Abstract 2059: Machine learning integration of transcriptome-wide spatial sequencing data and ultra-high plex spatial proteomic data enables the prioritization of cancer drug targets

Abstract

Abstract Spatial omics technologies are producing an unprecedented amount of ultra-high plex in situ data that are promising to revolutionize cancer prognoses and treatments. Analytical solutions to integrate big spatial data are, however, lagging relative to the rapid technology development and this hinders discoveries into the pathological processes underlying cancer initiation and progression. Here we present STimage, a machine learning approach to flexibly combine transcriptome-wide spatial sequencing data with single-cell protein-based spatial phenotyping (Phenocycler-Fusion), generated from same tissue samples. As a ground-truth for cell typing, Akoya’s single-cell spatial phenotyping technology enabled us to precisely define cell types and cell states that were then used to evaluate and deploy the data integration pipeline. With these data, STimage first maps cells onto tissue sections with reference an H&E image. This way multiple layers of molecular data are transferred into one common framework, which can then be analyzed together. Traditional pathology annotations on the H&E images are also integrated to add human understanding of morphological patterns in a cancer tissue. The integrated analysis improves cell neighborhood identification, which allows cell-cell interaction analysis based on spatial co-localization between cell types (using single-cell resolution protein data) and locally co-expressing ligand-receptor pairs (using transcriptome-wide spatial data). We applied STimage to head and neck and skin cancer samples, in order to demonstrate the broad applicability of this analysis pipeline for various cancer types. We demonstrate applications for both, diagnoses and prognoses. For diagnosis, ST image identified and cross-validated cell types whilst assessing the expression of markers for drug targets. We also present an in-depth case study for an oropharyngeal squamous cell carcinoma patient not responding to Nivolumab treatment (anti-PD-1). Based on Visium and PhenoCycler-Fusion data, we discovered a spatial signature for non-responsiveness. Furthermore, we used the predicted ligand-receptor interactions to rank the patient’s response potential to currently available drugs, with the top target being TF-TFRC. STimage is thus able to integrate multiple layers of spatial omics data to improve and solidify prognostic biomarkers for cancer treatments. This study highlights the power of machine learning integration to combine multiple spatial multi-omics data, in particular PhenoCycler-Fusion and Visium, for improving diagnosis, prognosis and treatment of diverse cancers. Citation Format: Xiao Tan, Andrew Causer, Jazmina Gonzales-Cruz, Ning Ma, Bassem Ben Cheikh, Oliver Braubach, Quan Nguyen. Machine learning integration of transcriptome-wide spatial sequencing data and ultra-high plex spatial proteomic data enables the prioritization of cancer drug targets [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 2059.