Topic:
 
Issue Section:
 Biomarkers, translational research and precision medicine
Background: Antiangiogenic therapy using bevacizumab has proven effective for a number of cancers; however, in breast cancer there is an unmet need to identify patients that benefit from such treatment.
Methods: In this phase II clinical trial (NeoAva-NCT00773695), patients (n = 132) with HER2 negative primary tumors of ≥ 25 mm were treated with neoadjuvant chemotherapy and randomized (1:1) to receive bevacizumab or chemotherapy only. Ratio of the tumor size before and after treatment was calculated to obtain a continuous scale reflecting the response to therapy. Tumor biopsies at week 0, 12 and 25 were analyzed by reverse phase protein arrays (RPPA) for expression levels of 210 proteins (of which 54 phospho-specific). Proteins with low variance across samples were filtered out and Lasso regression was then used to derive a predictor of tumor shrinkage from the expression of proteins prior to treatment. Leave-one-out cross-validation (LOOCV) was used to select the optimal Lasso model.
Results: From the Lasso analysis with LOOCV, we discovered a signature consisting of nine proteins which was capable to predict patients responding to treatment with bevacizumab in combination with chemotherapy with high accuracy. The corresponding protein score obtained as a weighted sum of the protein expressions was significantly different in patients with/without pathological complete response (pCR) or low/high residual cancer burden (RCB </ = > 2). The nine-protein signature was applied to corresponding mRNA data and the resultant score also showed significant separation in the above groups. Finally, the nine-protein signature was validated in an independent mRNA data set from a similar phase II clinical trial (PROMIX-NCT00957125) with scores significantly separating patients with/without pCR.
Conclusions: In this study we demonstrate that integration of multiple protein-expressions to create a signature is a promising approach for prediction of response to treatment with bevacizumab combined with chemotherapy in breast cancer patients. A prospective clinical trial is planned to confirm the potential clinical benefit of using the protein signature for treatment selection.
Clinical trial identification: NeoAva: NCT00773695 Promix: NCT00957125.
Legal entity responsible for the study: Oslo University Hospital HF (PI. Olav Engebråten).
Funding: Oslo University Hospital HF was sponsor and Roche Norway was co-sponsor of the clinical trial NeoAva.
Disclosure: G. Mills: Consultant, speaker, grant, research support: AZ/MedImmune, Tarveda, Myriad Genetics, AbbVie, Critical Outcomes Technology, Pfizer, Takeda/Millennium Pharm, Tesaro. A-L. Borresen-Dale: Shareholder, board member: Arctic Pharma AS; Consulting: PubGene AS, Saga Diagnostics AS. O. Engebråten: Research funds: Roche Norway was a co-sponsor of the NeoAva study. All other authors have declared no conflicts of interest.
© European Society for Medical Oncology 2019. Published by Oxford University Press on behalf of the European Society for Medical Oncology. All rights reserved. For permissions, please email: journals.permissions@oup.com.
This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)