TME infiltration

Quantify tumor microenvironment architecture — cell density patterns, immune infiltration phenotype, and spatial proximity metrics from segmented tissue slides.

Research question

How infiltrated is the tissue, what is the spatial organization of detected cells, and does the infiltration pattern suggest immune-desert, excluded, or inflamed phenotypes at the tissue level?

Who this is for

Translational and tumor biology researchers measuring TIL density and spatial immune organization
Immuno-oncology labs correlating histological infiltration with treatment response
Pharma tissue biomarker teams exploring histological endpoints in trial cohorts

Data requirements

Data	Required	Purpose
H&E or IHC slide	Yes	Tissue and cell detection input
Completed segmentation run	Yes	Spatial quantification prerequisite
Sample metadata with group labels	No (for cohort)	Responder/non-responder or treatment arm comparison

Run the full Slide quantification pipeline first, or ensure a completed Segment cells run exists.

Workflow

Upload H&E/IHC slide → Tiles → Tissue detect → Segment cells → Spatial quantification → Review infiltration metrics

Step 1 — Segment tissue and cells

Follow the slide quantification pipeline through completed Segment cells. IHC slides with immune markers support morphology-based detection; multiplex IF phenotyping is a planned extension.

Step 2 — Run spatial quantification

Launch Spatial quantification from the slide detail page:

curl -X POST http://localhost:8001/pathology/jobs/spatial-quantification \
  -H "Content-Type: application/json" \
  -d '{
    "study_id": "your-study-id",
    "slide_id": "your-slide-id",
    "parameters": { "segment_run_id": "your-segment-run-id" }
  }'

Step 3 — Review TME metrics

Inspect the spatial quantification panels:

Metric	Purpose
Cell density heatmap	Spatial distribution of detected cells across tissue
Nearest-neighbor distances	Proximity and clustering of cell populations
Infiltration phenotype	Research score summarizing spatial immune organization
Region metrics	Compartment-level density when tissue regions are detected

Toggle the density heatmap layer in the Slide viewer to correlate metrics with morphology.

Step 4 — Export for downstream analysis

Download cell_spatial_metrics.csv and region_metrics.json for statistical analysis or correlation with clinical metadata. Link sample records when cohort comparison workflows are enabled.

Step 5 — Cohort comparison (planned)

The cohort_compare pipeline API exists for group-level pathology biomarker comparison. Full UI workflow is in development — see products/pathology/plan.md.

Expected outputs

Density heatmap overlay in the slide viewer
Infiltration phenotype classification summary
Nearest-neighbor distance statistics
Per-region and per-cell spatial metric tables
Exportable CSV/JSON for cohort-level aggregation

Integration with Oncology

The Oncology area adds molecular TME profiling — immune deconvolution, cell-cell communication, and clinical endpoint correlation — on top of compbio single-cell pipelines. Pathology contributes the histological quantification layer: cell density, tissue boundaries, and morphology-guided spatial metrics.

Typical analyses

Analysis	Context	Question
TIL density	Pre-treatment biopsy	Is immune infiltration high at the tumor margin?
Treatment comparison	Pre vs. post IO therapy	Did cell density patterns change after treatment?
Stroma interface	H&E tumor-stroma boundary	Are immune cells excluded from the tumor core?
Trial endpoint	Responder vs. non-responder slides	Do histological metrics differ by response group?

Menu

Documentation