Back

LIMD1 Loss as an Early Driver of PD-L1 Upregulation and Immune Evasion in Lung Cancer

Shah, K. M.; Kennedy, P. T.; Black, J.; Litchfield, K.; Thakkar, K.; Contreras-Gerenas, M. F.; Brooksbank, K.; Yuan, O.; Grevitt, P.; Charrot, S.; Davies, J.; Dahal, L. N.; Lagos, D.; McGranahan, N.; Sharp, T. V.

2025-02-08 cancer biology
10.1101/2025.02.06.636805 bioRxiv
Show abstract

LIMD1 is a tumour suppressor gene frequently lost in non-small cell lung cancer (NSCLC), but its role in cancer-immune cell interactions remains unexplored. Here, we demonstrate that LIMD1 loss results in upregulation of the key immune checkpoint protein PD-L1. Using multi-region sequencing from the TRACERx dataset, we identify that LIMD1 loss is clonal in over 80% of squamous cell carcinoma (LUSC) and 40% of lung adenocarcinoma (LUAD) cases, correlating with increased PD-L1 expression. LIMD1 deficiency results in upregulation of basal and IFN{gamma}-induced PD-L1 expression in NSCLC cells and, consistent with its early loss during oncogenesis, in primary human small airway epithelial cells. Mechanistically, we demonstrate that LIMD1 interacts with the E3 ubiquitin ligase ARIH1 to mediate efficient PD-L1 ubiquitination and degradation, a process that is significantly impaired in LIMD1-deficient cells, resulting in increased PD-L1 stability. As a consequence, LIMD1-deficient tumour cells suppressed CD8+ T cell activation in vitro, and blockade of PD-L1 reversed this suppression. Clinically, we show that LIMD1 loss is associated with enhanced response to immune checkpoint inhibitors (ICIs) in NSCLC patient cohorts, revealing a novel cancer cell-intrinsic correlation of ICI efficacy. Our results uncover a tumour suppressor-mediated mechanism of PD-L1 expression and pave the way for stratified immunotherapy approaches in LIMD1-/- NSCLC.

Matching journals

The top 4 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.