Abstract PS5-03: Celtil score and long-term survival outcome in early stage HER2-positive (HER2+) breast cancer treated with anti-HER2-based chemotherapy: A correlative analysis of neoALTTO trial

Abstract

Abstract Background: Biomarkers to help escalate or de-escalate systemic therapies are urgently needed in early-stage HER2+ breast cancer patients. The combination of high stromal tumor infiltrating lymphocytes (TILs) and low tumor cellularity at 2 weeks of anti-HER2 therapy (CelTIL score) has been associated with high rates of pathologic complete response (pCR) after completion of neoadjuvant therapy. However, the value of CelTIL as a prognostic biomarker is unknown. Here, we present an independent validation of CelTIL in the NeoALTTO phase III trial. Methods: NeoALTTO randomized 455 patients (pts) with HER2+ early breast cancer to receive lapatinib (arm A), trastuzumab (Arm B) or lapatinib and trastuzumab (Arm C) for 6 weeks, followed by the assigned anti-HER2 treatment combined with paclitaxel weekly for 12 weeks. After surgery, patients received 3 cycles of fluorouracil, epirubicin and cyclophosphamide, and then continued the assigned anti-HER2 treatment for 34 weeks. CelTIL was centrally evaluated in tumor biopsies performed at week 2 in arms B and C using the pre-established formula (CelTIL score = -0.8 × tumor cellularity [%] + 1.3 ×TILs [%]). The primary objective was to evaluate the association of CelTIL (as a continuous variable and using the pre-established 33.59% cut-off as defined by Nuciforo P, et al; Ann Oncol. 2018) with event-free survival (EFS). Secondary objectives were to evaluate the association of CelTIL with overall survival (OS) and pCR. Univariable and multivariable analyses were performed adjusting for hormone-receptor status, pre-treatment tumor size and nodal status, planned type of surgery, pCR status, and treatment arm. Results: The CelTIL score was evaluable in 196 samples (108 samples in arm B and 88 samples in arm C), of which 45.4% (89/196) had low CelTIL score and 54.6% (107/196) had a high CelTIL score. As a continuous score, higher CelTIL levels were independently associated with improved EFS (hazard ratio [HR]=0.84 per 10% increment; 95% CI 0.73-0.97; P=0.006), but not OS (HR=0.85; 95% CI 0.71-1.03; P=0.094). Using the pre-established cutoff, the 5-year EFS estimate was 76% (95% CI 68-85%) and 60% (95% CI 50-72%) in pts with high- and low-CelTIL, respectively (adjusted HR=0.53; 95% CI 0.30-0.94; p=0.030). Moreover, the 5-year OS rate was 86% (95% CI 80-94%) and 73% (95 CI 64-84%) in high- and low-CelTIL respectively (adjusted HR=0.43; 95% CI 0.20-0.92; p=0.029). CelTIL as a continuous score was also independently associated with higher rates of pCR (odds ratio [OR]=1.18; 95% CI 1.02-1.36; p=0.024). The pCR rate in the high-CelTIL group was 37% (40/107) versus 18% (16/89) in the low-CelTIL group (adjusted OR=2.60; 95% CI 1.31-5.14; p=0.006). Conclusions: CelTIL score measured at week 2 of anti-HER2 therapy is significantly associated with long-term survival outcomes in early-stage HER2+ breast cancer, independently of pCR status and other prognostic variables. Further validation of this biomarker could help select early-on pts candidates for escalation or de-escalation of systemic therapy. Citation Format: Nuria Chic, Stephen Luen, Paolo Nuciforo, Roberto Salgado, Debora Fumagalli, Florentine Hilbers, Yingbo Wang, Evandro de Azambuja, Itsvan Lang, Serena Di Cosimo, Cristina Saura, Jens Huober, Aleix Prat, Sherene Loi. Celtil score and long-term survival outcome in early stage HER2-positive (HER2+) breast cancer treated with anti-HER2-based chemotherapy: A correlative analysis of neoALTTO trial [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PS5-03.