Back

Deep phenotyping of the prostate tumor microenvironment reveals molecular stratifiers of relapse and links inflammatory chemokine expression to aberrant metabolism

Krossa, S.; Andersen, M. K.; Midtbust, E.; Wess, M.; Kiviaho, A.; Sharma, A.; Viset, T.; Storkersen, O.; Giskeodegard, G. F.; Nykter, M.; Urbanucci, A.; Rye, M. B.; Tessem, M.-B.

2024-07-11 cancer biology
10.1101/2024.05.13.593822 bioRxiv
Show abstract

Understanding the molecular characteristics and changes of the tumor microenvironment (TME) associated with aggressive prostate cancer (PCa) is essential for precise diagnosis and treatment. We interrogated spatially resolved integrated transcriptomics and metabolomics data to build molecular strafiers discriminating patients with aggressive, potentially relapsing, and metastasizing PCa. We report a relapse associated (RA) gene expression signature characterized by upregulated immune response related gene expression scoring high in cancer, stroma, and glandular tissue of relapsing patients. Further, we identified a signature specific to a distinct sub-group of morphologically non-cancerous glands in prostate tissue from patients with relapsing cancer. This signature, named chemokine-enriched-gland (CEG) signature, was characterized by upregulated gene expression of pro-inflammatory chemokines. Glands with a high CEG score were enriched for club-like cells and surrounding stroma was infiltrated by immune cells. Tissue regions scoring high for both CEG and RA signatures were associated with reduced levels of citrate and zinc and loss of normal prostate secretory gland functions via reduced expression of genes necessary for citrate secretion. In summary we report that aggressive PCa is associated with an increased inflammatory status linked to chemokine production and club-like cell enrichment in potentially pre-cancerous prostate glands displaying an aberrant metabolism.

Matching journals

The top 11 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.