227P - Biomarkers for novel NK checkpoint inhibitor anti LLT1 antibody, ZM008: Patient transcriptome analysis

Background

C-type lectin superfamily protein LLT1 interacts with CD161 on NK cells and facilitates tumor immune escape. Disruption of LLT1-CD161 with anti LLT1 antibody, ZM008 activates immune cells and kills cancer cells. This study aims to identify new biomarkers to classify specific patient population in ZM008 clinical studies.

Methods

TCGA transcriptome data was used to understand expression of LLT1 in 33 cancers. Correlation analysis was performed to compare LLT1 expression with immune gene signatures. Genomic cancer markers (TMB, MATH, MMR and MSI) status were determined using “Maftools” and “MSI sensor”. Univariate Cox proportional hazard regression analysis (SURVIVAL) was used to measure patient survival.

Results

High expression of LLT1 was found in 11 different cancers. CIBERSORT analysis revealed significant immune cell infiltration (activated NK or CD8+ T) in TME. However, LLT1 expression in multiple cancers is negatively correlated with proinflammatory gene signatures (IL-2, IL-6, EOMES, and LAMP1 etc.), positively correlated with immune exhaustion markers PD1, LAG3, TIM3, TIGIT, ICOS and immune suppressive signals (high Treg cells and CD33+ MDSC; p <0.05). These data suggest immune suppressive “Cold tumor” phenotype associated with LLT1 expressing cancers. Further, immune check point predictors analysis in LLT1 positive tumors suggests positive correlation with high TMB score in COAD, UCEC and KICH; MSI score is elevated in PRAD and COAD; MMR genes expression is negatively correlated in COAD, KICH, BRCA, HNSC, KIRC, LUAD and PRAD; lastly, MATH score is correlated for GBM patients only (p <0.05). Subsequently, significant hazard ratios of 1.115, 2.114 and 1.067 were observed only for COAD, KICH and KIRC, respectively (p < 0.05).

Conclusions

The transcriptome data analysis revealed LLT1 expression is associated with immune repressive TME in multiple cancers, suggesting therapeutic significance of ZM008. Multiple anti-inflammatory gene signatures could be useful in stratifying patients for ZM008 treatment. In addition, TMB & MSI scores and MMR gene expression could be monitored as potential biomarkers in phase I clinical trials with ZM008.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

Zumutor Biologics.

Funding

Zumutor Biologics.

Disclosure

I. Garcia Ribas: Other, Personal, Advisory role, Independent senior medical advisor: Zumutor. All authors have declared no conflicts of interest.