Abstract P2-16-35: Impact of neoadjuvant chemotherapy on survival in luminal B tumours: A propensity score weighted analysis

Abstract

Abstract BACKGROUND: Although no survival benefit has been demonstrated by treating early breast cancer (BC) with neoadjuvant chemotherapy (NAC), it is now widely accepted as first intention strategy in triple negative and HER2-positive BC. However, the benefit of the neoadjuvant strategy is controversial in luminal tumours. The latter are considered as poor responder to chemotherapy and it remains unknown if delaying time to surgery by providing NAC as first treatment might be deleterious. The aim of this study was to compare the prognosis of patients with Luminal B BC treated with either NAC, either adjuvant chemotherapy. MATERIALS AND METHODS: We retrospectively identified in our institutional database 188 patients with grade 3 luminal B BC treated with NAC and 1203 patients treated with the same adjuvant chemotherapy (AC). To deal with the selection bias, the contribution of each subject was weighted by a propensity score (PS): 1/PS for patient with the NAC strategy and 1/(1-PS) for patients with the AC strategy. This enables to generate a pseudo-population with a balanced covariates combination between groups. We evaluated the disease free survival (DFS), the distant metastasis free survival (DMFS) and the overall survival (OS) using a weighted Cox model. To analyse the effect of NAC in high-risk subgroups, we introduced in the model interaction effects with the initial tumour size and the initial clinical nodal status. RESULTS: By construction, patients, tumor characteristics and trastuzumab use were similar between the population treated with NAC and the population treated with adjuvant CT. After PS weighting and adjustment on confounding factors, DFS (HR = 2.56, 95%IC [1.73 - 3.78]; p< 0.001), DMFS (HR = 1.28, 95%IC [1 - 1.63]; p < 0.001) and OS were lower in the NAC group (HR = 1.51, 95%IC [1.10-2.09]]; p = 0.01) when compared with the group of patients treated with adjuvant chemotherapy.The interaction test between the initial clinical nodal status, initial clinical tumour size and survival endpoints was significant, highlighting a different prognostic effect of NAC versus AC in nodes-positive than in node-negative patients, and a different prognostic effect of NAC versus AC in small than in large tumours. No significant DFS nor OS difference was found in node-negative patients. In node-positive patients with tumour larger than 20mm, the risk of death was nearly doubled when comparing NAC strategy to AC strategy (HR=2.30, 95%CI [1.34-3.4]), while the highest deleterious impact on OS of the NAC strategy against the AC strategy was observed in node-positive patients with a tumour size smaller than ≤20mm (HR=3.52, 95%CI [1.59-7.8]). CONCLUSION: In node-positive, grade 3, luminal B tumours, receiving NAC seems associated with an impaired survival when compared with receiving the same treatment in the adjuvant setting. Our results suggest that surgery as first treatment could be the preferred option, even if breast-conserving surgery cannot be obtained. In tumors smaller than 20mm, NAC strategy should be avoided irrespective of the nodal status. Citation Format: Enora Laas, Thomas Gaillard, Jean Guillaume Feron, Jean-Yves Pierga, Florence Coussy, Beatriz Grandal, Audrey Bellesoeur, Eléonore De Guillebon, Sonia Rozette, Clemence Evrevin, Lucie Laot, Marick Lae, Anne Sophie Hamy, Fabien Reyal. Impact of neoadjuvant chemotherapy on survival in luminal B tumours: A propensity score weighted analysis [abstract]. In: Proceedings of the 2019 San Antonio Breast Cancer Symposium; 2019 Dec 10-14; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2020;80(4 Suppl):Abstract nr P2-16-35.