Search

BackgroundAndrogen receptor (AR) is considered marker associated with better prognosis within hormone receptor positive tumors. Its role in triple negative tumors however is controversial showing both better and worse prognosis and different methods are used for identification of AR driven tumors. Conflicting results could be due to intrinsic molecular differences or scoring method for AR positivity. We attempted to develop an AR driven gene score and examined its utility in subtypes of breast cancer (BC).MethodsA bioinformatic pipeline was constructed and applied on publicly available microarray data sets obtained from AR positive BC cell lines treated with dihydrotestosterone (DHT). Expression levels of genes identified through the pipeline along with set of epithelial mesenchymal transition (EMT) markers, proliferation associated genes and enzymes involved in intracrinal androgen metabolism was evaluated in a cohort of Indian Breast cancer patients. Tumors were divided into AR high and low based on the gene score and association with clinical parameters, circulating androgens, disease free survival, proliferation and EMT markers were examined, all results were further validated in external public datasets.ResultsAR driven gene score was calculated as average expression of the 6 genes selected through bioinformatic analysis. 53% (133/249) tumors were classified as AR high by the gene score and had significantly better clinical parameters such as higher age, smaller tumor size, lower grade and lower proliferation. Tumors with high AR driven gene score had significantly better disease-free survival (mean survival time of 86.13 vs 72.69 months, log rank p=0.032) when compared to the AR low tumors. A subset analysis within TNBC (N=66) showed 36% (24/66) were AR high and had significantly higher expression of EMT markers (p=0.024). Though circulating levels of total testosterone was not different between the groups, intratumoral levels of 5 alfa reductase (SRD5A1) was significantly high in tumors with high AR driven score.ConclusionRole of AR in breast cancer is debatable and difficult to decipher with protein detection alone. Our results support the context dependent function of AR in driving better prognosis, while identifying its role in driving EMT within TNBC tumors.

The complexity of the ER-negative subtype of breast cancer arises due to the heterogeneous nature of the disease rendering them more aggressive and this poses a challenge to effective treatment and eventually the prognosis of the patients. We have explored the miRNA regulation of altered molecular signatures and the effect on tumour progression in ER-negative breast cancer. Using breast tumour specimens, gene expression data from public datasets and in-vitro and in-vivo model systems we have shown that low-levels of miR-18a in ER-negative tumours drives enrichment of hybrid Epithelial/Mesenchymal (E/M) cells with luminal attributes. On inhibition of miR-18a in ER-negative breast cancer cell lines, the cells showed traits of increased migration, stemness and drug-resistance. miR-18a/low tumours were also associated with increased expression of genes associated with EMT, stemness, drug resistance and immune-suppression. Further analysis of the miR-18a targets pointed out at a possible HIF-1α mediated signalling in these tumours. HIF-1α inhibition reduced the enrichment of the hybrid E/M cells and decreased the migratory ability of miR-18a/low cells. Our study reports for the first time a dual role of miR-18a in breast cancer that is subtype specific based on hormone receptor expression and a novel association of low miR-18a levels and enrichment of hybrid E/M cells. The results highlight the possibility of stratifying the ER-negative disease into clinically relevant groups by analysing epigenetic signatures.

BackgroundTriple-negative breast cancer (TNBC) is a heterogeneous disease with a significant clinical challenge. TNBC alarmingly comprises 25-30% of breast cancers in India compared with only 10-15% in the West. However, immunotherapy was approved for high-risk early-stage TNBCs in the West. Hence, a long-standing question is whether Indian TNBCs immunologically and clinically resemble Western TNBCs such that they respond similarly to immunotherapies. Here we sought to elucidate the immune landscape of Indian TNBCs for the first time, compare them to Western disease and associate them with clinical parameters, cellular types/signaling, and immunotherapy response.MethodsWe profiled 730 immune genes in 88 retrospective Indian TNBC samples using NanoString platform, clustered them into subtypes using a machine-learning approach, and compared them with Western TNBCs (n=422; public datasets). Subtype-specific gene signatures were identified, followed by clinicopathological, immune cell type, and pathway (multiomics) analyses. We also assessed responses to (cross-cancer) immunotherapy. Tumor-infiltrating lymphocytes (TILs) and pan-macrophage marker were evaluated using hematoxilin-eosin staining and immunohistochemistry, respectively.ResultsWe identified three robust and similarly distributed TNBC immune transcriptome subtypes (Subtypes-1-3) in Indian women, and they are represented correspondingly in Western TNBCs and associated with well-known TNBC subtypes. Irrespective of the ethnicity, Subtype-1 harbored tumor microenvironmental and anti-tumor immune events associated with smaller tumors, younger age, and a better prognosis. Subtype-1 mainly represented basal-like/claudin-low breast cancer and immunomodulatory TNBC subtypes. Subtype-1 showed an increase in a cascade of events, including damage-associated molecular patterns, acute inflammation, Th1 responses, T-cell receptor-related and chemokine-specific signaling, antigen presentation, and viral-mimicry pathways. Subtype-1 was significantly (p+ regulatory cells, basal-like/mesenchymal subtypes, and an intermediate prognosis. Subtype-3 patients expressed innate immune genes/proteins, including those representing macrophages and neutrophils, and had poor survival.ConclusionWe identified three immune-specific TNBC subtypes in Indian patients with differential clinical and immune behaviors, which largely overlapped with the Western TNBC cohorts. This study suggests cancer immunotherapy in TNBC may work similarly in both populations. Hence, this may expedite pharmaceutical adoption of immunotherapy to Indian TNBC patients.