“The promise of spatial genomics technology is whole-genome, single-cell expression profiling that is mapped to a specific tissue feature,” she continued. However, the reality of the technology is slightly different. Often researchers encounter a compromised resolution and the analytical data produced is sometimes difficult to interpret. Nevertheless, spatial genomic technologies hold enormous potential for disease indications such as pancreatic cancers in their very complex tissue architecture.
Available Spatial Technologies
While there are many spatial technologies available, Lyubetskaya identified two predominant technologies: sequencing-based and image-based. Slide-Seq is one example of sequencing-based technology and involves profiling a small neighbourhood of cells on a tissue sample. Although Slide-Seq is relatively easy to apply and operates at a high throughput, the technique can be expensive to implement and does not offer single-cell resolution.
- Unlocking the Potential of Spatial Multi-Omics and Cytometry
- Quantitative Mass Spectrometry: a Clinical Tool for Biomarker Discovery and Qualification
- ImmunoID NeXT ®: The Comprehensive Immuno Genomic Profiling Assay
Imaged-based technologies such as MERFISH have an incredibly high sub-cellular resolution. However, in most cases, the number of analytes available for profiling is limited. “At most, you can only look at hundreds of genes out of the genome, meaning pre-selecting panel is necessary,” Lyubetskaya pointed out.
Lyubetskaya then touched on BMS’s use of the spatial genomic technology called 10x Visium, which can be used to “create a clinically grade HD image of the tissue section you spatially profile.” This means that the same tissue section has two separate outputs: transcriptomics and high-resolution image. For BMS, the ideal case of feature identification includes a concordance between expression and pathology.
Perspectives on Proteomics in Research and Clinical Trials
Once the presentation concluded, Ducret took to the floor to discuss the current landscape of proteomics, attesting to its status as a realistic biomarker for both research and clinical use. Ducret identified that “there is comprehensive coverage and accurate quantification in proteomics today.” Indeed, current workflows are becoming both standardised and increasingly automated.
The conversation then turned to several of the key factors that are required for a successful study using proteomics. These include having a clear context of use and well-characterised samples in sufficient abundance. According to Ducret, “whenever possible, you want to use relevant samples which will be collected in the same way as in your study.” Additionally, the number of samples appropriate for statistical evaluation and security is vital.
NanoString GeoMx Digital Spatial Profiling
Decman then took over to introduce GSK’s omics-based technology, NanoString GeoMx DSP. This digital spatial profiling technology enables the study of RNA and protein expression in the context of relevant tissue sections. According to Decman, NanoString GeoMx “preserves the spatial relationship of cells and tissue compartments while allowing profiling of RNA or protein expression within the region of interest.”
NanoString GeoMx “preserves the spatial relationship of cells and tissue compartments while allowing profiling of RNA or protein expression within the region of interest.”
Feature extraction is based on Geometric Segmentation and uses tissue morphology or morphology markers via Molecular Segmentation to collect specific cell populations. “The benefits of studying morphology and relevant cell populations include determination of protein or RNA expression in, for example, tumour nest versus tumour stroma, or epidermis versus dermis in skin biopsies,” Decman explained. NanoString GeoMx DSP is not suitable for single cell analysis, however, individual cells labelled with a morphology marker can be segmented.
Q and A
In response to the various insights divulged by the panel, the audience was keen to learn more about the best practices for ensuring data resolution and whether this can be applied to broader cell types. Lyubetskaya answered: “To me, spatial data is a low bulk technology, which means it requires analytical approaches we typically apply to large bulk RNA-Seq datasets (The Cancer Genome Atlas). Although single-cell style analytics could be useful to some extent, they should be applied with caution”.
“The rule we apply to understand the cell-type mixture within clinical samples also applies to spatial data,” she continued. “Many vendors keep promising an increase in resolution, but my question is whether this is actually necessary for us to succeed?” Decman concurred, explaining that scientists must be very careful in defining the types of questions they are asking using spatial technologies and understanding the degree of resolution which these spatial technologies can provide. Ducret reiterated the importance of relying on context analysis instead of desiring single cell analysis.
The discussion concluded with some final thoughts on the prospect of mass spectrometry and non-mass-spectrometry-based platforms for plasma proteomics. At Oxford Global, we couldn’t have been more pleased with the turnout for our June’s Biomarkers Discussion Group. The conversation was engaging, the debate stimulating, and the industry insights invaluable. We will continue our Discussion Group series in September with a session focusing on ‘Overcoming challenges of NASH in Clinical Development’. Learn more about the Oxford Global Discussion Group series at our Biomarkers Portal.
Want to stay up to date with the latest Biomarker news? Register now for Oxford Global’s flagship event, Biomarkers US: In-Person. Happening 03 – 04 October 2022 in San Diego, USA, this is a must-attend forum covering the latest trends transforming biomarker and translational research