DNA methylation as an epigenetic signature predictive of response to neodjuvant treatment in TNBC patients.
Article 2018 en
Authors
BP
Begoña Pineda
JF
José Alejandro Pérez Fidalgo
ÁD
Ángel Díaz‐Lagares
Abstract
2 min read
e12658 Background: Prediction of response to neoadjuvant chemotherapy (NAC) in triple-negative breast cancer (TNBC) patients is an unmet medical need. NAC with anthracyclines/taxanes achieves about 30-40% pathological complete response (pCR) in TNBC. The aim of our study was to identify a predictive epigenomic signature of pCR to NAC in TNBC Methods: A prospective study was performed in samples from biopsy of TNBC patients performed before NAC administration. Patients were classified as responders (RCB = 0) and non-responders (RCB > 0). A DNA methylome study (Infinium HumanMethylation450 array, Illumina) was performed from a discovery cohort (DC, n = 24). Differentially methylated genes obtained from this study were validated by pyrosequencing (PyroMark Q96 System version 2.0.6, Qiagen) in a validation independent cohort (VC, n = 30) and in TN cell. In the same way, expression studies (qPCR) of the validated genes were performed both in patients and cell lines. Sample size was calculated in order to find differences in the proportion of pCR from 10% to 45% in 52 patients (with 15% drop-out). A predictive score with methylated genes and a ROC curve was calculated. Results: Array analysis showed a different methylation pattern between responder and non-responder patients. Selecting those differentially methylated CpGs located only in promoter regions, 9 genes were identified: 6 presented higher methylation in non-responder patients (gene 1 to 6) and 3 increased methylated genes in responders (gene 7 to 9). These genes were related to epithelial-mesenchymal transition and cell migration. After validation, the predictive ability of methylation of these genes for response (RCB = 0) or no response (RCB ≠ 0) was evaluated using a statistical model obtaining a predictive capacity of pCR of 72.7% for gene 7 and 9 with a ROC curve (AUC) 0.898 (CI95% = 0.788- 1.000). Gene expression studies correlated with methylation levels in both patients and cell lines. Conclusions: The 2-gene methylation panel with genes 7 and 9 accurately predicted response to NAC in TNBC. Further validation in prospective trials is warranted.
Marina Zafrilla-López, Miriam Acosta-Díez, Marina Mitjans, Anna Giménez‐Palomo, Pilar A. Sáiz, Carme Barrot‐Feixat, Ester Jiménez, Sergi Papiol, Victoria Ruíz, Patricia Gavín, María Paz García‐Portilla, Leticia González-Blanco, Julio Bobes, Thomas G. Schulze, Eduard Vieta, Antoni Benabarre, Bárbara Arias
Jorge Gómez‐Miragaya, Sebastián Morán, María Eréndira Calleja-Cervantes, Alejandro Collado-Solé, Laia Paré, Antonio Gómez, Violeta Serra, Lacey E. Dobrolecki, Michael T. Lewis, Ángel Díaz‐Lagares, Pîlar Eroles, Aleix Prat, Manel Esteller, Eva González‐Suárez
Jorge Gómez‐Miragaya, Sebastián Morán, María Eréndira Calleja-Cervantes, Alejandro Collado-Solé, Laia Paré, Antonio Gómez, Violeta Serra, Lacey E. Dobrolecki, Michael T. Lewis, Ángel Díaz‐Lagares, Pîlar Eroles, Aleix Prat, Manel Esteller, Eva González‐Suárez
Discussion(0)
No comments yet. Be the first to comment.