The Gridded Profiling Strategy
In the fight against cancer, seeing the full picture isn't just an advantage—it's a revolution.
For decades, cancer research often treated tumors as uniform masses, much like trying to understand a city by looking at a blurry satellite image. We knew there were distinct neighborhoods—cancer cells, immune cells, and structural tissues—but lacked the tools to map their complex interactions. This limitation began to dissolve with the advent of spatial biology, a field dedicated to studying gene and protein expression within their native tissue context. Among its most powerful tools is a method called gridded tissue profiling, a systematic approach that is transforming how we uncover the hidden complexities of cancer and other diseases.
Traditional molecular biology methods often involve grinding up tissue samples, analyzing the genetic material, and obtaining an average reading of gene expression across all cells. This is like blending a fruit salad and analyzing the overall sugar content—you get useful data but lose all information about which fruits contributed what. In cancer, where tumor heterogeneity (variations within a single tumor) can dramatically influence treatment response and disease progression, this spatial information is critical 7 .
Single-cell RNA sequencing advanced the field by profiling individual cells but still required dissociating tissues, permanently severing the spatial connections between cells 4 . As one publication notes, this aggressive processing "results in cell death, skewing analysis to only a fraction or subset of cells within the tissue" 4 .
This need for spatial context led to the development of Digital Spatial Profiling (DSP). NanoString's GeoMx® DSP platform, one of the leading technologies in this space, allows researchers to profile proteins or RNAs directly from intact tissue sections while preserving their spatial location 6 . The platform works by using antibodies or RNA probes tagged with photocleavable oligonucleotides that release upon UV light exposure, enabling collection and quantification of molecular information from user-defined regions 7 .
Among various region-of-interest (ROI) selection strategies available in Digital Spatial Profiling, the gridded profiling strategy stands out for its comprehensive and unbiased nature. Unlike other methods that target specific morphological features, gridded profiling takes a systematic, territory-covering approach to spatial analysis 2 .
The gridded approach expands on the geometric profiling concept by placing geometric ROIs at regularly defined intervals across the entire tissue sample 2 . Imagine placing a transparent grid over a tissue section and analyzing each square of the grid individually—this is essentially what gridded profiling accomplishes, though with much more flexibility and precision.
Systematic grid overlay on tissue sample
This strategy is made possible by the programmable digital micromirror device (DMD) within the GeoMx instrument, which contains an array of approximately two million steerable reflective micromirrors that can be dynamically configured to profile customized regions of any shape or size 2 6 . The DMD directs UV light to illuminate precisely defined areas, releasing oligonucleotide tags from only the selected regions for collection and subsequent analysis 2 .
By sampling across the entire tissue section rather than pre-selecting specific areas based on visible morphology, researchers can capture a complete picture of molecular heterogeneity 2 . This is particularly valuable for discovering unexpected patterns or rare cell populations that might be missed with targeted approaches.
Traditional region selection often depends on a pathologist's identification of "interesting" areas, inevitably introducing selection bias. Gridded profiling removes this subjectivity, allowing the molecular data itself to reveal significant patterns 2 .
The systematic arrangement of ROIs enables researchers to reconstruct spatial relationships and gradients across the tissue, answering questions about how proximity to specific features (like blood vessels or tumor boundaries) influences gene expression 8 .
The grid pattern can be tuned to match the research question, with adjustable ROI size and spacing to balance resolution with practical considerations like sequencing depth and cost 8 .
To understand how gridded profiling works in practice, let's examine its application in breast cancer research, where tumor heterogeneity presents significant challenges for treatment prediction.
A study by the GeoMx Breast Cancer Consortium utilized gridded profiling to characterize molecular diversity in breast cancer tumors 7 . The workflow followed these essential steps:
Researchers obtained formalin-fixed, paraffin-embedded (FFPE) breast cancer tissue sections cut at 5μm thickness, ensuring they fell within the instrument's scanning area of 35.3 by 14.4 mm 4 .
The team hybridized biological targets within the tissue with UV-cleavable RNA probes designed to detect the whole transcriptome (all protein-coding genes) 7 .
They stained cell types of interest with fluorescent antibodies to identify key morphological features, though for pure gridded profiling, these markers primarily assisted with orientation rather than region selection 4 .
Using the GeoMx instrument interface, researchers overlaid a tunable grid pattern across the entire tissue section, defining numerous ROIs at regular intervals without targeting specific morphological features 2 .
The instrument sequentially exposed each grid segment to UV light, releasing oligonucleotide tags only within the boundaries of each selected ROI 2 . These tags were collected into a 96-well plate.
The collected oligos underwent library preparation and next-generation sequencing. The resulting data was processed through Nanostring's GeoMx NGS Pipeline software and analyzed with their Data Analysis Suite 4 .
The gridded approach revealed remarkable heterogeneity within individual breast tumors that would have been obscured in bulk analyses. The data showed distinct gene expression signatures in different grid segments, corresponding to variations in tumor cell subtypes, immune cell infiltration, and stromal composition across the tissue 7 .
| Method | Spatial Context | Multiplex Capability | Bias in Region Selection | Best Use Case |
|---|---|---|---|---|
| Bulk RNA Sequencing | Lost during processing | Whole transcriptome | N/A | Average expression of homogeneous tissues |
| Single-Cell RNA Seq | Lost during dissociation | Whole transcriptome | N/A | Cell type identification without spatial needs |
| Traditional IHC/IF | Preserved | Low-plex (typically <7 markers) | High, based on visible features | Confirming suspected targets in defined areas |
| Geometric DSP | Preserved | High-plex (whole transcriptome) | Moderate, based on large areas | Comparing predefined tissue compartments |
| Gridded DSP | Preserved | High-plex (whole transcriptome) | Minimal, systematic sampling | Unbiased discovery of heterogeneity and patterns |
Perhaps most importantly, the GeoMx Breast Cancer Consortium found that this method could identify distinct spatial compartments within tumors that correlated with different clinical outcomes 7 . For instance, certain spatial patterns of immune cell distribution were associated with better responses to therapy, suggesting potential biomarkers for treatment selection.
| Grid Region | Key Upregulated Genes | Cell Types Present | Therapeutic Implications |
|---|---|---|---|
| Tumor Core | EGFR, KRAS, HER2 | Tumor cells, few immune cells | May respond to targeted therapies |
| Invasive Margin | CD8, IFNG, GZMB | Cytotoxic T-cells, tumor cells | Likely responsive to immunotherapy |
| Stromal Region | COL1A1, FN1, ACTA2 | Cancer-associated fibroblasts | May indicate barrier to drug delivery |
| Immune-Rich | PTPRC, CD68, CD19 | Diverse immune populations | Potential for immunotherapy response |
| Hypoxic | VEGF, HIFI1A, CA9 | Tumor cells, endothelial cells | May benefit from anti-angiogenics |
Implementing a successful gridded profiling experiment requires careful selection of reagents and tools. Here are the essential components:
| Reagent/Tool | Function | Considerations for Gridded Profiling |
|---|---|---|
| Whole Transcriptome Atlas | Pre-designed RNA probe panel targeting ~18,000 protein-coding genes | Ideal for discovery-phase gridded profiling where targets are unknown 7 |
| Cancer Transcriptome Atlas | Focused panel covering cancer-related genes | More cost-effective when cancer-specific signatures are the primary interest 7 |
| Morphology Markers | Fluorescent antibodies for tissue visualization | While gridded profiling minimizes selection bias, markers still aid orientation and interpretation 6 |
| Quality Control Tools | RNAscope, histopathology evaluation | Critical for verifying tissue quality, especially with archival samples 4 |
| NGS Library Prep Kits | Prepare collected oligos for sequencing | Must be compatible with the platform's unique photocleavable oligo system 4 |
While particularly valuable in cancer research, gridded profiling's unbiased approach shows promise across biomedical research. In inflammatory skin diseases like psoriasis and cutaneous lupus, where immune cell organization is critical to disease pathology, this method has helped researchers map spatially distinct immune niches that drive chronic inflammation 4 . Similarly, in neuroscience, gridded approaches could help characterize the complex architecture of neurodegenerative diseases 6 .
The non-destructive nature of Digital Spatial Profiling means that the same tissue section can be preserved for additional analyses after gridded profiling—a significant advantage over methods like laser capture microdissection that physically remove tissue regions 2 . This allows for validation studies or complementary analyses on the exact same tissue regions that showed interesting molecular signatures in the initial gridded analysis.
Gridded tissue profiling represents a significant shift in how we approach tissue analysis—from targeted interrogation of suspected areas to systematic mapping of entire tissue ecosystems. As spatial technologies continue to evolve, combining gridded profiling with even higher resolution methods will likely provide unprecedented insights into the architectural principles of health and disease.
The GeoMx Breast Cancer Consortium has emphasized that proper experimental design remains crucial—"There is no wrong way or right way to develop your ROI selection strategy... it depends on the types of questions you are looking to answer" 2 . For questions requiring comprehensive, unbiased characterization of tissue heterogeneity, gridded profiling stands as an exceptionally powerful strategy in the spatial biologist's toolkit.
As these methods become more accessible and standardized, they hold the potential not only to advance basic research but eventually to impact clinical diagnostics, enabling pathologists to make more precise prognostic predictions and treatment recommendations based on a tumor's spatial architecture. In the evolving landscape of precision medicine, seeing the full picture may make all the difference.