- Email: [email protected]
Brain Metastases After Breast-conserving Therapy for Early-stage Breast Cancer: Incidence and Characteristics by Subtype
S188
International Journal of Radiation Oncology Biology Physics
guidelines. Fifteen percent of all plans failed the constraint that 90% of the Lumpectomy PTV Eval received 90% of the boost dose, primarily due to inadequate deep coverage of the expansion volume. All other target and OAR constraints were met “Per Protocol” or “Variation Acceptable” in 95% or more of the plans. Conclusions: Overall, our historical plans showed poor compliance with RTOG 1005 dosimetric guidelines. Although they failed to consistently provide complete coverage to the Breast PTV Eval and Lumpectomy PTV Eval, they efficiently limited the dose delivered to OARs. Our original breast fields were set up using classical tangent field design, which historically results in relatively low rates of recurrence and toxicity. It may require long-term results of RTOG 1005 to assess whether adherence to contour and planning guidelines results in meaningful clinical improvement over historical controls. Author Disclosure: E. Kane: None. K. Langen: E. Research Grant; Accuray Inc, Decimal Inc, Philips Medical. F. Honoraria; Accuray Inc. G. Consultant; Advisory board: Accuray Inc. D. Buchholz: None. T. Dvorak: Q. Leadership; ASTRO Vice-Chairman of the Communications Committee.
exceeding 50%. Our data indicate that use of NCT does not compromise salvage success after a LRR, providing further assurance that this treatment strategy is safe for appropriately selected breast cancer patients. Author Disclosure: D.C. Weksberg: None. P.K. Allen: None. K.E. Hoffman: None. E.A. Strom: None. R. Shah: None. H.M. Kuerer: None. K.K. Hunt: None. T.A. Buchholz: None. E.A. Mittendorf: None.
1054 Outcomes and Predictive Factors for Salvage Therapy After Locoregional Recurrence Following Breast Conserving Therapy With Neoadjuvant Chemotherapy D.C. Weksberg, P.K. Allen, K.E. Hoffman, E.A. Strom, R. Shah, H.M. Kuerer, K.K. Hunt, T.A. Buchholz, and E.A. Mittendorf; The University of Texas MD Anderson Cancer Center, Houston, TX Purpose/Objective(s): While large clinical trials have shown equivalent overall survival (OS) for pre-operative versus post-operative chemotherapy in patients treated with breast conserving therapy (BCT), there are few data addressing locoregional recurrence (LRR) and salvage therapies in patients treated with neoadjuvant chemotherapy (NCT) as compared with those treated with surgery first. Here we characterize the clinical course and predictive features of salvage treatment for LRR after BCT, analyzed by initial treatment. Materials/Methods: A prospectively maintained database was used to identify 1589 patients who underwent BCT at our institution from 19872006; 1,141 patients underwent upfront surgery, and 448 received NCT. Kaplan-Meier and Cox regression analyses were performed for patients who experienced a LRR to analyze factors associated with OS, local control (LC) of recurrence, and distant metastasis-free survival (DMFS). Time zero was set as the date of LRR. Results: Fifty-six patients (30 initial surgeries, 26 NCT) developed a LRR, for a crude recurrence rate of 3%. The 10-year OS, subsequent LC after salvage, and DMFS from time of LRR were 39%, 81%, and 63%. Despite the fact that NCT patients had more advanced initial clinical stage (100% stage IIB vs. 0%, p < 0.01), the rates of OS, LC and DMFS were not different between groups. We next analyzed outcomes in those patients with potentially curable recurrence (n Z 43; less 13 patients with DM within 3 months of LRR). The 10-year OS, LC, and DMFS rates were 52%, 77%, and 69%. Eighteen of 24 initial surgery and 14 of 19 NCT patients had salvage surgery. Factors predictive of improved OS included initial clinical stage I-II disease, grade 1-2 disease, initial pathologically negative nodes, disease-free interval >36 months and use of surgery, chemotherapy or endocrine therapy (p 0.05 for all). On multivariate analysis, initial pathologically negative nodes (p < 0.05) and use of surgery for salvage (p < .01) were significant for predicting improved OS. These factors, and older age, were also associated with higher LC rates after salvage. Analysis of DMFS revealed estrogen receptor-positive disease and surgery for LRR as factors associated with reduced risk of DM, regional recurrence and initial adjuvant chemotherapy were associated with increased risk of DM. For each of these 3 endpoints, the addition of NCT to the multivariate model did not approach significance (p Z 0.87, 0.97, and 0.90). Conclusions: LRR is a rare event after BCT for patients treated with neoadjuvant chemotherapy and those undergoing surgery first. The majority who develop LRR remain curable with10-year OS rates
1055 Brain Metastases After Breast-conserving Therapy for Early-stage Breast Cancer: Incidence and Characteristics by Subtype N.D. Arvold,1 A. Niemierko,2 K.S. Oh,2 A.G. Taghian,2 N.U. Lin,3 J.R. Harris,3 and B.M. Alexander3; 1Harvard Radiation Oncology Program, Boston, MA, 2Massachusetts General Hospital, Boston, MA, 3 Dana-Farber Cancer Institute, Boston, MA Purpose/Objective(s): To characterize the risk and presentation of brain metastases (BM) that develop after breast-conserving therapy (BCT) for early stage breast cancer (BC), according to BC subtype approximation. Materials/Methods: We examined 1,434 consecutive patients with clinical stage I or II invasive BC who received BCT from December 1997 to July 2006. Five BC subtypes were approximated: estrogen receptor (ER) or progesterone receptor (PR) positive, HER2 negative, and grade 1/2 Z luminal A; ER+ or PR+, HER2-, and grade 3 Z luminal B; ER+ or PR+, and HER2+ Z luminal-HER2; ER-, PR-, and HER2+ Z HER2; and ER-, PR-, and HER2- Z triple negative (TN). Actuarial rates of BM were calculated using the Kaplan-Meier method. Results: Median follow-up was 85 months, and 36 women developed BM (crude rate, 2.5%). Overall 5-year cumulative incidence of BM was 1.7%; 0.1% for luminal A, 3.3% for luminal B, 3.2% for luminal-HER2, 3.7% for HER2, and 7.4% for TN. Women who developed BM were significantly more likely at initial BC diagnosis to be younger (p < 0.0001) and have tumors that were node-positive (p < 0.0001), grade 3 (p < 0.0001), hormone receptor-negative (p Z 0.006), and HER2+ (p Z 0.01). Median time from BC diagnosis to BM diagnosis was 51.4 months (range, 7.6-108 months), which was longer among luminal versus non-luminal subtypes (median, 61.4 months vs. 34.5 months). Seventy-eight percent of BM patients first developed extracranial distant metastases (DM) prior to BM. Median time from DM diagnosis to BM diagnosis was 12.8 months but varied substantially by subtype, including 7.4 months for TN, 9.6 months for luminal B, and 27.1 months for HER2. Nearly one-third (31%) of patients with DM developed BM. Eighty-one percent of all BM patients presented with neurologic symptoms, and 25% had leptomeningeal disease at diagnosis. Median number of BM at diagnosis was 2, with only luminal A tumors having a median of more than 3 BM at diagnosis. Median BM size in largest dimension was 15 mm, with TN (27 mm) and luminal B (16 mm) subtypes exhibiting the largest median BM sizes. Conclusions: The risk of BM after BCT for early-stage BC varies significantly by subtype. Given the symptomatic presentation of most patients and the short duration from extracranial DM to BM diagnosis, women with luminal B and TN tumors may benefit from routine brain MRI surveillance after diagnosis of DM, and this strategy should be prospectively evaluated. This may decrease the number of women presenting with symptomatic and/or large BM and increase treatment options, including stereotactic radiosurgery. Author Disclosure: N.D. Arvold: None. A. Niemierko: None. K.S. Oh: None. A.G. Taghian: None. N.U. Lin: None. J.R. Harris: None. B.M. Alexander: None.
1056 Breast-conserving Therapy Achieves Equivalent Locoregional Outcomes Compared to Mastectomy in Women With T1-2N0 Triple Negative Breast Cancer Z.S. Zumsteg, M. Morrow, B. Arnold, J. Zheng, Z. Zhang, M. Robson, E. Brogi, B. McCormick, S.N. Powell, and A.Y. Ho; Memorial Sloan-Kettering Cancer Center, New York, NY
International Journal of Radiation Oncology Biology Physics
guidelines. Fifteen percent of all plans failed the constraint that 90% of the Lumpectomy PTV Eval received 90% of the boost dose, primarily due to inadequate deep coverage of the expansion volume. All other target and OAR constraints were met “Per Protocol” or “Variation Acceptable” in 95% or more of the plans. Conclusions: Overall, our historical plans showed poor compliance with RTOG 1005 dosimetric guidelines. Although they failed to consistently provide complete coverage to the Breast PTV Eval and Lumpectomy PTV Eval, they efficiently limited the dose delivered to OARs. Our original breast fields were set up using classical tangent field design, which historically results in relatively low rates of recurrence and toxicity. It may require long-term results of RTOG 1005 to assess whether adherence to contour and planning guidelines results in meaningful clinical improvement over historical controls. Author Disclosure: E. Kane: None. K. Langen: E. Research Grant; Accuray Inc, Decimal Inc, Philips Medical. F. Honoraria; Accuray Inc. G. Consultant; Advisory board: Accuray Inc. D. Buchholz: None. T. Dvorak: Q. Leadership; ASTRO Vice-Chairman of the Communications Committee.
exceeding 50%. Our data indicate that use of NCT does not compromise salvage success after a LRR, providing further assurance that this treatment strategy is safe for appropriately selected breast cancer patients. Author Disclosure: D.C. Weksberg: None. P.K. Allen: None. K.E. Hoffman: None. E.A. Strom: None. R. Shah: None. H.M. Kuerer: None. K.K. Hunt: None. T.A. Buchholz: None. E.A. Mittendorf: None.
1054 Outcomes and Predictive Factors for Salvage Therapy After Locoregional Recurrence Following Breast Conserving Therapy With Neoadjuvant Chemotherapy D.C. Weksberg, P.K. Allen, K.E. Hoffman, E.A. Strom, R. Shah, H.M. Kuerer, K.K. Hunt, T.A. Buchholz, and E.A. Mittendorf; The University of Texas MD Anderson Cancer Center, Houston, TX Purpose/Objective(s): While large clinical trials have shown equivalent overall survival (OS) for pre-operative versus post-operative chemotherapy in patients treated with breast conserving therapy (BCT), there are few data addressing locoregional recurrence (LRR) and salvage therapies in patients treated with neoadjuvant chemotherapy (NCT) as compared with those treated with surgery first. Here we characterize the clinical course and predictive features of salvage treatment for LRR after BCT, analyzed by initial treatment. Materials/Methods: A prospectively maintained database was used to identify 1589 patients who underwent BCT at our institution from 19872006; 1,141 patients underwent upfront surgery, and 448 received NCT. Kaplan-Meier and Cox regression analyses were performed for patients who experienced a LRR to analyze factors associated with OS, local control (LC) of recurrence, and distant metastasis-free survival (DMFS). Time zero was set as the date of LRR. Results: Fifty-six patients (30 initial surgeries, 26 NCT) developed a LRR, for a crude recurrence rate of 3%. The 10-year OS, subsequent LC after salvage, and DMFS from time of LRR were 39%, 81%, and 63%. Despite the fact that NCT patients had more advanced initial clinical stage (100% stage IIB vs. 0%, p < 0.01), the rates of OS, LC and DMFS were not different between groups. We next analyzed outcomes in those patients with potentially curable recurrence (n Z 43; less 13 patients with DM within 3 months of LRR). The 10-year OS, LC, and DMFS rates were 52%, 77%, and 69%. Eighteen of 24 initial surgery and 14 of 19 NCT patients had salvage surgery. Factors predictive of improved OS included initial clinical stage I-II disease, grade 1-2 disease, initial pathologically negative nodes, disease-free interval >36 months and use of surgery, chemotherapy or endocrine therapy (p 0.05 for all). On multivariate analysis, initial pathologically negative nodes (p < 0.05) and use of surgery for salvage (p < .01) were significant for predicting improved OS. These factors, and older age, were also associated with higher LC rates after salvage. Analysis of DMFS revealed estrogen receptor-positive disease and surgery for LRR as factors associated with reduced risk of DM, regional recurrence and initial adjuvant chemotherapy were associated with increased risk of DM. For each of these 3 endpoints, the addition of NCT to the multivariate model did not approach significance (p Z 0.87, 0.97, and 0.90). Conclusions: LRR is a rare event after BCT for patients treated with neoadjuvant chemotherapy and those undergoing surgery first. The majority who develop LRR remain curable with10-year OS rates
1055 Brain Metastases After Breast-conserving Therapy for Early-stage Breast Cancer: Incidence and Characteristics by Subtype N.D. Arvold,1 A. Niemierko,2 K.S. Oh,2 A.G. Taghian,2 N.U. Lin,3 J.R. Harris,3 and B.M. Alexander3; 1Harvard Radiation Oncology Program, Boston, MA, 2Massachusetts General Hospital, Boston, MA, 3 Dana-Farber Cancer Institute, Boston, MA Purpose/Objective(s): To characterize the risk and presentation of brain metastases (BM) that develop after breast-conserving therapy (BCT) for early stage breast cancer (BC), according to BC subtype approximation. Materials/Methods: We examined 1,434 consecutive patients with clinical stage I or II invasive BC who received BCT from December 1997 to July 2006. Five BC subtypes were approximated: estrogen receptor (ER) or progesterone receptor (PR) positive, HER2 negative, and grade 1/2 Z luminal A; ER+ or PR+, HER2-, and grade 3 Z luminal B; ER+ or PR+, and HER2+ Z luminal-HER2; ER-, PR-, and HER2+ Z HER2; and ER-, PR-, and HER2- Z triple negative (TN). Actuarial rates of BM were calculated using the Kaplan-Meier method. Results: Median follow-up was 85 months, and 36 women developed BM (crude rate, 2.5%). Overall 5-year cumulative incidence of BM was 1.7%; 0.1% for luminal A, 3.3% for luminal B, 3.2% for luminal-HER2, 3.7% for HER2, and 7.4% for TN. Women who developed BM were significantly more likely at initial BC diagnosis to be younger (p < 0.0001) and have tumors that were node-positive (p < 0.0001), grade 3 (p < 0.0001), hormone receptor-negative (p Z 0.006), and HER2+ (p Z 0.01). Median time from BC diagnosis to BM diagnosis was 51.4 months (range, 7.6-108 months), which was longer among luminal versus non-luminal subtypes (median, 61.4 months vs. 34.5 months). Seventy-eight percent of BM patients first developed extracranial distant metastases (DM) prior to BM. Median time from DM diagnosis to BM diagnosis was 12.8 months but varied substantially by subtype, including 7.4 months for TN, 9.6 months for luminal B, and 27.1 months for HER2. Nearly one-third (31%) of patients with DM developed BM. Eighty-one percent of all BM patients presented with neurologic symptoms, and 25% had leptomeningeal disease at diagnosis. Median number of BM at diagnosis was 2, with only luminal A tumors having a median of more than 3 BM at diagnosis. Median BM size in largest dimension was 15 mm, with TN (27 mm) and luminal B (16 mm) subtypes exhibiting the largest median BM sizes. Conclusions: The risk of BM after BCT for early-stage BC varies significantly by subtype. Given the symptomatic presentation of most patients and the short duration from extracranial DM to BM diagnosis, women with luminal B and TN tumors may benefit from routine brain MRI surveillance after diagnosis of DM, and this strategy should be prospectively evaluated. This may decrease the number of women presenting with symptomatic and/or large BM and increase treatment options, including stereotactic radiosurgery. Author Disclosure: N.D. Arvold: None. A. Niemierko: None. K.S. Oh: None. A.G. Taghian: None. N.U. Lin: None. J.R. Harris: None. B.M. Alexander: None.
1056 Breast-conserving Therapy Achieves Equivalent Locoregional Outcomes Compared to Mastectomy in Women With T1-2N0 Triple Negative Breast Cancer Z.S. Zumsteg, M. Morrow, B. Arnold, J. Zheng, Z. Zhang, M. Robson, E. Brogi, B. McCormick, S.N. Powell, and A.Y. Ho; Memorial Sloan-Kettering Cancer Center, New York, NY