Back

Tumor cell-intrinsic mtDNA instability drives cGAS-dependent temporal remodeling of the melanoma immune microenvironment

Bashaw, A.; Winters, A.; Lei, Y.; Araujo, L. F.; Vo, D. G. T.; Sekar, A.; Bachman, J. F.; Graves, P. S.; Pineda, M.; Rubinstein, J. C.; Shadel, G.; Bosenberg, M.; West, L. C.; West, P.

2026-09-30 immunology
10.64898/2026.09.28.755003 bioRxiv
Show abstract

Mitochondrial DNA (mtDNA) instability is an emerging feature of cancer, yet its biological effects on the tumor microenvironment (TME) and tumor progression remain incompletely defined. Using melanocyte-specific conditional silencing of mitochondrial transcription factor A (TFAM) in an autochthonous BrafV600E/Pten-/- melanoma model, we demonstrate that tumor cell-intrinsic mtDNA instability accelerates melanoma growth and metastatic dissemination in a cGAS-dependent manner. Transcriptomic profiling of early and late tumor stages revealed that mtDNA instability drives temporal remodeling of the tumor microenvironment (TME), characterized by an early type I interferon (IFN)-rich inflammatory state that licenses immunoregulatory programs as tumors progress. Tumor-associated macrophages emerged as the principal immune cell type reshaped by tumor cell-intrinsic mtDNA stress signaling, with early enrichment of inflammatory states and a shift toward efferocytosis-associated, tissue-adaptive states. The early mtDNA instability-associated macrophage signatures mapped to human melanoma states associated with favorable survival and therapy response, whereas the late signatures mapped to states associated with poor prognosis and therapy resistance across independent cohorts. Anti-PD-1 treatment delayed progression of mtDNA-instable tumors, consistent with a therapeutic sensitivity window associated with the early IFN-rich immune state. These findings establish tumor cell-intrinsic mtDNA instability as a driver of temporal immune-stromal remodeling that promotes melanoma progression and shapes therapeutic response.

Matching journals

The top 6 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.