Stellaromics Dives into the Thick of Spatial Genomics

2025 has sometimes seemed like an unending dark winter of biotech, but just before AGBT spatial genomics company Stellaromics announced they had secured $80 million in Series B funding to advance their Pyxa platform.  Around that time, CEO Todd Dickinson and CTO Ye Fu chatted with me by phone, which any diligent and responsible scribe would have written up immediately.  Alas, that is not me.  But even two months of delay in writing cannot change the fact that Stellaromics is an interesting new entrant to the spatial genomics field.

Pyxa is based on the published STARmap approach.  A key differentiator from many spatial genomics techniques is enabling working with very thick tissue sections - most spatial approaches require thin slices of tissue.  We biologists like to pretend such sections are neatly stackable, but the reality is significant material is lost to microtome blade for each section - perhaps 5 microns per cut.   So an accurate stacking of such sections would have gaps where the blade cut through, with the lost material being around 1/2X to 1X the thickness of the recovered section.

Probing thick tissue sections is not an easy task, requiring very bright chemistry and tissue clearing.  Tissue clearing at times seems akin to alchemy - via embedding in hydrogel  - removing optically interfering molecules (particularly lipids) without disturbing the three-dimensional placement of the molecules of interest.  Stellaromics has now achieved tissue clearing of 100 micron sections versus the typically 4-10 micro section.  Fewer sections reduces batch effects and enables greater consistency which generating 10-50X more data per run than a thin section.  Plus it enables actually looking at cell-cell interactions in that vertical stack rather than having that information confounded by the zones lost to the microtome blade. Samples are loaded using a 12 well microwell plate, not slides, with each well capacious enough for two mouse brains.

RNA targets are converted to DNA amplicons using a two-probe system developed and patented at Stanford.  A RIBOmap assay is also under development, specifically labeling RNAs which are being actively translated.  Further multiomic readouts are planned for the future.  Sequencing of the amplicons occurs on instrument, with data processing onboard, but at this time Stellaromics is not revealing the compute architecture.  Data output ranges from tabular output with rows of genes, their expression level and three-dimensional coordinates to JPEG cell masks and subcellular segmentation information.  Resolution approaches the diffraction limit.

Initially the offering is a core panel of 225 genes with user option to add 25 more.  The design pipeline for the probes takes into account expected expression levels so that highly abundant genes don't swamp out the signal from rare genes.  Full custom panels are also possible, but have a longer design time.  Probes can even be designed to even targets as short as CRISPR guide RNAs.  Stellaromics doesn't plan to stay at 250 genes; the original Science publication probed 5K targets.   Data acquisition is a sequencing-by-ligation approach with error reduction as described in the STARmap publications.  

Stellaromics has had alpha instruments at Stanford and The Broad Institute for over a year.  A fully subscribed early access program should launch in the third quarter of this year.  The floor-standing instrument occupies a bit less than a meter square and just over 1.5 meters high and will be priced at $485K.  Stellaromics also has a services division enabling researchers to send samples for 3D spatial analysis.

Spatial genomics is an exciting technology, but has yet to find its killer app.  Much of the effort to date has been aimed at late stage translational work with a corresponding emphasis on being able to read samples stored as Formalin Fixed, Paraffin Embedded (FFPE) material.  FFPE is on Stellaromics' roadmap, but they're definitely looking in other directions.  Much like Element Biosciences, they see great opportunity in inserting spatial profiling into early drug discovery, such as profiling effects of drugs on organoids, mapping the fate of Anti-Sense Oligonucleotides (ASOs) as gene silencing therapeutics or as mentioned tracking guide RNAs for CRISPR therapies.  They also see great opportunities for using RIBOmap translational profiling early in drug discovery.

Spatial profiling remains a highly dynamic space.  The past year has seen multiple acquisitions in this space and extensive patent litigation.  Indeed, the litigation firm of 10X & Genomics drove Nanostring into bankruptcy and then acquisition by Bruker.  Singular Genomics basically shuttered their badly limping sequencing business and threw all their effort into spatial profiling and was acquired by investment firm Deerfield Management.  Akoya was acquired by Quanterix.  Vizgen merged with fellow spatial company Ultivue.  Too many of those feel like highly defensive maneuvers in a hostile investment environment; it's a breath of fresh air seeing significant investment going to Stellaromics and their new Pyxa platform.