Abstract 72P
Background
Tumor-infiltrating immune cells could serve as biomarkers and provide information for potential combined immunotherapy strategies. This study characterized tumor immune microenvironment patterns associated with PD-1 blockade resistance among skin cancer subtypes.
Methods
We analyzed 18,238 longitudinal pre- and post-treatment immune cells, including 11,969 CD8 T cells, 3,678 regulatory T cells (Tregs), and 2,591 macrophages from dissociated tumor samples of 12 skin cancer patients (5 melanoma, 5 basal cell carcinoma [BCC] and 2 squamous cell carcinoma [SCC] patients) who did not respond to anti-PD-1 therapy. Well-integrated gene-barcode matrix was subjected to PCA and the first 30 PCA components was used for Uniform Manifold Approximation and Projection dimension reduction. We determined high frequency genes (HFGs) as the top 200 frequently genes expressed in an immune cell type of each cancer subtype.
Results
The comparison between melanoma and BCC (non-melanoma skin cancer) revealed clear clustering of CD8 T cells, Tregs, and macrophages by cancer type at both pre- and post-treatment timepoints. Pseudo-time analysis showed proximity for phenotypic transition of immune cells within the same cancer subtype but distinct trajectories between two subtypes. We identified differentially expressed genes (DEGs) between melanoma and BCC for each immune cell type, for instance CD38, IFNG, SIRPG, PTPN6 for CD8+ T cells, CD200, CD38, ELMO2, HEXB for Tregs, and CCL17, IL23A, MMP9, CCL20, POLD2 for macrophages. Gene set enrichment analysis found the DEGs were enriched for positive or negative immune response and multiple metabolic pathways. For each immune cell type, around 35% of HFGs were not co-expressed in melanoma and BCC. Similar analyses were performed between non-melanoma skin cancers BCC and SCC, which also showed cancer-specific effects but to a lesser degree.
Conclusions
The PD-1 blockade resistance landscapes for CD8 T cells, Tregs, and macrophages varied among skin cancer subtypes, highlighting the need to explore cancer-specific resistance mechanisms and to develop the optimal combination strategy for each subtype.
Legal entity responsible for the study
Anlin Li.
Funding
National Students’ Innovation and Entrepreneurship training program (Grant n. 201910571001); Special Funds for the Cultivation of Guangdong College Students’ Scientific and Technological Innovation (Grant n. PDJH2019A0212); Guangdong Medical University College Students’ Innovation Experiment Project (Grant n. ZZZF001).
Disclosure
All authors have declared no conflicts of interest.