High-plex proteomic prognostic marker discovery for patients with pancreatic cancer adenocarcinoma using digital spatial profiling.

Abstract

611 Background: The long-term outcomes following surgical resection for Pancreatic Ductal Adenocarcinoma (PDAC) remain poor, with only 20% of patients surviving 5 years after pancreatectomy. With its immune-privileged nature, starting from the early pre-neoplastic state, it appears to escape easily from the antitumor immune response. Despite rationale for targeting immune pathways in PDAC, there has been little benefit observed at this point. The aim of the current study was to interrogate immune landscape of PDAC utilising the Nanostring GeoMx Digital Spatial Profiler (DSP), a state-of-the art analysis platform enabling Hi-plex proteomic characterisation whilst maintaining tumor microenvironment (TME) topographical features. Methods: We assessed Formalin Fixed Paraffin Embedded (FFPE) tumor samples from 28 treatment-naive PDAC cases represented in a multi-regional tissue microarray (TMA) for which extensive IHC, molecular, genomic characterisation and clinicopathological follow-up data is available. Following multiplex IHC staining for DAPI, panCK, aSMA and CD3 to guide region selection, we employed the GeoMx DSP system (NanoString) to select regions within multiple TMA cores. We quantified 60 immune markers simultaneously in multiple tissue compartments defined by immunofluorescence co-localization including (tumor [panCK+ve], immune stroma (PanCK-ve]. Data analysis was performed by a combination of DSP analysis suite and custom R pipeline. Results: The spatially informed variable assessment by DSP was validated by both regression and variable prognostication compared with IHC for stromal CD3, CD8 CD68 in near serial TMA PDAC sections. Unsupervised analysis of DSP proteome data in the panCK-negative regions identified an immune poor group associated with shorter Overall Survival (OS) (13.0 versus 31.1months, P = 0.005). When transcriptomic subtype was considered, the checkpoint inhibitor B7-H3 was significantly upregulated in the squamous subtype tumours versus the classical group (Log2: 1.63, P = 0.001). Patients with high B7-H3 expression, using a median expression cut-off, were associated with shorter OS on multivariate analysis (Hazard Ratio: 4.16, P = 0.01) including lymph node and resection margin status, a finding that was validated in an external cohort at the transcriptome level. Conclusions: This pilot scale discovery study shows the potential of the Nanostring DSP technology in the identification of spatially-informed biomarkers with prognostic relevance in biopsy sized samples from treatment-naive PDAC. We identified a number of relevant candidate immune predictors in spatial context that are currently undergoing validation in larger independent cohorts and the neoadjuvant setting. Future studies will apply this technology to pre- and post-treatment biopsy samples.