Spatially resolved multi-omics single-cell analyses inform mechanisms of immune-dysfunction in pancreatic cancer

As pancreatic ductal adenocarcinoma (PDAC) continues to be recalcitrant to therapeutic interventions including poor response to immunotherapy, albeit effective in other solid malignancies, a more nuanced understanding of the immune microenvironment in PDAC is urgently needed. Using a spatially-resolved multimodal single cell approach we unveil a detailed view of the immune micromilieu in PDAC with specific emphasis on the correlation of immune subtypes with patient survival. We substantiate the exhausted phenotype of CD8 T cells and immunosuppressive features of myeloid cells, and highlight immune subpopulations with potentially underappreciated roles in PDAC, particularly CD4 T cell subsets presenting immunosuppressive phenotypes with varying modes of exhaustion. We also demonstrate the dynamic changes associated with transcriptional reprogramming of immune subtypes within adjacent normal tissue and tumor surrounding stroma, and further reveal striking differences between immune phenotypes in PDAC and lung adenocarcinoma, which at least partially explain their differential responsiveness to current immunotherapies and might have implications for the development of novel treatment strategies. Eight tissue sections were subjected to spatial transcriptomic analysis by using the Visium technology(30) (10x Genomics) according to the protocol of the manufacturer. Briefly, tissue slices approximately 10um in size were cut from fresh frozen OCT embedded tissue sections and placed on the Visium slide. The tissue sections were lysed in situ on the slide, mRNA was released which was captured by the oligonucleotides attached to the slide. Double stranded cDNA was prepared and finally eluted out from the slide into a tube. cDNA was amplified, cleaned and a dual indexed library was prepared. The library was sequenced at the DKFZ sequencing core facility using a NextSeq platform. Cell types in different regions of the tissue sections were annotated by pathologists and the annotation was transferred to the spatial transcriptomic experiments in the array of spots.