an accelerated whole-breast irradiation regimen. A boost to the tumor bed is recommended for higher-risk whole-breast radiotherapy patients, i.e., those who are younger than 50 years old and have positive axillary nodes, lymphovascular invasion, or close margins. The American Society of Breast Surgeons and ASTRO have issued guidelines for the selection of patients for APBI, which are summarized in Table 2. The ASTRO guidelines (Smith et al. 2009) were spurred by in part by the increasing use of APBI outside of clinical trials. Recent studies have confirmed this trend. One analysis of data from the Surveillance, Epidemiology, and End Results (SEER) database found that of the women with early stage breast cancer who underwent breast-conserving surgery followed by radiotherapy, the use of implantable APBI increased from 0.4% in 2000 to 6.8% in 2007 (p<0.001). This increase was particular notable among women aged 70 to 79. Among Medicare patients aged 67 years or older, Smith et al. (2012) found that the use of brachytherapy rose from 3.5% in to 12.5% in Blue Cross and Blue Shield Association. Reproduction without prior authorization is prohibited.

13 Accelerated Radiotherapy after Breast-Conserving Surgery for Early Stage Breast Cancer (p<0.001 for trend). Another analysis (Husain et al. 2011) of SEER data from 2002 to 2007 concluded that 38.1% of breast brachytherapy cases were suitable, according to the ASTRO guidelines, 43.5% of the cases were cautionary, and 18.3% of the cases were unsuitable. During the period, the percentage of unsuitable cases declined, while the percentage of cautionary cases rose (p< for both trends). ASTRO released guidelines on fractionation for whole-breast irradiation in (Smith et al. 2010) They rely on 4 trials described below. They conclude that Data are sufficient to support the use of HF-WBI [hypofractionated or accelerated, whole breast irradiation] for patients with early breast cancer who meet all of the aforementioned criteria, including aged 50 years or older, disease Stage pt1-2 pn0, no chemotherapy, and treatment with radiation dose homogeneity within + 7% in the central axis plane. The task force did not agree on whether it is recommended to use HF-WBI when receiving a tumor boost. FDA Status. The newer radiotherapy modalities presented in this Assessment have been approved or cleared for marketing by the U.S. Food and Drug Administration (FDA; see Appendix Table A). All brachytherapy devices have been approved through the 510(k) process and are either balloon brachytherapy or hybrid balloon-interstitial brachytherapy devices. The FDA has required a black box warning on each stating that The safety and effectiveness of the [brachytherapy device] as a replacement for whole breast irradiation in the treatment of breast cancer has not been established. Between 1 January 2009 and 1 June 2012 (last update scheduled 4/30/12), 5 malfunctions and 2 injuries related to the SAVI were reported to the MAUDE (Manufacturer and User Facility Device Experience; gov/scripts/cdrh/cfdocs/cfmaude/search.cfm) adverse event list. The malfunctions and one of the injuries were due to problems with removal or catheter rupture or use. One injury was reported by a patient, and other than having the device reportedly implanted for 13 days (versus the usual 5), the cause was not clear. The MAUDE data are generally not believed to be complete, since they rely on voluntary reports. Also, the context is not clear (e.g., reasons for difficulties in inserting radioactive material, which resolve when the device was moved). The connection between injuries and a device may not be clear in some cases. During the same time period, there were 29 MAUDE reports on MammoSite. Four of the reports appear to be duplicates, and 2 are apparently unrelated. Of the remaining 23, 12 were classed as injuries, 10 reported on malfunctions, and 1 was classified as other, although these categories are not used uniformly. The most common malfunction was rupture (7 cases) or a hole (1 case) in the MammoSite balloon. There were several cases each of fluid leaking from the balloon or problems in inflating or deflating the balloon. A few of these cases may have been associated with higher than planned radiation doses. The injuries reported included seromas, infections, cellulitis, abscess, opening of incision years later, and radioactive burn. Some of these are reported as adverse events associated with use of MammoSite, e.g., seromas. One case of angiosarcoma was reported 4 years after treatment with the MammoSite device. There were 32 MAUDE reports to date on the Contura Multi-Lumen Balloon Source Applicator for Brachytherapy device, 1 on the Axxent balloon brachytherapy device, and none on the Adjustable Multi-Catheter Source Applicator or ClearPath device. Methods Search Methods To update the literature, a search was performed on MEDLINE and Embase in May 2012 using the following strategy: accelerated whole breast irradiation OR (hypofractionated AND (radiotherapy OR radiation therapy OR irradiation) AND breast). For accelerated partial-breast irradiation, searches were performed on MEDLINE and Embase in late April 2012 using the same terms listed in the 2007 TEC Assessment and the literature search cited in the consensus statement of the American Society of Radiation Oncology (Smith et al. 2009). Selection Criteria All comparative English-language articles in peer-reviewed journals on accelerated radiotherapy following breast-conserving surgery were included. Uncontrolled (i.e., single-arm) studies were excluded, because indirect comparisons in this setting are not reliable. Initial patient differences may account for any perceived differences in treatments and cannot be controlled for adequately to ensure the validity of the conclusions. Abstracts were excluded, 2013 Blue Cross and Blue Shield Association. Reproduction without prior authorization is prohibited. 13

14 Table 2. Professional Medical Society Criteria for Performing APBI Blue Cross and Blue Shield Association. Reproduction without prior authorization is prohibited. ASTRO Criteria: ASTRO Criteria: ASTRO Criteria: Categories Factor Suitable Cautionary Unsuitable Criteria for APBI from American Society of Breast Surgeons Patient factors Age >60 years years <50 years 45 years for invasive cancer; >50 years for DCIS BRCA1/2 mutation Not present NR Present NR Pathologic factors Tumor size 2 cm cm >3 cm 3 cm T stage T1 T0 or T2 T3-4 NR Margins Negative >2 mm Close (<2 mm) Positive Microscopically negative Grade Any NR NR NR LVSI No Limited/focal Extensive NR ER status Positive Negative* NR NR Multicentricity Unicentric NR Present NR Multifocality Clinically unifocal, total Clinically unifocal, total Clinically multifocal or NR size 2.0 cm size= cm microscopically multifocal, total size 3 cm Histology Invasive ductal or other favorable subtypes Invasive lobular NR Invasive ductal carcinoma or DCIS Pure DCIS Not allowed 3 cm >3 cm 3 cm EIC Not allowed 3 cm >3 cm NR Associated LCIS Allowed NR NR NR Nodal factors N stage pn0 (i, i + ) NR pn1, pn2, pn3 SN pn0 Nodal surgery SN Bx, ALND NR None performed Technology Evaluation Center Treatment factors Neoadjuvant therapy Not allowed NR If used NR NR=not reported *Strongly encouraged to enroll in NSABP B-39/RTOG trial. Abbreviations: ALND: axillary lymph node dissection; DCIS: ductal carcinoma in situ; EIC: extensive intraductal component; ER status: estrogen receptor status; LCIS: lobular carcinoma in situ; LVSI: lymphovascular space invasion; SN Bx: sentinel node biopsy; N stage: nodal stage; T stage: tumor stage Sources: Smith et al [ASTRO criteria; see also Prosnitz et al. 2009]; American Society of Breast Surgeons 2011.

15 Accelerated Radiotherapy after Breast-Conserving Surgery for Early Stage Breast Cancer unless copies of the slide presentation used at a national professional meeting could be obtained or they provided follow-up data for a previously published comparative article that described the study in detail. Medical Advisory Panel Review This Assessment was reviewed by the Blue Cross and Blue Shield Association Medical Advisory Panel (MAP) on June 12, To maintain the timeliness of the Assessment s scientific information, literature search updates were performed subsequent to the Panel s review (see Search Methods ). If the search updates identified any additional studies that met the criteria for detailed review, the results of these studies were included in the tables and text where appropriate. Part I: Accelerated Whole-breast Irradiation Health Outcomes Benefits. The potential health benefits of accelerated whole-breast irradiation compared to whole-breast irradiation include a reduced number of treatments and a shorter duration of total treatment. These advantages may encourage adherence with the standard of care that women undergo radiation therapy following breast-conserving surgery, Harms. The potential harms associated with accelerated whole-breast irradiation compared to whole-breast irradiation are primarily increased toxicity from the higher doses of radiotherapy per treatment and a possibility of greater recurrence. Specific Assessment Question For women undergoing breast-conserving therapy, does accelerated whole-breast irradiation improve the net health outcome by at least as much as whole-breast irradiation? Formulation of the Assessment Patient Indications Relevant patients are those who select breastconserving surgery and radiation therapy for initial treatment of stage I or II breast cancer and are at low risk of local and regional recurrence. Technologies to be Compared Conventional Whole-breast External-beam Radiotherapy Following Breast-conserving Surgery. It can be delivered in the form of photons (megavoltage X-rays or g rays from a cobalt-60 source) or high-energy electrons (6 15 MeV) from a linear accelerator. It is usually given in fractions of Gy per day, 5 days weekly, with a total whole-breast irradiation dose of Gy and a boost dose of Gy when included. Accelerated, Hypofractionated Whole-breast External-beam Radiotherapy Following Breast-conserving Therapy. The precise specifications vary, but it is generally given in fractions of Gy over the course of 3 weeks with a total whole-breast irradiation dose of Gy and a boost dose of Gy when included. Review of Evidence One means of accelerating radiotherapy following breast-conserving surgery is to increase the dose per treatment and reduce the total number of treatments. This option has been explored primarily in Canada and Europe. The challenge has been to identify the radiation dose per fraction that balances the impact of radiation on tumor control versus the rate of late adverse events (see, for example, Yarnold et al. 2005; START 2008a). The initial concern with breast radiotherapy was that treating a patient with more than 2 Gy per fraction (i.e., treatment episode) would lead to a steeper rise in late adverse events than in tumor control, thereby worsening outcomes. Some cancers, such as squamous cell carcinomas of the head and neck, are subject to this constraint, while higher doses per fraction may be tolerated for other cancers. The studies evaluating accelerated wholebreast radiotherapy have focused both on identifying the optimal dose per fraction and on comparing an accelerated whole-breast regimen to the traditional regimen. As a result, some studies have compared the use of fewer, higher dose fractions to the traditional regimen while keeping the duration of treatment constant, e.g., twenty-five 2-Gy fractions over 5 weeks versus thirteen 3-Gy fractions over 5 weeks. These studies demonstrate the 2013 Blue Cross and Blue Shield Association. Reproduction without prior authorization is prohibited. 15

16 Technology Evaluation Center impact of a larger fraction, while controlling for the duration of treatment. Others also reduce the duration of treatment for the accelerated regimen while increasing the dose per treatment, i.e., fifteen 2.67-Gy fractions over 3 weeks. While both types of studies are important, the latter are more directly applicable to assessing the use of an accelerated wholebreast regimen in clinical practice. Four randomized trials of accelerated wholebreast radiotherapy versus whole-breast irradiation have been published. Patients and health care providers were not blinded to treatment allocation, unless otherwise indicated. Two articles report on one trial (Yarnold et al. 2005; Owen et al. 2006). Two other trials (START 2008a, b) were part of the same larger effort in the U.K., and their design was based in part of the preliminary results of the first trial. Also, the tumor control data from the Yarnold et al. (2005) and Owen et al. (2006) study were incorporated into parts of the START A trial. The fourth trial was conducted in Canada (Whelan et al. 2002, 2010). The Canadian study and one of the British studies (START B) are particularly useful, as they directly compare a 5-week to a 3-week regimen. They are both prospectively designed noninferiority trials. Both trials were planned to accept a maximum loss of efficacy of 5 percentage points in the accelerated group (one-sided α=0.025 or 0.05).The key features and summary results of all of these studies, as well as a quasi-experimental comparative study conducted in 1999 (Yamada et al. 1999), are reported in Tables 3 and 4. Whelan et al. (2002, 2010) conducted a randomized trial of accelerated whole-breast irradiation versus whole-breast irradiation. Of 2,429 eligible patients, 51% agreed to participate in the trial. During the study, 20 of 1,234 patients did not complete the assigned treatment, but intention-to-treat analysis was used. The 5-year local recurrence rates were 2.8% (calculated by subtracting local recurrence-free survival reported in paper from 100%) in the 42.5 Gy accelerated whole-breast irradiation arm versus 3.2% in the conventional whole-breast irradiation arm (difference: 0.4%; 95% CI: -1.5%, 2.4%). The local recurrence rate with 42.5 Gy accelerated whole-breast radiotherapy was not worse than conventional whole-breast irradiation, when applying a noninferiority margin of 5%. Data are presented on 5-year local recurrence by age, tumor size, and adjuvant systemic therapy. Results were reported by treatment group according to stratification factors. The largest absolute difference in recurrence between treatment arms was 3.6% (95% CI: -1.7 to 9.0) for patients younger than 50 years old (accelerated whole-breast irradiation, 3.6%; whole-breast irradiation, 7.2%). No statistically significant differences between treatment arms were found for disease-free survival (p=0.37) or overall survival (p=0.78). No statistically significant differences in cosmetic outcomes or adverse events were reported. An update of this study was published in 2010 (Whelan et al. 2010). The results reported on a median 144 months of follow-up. The 10-year local recurrence was 6.2% for the 42.5-Gy arm accelerated whole-breast irradiation arm and 6.7% for the conventional whole-breast irradiation (absolute difference: 0.5%; 95% CI: 2.5%, 3.5%). The percentage of patients with a good or excellent cosmetic outcome was 70% for accelerated whole-breast irradiation and 71% for whole-breast irradiation (absolute difference: 1.5%, 95% CI: 6.9%, 9.8%). At 10 years, 8.9% of accelerated whole-breast irradiation and 7.7% of whole-breast irradiation patients had Grade 2 or 3 late skin toxicity; 11.9% and 10.4%, respectively, had Grade 2 or 3 late subcutaneous tissue toxicity, although the sample size declined from 1,220 at baseline to 451 at 10 years. In an exploratory subgroup analyses, treatment effects were similar by age, tumor size, estrogen-receptor status, and chemotherapy use. However, local recurrence at 10 years for patients with high-grade tumors was 4.7% for the conventional whole-breast irradiation arm and 15.6% for the 42.5 Gy accelerated whole-breast irradiation arm. The absolute difference equals 10.9 percentage points (95% CI: 19.1, 2.8, p=0.01). Strong points of the study design include randomization; histopathologic confirmation of local recurrence and, if possible, first recurrence at other sites; use of standardized instruments (European Organisation for Research and Treatment of Cancer [EORTC] Cosmetic Rating System and Radiation Therapy Oncology Group [RTOG]/EORTC late radiation morbidity scale); description of sample size calculations; and small loss to follow-up (<5%). Using the modified U.S. Preventive Services Task Force criteria for rating study quality (Harris et al. 2001), the Whelan et al. (2002, 2010) trial is assigned a good rating. As for generalizability, the patients in this trial all had invasive carcinoma of the breast with negative lymph nodes Blue Cross and Blue Shield Association. Reproduction without prior authorization is prohibited.

17 2013 Blue Cross and Blue Shield Association. Reproduction without prior authorization is prohibited. 17 Table 3. Key Features, Controlled Studies on Accelerated Whole-breast Irradiation in Patients Undergoing BCT for Breast Cancer Citation, Study Design (Modified USPSTF Rating) Surgical Methods Radiation Therapy Methods Other Treatments Patient Characteristics Whelan et al. 2002, 2010 Multicenter (10 sites), randomized trial (Good) Breast-conserving surgery Accelerated WB 42.5: 42.5 Gy in 16 fractions over 22 days WB: 50 Gy in 25 fractions over 35 days No boost, no RT to lymph nodes Supine position Yarnold et al 2005; Breast-conserving Accelerated WB 39: 39 Gy in Owen et al surgery 13 fractions over 5 wks Multicenter (2 sites), Accelerated WB 42.9: 42.9 Gy in randomized trial 13 fractions over 5 wks (Fair) WB: 50 Gy in 25 fractions over 5 wks 14-Gy tumor bed boost originally assigned by randomization; later, all received. Use of boost equally distributed across treatment groups: 1,051 received boost; 359, no boost Supine position Systemic (same % % 60 yr: Accelerated in each arm): None, WB 42.5, 45, WB, 51; 48%; tamoxifen, 41%; 19 20% tumors >2 cm chemotherapy, 11% (100% <5 cm); 73 74% Grade I, II; 71% ER+; all pn0, negative margins Adjuvant tamoxifen, 64% Mean age, 54.5 yrs Adjuvant chemotherapy, (SD 9.7); 93.9% ct1 13.9% or ct2; 67.3%, node No adjuvant systemic negative (of those with therapy, 20.5% axillary staging), 24.1%, 1-3 positive nodes Axillary treatment: Surgery, 60.1% Radiotherapy, 20.6% Characteristics Used for Matching or Stratification Age (<50 or 50), tumor size ( 2 cm or >2 cm), adjuvant systemic therapy (tamoxifen, any chemotherapy, or none), center Treatment center, presence of microscopic foci, 3 mm surgical margin Accelerated Radiotherapy after Breast-Conserving Surgery for Early Stage Breast Cancer

18 Blue Cross and Blue Shield Association. Reproduction without prior authorization is prohibited. Table 3. Key Features, Controlled Studies on Accelerated Whole-breast Irradiation in Patients Undergoing BCT for Breast Cancer (cont d) Citation, Study Design (Modified USPSTF Rating) Surgical Methods Radiation Therapy Methods Other Treatments Patient Characteristics START A (2008) Multicenter (17 sites), randomized trial (Good) Breast-conserving surgery ( %) or mastectomy Accelerated WB 39: 39 Gy in 13 fractions over 5 wks Accelerated WB 41.6: 41.6 Gy in 13 fractions over 5 wks WB: 50 Gy in 25 fractions over 5 wks 14% had RT to lymph nodes, decided before randomization Some received 10-Gy boost using electrons Supine position By treatment regimen By treatment regimen (accelerated WB 39, (accelerated WB 39, accelerated WB 41.6, WB) accelerated WB 41.6, WB) Anthracycline-based Median age: 57.1, 57.0, 57.6 chemotx: 71.1%, 70.1%, 68.7% % negative nodes: 67.4, 71.5, 68.6 % tamoxifen/no chemotx: 51.0, 55.7, 55.5 % tumor grade 1: 20.2, 20.0, 21.0 % tamoxifen + chemotx: 25.5, 25.0, 23.1 % tumor grade 2: 49.9, 50.5, 49.3 % chemotx/no tamoxifen: % tumor grade 3: 11.1, 10.3, , 27.6, 28.3 % no adjuvant therapy: % no adjuvant therapy: 9.1, 7.1, , 7.1, 6.9 Lymphatic treatment: % boost BCS pts only: % surgery/no RT: 60.5, 61.0, , 84.8, 81.4 About 78% pts had invasive % surgery + RT: ductal cancer 12.9, 12.7, 15.9 Characteristics Used for Matching or Stratification Hospital, type of surgery, and intention to use boost Technology Evaluation Center

19 2013 Blue Cross and Blue Shield Association. Reproduction without prior authorization is prohibited. 19 Table 3. Key Features, Controlled Studies on Accelerated Whole-breast Irradiation in Patients Undergoing BCT for Breast Cancer (cont d) Citation, Study Design (Modified USPSTF Rating) Surgical Methods Radiation Therapy Methods Other Treatments Patient Characteristics START B Multicenter (23 sites), randomized trial (Good) Breast-conserving surgery (accelerated WB 40, WB; 91.7%, 92.3%) or mastectomy Accelerated WB 40: 40 Gy in 15 fractions over 3 wks WB: 50 Gy in 25 fractions over 5 wks Prone position 99.0% compliance with allocated treatment By treatment regimen (accelerated WB 40, WB) Systemic therapy: % tamoxifen/no chemotx: 73.0, 70.8 % tamoxifen + chemotx: 14.0, 16.4 % chemotx/no tamoxifen: 7.0, 7.0 % no adjuvant therapy: 4.2, 3.3 Lymphatic treatment: % surgery/no RT: 88.6, 88.7 % surgery + RT: 7.1, 6.7 By treatment regimen (accelerated WB 40, WB) Median age: 57.8, 57.0 % negative nodes: 72.4, 75.2 % tumor grade 1: 28.0, 27.7 % tumor grade 2: 47.9, 46.9 % tumor grade 3: 22.3, 23.6 % boost BCS pts only: 43.8, 41.4 % invasive ductal cancer: 75.9, 78.3 Characteristics Used for Matching or Stratification Hospital, type of surgery (BCS or mastectomy), intention to use boost or not Accelerated Radiotherapy after Breast-Conserving Surgery for Early Stage Breast Cancer

20 Blue Cross and Blue Shield Association. Reproduction without prior authorization is prohibited. Table 3. Key Features, Controlled Studies on Accelerated Whole-breast Irradiation in Patients Undergoing BCT for Breast Cancer (cont d) Citation, Study Design (Modified USPSTF Rating) Surgical Methods Radiation Therapy Methods Other Treatments Patient Characteristics Yamada et al Quasi-experimental (Poor) Breast-conserving surgery Accelerated WB 40: 40 Gy in 16 fractions in just over 3 wks WB: 50 Gy in 25 fractions over 5 wks No boost Pts with 3+ positive lymph nodes had regional radiotherapy to ipsilateral axilla and supraclavicular lymph node bed 100% completed prescribed therapy Chemo for node-positive, Matched group: premenopausal Age <40, 14.4% Tumor <2 cm, 55.1% Tamoxifen for Receptor-positive, 61.9% node-positive, Grade: well-differentiated, post-menopausal 5.9% Grade: poorly differentiated, 39.8% Infiltrating ductal, 90.1% Extensive intraductal component, 7.6% Positive axillary node, 30% Lymphvascular space invasion, 57.6% Positive surgical margin, 1.7% Cytotoxic chemotherapy, 7% Hormonal therapy, 33% Characteristics Used for Matching or Stratification Age, tumor size, histology, lymphovascular space invasion, extensive intraductal component, lymph node status, final surgical margins, chemotherapy, hormonal therapy Technology Evaluation Center