Advancing technologies in Digital Spatial Profiling (DSP) enable Whole Transcriptome Analysis (WTA) within the context of specific histopathologic features. These DSP approaches are well suited to characterizing solid tumors, which are composed of heterogeneous neoplastic cells, immune cells, and stromal populations. Using DSP, morphologically and spatially distinct cell populations can be separately evaluated for differences in transcriptome or protein expression.
Mouse models of cancer are particularly attractive to DSP approaches, due to their well-characterized and repeatable disease progression. The KPC model of pancreatic ductal adenocarcinoma (PDA), driven by the Kras and Trp53 transgenes, follows a consistent, stepwise process of neoplastic transformation, separable into various stages based on well-described histopathologic features. Pancreatic lesions occurring in KPC mice closely recapitulates the biology of human PDA in terms of histopathology and clinical features. Depending on the age of a KPC mouse, normal pancreatic acini and ducts are often present alongside a spectrum of benign PanIN lesions, which progress along successive grades based on cytologic and architectural atypia (PanIN 1a, 1b, 2, 3); this closely parallels human histopathology. PanIN lesions are thought to progress to invasive PDAC because PanIN appears before PDA in KPC mice. PDA tumors in KPC mice also harbor desmoplastic stroma and aggressively metastasize to regional and distant organs. By leveraging DSP, differences in transcriptome can be compared for each progressive grade. Further, differences within malignant, locally invasive, and metastatic lesions can be evaluated.
The NanoString GeoMX DSP technology became commercially available in 2019 . This platform provides spatial analysis to quantify either (1) the expression of a panel of proteins or (2) Whole Transcriptome Analysis (WTA) within regions of interest as defined by histomorphologic features. We are interested in WTA, which will be discussed briefly. First, 5um thick sections are permeabilized to expose RNA, which is hybridized in situ to gene-specific probes containing a photocleavable molecular barcode. Second, the tissue is stained with fluorescent antibodies, typically an epithelial marker and stromal marker, to reveal morphology or markers of interest. Third, annotations are made digitally, which represent unique histopathological or spatial features. Fourth, each separate annotation is individually exposed to UV light, which liberates the photocleavable barcodes from the bene-specific probes. Fifth, the cleaved barcodes are aspirated and quantified with the NanoString nCounter system or using next-generation sequencing. Panels of probes are available from Nanostring that enable the detection of up to 18,000 genes simultaneously.
Some limitations to keep in mind with this technology are that 200 cells – within a single ROI – are recommended to clear the limit of detection, although highly expressed transcripts have been detected in as few as 21 cells (Merritt et al, 2020). Also, the resolution of detection is approximately 10-30 um. Whole-exome sequencing and array comparative genomic hybridization have provided some insight into subclonal evaluation of PDA tumors within KPC mice (Niknafs, 2019); however, such analyses are limited due to the inclusion of whole tumor without spatial information.