1761MO - Defining subset-wise myeloid responses to immune checkpoint blockade in melanoma

Background

Immune checkpoint blockade (ICB) is associated with improved survival in patients with metastatic melanoma (MM). Despite this, many patients have limited clinical benefit, and there is ongoing attrition amongst responders. Moreover, ICB is associated with auto-immune toxicity. Peripheral biomarkers can help early identification of non-responders, as well as assist in treatment stratification. Recent evidence suggests that myeloid populations may play a role in modulating ICB response, but this has yet to be full defined at the transcriptomic level.

Methods

We performed bulk RNA-seq of peripheral CD14⁺ cells across 114 pre-treatment, 96 post-treatment MM patient samples and 45 healthy donors. Single-cell RNA-sequencing (scRNA-seq) across a total of 24,457 peripheral CD14⁺ cells from eight paired MM patient samples and three healthy donors was performed.

Results

Bulk RNA-seq data demonstrates distinct cancer-associated transcriptional profiles in monocytes. ICB is associated with induction of pathways including IFNγ signalling and antigen presentation, with more pronounced responses with combination ICB (cICB: anti-PD1 and anti-CTLA-4). Transcriptional CD14⁺ profiles are correlated with clinical outcome. At the single cell level, ICB leads to maturation within the classical monocyte compartment, and acquisition of a non-classical monocyte phenotype. Subset-wise monocyte signatures at baseline and following ICB are associated with clinical outcome. In particular, a myeloid-derived suppressor cell (MDSC) signature is negatively associated with the development of auto-immune adverse effects, and is modulated by cICB.

Conclusions

This work confirms ICB-dependent modulation of circulating monocytes, with novel observations regarding subset-wise myeloid responses. This demonstrates the potential clinical utility of circulating immune populations as diagnostic, predictive and prognostic markers in patients with MM.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

Cancer Research UK, Wellcome Trust, University of Oxford.

Disclosure

All authors have declared no conflicts of interest.