Research work on An Epigenetics-based Signature Prognostic Model For Breast Cancer Survival by Dr. Nita Parekh and her students were awarded the  Best Paper Award (Track 7 Online) at the 8th IEEE Pune Section International Conference (IEEE PUNECON 2025) held from 12 – 15 December 2025. The theme of the conference was Generative Futuristic Systems. The conference was jointly organized by Savitribai Phule Pune University and the IEEE Pune Section in Pune, Maharashtra. 

Here is the summary of the paper as explained by the authors:

Epigenetic alterations have gained significant attention as biomarkers for diagnostic, prognostic and predictive purposes in various cancers. While numerous studies have investigated the role of DNA methylation and microRNAs as regulators of gene expression independently in cancer, integrative analyses of the two major epigenetic factors remain limited. In this study, we considered differentially methylated positions and differentially expressed microRNAs to propose diagnostic and prognostic markers for breast cancer. Analysis of the TCGA-BRCA data based on the status of Estrogen, Progesterone, and HER2 receptors identified 63 epigenetic biomarkers (39 CpG sites, 24 miRNAs). Classification of breast cancer into Luminal, Her2-enriched and Triple Negative molecular subtypes using the 63 markers achieved a high accuracy of 98.1% and Mathew’s correlation coefficient of 0.95. To investigate their role in prognosis, we constructed a multivariate Cox proportional hazard regression model for overall survival. This resulted in a novel 17- epigenetic-based prognostic signature. Time-dependent ROC analysis showed robust predictive performance, with 3- and 5-year AUCs of 0.75 and 0.72 on the test set for stratifying the patients in high-risk and low-risk groups. The proposed signature includes oncogenes and tumor suppressor genes involved in estrogen-mediated signaling, AKT/mTOR, WNT/PCP, FAK/Src/PI3K/AKT, MAPK/ERK, and estrogen-driven pathways, reflecting roles in proliferation, migration, invasion, and survival. Thus, using the multi-omics integrative approach, we find the prognosis prediction with risk stratification aligning with clinical parameters such as age, tumor stage, receptor status, and metastasis, thereby providing early diagnostic and prognostic markers for precision therapy.

January 2026