Gene expression analysis on core biopsy samples is feasible and identifies candidate genes that correlate with pCR to primary chemotherapy. Gene expression in IBC differs significantly from noninflammatory breast cancer.

This paper compares performance of redundant representation and sparse coding against classical kernel methods for classifying histological sections. Sparse coding has been proven to be an effective technique for restoration, and has recently been extended to classification. The main issue with classification of histology sections is inherent heterogeneity as a result of technical and biological variations. Technical variations originate from sample preparation, fixation, and staining from multiple laboratories, where biological variations originate from tissue content. Image patches are represented with invariant features at local and global scales, where local refers to responses measured with Laplacian of Gaussians, and global refers to measurements in the color space. Experiments are designed to learn dictionaries, through sparse coding, and to train classifiers through kernel methods with normal, necorotic, apoptotic, and tumor with with characteristics of high cellularity. Two different kernel methods of support vector machine (SVM) and kernel discriminant analysis (KDA) are used for comparative analysis. Preliminary investigation on histological samples of Glioblastoma multiforme (GBM) indicates that kernel methods perform as good if not better than sparse coding with redundant representation.

Introduction: The SEER Program of the National Cancer Institute (NCI) is an authoritative source of cancer incidence and survival statistics, collecting population-based data for ∼28 percent of the US. As innovations in molecular testing are validated and recommended in guidelines, new genomic research models are needed to characterize their use and impact on patient outcomes in clinical practice. To that end, Genomic Health and SEER have collaborated to electronically supplement the SEER registries with Oncotype DX results. This first report characterizes breast cancer specific survival (BCSS) in node (N)- hormone receptor (HR)+ HER2- invasive breast cancer. Methods: The prospectively-defined population-biomarker-outcomes linkage had objectives, analysis cohort, standardized biomarker, and outcome defined prior to data linkage. Patients were eligible if N-, HR+, HER2- (by RT-PCR), had no prior malignancy, were 40-85 years of age, and were diagnosed between Jan 2004 (Oncotype DX available in Jan 2004) and Dec 2011 (SEER survival analysis complete through 2012). BCSS was defined as previously described and assessed rigorously (Howlader et al, JNCI 2010). Oncotype DX Recurrence Score results were provided to SEER as mandated by their registry operations. Analysis by treatment was interpreted in the context of the lack of randomization and well known limitations of registry reporting of chemotherapy (CT) treatment (Noone et al, Medical Care 2014). Results: Among 169,158 eligible patients, 38,568 (23%) had a Recurrence Score (RS), increasing from 2% of 2004 diagnoses to 35% of 2011 diagnoses. Patients with RS had median age of 57 yr, were 99.4% female, 84% white, 29% grade 1 and 54% grade 2, 25% 1cm or less and 53% between 1 and 2cm. Median follow-up was 39 mo with 8,239 patients having >5 yrs follow-up. Among low risk RS <18 (N=21,023), intermediate risk RS 18-30 (N=14,494) and high risk RS ≥31 (N=3,051) patients, chemotherapy use was reported in 7%, 34%, and 69%, respectively. 5-year BCSS (with 95%CI) are provided in Table. Continuous RS was significantly associated with BCSS unadjusted (p<0.001), and after adjusting for age, grade, and tumor size (p<0.001), and when stratified by treatment (p<0.001). Conclusions: 5 yr BCSS outcomes are excellent (99.6%) in the over 21,000 patients with low RS disease. As expected, high RS disease is associated with lower 5 yr survival despite addition of chemotherapy. More rigorous ascertainment of treatment is planned to draw conclusions regarding chemotherapy treatment benefit. Analyses that examine test utilization as well as annual survival updates to obtain longer term event rates will be important to gain additional insights into patient subgroup outcomes that can be characterized by large population-based genomic studies.  RS<18RS 18-30RS≥31 NBCSS (95%CI)NBCSS (95%CI)NBCSS (95%CI)Analysis all pts2102399.6 (99.4-99.7)1449498.6 (98.3-98.9)305195.6 (94.4-96.6)Analysis by CT reported      No/Unknown CT1955499.6 (99.5-99.7)957098.6 (98.2-98.9)93694.0 (91.1-96.0)Yes CT146999.3 (98.4-99.7)492498.6 (98.0-99.0)211596.4 (95.0-97.4) Citation Format: Shak S, Petkov VI, Miller DP, Howlader N, Gliner N, Howe W, Schussler N, Cronin K, Baehner FL, Penberthy L. Breast cancer specific survival in 38,568 patients with node negative hormone receptor positive invasive breast cancer and oncotype DX recurrence score results in the SEER database. [abstract]. In: Proceedings of the Thirty-Eighth Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2015 Dec 8-12; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(4 Suppl):Abstract nr P5-15-01.

#53 Background: The Recurrence Score (RS) has been validated for estimating residual risk of distant recurrence (DR) in ER+ node negative (N-) primary breast cancer patients receiving adjuvant tamoxifen. Aromatase inhibitors (AIs) are widely used for adjuvant treatment of ER+ disease in postmenopausal women. The RS has not been evaluated in patients treated with an AI.
 Method: Tumour blocks were collected retrospectively from patients in the monotherapy arms of the ATAC trial under the TransATAC protocol yielding 1308 hormone receptor positive (HR+) blocks in UK patients for analysis (1). Cox proportional hazards model was used to test the significance of adding RS to a clinical model (age, tumour size, grade, treatment) and to Adjuvant! Online.
 Results: 65 patients received chemotherapy, 4 were centrally HR negative and 8 did not start adjuvant therapy leaving 1231 evaluable patients of which 872, 306 and 53 were N-, N+ and N unknown in whom there were 72, 74 and 6 DRs, respectively. In the prospectively-defined primary multivariate analysis tumour size, grade and RS were each separately statistically significant in predicting time to DR in N- patients (p<0.001, 0.003 and <0.001, respectively). Similar results were seen in N+ patients. The 9-year rates of DR by nodal status for the pre-specified RS groups are shown in the table. RS showed statistically significant prognostic value beyond that provided by Adjuvant! Online in both N- (p<0.001) and N+ patients (p=0.003).
 
 Conclusions: Oncotype DX RS is an independent predictor of the risk of distant recurrence in N- and N+ HR+ patients treated with anastrozole or tamoxifen. The data are not predictive of a differential benefit between anastrozole and tamoxifen.
 (1) Dowsett et al JCO 2008, 26, 1059. Citation Information: Cancer Res 2009;69(2 Suppl):Abstract nr 53.

Background: We have previously reported the results of surgical excision without irradiation for selected patients with DCIS in ECOG E5194, where the 5-year rates of local recurrence varied with age, grade, and lesion size (Hughes et al. J Clin Oncol 2009; 27:5319). New methods are needed to provide more accurate and reproducible assessment of recurrence risk.