Is post-mastectomy radiation therapy contributive in pN0-1mi breast cancer patients? Results of a French multi-centric cohort

Publication date: December 2017
Source:European Journal of Cancer, Volume 87
Author(s): Violaine Forissier, Agnès Tallet, Monique Cohen, Jean-Marc Classe, Fabien Reyal, Nicolas Chopin, Chafika Mazouni, Pierre Gimbergues, Emile Daraï, Pierre Emmanuel Colombo, Pierre Azuar, Eric Lambaudie, Gilles Houvenaeghel
AimTo assess the value of post-mastectomy radiation therapy (PMRT) to breast cancer (BC) patients with no or minimal lymph node (LN) involvement.Materials and methodsWe retrospectively analysed a French multi-centric cohort of 4283 patients treated by mastectomy and axillary dissection, with or without PMRT, between 1980 and 2013. Practices were analysed for three treatment periods (1980–1999, 2000–2005 and 2006–2013). The impact of PMRT on loco-regional recurrence (LRR), disease-free survival (DFS), BC-specific survival and overall survival was assessed in pN0-1mi patients using multivariate analyses (logistic regression and Cox model). It was subsequently assessed based on the number of clinicopathological recurrence-risk factors, generating a prognostic index (French-PMRT index), to isolate a pN0-1mi patients subgroup that might derive a benefit from PMRT. We tested the accuracy of the Cambridge-PMRT (c-PMRT) index to discriminate between patients with significantly different outcomes and the value of PMRT in each c-PMRT prognostic group.ResultsMore than half of the pN0-1mi patients of our cohort underwent PMRT, which almost significantly improved LRR-free survival and DFS. Matching pN0-1mi patients based on the number of clinicopathologic recurrence-risk factors identified a higher risk subpopulation (≥3 recurrence-risk factors), but PMRT did not improve patient outcomes. Although the c-PMRT index had the potential to predict patient outcomes, its use did not help in making the decision of whether or not to use PMRT.ConclusionWe failed to isolate a subgroup of early BC patients without LN involvement suitable for PMRT, despite studying a large cohort.



http://ift.tt/2iymc3t