Abstract P6-08-36: SA02 trial: Results of a genomics-based prospective cohort in node-positive early breast cancer with « good-prognosis signature » treated with adjuvant chemotherapy

Background . Adjuvant chemotherapy (CT) for node-positive (N+) early breast cancer (EBC) is based upon an anthracycline-taxane combination. However, the benefit of taxane is limited to a small population, but associated to morbid and financial costs, making crucial the identification of patients likely to benefit or not from anthracycline-based regimen without taxane. Using DNA microarrays to profile a retrospective series of 498 patients (pts) treated with adjuvant anthracycline-based CT without taxane, we had identified and validated a gene expression signature (GES) associated with metastatic relapse. The corresponding Relapse Score (RS) sorted the patients (pts) in two groups: the "good-prognosis" group (75% of pts) with a 5-year metastasis-free survival (MFS) of 82%, and the "poor-prognosis" group (25% of pts) with a 56% 5-year MFS. We present here the results of a prospective multicentric national cohort of 175 pts, SA02, initiated to analyze anthracycline-based adjuvant CT without taxane in N+ EBC pts with a "good-prognosis" RS, with the aim of confirming their good prognosis in term of 5-year MFS. Methods. Women with surgical EBC were screened for inclusion in 4 French hospitals. After diagnosis of lymph node involvement, frozen tumor samples were used for hybridization on Affymetrix U133 Plus 2.0 microarrays, and the RS was defined. RS-based "good-prognosis" pts were treated in the SA02 cohort and received 6 FEC100 cycles (Fluoro-uracile 500 mg/m 2 , Epirubicin 100 mg/m 2 , Cyclophosphamide 500 mg/m 2 , every 21 days), followed by adjuvant radiotherapy, hormone therapy and/or trastuzumab according to standard guidelines. Pts with a "poor-prognosis" or non-evaluable RS were not included in the cohort. We present here an analysis with a median follow-up of 50 months. Results : Between May 2007 and May 2010, samples from 175 eligible N+ EBC pts were collected for gene profiling. The profiling failed for 54 samples (31%), due to insufficient RNA amount, poor RNA quality, poor labeling performance, or forgotten. 102 pts (84%) were defined as "good-prognosis", while only 5 pts (4%) were defined as "poor-prognosis", and 14 pts (12%) could not be assigned a prognostic group. The percentage of "good-prognosis" RS was higher than observed in the initial retrospective series. On the 102 "good-prognosis", 88 were included in the SA02 cohort to receive 6 FEC100 cycles, 14 pts were not included in the cohort due to investigator or patient’s decision. The mean time from the date of surgery to the onset of chemotherapy was 5.4 +/-1.5 weeks. With a median follow-up of 50 months, the 2-year and 4-year MFS are 98% (95% CI 91-99) and 96% (95% CI 89-99). The 2-year and 4-year overall survival are 100% and 99% (95% CI 92-100). Conclusion. This analysis confirms that genomic analyses are feasible in clinical practice. The MFS results with a median follow-up of 50 months are within the expected hypothesis; follow-up is needed to confirm or not these results at 5 years. Final follow-up data will be available in 2015. Citation Format: Bertucci Francois, Extra Jean-Marc, Ferrero Jean-Marc, Bachelot Thomas, Autret Aurelie, Boyer-Chammard Agnes, Viens Patrice. SA02 trial: Results of a genomics-based prospective cohort in node-positive early breast cancer with « good-prognosis signature » treated with adjuvant chemotherapy [abstract]. In: Proceedings of the Thirty-Seventh Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2014 Dec 9-13; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2015;75(9 Suppl):Abstract nr P6-08-36.