3D spatial biology at Laboratory of Systems Pharmacology

3D imaging starts long before a scan is taken. Clinical collaborators annotate regions of interest such as early cancer or vertical growth phase on H&E sections based on their expert medical knowledge. This targeted approach ensures that only the most biologically meaningful tissue regions receive high‑resolution 3D coverage with minimal risk of false positives.

• Academia – exploration of diverse cell states in a resource-efficient manner.
• Industry – reproducible, focused acquisition aligned with quality controlled workflows.

The thick section CyCIF process uses iterative cycles of staining, imaging, and bleaching using all open-source reagents and protocols. Each round includes a nuclear marker for reliable 3D registration. Tile based segmentation and reconstruction handle datasets of 200–900 terabytes (depending on size of region to be imaged) across dozens of channels.

• Academia – flexibility to develop antibody panels of novel markers and incorporate different image analytical methods.
• Industry – standardized protocols for reproducible signal quantification to reduce interpretive risk.

Clarence emphasizes that achieving sufficient contrast in 35–50 micron FFPE tissue required optical sectioning, high-NA oil immersion objectives, sensitive detectors, and flexible software to define regions of interest and align cycles.

Throughout development, Clarence works with microscopy specialists and computational biologists to refine contrast, reduce crosstalk, integrate second harmonic generation or adopt new segmentation tools like uSegment3D and u-Unwrap3D.

• Academia –teaching and methodological transparency.
• Industry – validated, repeatable workflows for scale.

It also reflects the LSP’s belief that thick section imaging must become accessible beyond a small number of expert labs.

Examples that Clarence and his colleagues highlight include:

• intact immune and tumor cells with extended processes in melanoma tissue
• PD1–PDL1 punctate distribution at actual membrane contacts
• vascular segments spanning depth ranges hard to interpret in 2D
• rare cell populations such as T PEX cells in preserved spatial niches
• multicellular communities where several immune cells interact to engage a single tumor cell effectively

The LSP’s work reflects a shift in spatial biology: from incomplete fragments toward whole cell volumes, context rich in increasing our understanding of disease. This enables both new scientific questions and more informed decisions in translational environments.