ESTRO 2020 Abstract Book

S173 ESTRO 2020

OC-0329 Patterns of nodal metastases and locoregional recurrences in locally advanced breast cancer C. Hvid 1 , B. Offersen 2 , H.M. Nielsen 1 1 Aarhus University Hospital, Dept of Oncology, Aarhus C, Denmark ; 2 Aarhus University Hospital, Dept of Experimental Clinical Oncology, Aarhus C, Denmark Purpose or Objective With advances in management, loco-regional recurrence (LRR) of early breast cancer in Denmark is historically low with a 5 year risk of 1.6% for local and 0.8% for regional recurrence (DBCG-IMN study). However, much higher LRR rates have been reported for locally advanced breast cancer (LABC) and inflammatory breast cancer (IBC). For early breast cancer, ESTRO consensus guidelines for target volume delineation have recently been published. The purpose of this study is to determine whether these guidelines are appropriate for use in LABC and IBC. Furthermore, to gain information on the extent of nodal involvement in patients presenting with N3 disease. Material and Methods A total of 229 consecutive patients have been treated for LABC and IBC at our institution from September 2006 to May 2018. Median follow up time was 44 months. Patients were identified by chemotherapy prescription lists and verified by review of treatment records. All patients received neoadjuvant taxane-containing chemotherapy followed by surgery and radiation therapy (RT) with 48/50 Gy in 24/25 fractions or 40 Gy in 15 fractions. Patients with pathologically verified LRR were identified by review of treatment records. Localization of LRR was determined from records, including clinical photography, imaging studies and RT treatment plans and charted in accordance with the ESTRO consensus guidelines. In a subset of 73 patients with N3 disease, treatment files and imaging studies were reviewed to perform a more detailed assessment of nodal disease extension. Seven patients with N3 disease received a boost to residual N- site gross tumor, median dose 14 Gy in 7 fractions (range 10-16 Gy). Results Fourteen LRRs were identified in 13 patients. The locations of LRRs are shown in relation to ESTRO consensus guidelines in figure 1. One patient developed simultaneous LRR locally and in CTVn_L2 while one patient developed an isolated regional recurrence in CTVn_L4, where she had received a boost. The remaining 12 LRRs were isolated chest wall recurrences. They were characterized by being localized close (less than 1 cm) to the mastectomy scar (12/13) and presenting with skin involvement (11/13). All patients with N3 disease had nodal involvement of the axilla level 1/2 as well. Forty patients (55%) only had involvement of level 3, 26 (35%) of both level 3 and 4, while a minority (10%) had involvement of level 4, but no apparent involvement of level 3.

work is to estimate the true benefit in terms of live saved for various radiotherapy techniques, accounting for the usage of radiotherapy over time for various stages, and the patient’s long-term survival. Material and Methods The study used data from the SEER 18 registry database. Non-metastatic breast cancer patients treated between 1988 and 2012 were included. Information on diagnosis date, cancer stage, patient age, use of RT, length of follow-up, status at study cut-off, and date of SLC were extracted. Kaplan Meyer statistic was used to calculate overall and SLC free survivals for various stages. Multivariate competing risk analysis (stepwise PHREG procedure in SAS) was used to assess factors associated with SLC incidence, accounting for interaction between the delivery of RT with date of diagnosis and age at diagnosis. A modified BEIR VII model was used to calculate the absolute number of SLC for each stage and age, accounting for survival and use of RT. Results

A total of 765,639 patients were identified, 49.8% received RT and 50.2% did not. There was a 29.2% increased use of RT after 1998 (p<0.001), and also an increased use for younger patients (p<0.001). There were large differences in overall survival (OS) between stages, with a 22.9 years median OS for DCIS and 4.6 years for advanced stages. The OS improved over the last 3 decades (p<0.001). Risk factors for developing SLCs included the delivery of RT (HR=1.579, p<0.001), earlier stages (HR=1.112, p<0.001), younger age (HR=1.061, p<0.001), and longer follow-up (HR=1.034 p<0.001). For all women diagnosed in the US in 2019 with DCIS or localised breast cancers who are eligible for accelerated partial breast irradiation (APBI), we estimated to 18,179 SLCs and 14,543 deaths due to standard RT using Darby reported a mean lung dose of 5.7 Gy. We calculated that 11,619 live could be spared using 3D-CRT APBI, 11,862 using HDR brachytherapy, 13,574 using robotic SBRT, and 14,288 using LDR brachytherapy. Conclusion With the improved breast cancer patient survival and the increased use of RT, there is an increased and significant risk SLC. APBI could reduce this risk and has therefore a clinically relevant survival advantage compared to whole breast radiotherapy for early stage breast cancers. Our results show that the use of APBI can save more than 10,000 lives annually in the US.