263P - Predictors of effectiveness of the use of steroidal aromatase inhibitors after progression on non-steroidal aromatase inhibitors in advanced breast...

Date 10 October 2016
Event ESMO 2016 Congress
Session Poster display
Topics Breast Cancer, Metastatic
Presenter Maria Madalena Ulrich de Menezes Pereira dos Santos
Citation Annals of Oncology (2016) 27 (6): 68-99. 10.1093/annonc/mdw365
Authors M.M. Ulrich de Menezes Pereira dos Santos, C. Cardoso, M. Sousa, S. Esteves, M. Brito, A. Moreira
  • Oncology, IPO LFG, 1250-255 Lisboa - Lisboa/PT



Clinical evidence suggests that steroid aromatase inhibitors (SAI) lack cross resistance with non-steroid aromatase inhibitors (nSAI). Therefore, the use of SAI in advanced breast cancer (ABC) patients (pts) after disease progression on nSAI has been considered a reasonable option in several centers. Our aim is to confirm the effectiveness of using this strategy in ABC and identify predictors of response.


Retrospective analysis of a cohort of consecutive ABC female postmenopausal hormone receptor patients treated in a single cancer center with SAI (exemestane) after progression with nSAI (letrozol or anastrozole), between 2009 and 2013. Effectiveness outcomes were time to progression (TTP) and clinical benefit (CB), defined as complete response or partial response or stable disease for more than 6 months. We used logistic and Cox regression models.


160 ABC pts were identified, with a median age of 60 (range 32-87). In 74% pts estrogen receptor was high (≥75%) and in 16% HER2 was overexpressed. Visceral disease was present in 25% pts and 67% had been previously treated with adjuvant endocrine therapy. In ABC, 57% were initially treated with nSAI. The average number of therapeutic lines until exemestane was 2.3 (1-8) and 33% were treated with exemestane immediately after progression under nSAI. The exemestane CB rate was 64%. In univariable analysis, the factors associated with CB were visceral disease (p = 0.047), previous CB to nSAI (p = 0.047) and number of treatment lines between nSAI and exemestane (p = 0.011). In multivariable analysis, visceral disease and CB to nSAI were the only independent factors significantly associated with CB to exemestane (OR 0.38 and 2.15, respectively). Median TTP was 7.8m (95%CI 6.8-9.2). In multivariable analysis, visceral disease and the number of treatments between AI were the only independent factors significantly associated with TTP (HR 1.35 and 1.56, respectively).


The use of exemestane after progression on nSAI seems to be a valid option, especially in ABC patients with non-visceral disease, few treatment lines between nSAI and SAI and with previous CB to nSAI.

Clinical trial identification

Legal entity responsible for the study

IPO Lisboa


IPO Lisboa


All authors have declared no conflicts of interest.