- Email: [email protected]
P104 Patterns of recurrence and survival in HER2+ patients relapsing after receiving adjuvant trastuzumab
14th St.Gallen International Breast Cancer Conference / The Breast 24S1 (2015) S26–S86
receiving 18 H SC cycles; 193 patients (10%) withdrew and 121 (7%) were still on treatment (4 did not receive treatment). Concurrent chemotherapy was received by 1074 patients (58%) and sequential chemotherapy by 604 (32%). One hundred and eighty-six patients (10%) did not receive any chemotherapy and these data will be reported later. Main safety results during the treatment period are shown in the table. Conclusion: SafeHer Cohort A interim results confirm the safety and tolerability of H SC 600 mg q3w with concurrent and sequential chemotherapy for HER2-positive EBC. No new safety signals were identified and H SC results are consistent with the known safety profile of adjuvant H. Disclosure of Interest: Consultant/advisor: Roche/GNE (JG, MDL, MV, HAA, BA, XP); Novartis/Celgene/Amgen (MDL). Stock/royalties/ equity: Roche/GNE (NAS, DH). Research/travel: Roche/GNE (JG, MDL, MV, HAA, BA); Novartis/Celgene/Amgen (MDL); Eisai KR (KHJ). Employment: Roche/GNE (NAS, DH, MS). P104 Patterns of recurrence and survival in HER2+ patients relapsing after receiving adjuvant trastuzumab M. Hattori *, T. Fujita, M. Sawaki, N. Kondou, A. Yoshimura, M. Ichikawa, J. Ishiguro, H. Iwata. Department of Breast Oncology, Aichi Cancer Center, Nagoya, Japan Goals: Although 1 year of adjuvant trastuzumab improves survival in patients with HER2+ breast cancer, some patients will develop recurrent disease. The objective of this study was to examine the association between the patterns of recurrence and survival after relapse in HER2+ breast cancer patients who have relapsed after receiving adjuvant trastuzumab. Methods: We utilized data from 321 HER2+ breast cancer patients identified as having received adjuvant trastuzumab at our institution between 2003 and 2012. Patients with ipsilateral breast tumor recurrence were excluded from analysis. The impact of patterns of recurrence and clinicopathologic factors on overall survival (OS) after relapse were analyzed with Cox regression model, and OS after relapse was estimated by Kaplan–Meier method. Results: At a median follow-up of 45.5 months, 29 patients had developed recurrent disease and 17 patients had died. The median treatment free interval from the end of adjuvant trastuzumab in patients with recurrent disease was 10.3 months, response rate to first anti-HER2 therapy for metastatic disease was 53.8%, and the median OS after relapse was 25.4 months. In univariate analysis, hormone receptor status, clinical stage at first diagnosis, treatmentfree interval (≤6 months) and visceral metastases as first site of relapse had no significant influence on OS after relapse. Multivariate analysis revealed that response to first anti-HER2 therapy for metastatic disease (41.6 vs. 22.6 months, HR 2.8; p = 0.059) and brain metastasis as first site of relapse (22.5 vs. 32.9 months, HR 4.3; p = 0.086) were associated with OS after relapse but low in statistical significance. Conclusion: Our findings indicate response to initial anti-HER2 therapy after relapse and brain metastasis as first site of relapse were prognostic factors for survival after relapse in HER2+ breast cancer patients who have relapsed after receiving adjuvant trastuzumab. On the other hand, treatment-free interval (≤6 months) did not impact on survival after relapse. Disclosure of Interest: No significant relationships.
S61
P105 The MAGIC survey in HR+, HER2− breast cancer (BC): when might multigene assays be of value? M. Aapro1 *, M. De Laurentiis2 , B. Linderholm3 , E. Mamounas4 , C. Markopoulos5 , M. Martin6 , P. Neven7 , D. Rea8 , R. Rouzier9 , C. Thomssen10 . 1 Medical Oncology, Institut Multidisciplinaire d’Oncologie, Clinique de Genolier, Genolier, Switzerland, 2 Department of Senology, National Cancer Institute G. Pascale Foundation, Naples, Italy, 3 Department of Oncology, Sahlgrenska Academy and University Hospital, Gothenburg, Sweden, 4 University of Florida Health Cancer Center at Orlando Health, Orlando, United States of America, 5 Department of Surgery, Athens University Medical School, Athens, Greece, 6 Medical Oncology Service, Hospital General Universitario Gregorio Mara˜ no ´n, Madrid, Spain, 7 Multidisciplinary Breast Centre and Gynaecological Oncology, UZ Leuven, Leuven, Belgium, 8 School of Cancer Sciences, University of Birmingham, Birmingham, United Kingdom, 9 Department of Surgery, Institut Curie-Universit´e Versailles-Saint-Quentin, Paris-Saint-Cloud, France, 10 Department of Gynecology, Martin-Luther-University Halle-Wittenberg, Halle (Saale), Germany Goals: The MAGIC survey evaluated which criteria clinicians use regarding the need for adjuvant chemotherapy (AdjCT) and showed that there was substantial heterogeneity across clinicians and countries in treatment decisions (Aapro et al, EBCC 2014, abstract 24). Multigene assays (MGA) can help to make more-informed decisions by providing prognostic and predictive information beyond traditional parameters. We here present data on BC patient profiles with a high heterogeneity in treatment recommendations based on traditional parameters. Methods: From August 2013 until January 2014, physicians with ≥5 years’ experience in BC treatment and participating in multidisciplinary teams were invited for the MAGIC survey capturing treatment recommendations for randomly generated early BC patient profiles (n = 672). A conjoint analysis was used to assess which patient attributes were considered for treatment decisions. Results: The survey was completed by 911 physicians from 52 countries, of whom 72% had >10 years’ experience. Their treatment recommendations showed that for BC patient profiles with only high-risk or only low-risk characteristics, there was a high consensus to recommend AdjCT or no AdjCT (endocrine treatment alone); 42% of the profiles had >75% probability of being recommended AdjCT and 6% had >75% chance of being recommended no AdjCT. There was substantial uncertainty for 52% of patient profiles with at least every fourth patient likely to receive a different treatment recommendation if visiting a different physician. 15% of patient profiles had a very high uncertainty with <50% probability to be recommended either chemotherapy and endocrine treatment or the latter alone. These patient profiles tended to have the following characteristics: >50 years old, tumor size <3 cm, Grade 1 or 2 tumor, high ER expression, and Ki67 expression <20%. Conclusion: There was substantial heterogeneity in treatment recommendations and an overall tendency to give chemo-endocrine rather than endocrine treatment alone. The highest uncertainty in treatment decisions was seen in patients with intermediate risk by clinical and pathological parameters. MGAs may facilitate decisionmaking in these situations. Disclosure of Interest: MA: consultant for Amgen, Astellas, BMS, Celgene, GSK, Helsinn, Hospira, Novartis, Merck, Merck Serono, Pfizer, Pharmacosmos, Pierre Fabre, Roche, Sandoz, Teva, Vifor; honoraria for lectures at symposia received from Amgen, Bayer Schering, Cephalon, GSK.
receiving 18 H SC cycles; 193 patients (10%) withdrew and 121 (7%) were still on treatment (4 did not receive treatment). Concurrent chemotherapy was received by 1074 patients (58%) and sequential chemotherapy by 604 (32%). One hundred and eighty-six patients (10%) did not receive any chemotherapy and these data will be reported later. Main safety results during the treatment period are shown in the table. Conclusion: SafeHer Cohort A interim results confirm the safety and tolerability of H SC 600 mg q3w with concurrent and sequential chemotherapy for HER2-positive EBC. No new safety signals were identified and H SC results are consistent with the known safety profile of adjuvant H. Disclosure of Interest: Consultant/advisor: Roche/GNE (JG, MDL, MV, HAA, BA, XP); Novartis/Celgene/Amgen (MDL). Stock/royalties/ equity: Roche/GNE (NAS, DH). Research/travel: Roche/GNE (JG, MDL, MV, HAA, BA); Novartis/Celgene/Amgen (MDL); Eisai KR (KHJ). Employment: Roche/GNE (NAS, DH, MS). P104 Patterns of recurrence and survival in HER2+ patients relapsing after receiving adjuvant trastuzumab M. Hattori *, T. Fujita, M. Sawaki, N. Kondou, A. Yoshimura, M. Ichikawa, J. Ishiguro, H. Iwata. Department of Breast Oncology, Aichi Cancer Center, Nagoya, Japan Goals: Although 1 year of adjuvant trastuzumab improves survival in patients with HER2+ breast cancer, some patients will develop recurrent disease. The objective of this study was to examine the association between the patterns of recurrence and survival after relapse in HER2+ breast cancer patients who have relapsed after receiving adjuvant trastuzumab. Methods: We utilized data from 321 HER2+ breast cancer patients identified as having received adjuvant trastuzumab at our institution between 2003 and 2012. Patients with ipsilateral breast tumor recurrence were excluded from analysis. The impact of patterns of recurrence and clinicopathologic factors on overall survival (OS) after relapse were analyzed with Cox regression model, and OS after relapse was estimated by Kaplan–Meier method. Results: At a median follow-up of 45.5 months, 29 patients had developed recurrent disease and 17 patients had died. The median treatment free interval from the end of adjuvant trastuzumab in patients with recurrent disease was 10.3 months, response rate to first anti-HER2 therapy for metastatic disease was 53.8%, and the median OS after relapse was 25.4 months. In univariate analysis, hormone receptor status, clinical stage at first diagnosis, treatmentfree interval (≤6 months) and visceral metastases as first site of relapse had no significant influence on OS after relapse. Multivariate analysis revealed that response to first anti-HER2 therapy for metastatic disease (41.6 vs. 22.6 months, HR 2.8; p = 0.059) and brain metastasis as first site of relapse (22.5 vs. 32.9 months, HR 4.3; p = 0.086) were associated with OS after relapse but low in statistical significance. Conclusion: Our findings indicate response to initial anti-HER2 therapy after relapse and brain metastasis as first site of relapse were prognostic factors for survival after relapse in HER2+ breast cancer patients who have relapsed after receiving adjuvant trastuzumab. On the other hand, treatment-free interval (≤6 months) did not impact on survival after relapse. Disclosure of Interest: No significant relationships.
S61
P105 The MAGIC survey in HR+, HER2− breast cancer (BC): when might multigene assays be of value? M. Aapro1 *, M. De Laurentiis2 , B. Linderholm3 , E. Mamounas4 , C. Markopoulos5 , M. Martin6 , P. Neven7 , D. Rea8 , R. Rouzier9 , C. Thomssen10 . 1 Medical Oncology, Institut Multidisciplinaire d’Oncologie, Clinique de Genolier, Genolier, Switzerland, 2 Department of Senology, National Cancer Institute G. Pascale Foundation, Naples, Italy, 3 Department of Oncology, Sahlgrenska Academy and University Hospital, Gothenburg, Sweden, 4 University of Florida Health Cancer Center at Orlando Health, Orlando, United States of America, 5 Department of Surgery, Athens University Medical School, Athens, Greece, 6 Medical Oncology Service, Hospital General Universitario Gregorio Mara˜ no ´n, Madrid, Spain, 7 Multidisciplinary Breast Centre and Gynaecological Oncology, UZ Leuven, Leuven, Belgium, 8 School of Cancer Sciences, University of Birmingham, Birmingham, United Kingdom, 9 Department of Surgery, Institut Curie-Universit´e Versailles-Saint-Quentin, Paris-Saint-Cloud, France, 10 Department of Gynecology, Martin-Luther-University Halle-Wittenberg, Halle (Saale), Germany Goals: The MAGIC survey evaluated which criteria clinicians use regarding the need for adjuvant chemotherapy (AdjCT) and showed that there was substantial heterogeneity across clinicians and countries in treatment decisions (Aapro et al, EBCC 2014, abstract 24). Multigene assays (MGA) can help to make more-informed decisions by providing prognostic and predictive information beyond traditional parameters. We here present data on BC patient profiles with a high heterogeneity in treatment recommendations based on traditional parameters. Methods: From August 2013 until January 2014, physicians with ≥5 years’ experience in BC treatment and participating in multidisciplinary teams were invited for the MAGIC survey capturing treatment recommendations for randomly generated early BC patient profiles (n = 672). A conjoint analysis was used to assess which patient attributes were considered for treatment decisions. Results: The survey was completed by 911 physicians from 52 countries, of whom 72% had >10 years’ experience. Their treatment recommendations showed that for BC patient profiles with only high-risk or only low-risk characteristics, there was a high consensus to recommend AdjCT or no AdjCT (endocrine treatment alone); 42% of the profiles had >75% probability of being recommended AdjCT and 6% had >75% chance of being recommended no AdjCT. There was substantial uncertainty for 52% of patient profiles with at least every fourth patient likely to receive a different treatment recommendation if visiting a different physician. 15% of patient profiles had a very high uncertainty with <50% probability to be recommended either chemotherapy and endocrine treatment or the latter alone. These patient profiles tended to have the following characteristics: >50 years old, tumor size <3 cm, Grade 1 or 2 tumor, high ER expression, and Ki67 expression <20%. Conclusion: There was substantial heterogeneity in treatment recommendations and an overall tendency to give chemo-endocrine rather than endocrine treatment alone. The highest uncertainty in treatment decisions was seen in patients with intermediate risk by clinical and pathological parameters. MGAs may facilitate decisionmaking in these situations. Disclosure of Interest: MA: consultant for Amgen, Astellas, BMS, Celgene, GSK, Helsinn, Hospira, Novartis, Merck, Merck Serono, Pfizer, Pharmacosmos, Pierre Fabre, Roche, Sandoz, Teva, Vifor; honoraria for lectures at symposia received from Amgen, Bayer Schering, Cephalon, GSK.