Back

Liquid-biopsy transcriptomic profiling uncovers molecular mediators of resistance to androgen receptor signaling inhibition in lethal prostate cancer

Zhang, J.; Zimmermann, B.; Galletti, G.; Halabi, S.; Gjyrezi, A.; Yang, Q.; Gupta, S.; Verma, A.; Sboner, A.; Anand, M.; George, D. J.; Gregory, S. G.; Hong, S.; Pascual, V.; Mavragani, C. P.; Antonarakis, E. S.; Nanus, D. M.; Tagawa, S. T.; Elemento, O.; Armstrong, A. J.; Giannakakou, P.

2021-11-02 oncology
10.1101/2021.11.01.21265757 medRxiv
Show abstract

Androgen receptor signaling inhibitors (ARSi) are a mainstay for patients with metastatic castration-resistant prostate cancer (mCRPC). However, patient response is heterogeneous and the molecular underpinnings of ARSi resistance are not well elucidated. Here we performed transcriptome analysis of circulating tumor cells (CTCs) and peripheral blood mononuclear cells (PBMC) in the context of a prospective clinical trial of men with mCRPC treated with abiraterone (Abi) or enzalutamide (Enza). CTC RNA-sequencing identified that RB loss and enhanced E2F signaling along with BRCA loss transcriptional networks were associated with intrinsic ARSi resistance, while an inflammatory response signature was significantly associated with acquired resistance. Transcriptomic analysis of matching PBMCs identified enrichment of inflammasome gene signatures indicative of activated innate immunity at progression, with concurrent downregulation of T and NK cells. Importantly, CTC gene signatures had a significant positive association with circulating immune macroenvironment (CIME) signatures. Taken together, these data demonstrate that liquid biopsy transcriptomics can identify molecular pathways associated with clinical ARSi resistance paving the way for treatment optimization in patients with mCRPC.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.