1948P Characterising the peripheral myeloid response to immune checkpoint blockade

Immune checkpoint blockade can offer survival benefit for those with advanced or metastatic melanoma. However, only a proportion of patients will achieve clinical response and there is risk of side effect-associated morbidity. The circulating myeloid population is associated with a poor prognosis in the context of melanoma and CTLA4 blockade. Recent evidence has shown a key role for PD-1 on myeloid cells in inhibiting the anti-tumour immune response. Despite this, the myeloid response to checkpoint inhibitor therapy has not been well-characterised. We have characterised transcriptomic expression using bulk RNA-seq in CD14+ cells in response to checkpoint inhibitor therapy across 97 patients. To further characterise the response at a single cell level we performed single-cell sequencing (scRNA-seq) of peripheral monocytes across eight patients prior to and after treatment. Transcripts differentially expressed in response to single versus combination immune checkpoint blockade are identified, and a distinct transcriptomic profile is characterised in patients versus healthy controls. Notably, a number of expression associations are made with clinical outcome in terms of treatment associated side effects, confirming the importance of monocytes in predicting clinical response. scRNA-seq reveals distinct monocyte subsets, of which one is characterised by over-expression of human leukocyte antigen (HLA) and interferon-induced genes, is specific to the post-treatment state. We use deconvolution to identify these monocyte subsets in bulk RNA sequencing (RNA-seq) data across a large patient cohort. In summary, characterisation of both bulk RNA-seq and scRNA-seq reveals novel insights into inter-individual variation within the peripheral monocyte population in response to checkpoint inhibitor therapy and has potential ramifications for early identification of non-responding patients. Application of these subsets to bulk RNA-seq data may provide opportunities to explore association between monocyte subsets with clinical outcome.