Search

Abstract. Long-term adherence to adjuvant endocrine therapy (ET, tamoxifen and aromatase inhibitors) is paramount for patients with early-stage breast cancer. Adherence to adjuvant endocrine therapy is hampered by numerous side effects associated with sustained estrogen deprivation. We aimed to describe recent real-world patterns of therapy, patients’ discontinuations of ET, toxicities, quality of life (QoL) and to develop a predictive model of early ET discontinuation. Methods. We used the first 9595 patients of the French CANTO cohort (NCT01993498), to evaluate among 6238 premenopausal and postmenopausal patients with HR+/HER2- stage I-III BC, who were prescribed adjuvant ET: a. treatment patterns of adjuvant ET including change of ET prescription during the follow-up course b. ET-associated toxicities and c. impact on QoL. Independent variables included medical history and toxicities as measured by : Common Toxicity Criteria Adverse Events (CTCAE) v4, Patient-Reported Outcomes (PROs) including European Organization for Research and Treatment of Cancer (EORTC) Quality of Life Questionnaires (QLQ-C30) and Breast Cancer Module (BR23), and Hospital Anxiety and Depression Scale (HADS). Treatment discontinuation and treatment change were determined on the basis of patient’s declaration and medical decisions reported in the CANTO eCRF. We used patient data at 4 months from therapy initiation to train and evaluate on a held-out test set a machine-learning model (gradient-boosted trees) that is predictive of time to early discontinuation i.e. permanent discontinuation before four years of additional therapy. Results. 4052 post-menopausal patients and 2186 premenopausal patients were included in this analysis. Median follow-up after ET initiation is 3 years and 2 months. 86% of post-menopausal patients were prescribed a non-steroidal AI initially and 92% of premenopausal patients received tamoxifen first. Discontinuation rate of the first adjuvant endocrine therapy at 1 year was 14% and 10% in premenopausal and post-menopausal patients, respectively. Among 741 post-menopausal and 340 premenopausal patients who started a second ET, discontinuation of the second prescribed adjuvant ET at 1 additional year of therapy is 30% in both populations. Patients who switched from a first adjuvant ET to a second or further one continued to have more treatment-related toxicities and associated decrements in QoL. Exclusions due to data completeness and outcome definition led to 5331 patients being used for the model (4264 in the training set and 1067 in the validation set). In that population, the permanent discontinuation rate at 3 years is 6%. Our prediction model of time to early discontinuation obtains a C-index of 0.78 in the held-out validation set. Conclusion. Tolerability and continued adherence to ET remains a challenge for many patients. Early discontinuation models may assist in identifying patients who are likely to interrupt their adjuvant ET. Adapted clinical management, including robust support and management of toxicities, as well as new and more tolerable adjuvant endocrine therapies may improve the clinical outcomes of these patients. Citation Format: Felix Balazard, Aurélie Bertaut, Élise Bordet, Stéphane Mulard, Julie Blanc, Nathalie Briot, Gautier Paux, Asma Dhaini Merimeche, Olivier Rigal, Charles Coutant, Marion Fournier, Christelle Jouannaud, Patrick Soulie, Florence Lerebours, Paul-Henri Cottu, Olivier Tredan, Laurence Vanlemmens, Christelle Levy, Marie-Ange Mouret-Reynier, Mario Campone, Keri J. S. Brady, Medha Sasane, Megan Rice, Catherine Coulouvrat, Anne-Laure Martin, Alexandra Jacquet, Ines Vaz-Luis, Christina Herold, Barbara Pistilli. Patterns of adjuvant endocrine therapy, discontinuations, toxicities and quality of life: Development of a model for early discontinuation using the CANTO cohort [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr P1-13-08.

Background: Determination of HER2 status in invasive breast carcinomas is now mandatory for the optimal treatment of the patient. A vast majority of initial published studies described positivity in about 20 to 25 % of breast carcinomas, but the proportion of HER2 positive gene expression in women with breast cancer varies in the literature from 9 to 74%. The population of the tumor analyzed (for example histologic types, stage, age), the type of the assays used and the lack of uniformity or quality controls in detection assays may partly explain these differences. Our study describes the evolution of HER2 positive invasive breast cancer proportions during a period of 10 years and identifies clinical and pathological characteristics of these positive tumors. Methods: Based on the regional, population-based breast cancer registry, we identify 2396 patients diagnosed with invasive breast cancer in the Côte-d'Or department and treated in a single institution between 1998 and 2008. The HER2 status was determined by immunohistochemistry and by fluorescence in situ hybridization by a single team that performed regular internal quality controls and participated to multiannual external international quality control programs. Results: During the 1998–2008 period, 12.8% of invasive breast cancers were identified as HER2 positive, with no trend to annual variations (p = 0.27). During the same period, despite an increase of 45% of the annual diagnosed cancers in the institution, the clinical and pathological characteristics such as age, tumor size, grade, hormonal receptors expression remain stable. We only noticed a slight but significant increase of lobular carcinomas (p = 0.036). In multivariate analysis, HER2 positivity was as expected associated with: hormonal receptor negative status (OR = 1.7 [IC95%: 1.2–2.4], p = 0.0035), metastatic presentation at diagnosis (OR = 2 [IC95%: 1.2–3.4], p = 0.0092), ductal subtype (OR = 1.9 [IC95%: 1.1–3.1], p = 0.0178) and high SRB grade (OR = 2.8 [IC95%: 1.7–4.5] for SBR II, OR = 5.4 [IC95%: 3.2–9.1] for SBR III, p = 0.0035). Conversely, patients older than 75 years presented mostly HER2 negative tumors (OR = 0.5 [IC95%: 0.3–0.8], p < 0.0001). Conclusions: This population based study on HER2 breast cancer status emphasizes that in a population with stable clinical and tumoral characteristics, when standardized and controlled assays are used, the proportion of HER positive tumors remains stable over a large period and is lower than usually reported (12.8%). Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr P3-08-07.

Background: Breast tumors with negative estrogen receptor (ER-) and positive progesterone receptor (PR+) are rare (from 0 to 3.4 % according to the studies), and their existence is contested. These markers determine cancer molecular subtypes which play a determinant role for both the management and the prognosis of breast cancers. It is then essential to document the real existence of ER-/PR+ tumors. The present study aimed at determining if ER-/PR+ tumors share more basal or luminal characteristics. Methods: Between 2000 and 2015, 50 patients with ER-/PR+ breast tumors, representing 0.6 % of all breast cancers diagnosed in our institution, were included in this study. Their clinical (age, node status), morphological (size, histological type, Elston and Ellis (EE) grade, necrosis, inflammation, pushing margins, central scar, mitotic index) and immunohistochemical characteristics (ER, PR, HER2, CK5/6 and EGFR status) were compared with those of 50 luminal and 50 triple negative (TN) tumors randomly selected in our lab database. At the time this abstract is written, the Ki67 index determination is still in progress. Five of these ER-/PR+ tumors were also given a molecular test (Nanostring). Qualitative variables were compared using Chi2 or Fisher test, quantitative variables were compared using Student or Mann & Whitney tests. To take into account for multiple comparisons, p-values less than 0.025 were considered as significant. Results: The results are summarized in table 1. For almost all the analyzed criteria, ER-/PR+ tumors present statistical difference with luminal ones. On the contrary, they share most of TN tumors characteristics. The 5 molecular analyzes performed on ER-/PR+ samples showed the following phenotypes: 3 basal, 1 HER2 enriched and 1 luminal. For this last one, new immunohistological analyzes reveal in fact an ER+ staining. Table 1: Study results ER-/PR+ tumorsER+ tumorsTN tumorsp ER-/PR+ vs ER+p ER-/PR+ vs TNSize (mm)23.12525.20.00440.4822Histological type0.00120.2424 ductal9480.4100 lobular219.60 neuro-endocrine40 Differentiation Conclusion: This study tends to support that ER-/PR+ tumors may not exist and are likely to be TN cancers or less frequently false negative ER+ tumors. In a practical point of view, we think that 1) when a tumor shows ER-/PR+ and TN characteristics, it is probably a false positive PR staining, and the tumor has to be considered as a TN one, 2) when an ER-/PR+ tumor don't fit the triple negative tumors characteristics, ER must be retested in order to exclude a true luminal tumor. Citation Format: Molly D, Bertaut A, Blanchet C, Beltjens F, Charon-Barra C, Loustalot C, Desmoulins I, Arnould L. Do estrogen receptor negative and progesterone receptor positive breast tumors really exist? Attitude for not making them real. [abstract]. In: Proceedings of the Thirty-Eighth Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2015 Dec 8-12; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(4 Suppl):Abstract nr P1-01-14.