Cancer cachexia, affecting up to 80% of patients with advanced cancer, is characterized by metabolic and inflammatory dysregulation driven by tumour- and host-derived factors. Although cytokines are central to cachexia pathogenesis, their circulating profiles and transcriptional relevance across models and sexes remain incompletely defined.

We characterized plasma cytokines and skeletal muscle inflammatory signatures in multiple cachexia models, including Lewis Lung Carcinoma (LLC), Colon-26 (C26) and Apc Min/+ mice. Animals were monitored for 4 weeks (LLC), 25 days (C26) or until ∼20 weeks of age (Apc Min/+) per model appropriate endpoints.

Publicly available RNA-sequencing datasets from gastrocnemius (LLC) and tibialis anterior (C26 and Apc Min/+) muscles were integrated with plasma profiling using a Mouse Cytokine/Chemokine 32-Plex assay. To enhance accessibility and reproducibility, we developed an interactive Shiny application – thecachexiatlas/Cytokine Explorer – enabling dynamic visualization of cytokine levels, transcriptional signatures and phenotype correlations across models and sexes.

Cytokine receptor encoding genes (Il6ra, Il4ra, Csf3, Osmr) and inflammatory pathways were consistently enriched in skeletal muscle of cachectic animals across models and sexes. Among circulating cytokines, granulocyte colony-stimulating factor (G-CSF) was uniquely elevated in all models and both sexes.

Elevated G-CSF levels exhibited stronger negative correlations with tibialis anterior muscle mass than the commonly used cytokine interleukin (IL)-6, particularly when expressed as the ratio of G-CSF to the anti-inflammatory cytokine IL-13. The G-CSF:IL-13 ratio may represent a robust global biomarker of cachexia severity.

The Cytokine Explorer Shiny app provides an open, interactive platform to facilitate hypothesis generation and advance research in cancer cachexia.