Abstract P3-06-12: Predicting residual risk of recurrence after neoadjuvant chemotherapy- a retrospective analysis of EndoPredict® in the GeparTrio trial

Abstract

Abstract Background: Patients (pts) with breast cancer who do not achieve a pathological complete response (pCR) after neoadjuvant chemotherapy (NACT) have a poor prognosis and might be candidates for post-neoadjuvant clinical trials investigating novel agents. The CPS-EG score (Mittendorf et al. JCO 2011) – a combination of clinical/pathological parameters - is currently used as criterion for selecting patients with highest risk of recurrence after NACT. Here, we examined whether the gene expression test EndoPredict (EP) performed on surgical specimen after NACT provides independent prognostic information for predicting the likelihood of recurrence in breast cancer patients with ER-positive, HER2-negative (ER+/HER2-) disease. Methods:The molecular EP score was determined by qRT-PCR in 76 available surgical specimen classified as ER+/HER2- from breast cancer pts with residual disease after NACT participating in the neoadjuvant GeparTrio trial. The pre-specified clinical/molecular hybrid score EPclin was determined using ypT and ypN after NACT as clinical variables. Patients were classified as having low or high risk according to pre-defined cut-off values. CPS-EG score was calculated based on stage before and after NACT and pretreatment grade and ER status. Primary endpoint was disease-free survival (DFS). Associations were assessed with uni- and multivariate Cox proportional hazard models. The unbiased c-index was estimated for common clinical/pathological parameters and EP/EPclin scores. Results: Among the 76 ER+/HER2- breast cancer pts evaluated, EP classified 50% of all pts (n=38) as low risk. EP high-risk patients had a significantly increased risk for relapse compared to the low-risk group (continuous EP HR 1.14 [95% CI 1.05-1.27] p-value 0.02; log-rank p=0.015). Multivariable Cox regression and unbiased c-index estimates (using EP and CPS-EG score as continuous variable) showed that the EP-score (HR 1.15 [95% CI 1.03-1.29 p=0.014] and CPS-EG (HR 1.51 [95% CI 1.07-2.13] p=0.019) provide independent prognostic information. Using the composite EPclin score 13 pts (17%) were classified as being EPclin low. The risk of relapse was significantly higher for EPclin high compared to low (continuous EPclin HR 1.78 [95% CI 1.29-2.46] p<0.001; log-rank p=0.047). Bivariate Cox regression analysis including the CPS-EG and EPclin score (both as continuous variables) showed that only EPclin (HR 1.63 [95% CI 1.14-2.31] p=0.0068) but not the CPS-EG (HR 1.25 [95%CI 0.85-2.31] p=0.26) was significantly associated with worse DFS. Results were similar for overall survival. Conclusions: Our study shows that EPclin performed on surgical specimen of ER+/HER2- pts after NACT is an independent predictor of recurrence in pts not achieving a pCR after NACT. EP/EPclin low risk patients are probably sufficiently treated with (neo-)adjuvant chemo-endocrine treatment alone, whereas EP(clin) high risk patients have an increased risk of recurrence, despite receiving standard (neo-)adjuvant chemo-endocrine therapy. The identification of molecular luminal high-risk patients could help to identify high risk patients as candidates for novel drug-based approaches in addition to endocrine therapy to overcome resistance in post-neoadjuvant trials. Citation Format: Sibylle Loibl, Jan C Brase, Stephan Gade, Jens Huober, Kristin Krappmann, Knut Engels, Falko Fend, Berit Maria Pfitzner, Joern Hilfrich, Christoph Thomssen, Hans Juergen Holzhausen, Silvia Darb-Esfahani, Christian Schem, Keyur Mehta, Ralf Kronenwett, Gunter von Minckwitz, Carsten Denkert. Predicting residual risk of recurrence after neoadjuvant chemotherapy- a retrospective analysis of EndoPredict® in the GeparTrio trial [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 P3-06-12.