Publications by authors named "M A Troester"

The ternary complex of PGRMC1-σ2R/TMEM97-LDLR has recently been discovered and plays a role in cholesterol transport. This study investigated whether individual components of that complex are prognostic breast cancer biomarkers and defined expression in established molecular subtypes. 4,463 invasive breast cancers were analyzed as a function of molecular and phenotypic markers, estimates of cellular proliferation, and recurrence-free survival.

View Article and Find Full Text PDF

Genome-wide association studies have identified approximately 200 genetic risk loci for breast cancer, but the causal variants and target genes are mostly unknown. We sought to fine-map all known breast cancer risk loci using genome-wide association study data from 172,737 female breast cancer cases and 242,009 controls of African, Asian and European ancestry. We identified 332 independent association signals for breast cancer risk, including 131 signals not reported previously, and for 50 of them, we narrowed the credible causal variants down to a single variant.

View Article and Find Full Text PDF
Article Synopsis
  • The SEER registry uses criteria like laterality and histology to distinguish between second primary breast cancers, affecting how accurately tumors are classified.
  • Analysis of receptor status in a large sample of contralateral and ipsilateral tumors indicated that ipsilateral tumors may actually be recurrences due to their higher receptor dependence and younger patient demographics.
  • While SEER's criteria enhance specificity, they may also lead to inaccuracies in classifying tumors, highlighting the need for better datasets to compare classification methods.
View Article and Find Full Text PDF

In the Carolina Breast Cancer Study (CBCS), clustering census tracts based on spatial location, demographic variables, and socioeconomic status is crucial for understanding how these factors influence health outcomes and cancer risk. This task, known as spatial clustering, involves identifying clusters of similar locations by considering both geographic and characteristic patterns. While standard clustering methods such as K-means, spectral clustering, and hierarchical clustering are well-studied, spatial clustering is less explored due to the inherent differences between spatial domains and their corresponding covariates.

View Article and Find Full Text PDF
Article Synopsis
  • Researchers focused on breast cancer subtypes Luminal A and Luminal B, using machine learning to analyze H&E images, aiming to identify tumor characteristics linked to higher recurrence risks.
  • The study involved training models on segmented images of tumors, finding that an image-based protocol effectively predicted recurrence times, comparable to traditional genomic testing methods (PAM50).
  • Results indicated that while adjusting for tumor grade didn't significantly improve subtype prediction, the image analysis provided a viable alternative in identifying patients in need of genomic testing, potentially increasing testing rates among ER+/HER2-patients.
View Article and Find Full Text PDF