Application of the AMOCATI R workflow to tumor transcriptomic data delineates the adverse effect of immune cell infiltration in immune-privileged organs
Regnier, P.; Cagnard, N.; Podsypanina, K.; Darrasse-Jeze, G.
Show abstract
Immune cells are present inside tumor tissue and can alter tumor growth. Expression profiles of human tumors hold transcripts from cancer cells and their microenvironment, including the infiltrating immune cells. Few standardized methods examine tumor immunobiology relying only on tumor transcriptome data. Using a new in-house developed R analysis workflow called AMOCATI, we classified 43 cancer types from 11,176 patients according to the degree of infiltration by 18 distinct immune cell subsets, measured by the abundance of their transcriptomic signature, and calculated its effect on the disease outcome. In about half of cancers affecting organs without immune privilege, immune cell infiltration has beneficial effects. In contrast, immune infiltration in cancers of immune-privileged organs (eye, testis and brain) confers poor prognosis. Moreover, transcriptional evidence of increased immune cell activity in immune-privileged cancer sites is associated with bad prognosis. Thus, our results suggest that the effect of immune infiltration may depend on the origin of the primary tumor. SIGNIFICANCEOur in-house developed computational R approach AMOCATI allows to easily download public transcriptomic and clinical data, classify and analyze them. AMOCATI permitted us to define gene expression signatures associated with short- or long-term survival from 11,176 untreated patient unsorted biopsies in 43 types of cancer. We present the level of infiltration of 18 types of immune cell subsets transcriptomic signatures and 50 immune-related pathways in all these cancers Correlation between immune infiltration of the tumor and survival establishes a link between tumor tissue of origin and the overall effect of immune infiltration on survival. Immune cell infiltration in tumors from immune privileged organs correlate with shorter survival. Contrary to what we observe in hot tumors, biological pathways of immune response are associated with a short-term survival profile in these cancers.
Matching journals
The top 11 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Self-Supervised Learning Reveals Clinically Relevant Histomorphological Patterns for Therapeutic Strategies in Colon Cancer 96%
- Dissecting tumor cell programs through group biology estimation in clinical single-cell transcriptomics 96%
- Tumour gene expression signature in primary melanoma predicts long-term outcomes: A prospective multicentre study 96%
Similar papers in this journal
Similar papers in this journal
- Pan-cancer identification of clinically relevant genomic subtypes using outcome-weighted integrative clustering 96%
- DeepProg: an ensemble of deep-learning and machine-learning models for prognosis prediction using multi-omics data 96%
- Beyondcell: targeting cancer therapeutic heterogeneity in single-cell RNA-seq 94%
Similar papers in this journal
- Overcoming Resistance to BRAFV600E Inhibition in Melanoma by Deciphering and Targeting Personalized Protein Network Alterations 94%
- Explainable, federated deep learning model predicts disease progression risk of cutaneous squamous cell carcinoma 94%
- Multimodal single-cell profiling reveals cancer crosstalk between macrophages and stromal cells in poor prognostic cholangiocarcinoma patients. 94%
Similar papers in this journal
- The Epithelial and Stromal Immune Microenvironment in Gastric Cancer: A Comprehensive Analysis Reveals Prognostic Factors with Digital Cytometry 95%
- Omics integration analyses reveal the early evolution of malignancy in breast cancer 94%
- Use of high-plex data reveals novel insights into the tumour microenvironment of clear cell renal cell carcinoma 93%
"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.