Search

Background: With the rise of genomic testing, more clinicians are using panels to understand the genetic profile of breast cancer to help aid in clinical management. However, little is known about the relationship between the results of genomic tests and the likelihood of identifying an underlying germline variant, and how this should integrate into clinical decision making. Methods: Data was obtained from the Informed Genetics Annotated Patient Registry (iGAP), an IRB-approved, multi-centered longitudinal registry designed to capture biomarker test results and their impact on treatment practices and outcomes. Two genomic tumor profiling tests were studied - MammaPrint recurrence risk and Blueprint molecular subtypes, including Luminal type A, Luminal type B, Basal, and HER 2 type. Of the 3400 patients currently enrolled in the registry, 528 have been diagnosed with breast cancer and underwent tumor profiling by both MammaPrint and BluePrint as well as germline genetic testing, including analyses of 24 cancer susceptibility genes (ATM, BARD1, BRCA1, BRCA2, BRIP1, CDH1, CHEK2, NBN, PALB2, PTEN, STK11, TP53, APC, BMPR1A, CDK4, EPCAM, MLH1, MSH2, MSH6, MUTYH, PMS2, RAD51C, RAD51D, SMAD4). Differences in positive germline variant rates were tested for with two-sided, Chi-Square tests using the prop.test function in R. Results: 231 (44.17%) were classified as High-Risk for recurrence on MammaPrint, with a 0.13 PVs detected per patient tested (positive germline variant (PV) Rate), 269 (51.34%) were identified as having a Low-Risk, with a 0.0849 PV rate, and 23 (4.4%) Ultra-Low-Risk, with a 0.0455 PV rate. There is not a significant difference between the High-Risk and Low-Risk for recurrence (p=0.09). 45 (8.54%) Basal molecular subtype identified by BluePrint panel, with a 0.1778 PV rate, 292 (55.41%) classified as Luminal A type, with a 0.0819 PV rate, 171 (32.45%) Luminal B with 0.1078 PV rate, and 13 (2.47%) HER2 Type with 0.07 PV rate. There was a significant difference between Basal and Luminal A PV rates (p=0.042), but no other statistically significant differences were found. Conclusions: Patients with a Basal molecular subtype have a significantly higher likelihood of having a germline pathogenic variant compared to Luminal A subtype. There was a trend that did not reach statistical significance for MammaPrint High Risk to have a higher likelihood of germline pathogenic result compared to MammaPrint Low Risk. This data adds another parameter for germline testing in those breast cancer patients who fall outside of current NCCN testing criteria. Citation Format: Peter Beitsch, Chloe Wernecke, Rakesh Patel, Barry Rosen, Eric Brown, Gia Compagnoni, Ian Grady, Lindsay Gold, Pat Whitworth, Linda Ann Smith, Mariusz Wirga, Richard Reitherman, Steven Cai, Toan Nguyen, Valerie Traina, Dennis Holmes, Paul Baron, Brittany Krautheim, Anne Peled, Walt Taylor, Kelly Bontempo, Brenna Bentley, Krista Ortega, Pouyan Ahmadi. Germline Testing Results in Patients with Genomic Tumor Profiling [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr P3-05-43.

Background: Genetic resources are underutilized when it comes to being incorporated into a breast cancer patient’s treatment, but that isn’t the only piece being overlooked. The Ki-67 proliferation index expressed (Ki-67%) is an established marker of tumor proliferation and aggressive behavior. We hypothesized that Ki-67% could have increased clinical utility when correlated with genomic testing results. Methods: Data was obtained from the Informed Genetics Annotated Patient Registry (iGAP), an IRB-approved, multi-center longitudinal registry designed to capture biomarker test results and their impact on treatment practices and outcomes. Tumor grades and Ki-67% were taken from patient pathology reports. The average Ki-67% was then calculated and compared for each tumor grade, MammaPrint genomic recurrence risk category (ultra low risk, low risk, and high risk), and Blueprint molecular subtype (Luminal type A, Luminal type B, Basal, and HER 2 type). ANOVA statistical analysis was performed for significance values. Results: Of 3102 patients enrolled in the iGAP Registry, 733 were diagnosed with breast cancer and had available tumor grade and Ki-67% data. Among these patients, 357 had genomic recurrence risk (MammaPrint) and 220 genomic molecular subtyping (BluePrint) reports. As expected, tumor grades were significantly positively correlated with Ki-67% (p< 0.0001 between all 3 tumor grade groups). Average Ki-67% in each genomic recurrence risk revealed a significant difference between Low Risk (14%, range 1-70%) and High Risk (36%, range 1-95%, p< 0.0001). Among the genomic molecular subtypes, there were significant differences in Ki-67% between Basal (avg 69%, 17-95%) and Luminal type A (avg 13%, 1-70%, p< 0.0001)) and all other subtypes, while Luminal type B (avg 24%, 1-80%) and HER 2 (avg 38%, 29-45%) were not significantly different from each other (p=0.2817), but still significantly different to all other subtypes. Conclusion: From these results, we can deduce that molecular subtype correlates with, but is clinically distinct from, Ki-67 proliferation index. These results also indicate that molecular subtype correlates with higher tumor grades, possibly due to increased cell proliferation. Achieving truly personalized clinical decision making requires utilizing multiple modalities and biomarkers, integrating the results into management. Citation Format: Chloe Wernecke, Krista Ortega, Kelly Bontempo, Brenna Bentley, Christina Hoyer-Kimura, Peter Beitsch, Rakesh Patel, Barry Rosen, Gia Compagnoni, Ian Grady, Eric Brown, Lindsay Gold, Pat Whitworth, Linda Ann Smith, Richard Reitherman, Mariusz Wirga, Steven Cai, Toan Nguyen, Valerie Traina, Dennis Holmes, Paul Baron, Brittany Krautheim, Anne Peled, Walt Taylor. Genomic testing and Ki-67 Percentage: Two puzzle pieces being undervalued in breast cancer treatment [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr P3-05-44.

BACKGROUND: It is known that breast cancer subtype (e.g., luminal vs. triple negative (TN)) can affect response to systemic therapy and prognosis; however, it is less well-understood whether these subtypes affect margin status and should therefore alter surgical management. METHODS: Data from two randomized trials evaluating cavity shave margins (CSM) on margin status in patients undergoing partial mastectomy (PM) were used for this analysis. The data were restricted to patients who had invasive carcinoma present in the PM specimen, and in whom data for all three receptors (ER, PR and HER-2) were known. Patients were classified as luminal if they were ER and/or PR+, HER-2 enriched if they were ER and PR negative but HER-2 positive, and TN if they were negative for all three receptors. We evaluated the impact of subtype on the margin status at the time the surgeon had completed their standard PM, prior to randomization to CSM vs. no CSM. Non-parametric statistical analyses were performed using SPSS Version 26. RESULTS: 350 patients were included in this cohort for analysis. The median patient age was 64 (range; 32-94 years) and the median invasive tumor size was 1.2 cm (range; 0.6-8.0 cm). 326 (93.1%) were luminal type, 22 (6.3%) were triple negative, and 2 (0.6%) were HER-2 enriched. Subtype was significantly correlated with race (black patients were more likely to have TN disease than white patients, 22.2% vs. 3.8%, p=0.001), palpability (TN tumors were more likely to be palpable than luminal cancers 54.5% vs. 29.8%, p=0.007) and grade (78.9% of TN cancers were high grade vs. 13.5% of luminal cancers p0.05 for all). While patients with TN and HER-2 enriched tumors were more likely to receive neoadjuvant therapy, this did not reach statistical significance (p=0.117). Surgeons were no more likely to take selective margins on the basis of molecular subtype (p=0.413). In this cohort, the overall positive margin rate was 33.7%. This did not vary based on molecular subtype (positive margin rate: 33.7% for patients with luminal tumors vs. 36.4% for those with TN tumors, p=0.425). On multivariate regression controlling for molecular subtype, race, grade and palpability, the only factor which predicted positive margin status was grade (p=0.005), with high grade tumors being significantly more likely to have a positive margin than low grade tumors, independent of other factors (OR=3.503, 95% CI: 1.638-7.494, p=0.001). CONCLUSION: While molecular subtype correlates with race, tumor grade and palpability, it does not predict margin status. Therefore, molecular subtype should not, independent of other factors, influence surgical decision-making. Citation Format: Andrew Fenton, Elisabeth Dupont, Theodore Tsangaris, Carlos Garcia-Cantu, Marissa Howard-McNatt, Akiko Chiba, Adam Berger, Edward Levine, Jennifer Gass, Kristalyn Gallagher, Sharon Lum, Ricardo Martinez, Alliric Willis, Sonali Pandya, Eric Brown, Amanda Mendiola, Mary Murray, Naveenraj Solomon, Maheswari Senthil, David Ollila, David Edmonson, Melissa Lazar, Jukes Namm, Fangyong Li, Meghan Butler, Noreen McGowan, Maria Herrera, Yoana Avitan, Brian Yoder, Laura Walters, Tara McPartland, Victor Haddad, Hongwei Ma, Ming Xie, Anees Chagpar. Does breast cancer subtype impact margin status in patients undergoing partial mastectomy? [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PS1-15.

Background: Racial/ethnic disparities have been well-documented in access to cancer screening and treatment, as well as treatment outcomes. Less is known regarding the yield of genetic pathogenic variants (PVs) in non-white populations. Methods: Patient data was obtained from the Informed Genetics Annotated Patient Registry (iGAP), an IRB-approved multi-center longitudinal, observational study, in which 2148 patients self-declared race/ethnicity and underwent germline genetic testing at any lab. Analyses were limited to 24 cancer susceptibility genes (ATM, BARD1, BRCA1, BRCA2, BRIP1, CDH1, CHEK2, NBN, PALB2, PTEN, STK11, TP53, APC, BMPR1A, CDK4, EPCAM, MLH1, MSH2, MSH6, MUTYH, PMS2, RAD51C, RAD51D, SMAD4), 21 of which have clinical management guidelines from the NCCN (excluding NBN, BARD1, CDK4).1 Descriptive statistics were used to assess and compare data from these populations and germline genetic testing results. Results: The Registry included 2148 patients, 1662 (77.37%) with a personal history and 1536 (71.51%) with a family history of cancer. The patients were 74.39% White, 6.33% Hispanic, 5.59% African/Black, 5.03% Asian, 1.63% Other, 1.35% Ashkenazi, and 5.68% Unknown. The overall germline PV rate in the cohort was 0.1089 PVs/patient tested, with 234 PVs detected in 227 patients. The PV rate among racial/ethnic groups were as follows: White 170/1598 (0.1064), Asian 8/108 (0.0741), Hispanic 27/136 (0.1985), African/Black 11/120 (0.0917), Ashkenazi 6/29 (0.2069). In patients self-reporting as Hispanic, the PV rate was similar to PV rate in those self-reporting as Ashkenazi, and significantly higher (p=0.00027) than PV rate in those of other self-reported race/ethnicity. Gene level PV rates are shown in Table 1. Conclusions: Those who reported being Hispanic had an increased overall PV rate. This could be due to the greater representation of Hispanics from New Mexico who may have Ashkenazi ethnicity. Further studies are needed to understand whether these differences are a result of disparate access to testing, true population differences, lack of data in non-White populations skewing variant classification or other factors. Table 1. Gene PV Rates by Racial/Ethnic Category. Citation Format: Peter Beitsch, Chloe Wernecke, Rakesh Patel, Barry Rosen, Gia Compagnoni, Ian Grady, Eric Brown, Lindsay Gold, Pat Whitworth, Linda Ann Smith, Mariusz Wirga, Richard Reitherman, Steven Cai, Toan Nguyen, Valerie Traina, Dennis Holmes, Paul Baron, Brittany Krautheim, Anne Peled, Walt Taylor, Kelly Bontempo, Brenna Bentley, Krista Ortega, Pouyan Ahmadi. Racial/Ethnic Groups Have Different Rates of Pathogenic Variants in Common Cancer Genes [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr P6-02-09.

Activating RAS pathway mutations are found in >30% of human tumors, and often associated with poor outcomes. While RAS-pathway targeted drugs have been approved, their activity as single-agents remains modest, supporting the need for rational targeted combinations to improve patient outcomes. Multiple studies have proposed a chemical synthetic lethality between Class I histone deacetylase inhibitors (HDACi), and RAS-pathway inhibitors in RAS-pathway mutated models. This synthetic lethality occurs due to a combined effect to inhibit double-stranded (ds) DNA repair and other survival pathways to drive apoptosis and tumor regressions. The use of HDACi in solid tumors and combinations has been challenged by poor potency, selectivity, safety and convenience, highlighting the need for a better HDACi in the clinic. OKI-179, a novel largazole derivative, is a potent, Class I-selective, oral HDACi that has completed Ph 1 clinical trials in patients with solid tumors. The clinical profile shows the potential to achieve exposure consistent with preclinical activity with strong pharmacodynamic activity at tolerated doses, supporting the development of OKI-179 in solid tumor combinations. Here we show synergistic activity of combining OKI-179 with the MEK inhibitor binimetinib (bini) or bini + RAF-inhibitor encorafenib (enco/bini) in both CDX and PDX models of NRAS-mutated melanoma and BRAF-mutated colorectal cancer (CRC). In vitro, combining OKI-179 with bini in SKMEL-2 NRASMT melanoma cells shows strong synergy, inducing a significant increase in cell death compared to single agents at clinically achievable drug concentrations. This activity was associated with a significant increase in gH2AX, a marker of ds DNA breaks, and apoptosis as measured by cleaved PARP. The activity of this combination was further investigated in vivo in established PDx models of NRAS-mutant melanoma, MEL278 and MM415. As single agents, both bini (3.5 mg/kg PO bid,21 d) and OKI-179 (80 mg/kg PO; 5/wk x 3 wk), doses that mimic the clinical exposure of these molecules, showed modest tumor growth inhibition (TGI). Combining bini and OKI-179 induced significantly greater TGI, including regression in 28-50% of the tumors, as compared to 0-6% for bini alone. A similar outcome was also observed in HT29 tumors (BRAFmt CRC) where the combination of OKI-179 and enco/bini showed regression in 50% of tumors, compared to 0% for OKI-179 or enco/bini alone. This supports the potential for additive efficacy of HDACi combined with RAS-pathway inhibitors across RAS-pathway mutations and tumor types. No additivity was observed in only one model (CRC563 - BRAFmt). These data show the potential for OKI-179 to synergize with RAS-pathway inhibitors across multiple indications. To investigate, a Phase1b/2 trial of OKI-179 with binimetinib in NRAS melanoma will initiate in early 2022. Citation Format: Richard Woessner, Patrice Lee, Eric Brown, Duncan H. Walker, James Winkler, Tony Piscopio. Theclass Iselective, oral HDAC inhibitor OKI-179 increases tumor regressions when combined with the MEK inhibitor binimetinib inmodels of NRAS melanoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr LB204.

This abstract was withdrawn by the authors.

Background: The COVID-19 pandemic rapidly altered health care worldwide. To save protective equipment and minimize exposures, many hospitals stopped some or all cancer surgery, leading oncologic providers to quickly adjust patient management. The goal of this study is to describe the breast cancer patient level changes which occurred during the initial months of COVID-19 in the United States. Methods: The American Society of Breast Surgeons developed a COVID-19 specific registry, within the established HIPPA compliant Mastery of Breast Surgery Program. Surgeons entered patient demographic data as well as their surgical and medical care (Neoadjuvant endocrine (NET) vs Neoadjuvant chemotherapy (NCT)). Data fields tracked whether decisions were usual for that practice, or modified due to COVID-19. Results: Between 3/1 and 6/17/2020, data from 1781 patients was entered by 154 surgeons. Mean age was 63, 78% Caucasian, 10% African American, 6% Hispanic; with geographic distribution ranging from 10.8% Northwest to 29.5% Northeast. Initial consultation took place in-person for 94.8% and only 5.2% (89) occurred via video/telephone. To date, just over 1% (14) of patients tested positive for COVID-19. Mean invasive tumor size was 21.2mm and 15.7% were node positive. Of 1445 invasive breast cancers 75% (1081) were ER +/HER2-, 13.5% (195) HER2+, 11.1% (160) triple negative (TNBC) (9/missing data). DCIS comprised 18.2% (325) of the cohort. Of 267 cases of ER+ DCIS, 49% (131) had primary surgery and 49% (130) received NET. The majority of NET use was due to COVID-19, 95% (124). Almost all (50/52) ER- DCIS underwent primary surgery (6/missing ER). Table 1 describes the management for the 1436 patients with invasive cancer with known biomarkers. NET due to COVID-19 was utilized in 45% (482), with only 5% (54) as part of usual practice. Increasing age was not a statistically significant factor in the use of NET (OR 0.99, 95% CI 0.97-1.01). In comparison to patients from the Northwest, patients from the Southwest and Northeast had the greatest use of NET(COVID-19) vs NET(usual) (ORs 14.4 and 4.6) Genomic assay testing was performed on the core biopsy in 216 patients, with 65% (141) due to COVID-19. Among the patients who had genomic testing due to COVID-19, 116 (82%) had NET, 18 (13%) had NCT, with the remaining having primary surgery. Of 472 patients treated with primary surgery for which the impact of COVID-19 was provided, surgery was delayed in 20% (96). Patients from the Northeast had a 2.1 x greater odds of having surgery delayed in comparison to those from the Midwest. Patients also experienced changes to their surgical plan with the most common changes being 6% (27) converting from mastectomy to breast conservation and 7% (34) from mastectomy with reconstruction to mastectomy without reconstruction. Conclusion: COVID-19 led to significant modifications in breast cancer treatment, including high rates of NET, genomic assay testing on core biopsies as well as delays in surgery; each of which were consistent with the prioritization and treatment recommendations from the COVID-19 Pandemic Breast Cancer Consortium. The majority of patients with TNBC and HER2+ disease received guideline concordant NCT. The ASBrS Mastery COVID-19 registry provides a snapshot into the rapid care changes caused by the pandemic, has ongoing data entry and analysis and will enable understanding of the impact on long term breast cancer outcomes. Table 1: Biomarker specific treatmentsER+/HER2- (1081)TNBC (160)HER2+/any ER (195)Mean Age656159African American (%)97(9%)21(13%)27(14%)Primary Surgery (usual)386 (36%)50 (31%)37 (19%)Primary Surgery (COVID-19)28 (2.6%)6 (3.7%)6 (3.1%)NET (usual)54 (5%)NA1(0.5%)NET (COVID-19)482 (45%)NA6 (3.6%)NCT (usual)90 (8.3%)98 (61%)137 (70%)NCT (COVID-19)39 (3.6%)6 (3.7%)8 (4.1%) Citation Format: Lee Gravatt Wilke, Toan T Nguyen, Qiuyu Yang, Bret M Hanlon, Kathryn A Wagner, Pamela Strickland, Eric A Brown, Jill Dietz, Judy C Boughey. Impact of the COVID-19 pandemic on the multidisciplinary management of breast cancer: Initial analysis of the american society of breast surgeons mastery COVID-19 registry [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr SS2-01.

INTRODUCTION: Routine resection of cavity shave margins has been shown in single center studies to result in a significant reduction in positive margin and re-excision rates. In this prospective multicenter randomized controlled trial, we sought to validate these findings across practice settings. METHODS: Nine centers across the United States, varying in practice setting and patient population, participated in this clinical trial of 398 stage 0-III breast cancer patients undergoing partial mastectomy (with or without resection of selective cavity margins). Participants were stratified by clinical stage and randomized 1:1 to either have routine cavity shave margins resected (“shave”) or not (“no shave”). Randomization group was revealed to the surgeon intraoperatively, after they had completed their standard partial mastectomy and were ready to close. Positive margins were defined as “tumor at ink” for invasive cancer or within 2 mm for ductal carcinoma in situ (DCIS). Adverse events were defined as seromas requiring percutaneous drainage, and/or hematomas or abscesses requiring operative intervention. RESULTS: Median patient age was 65 (range; 29-94). 116 patients had invasive disease, 74 had DCIS, 179 had both, and 29 had no residual cancer at the time of partial mastectomy. The median invasive cancer size was 1.2 cm (range; 0.05-8.00 cm); the median extent of DCIS was 0.9 cm (range; 0.05-6.40 cm). The “shave” and “no shave” groups were well matched at baseline for clinicopathologic and demographic factors. FactorShave (n=197)No Shave (n=201)p-valueAge (years); median (range)67 (36-94)64 (29-89)0.585Race 0.062-- White173 (87.8%)164 (81.6%) -- Black20 (10.2%)33 (16.4%) -- Asian2 (1.0%)2 (1.0%) -- Native American0 (0%)2 (1.0%) -- Unknown/Declined2 (1.0%)0 (0%) Hispanic ethnicity28 (14.2%)32 (15.9%)0.806Invasive tumor size (cm); median (range)1.30 (0.09-8.00)1.20 (0.05-7.50)0.282DCIS extent (cm); median (range)0.80 (0.10-6.40)1.00 (0.05-5.50)0.906Invasive histology 0.556-- Ductal177 (89.8%)186 (92.5%) -- Lobular16 (8.1%)13 (6.5%) -- Mucinous3 (1.5%)2 (1.0%) -- Other1 (0.5%)0 (0%) Neoadjuvant therapy15 (7.6%)19 (9.5%)0.592Palpable tumor57 (28.9%)56 (27.9%)0.825Node positive*24 (16.3%)16 (10.6%)0.175*Of the 298 patients who had lymph nodes evaluated Prior to randomization, positive margin rates were similar in the “shave” and “no shave” groups (38.1% vs. 37.3%, respectively, p=0.918). After randomization, however, those in the “shave” group were significantly less likely than those in the “no shave” group to have positive margins (8.6% vs. 37.3%, respectively, p CONCLUSION: Resection of cavity shave margins significantly reduces positive margin and re-excision rates in patients with stage 0-III breast cancer undergoing partial mastectomy. Citation Format: Chagpar AB, Tsangaris T, Garcia-Cantu C, Howard-McNatt M, Chiba A, Berger AC, Levine E, Gass JS, Gallagher K, Lum SS, Martinez RD, Willis AI, Pandya SV, Brown EA, Fenton A, Mendiola A, Murray M, Haddad V, Solomon NL, Senthil M, Bansil H, Ollila D, Snyder SK, Edmonson D, Lazar M, Namm JP, Li F, Butler M, McGowan NE, Herrera ME, Avitan YP, Yoder B, Dupont E. Does resection of cavity shave margins result in lower positive margin and re-excision rates in patients with stage 0-III breast cancer? Results from a prospective multicenter randomized controlled trial [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr PD8-07.

Abstract #5131 Purpose: Initial phase I/II results utilizing balloon brachytherapy to deliver accelerated partial breast irradiation (APBI) in only 2-days in patients with early stage breast cancer are presented. Materials and Methods: From 3/2004 – 8/2007, 45 patients received adjuvant radiation therapy after lumpectomy with balloon brachytherapy in a phase I/II trial delivering 2800 cGy in 4 fractions of 700 cGy to a depth of 1 cm from the applicator surface. Inclusion criteria for the protocol included: (1) patient age > 40 years, (2) pathological confirmation of stage 0/I/II breast carcinoma, (3) > 3 cm tumor size, (4) < 3 pathologically positive lymph nodes, and (5) negative margins (per NSABP criteria). Toxicities were evaluated using the NCI CTAE v3.0 scale and cosmesis (Harvard Criteria) was documented at > 6 months. Results: The median age was 66 years (48-83) and median skin spacing was 12 mm (8-24). The median follow-up was 11.4 months (5.4-48 mo) with 20 patients (44%) followed greater than one year, 11 (24%) greater than 2 years and 6 (13%) greater than 3 years. At < 6 mo (n=45), grade II toxicity rates were 9% radiation dermatitis (n=4), 13% breast pain (n=6), 2% edema (n=1), and 2% hyperpigmentation (n=1). Grade III breast pain was reported in 13% (n=6). At > 6 months (n=43), grade II radiation dermatitis, induration, and hypopigmentation were 2% (n=1), grade III breast pain was reported in 2% (n=1). Infection was 16% (n=7) at < 6 mo and 2% (n=1) at > 6 mo. Of these pts, only 4% (n=2) had a positive culture, whereas the others were treated empirically due to brisk erythema. Persistent seroma > 6 months was 30% (n=13). FN was seen in 9% (n=4), but only 1 pt required treatment. Rib fracture was seen in 5% (n=2). Cosmesis was good/excellent in 96% and fair in 4%. Conclusions: Treatment with balloon brachytherapy using a 2-day dose schedule resulted acceptable rates of grade II/III chronic toxicity rates and similar cosmetic results observed with standard 5-day APBI schedule. Citation Information: Cancer Res 2009;69(2 Suppl):Abstract nr 5131.