ESTRO 37 Abstract book

S78

ESTRO 37

had clinician and 583 patients had photographic assessments. Patients reported more NTEs than clinicians or via photographs (p<0.001 for most NTEs). Concordance was generally poor on an individual patient level; weighted kappa ranged from 0.03 (overall change in appearance) to 0.17 (breast shrinkage). % agreement was lowest between PROMs and photographic assessments for change in breast appearance (26.4%) and highest between PROMs and clinician assessment for breast oedema (90.6%). Concordance was not affected by the baseline characteristics tested and was similar between data ascertainment methods at 2 and 5 years. Conclusion Concordance was poor on an individual patient level however there was agreement on a population level as both PROMs and clinician assessments were able to distinguish between the radiotherapy schedules in IMPORT LOW. Poor concordance on an individual patient level may result from the data ascertainment methods having different comparators and patients reporting more NTEs. PROMs should be prioritised and further exploration of PROMs as a primary data ascertainment method is required. OC-0157 The impact of the boost technique on breast cancer recurrence in breast-conserving therapy I. Kindts 1 , K. Verhoeven 2 , A. Laenen 3 , H. Janssen 1 , E. Van Limbergen 1 , C. Weltens 1 1 University Hospitals Leuven, Department of Radiation Oncology, Leuven, Belgium 2 GROW - School for Oncology and Developmental Biology- Maastricht University Medical Centre, Department of Radiation Oncology MAASTRO, Maastricht, The Netherlands 3 KU Leuven University, Leuven Biostatistics and Statistical Bioinformatics Centre L-Biostat, Leuven, Belgium Purpose or Objective Adding a boost to the tumour bed after whole-breast irradiation in breast-conserving therapy reduces local recurrence rates. The purpose of the present study was to investigate whether the boost technique influences the magnitude of the effect. Material and Methods Patients treated with breast-conserving therapy for invasive breast cancer between 2000 and 2007 were included in the analysis. Breast-conserving therapy included breast-conserving surgery followed by whole- breast irradiation and a sequential boost to the tumour bed. All radiotherapy treatments were conducted at the same institute. An electron boost was applied for a superficial boost volume (less than 29 mm under the epidermis). A brachytherapy boost was proposed in all other cases. When patients refused a brachytherapy boost or when it was technically not possible, a photon boost was administered. Patients were divided into three groups according to the boost technique: electrons, brachytherapy or photons. The endpoints were local recurrence and any recurrence (local recurrence, regional recurrence or metastasis) in the three groups. Clinicopathological and treatment characteristics of the three groups were compared using the Chi-square test or Mann-Whitney U test. Cox regression models were used for the analysis of association between prognostic factors and outcome. Correction for the confounders in the association between boost technique and outcome was performed using multivariable models. Results 1920 tumours were eligible for analysis. 41 tumours were excluded because there was no information on the boost or no boost was given. 1448 patients (77.1 %) received an electron boost, 334 (17.8 %) a brachytherapy boost and

97 (5.1 %) a photon boost. Median follow-up was 13.1 years. Patients in the three groups significantly differed with respect to age, pT, pN, tumour type, (extensive) ductal carcinoma in situ, regional radiotherapy and year of diagnosis. 10-year local recurrence rates were 2.2 % and were significantly higher in younger, pN-positive or estrogen- receptor negative patients and when chemotherapy was administered. In multivariable analysis there was no significant difference between the three groups for the local recurrence risk (p 0.89) (Figure).

Figure: local recurrence-free estimates by boost technique 10-year any recurrence rate was 10.8 %. Factors influencing the any recurrence rate were: age, grade, pT, pN, section margins, lymphovascular invasion, estrogen- or progesteron receptor or Her2 status, regional radiotherapy, chemotherapy and year of diagnosis. In multivariable analysis there was no significant difference between the brachytherapy group and the electron group or the photon group (p 0.11 and p 0.28, respectively). The photon group had more recurrences compared to the electron group (HR 1.81, 95%CI 1.12;2.92, p 0.02). Conclusion The local recurrence risk reduction of the tumour bed boost in breast-conserving therapy is not influenced by the applied boost technique. OC-0158 Histologic heterogeneity of triple negative breast cancer: a National Cancer Center Database study M. Mills 1 , G. Yang 1 , D. Oliver 1 , C. Liveringhouse 1 , K. Ahmed 1 , A. Orman 1 , C. Laronga 2 , R. Diaz 1 1 Moffitt Cancer Center, Radiation Oncology, Tampa, USA 2 Moffitt Cancer Center, Surgical Oncology, Tampa, USA Purpose or Objective Triple-negative breast cancer (TNBC) has historically been associated with poor prognosis. Recent studies have suggested that molecular subtyping of TNBC has important clinical implications. However, the utility of histologic subtyping in rare histologies with TNBC has not yet been well-characterized. This study utilizes data from the National Cancer Center Database (NCDB) to complete the largest series to date investigating the prognostic importance of histology and treatment ramifications in TNBC. Material and Methods A total of 734,047 patients (pts) with histologic subtypes of invasive ductal carcinoma (IDC), metaplastic breast carcinoma (MBC), medullary breast carcinoma (MedBC), adenoid cystic carcinoma (ACC), invasive lobular carcinoma (ILC), or epithelial breast carcinoma (EBC) treated between 2004 and 2012 were identified in the NCDB for analysis. Of these, 89,512 pts with TNBC that received surgery were included for analysis. Proportion of TNBC and pt and treatment characteristics within MBC, MedBC, ACC, ILC and EBC subtypes were compared to IDC