Abstract Black and Brown Brazilian women have higher breast cancer mortality when compared to Whites, an association that remains after adjusting for age, level of education and stage at diagnosis. Limited studies have researched this population, which can benefit from the identification of molecular markers that can provide an insight into who is at greater risk of mortality. Epigenetic biomarkers represent the combined effect of environmental and genetic factors. Particularly, DNA methylation biomarkers can be valuable in providing information about breast cancer risk and prognosis. While studies in other populations have identified DNA methylation biomarkers associated with breast cancer and its clinicopathological outcomes, almost no research exists on Latin-American breast cancers. The goal of this work was to identify DNA methylation patterns associated with clinico-pathological characteristics in Afro-Brazilian breast cancer tumors. DNA methylation marks from 48 frozen breast tumors of these patients treated at the National Cancer Institute Breast Cancer Hospital in Rio de Janeiro, Brazil were measured with the Illumina MethylationEPIC array (850K). After filtering, 739,883 DNA methylated positions (DMPs) were used for analysis. We performed consensus cluster analysis with the top 1000 most variable positions by resampling 80% of the data 10,000 times to assess cluster stability. Using hierarchical clustering with k=2, we identified two distinctive DNA methylation clusters. These clusters had different subtype profiles, disease staging and survival. Cluster one was enriched for triple negative breast cancers (Cluster 1 44% vs Cluster 2 0%, p=0.0006) and had a higher percentage of later stage disease (Cluster 1 66% vs Cluster 2 37%, p=0.06). It also had lower survival when compared to Cluster 2 (Hazard ratio (HR)=4.74, 1.45-15.51, p=0.01). Cluster 1 also had higher African ancestry (Cluster 1: 0.44±0.23 vs Cluster 2: 0.33±0.16, Wilcoxon Rank Sum (WRS) p=0.016), and Cluster 2 higher Indigenous American ancestry (Cluster 1: 0.05±0.06 vs Cluster 2: 0.10±0.08, WRS p=0.030). Overall DNA methylation was higher in Cluster 1 than 2 (Cluster 1: 0.81±0.16 vs Cluster 2: 0.64±0.15, WRS p=6.11x105). We identified 2,812 DMPs between the clusters with adjusted-p0.0001 and for which the absolute difference between clusters was higher than 0.2. Most DPM (60%) were hypermethylated in Cluster 1 with top hits in genes already know to be involved in breast carcinogenesis including ATF6 (Δβ=0.39, p=4.15x10-9), ITPR1 (Δβ=0.47, p=5.17x10-9), LIMK1 (Δβ=-0.30, p=1.44x10-9), WNK1 (Δβ=0.40, p=4.05x10-9) and CXCL12 (Δβ=-0.20, p=5.11x10-7). Our small sample limits the generalization of our findings; however, in future studies with additional samples we are planning to test the ability of this DNA methylation signature in identifying those with a greater risk of disease mortality. The research presented here expands the current knowledge on the molecular patterns and mechanisms underlying breast cancer in Afro-descendant women, particularly those from Latin-America. Citation Format: Lissette Delgado-Cruzata, Andres Luiz da Silva. Christianes, Diego J. Gomes de Paula, Jennifer Vieira Gomes, Marcelino Rau, Tatiana A. Simão, Monique de Souza Almeida. Lopes, Cynthia Chester. Cardoso, Sheila C. Soares-Lima. Specific DNA methylation patterns are associated with lower survival in Afro-Brazilian breast cancers abstract. In: Proceedings of the 18th AACR Conference on the Science of Cancer Health Disparities; 2025 Sep 18-21; Baltimore, MD. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2025;34(9 Suppl):Abstract nr C086.
Delgado‐Cruzata et al. (Thu,) studied this question.