Abstract 4469: Twelve epigenetic marker panel for circulating DNA in plasma can detect early breast cancer

Introduction: Breast cancer is usually divided into several subtypes, and clinical courses are different by subtypes. To manage patients adequately, many biomarker studies are conducted. Circulating DNA (ciDNA) is one of the attracted biomarkers for cancer management, but there are little reports that ciDNA is useful for early detection of breast cancer. DNA methylation pattern is also different among subtypes, it has possibilities to be a biomarker. To detect early breast cancer, we developed epigenetic marker panel for ciDNA by using the subtype-specific methylation markers and common methylation markers among subtypes. Methods: Twelve novel methylation markers containing four common methylation markers and eight subtype-specific methylation markers were selected by methylation analysis. DNA samples for methylation analysis were obtained from 56 samples of microdissected cancer cells, 34 samples of cell lines and 29 samples of blood from healthy volunteers. These markers might detect circulating tumor DNA. We also selected four novel internal controls for methylation specific PCR (MSP) to estimate amount of ciDNA. Digital droplet MSP was used in this panel for estimation. Algorithms for discrimination between healthy volunteers and cancer patients were built by using ciDNA in plasma from 80 healthy volunteers and 87 cancer patient as a training set. All combination of 12 methylation markers and three parameters (geometric mean of internal controls, geometric mean of methylation markers, number of positive methylation markers) was estimated with Support Vector Machine (SVM). The best algorithm was selected by leave-one-out cross-validation method. We evaluated the algorithm with ciDNA in plasma from 53 healthy volunteers and 58 breast cancer patients as an independent validation set. The result of each samples was output as a diagnostic index. Results: The best algorithm adopted four methylation markers and two parameters. The area under the curve of receiver operating characteristic curve by SVM was 0.92 in the training set and 0.88 in the validation set. Sensitivity and specificity were 0.91 and 0.83 in the training set, 0.84 and 0.79 in the validation set, respectively. Sensitivity of stage0-I, IIA, IIB-III, and metastatic breast cancer were 0.86, 0.87, 0.84, and 0.97, respectively. This panel realized high sensitivity even in early breast cancer. There was statistically significant trend that higher stage category showed higher diagnostic index. Sensitivity of triple negative, luminal HER2 and HER2 positive breast cancer were all 1.0 whereas sensitivity of luminal breast cancer was 0.83. This panel also realized high sensitivity regardless of subtype. Conclusions: Our epigenetic marker panel could detect early breast cancer. It had high accuracy in every subtype, it could be a strong tool for blood-based screening of breast cancer. Citation Format: Natsue Uehiro, Fumiaki Sato, Shigehira Saji, Masakazu Toi. Twelve epigenetic marker panel for circulating DNA in plasma can detect early breast cancer. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 4469.