Back

Decoy Antibodies Block Extracellular HSP70, Prevent Self Signaling and Inhibit Melanoma Cell Survival

Ben-Shalom, N.; Parikh, S.; Abramovitz, L.; Yefet, R.; Manich, P.; Tussia-Cohen, D.; Moskowitz, O.; Parikh, R.; Melamed, S.; Polonsky, K.; Apcher, S.; Hagai, T.; Cohen, M.; Wolf, Y.; Shapira-Frommer, R.; Levy, C.; Freund, N.

2025-12-29 immunology
10.64898/2025.12.29.696857 bioRxiv
Show abstract

Melanoma cells actively secrete melanosomes-large, extracellular vesicles (EVs) enriched with oncogenic factors that reprogram the tumor microenvironment, enhance self-signaling, and promote tumor growth. Despite their abundance and immunogenic potential, humoral responses to melanoma-derived melanosomes remain unexplored. Here, we identify a novel immune surveillance mechanism in which melanosome-elicited decoy antibodies target melanoma-derived melanosomes by binding to the extracellular form of heat shock protein 70 (HSP70), a chaperone broadly implicated in cancer cell survival and stress adaptation. Anti-HSP70 decoy antibodies potently and effector-independently inhibit growth and survival of both murine and human melanoma cells and suppress key transcriptional programs involved in proliferation, cytoskeletal dynamics, and metabolism. In a preclinical B16 melanoma model, prophylactic administration of decoy monoclonal antibodies Mel322-34 and Mel321-35 conferred significant survival benefits of 27% and 48%, respectively. Strikingly, anti-HSP70 antibodies were enriched in the sera of melanoma patients achieving complete responses to immune checkpoint blockade, in contrast to non-responders with progressive disease. Collectively, our findings uncover a novel EV-antibody axis as a promising avenue to block cancer-promoting signaling pathways. Decoy autoantibodies targeting the extracellular form of HSP70 advance the understanding of tumor-intrinsic vulnerability and promote biomarker-driven immunotherapy in melanoma.

Published in Cell Reports (predicted rank #8) · training set

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.