Back

Anti-CMV IgG Titre Determines Organ-Specific Protection Towards Immune Checkpoint Blockade Induced Toxicities

Milotay, G.; Little, M.; Mackay, S.; Muldoon, D.; Huynh, A.-N.; Farkas, A.; Sun, S.; Niu, G.; Tong, O.; Taylor, C. A.; Middleton, M. R.; Klenerman, P.; Fairfax, B. P.

2025-11-09 oncology
10.1101/2025.11.07.25339789 medRxiv
Show abstract

Seropositivity for human cytomegalovirus (CMV) is associated with protection against severe (Grade 3+) immune-related adverse events (irAEs) post Immune Checkpoint Blockade (ICB). Here, in a prospectively recruited pan-cancer ICB-treated cohort (n=448 patients), we identify a novel relationship between the relative baseline titre of anti-CMV IgG antibody and organ-specific protection against irAEs. In CMV seropositive patients, whereas anti-CMV IgG antibody level is stable over years, increased pre-treatment titre is independently associated with reduced all-organ Grade 3+ irAEs. This pan-organ association sub-divides into organ-specific effects; protection against non-colitis irAEs being observed only in those with an above median titre of anti-CMV IgG antibody (P High titre vs. CMV-=2x10-5), whereas CMV-related protection against colitis is unrelated to titre (P Low titre=0.0016, P High titre=0.0012). We demonstrate that anti-CMV IgG antibody titre is robustly related to peripheral immune subset composition, with higher anti-CMV IgG titre associated with elevated CD4+ and CD8+ T cell cytotoxicity and effector cell expansion. Conversely, CMV seropositivity is associated with general reduced circulating Tregs irrespective of titre. This work reinforces the importance of CMV in modulating ICB-induced irAEs, revealing a complex relationship between degree of humoral anti-CMV immunity and organ-specific protection, whilst further highlighting the clinical utility of CMV serology in predicting ICB induced irAEs.

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.