Tumor associated macrophages as key contributors and targets in current and future therapies for melanoma

This study aims to rigorously explore the potential causal relationships among gut microbiota (GM), immune cells, and melanoma skin cancer among participants from Europe, where this disease exhibits significant prevalence and profound societal impact. Using the genome-wide association analysis database, a double-sample Mendelian randomization (MR) analysis was drawn upon to investigate GM, immune cells, and melanoma skin cancer. The inverse variance weighted approach was applied to estimate the causal connections among these variables. A two-step MR analysis was employed to quantitatively gauge the impact of immune cells mediated GM on melanoma skin cancer. To address potential sources of bias, such as pleiotropy and heterogeneity, multiple analytical techniques were integrated. The MR analysis pinpointed 6 GM taxa related to either an augmented or declined risk of late-stage melanoma skin cancer. In the same vein, 32 immune cell phenotypes were noticed as correlates with modified risk of melanoma skin cancer. Our study also implies that the probable association between GM and melanoma could be facilitated by 5 immune cell phenotypes. The findings of our study underline certain GM taxa and immune cells as potential influencers on the onset and development of melanoma skin cancer. Importantly, our results spotlight 5 immune cell phenotypes as potential agents mediating this association.

show abstract

High-accuracy prediction of 5-year progression-free melanoma survival using spatial proteomics and few-shot learning on primary tissue

Möller,

Ostalecki,

Hartebrodt

et al. 2024

Preprint

View full text Add to dashboard Cite

Emerging techniques in imaging-based spatial proteomics (ISP) enable in-depth insights into the architecture and protein abundance of tissue(s). Explainable machine learning (xML) models promise to yield substantial advances in ISP data-based diagnosis and prognosis. However, a clinical application of these new possibilities predicting the course of a tumor has not been suggested yet. Here, we use a few-shot learning workflow on histological multi-antigen images to predict 5-year progression-free survival (PFS) in melanoma. We address the problem of a relatively small cohort (n = 22), by utilizing a pre-trained convolutional neural network (CNN) model, which we further pre-train on a proxy task for which more samples were available (n = 39) before fine-tuning for PFS prediction. Our approach yielded a model achieving an accuracy of more than 90%, outperforming baseline models trained on clinical data by around 10%. Using an xML technique, we identified immune infiltration and proteins associated with tumor progression as crucial predictors. This indicated that our models' PFS predictions are not only highly accurate but also grounded in a relevant biological background.

show abstract

Tumor associated macrophages as key contributors and targets in current and future therapies for melanoma

Cited by 7 publications

References 157 publications

Focal adhesion kinase-mediated interaction between tumor and immune cells in the tumor microenvironment: implications for cancer-associated therapies and tumor progression

Focal adhesion kinase-mediated interaction between tumor and immune cells in the tumor microenvironment: implications for cancer-associated therapies and tumor progression

A two-step, two-sample Mendelian randomization analysis investigating the interplay between gut microbiota, immune cells, and melanoma skin cancer

High-accuracy prediction of 5-year progression-free melanoma survival using spatial proteomics and few-shot learning on primary tissue

Contact Info

Product

Resources

About