Abstract
Abstract Background A significant portion of breast cancer patients treated in the neoadjuvant setting does not achieve a pathological complete response (pCR) and has an increased risk of relapse after surgery. The current lack of reliable predictors of response to guide patient therapy in clinical practice requires the identification of predictive biomarkers that could discriminate between responders and non-responders. The purpose of this study was to investigate the potential of circulating microRNAs (miRNAs) as novel non-invasive predictive biomarkers of response to neoadjuvant treatment in HER2-positive breast cancer patients enrolled in the Neo-ALTTO study. Methods We performed a longitudinal miRNA monitoring of the well-annotated and high-quality plasma samples of patients treated with neoadjuvant lapatinib, trastuzumab or their combination. Before profiling, the overall plasma samples were randomly splitted in two sets, namely the training set and the testing set, with distribution of timing of plasma collection, arm of treatment, rate of pCR, and tumor characteristics resembling that of the entire Neo-ALTTO patient population. To this end a PCR-based high-throughput approach was firstly employed in the training set. After a proper normalization step, for each treatment arm and time-point, a panel of potential miRNAs associated to pCR was identified by resorting to univariate approaches. Multivariate penalized logistic regression models were implemented thereafter, in order to identify specific signatures for each time-point and treatment arm. Finally, the predictive capability of the aforementioned signatures was evaluated in the testing set. Results Plasma miRNA profiles were available for 435 of the 455 (96%) patients enrolled in the Neo-ALTTO study. In details, miRNA levels were analyzed in 141, 151 and 143 patients assigned to neoadjuvant trastuzumab, lapatinib, and their combination, respectively, at baseline and after 2 weeks of treatment. A total of 30 miRNAs (including both normalizators and specific miRNAs) were identified and 6 signatures were found predictive of pCR in terms of Area Under the ROC Curve (AUC) in the training population. The predictive capability of 4 of 6 of the identified signatures was confirmed in the testing set, specifically: lapatinib at baseline AUC (95%CI) = 0.86 (0.73-0.98) and after two weeks AUC (95%CI) = 0.71 (0.55-0.86), trastuzumab after two weeks AUC (95%CI) = 0.81 (0.70-0.92), and lapatinib + trastuzumab after two weeks AUC (95%CI) = 0.67 (0.51-0.83). Of note, the predictive value of the signature of trastuzumab after two weeks was confirmed after adjustment for hormonal receptor status - training set AUC (95%CI) = 0.90 (0.80-0.99), testing set, AUC (95%CI) = 0.84 (0.74-0.94). Conclusion These findings provide the first evidence of the potential of circulating miRNAs to predict treatment response in HER2-positive breast cancer patients treated with neoadjuvant therapy. Results obtained in the trastuzumab arms are of special value as in women with unfavorable miRNA signature it is anticipated an unfavorable response to paclitaxel plus trastuzumab after just 2 weeks of treatment. Data on correlation with elapse free survival will be presented at the meeting. Citation Format: Di Cosimo S, Appierto V, Tiberio P, Verderio P, Pizzamiglio S, Bottelli S, Iorio M, Baselga J, Piccart M, Huober J, Brase J, de la Pena L, Fumagalli D, de Azambuja E, de Braud F, Daidone MG. Plasma microRNA levels for predicting therapeutic response to neoadjuvant treatment in HER2-positive breast cancer: Results from Neo-ALTTO [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr S3-02.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call