CAR-T Tumor Microenvironment Modeling

ISEF Category: Translational Medical Science

Subcategory: Pre-Clinical Studies  ·  Difficulty: Advanced  ·  Setup: University Lab  ·  Time: Full Year

The Hook

A CAR-T cell can look amazing in one setting and stall in another. Solid tumors fight back with low oxygen, cramped space, and tired immune cells. Your project asks why that happens, and which CAR-T design choices help most. You can turn that question into a simulation study with real translational value.

What Is It?

This project uses a computer model to study how CAR-T cells, which are engineered immune cells, interact with a solid tumor. Think of it like a crowded battlefield on a screen. The tumor does not sit still. It grows, consumes oxygen, and creates a harsh microenvironment that can slow immune cells down.

PhysiCell is an open-source modeling tool that can simulate cell behavior over time. In this project, you can set up a virtual tumor-on-a-chip, then change factors like hypoxia, CAR binding strength, and T-cell exhaustion. Hypoxia means low oxygen. Exhaustion means immune cells lose power after repeated fighting. Your goal is to find patterns that suggest which CAR-T design choices work best in tough tumor settings.

Why This Is a Good Topic

This is a strong science fair topic because you can test clear variables, measure output from the model, and compare many scenarios without a wet lab. The project connects to a real medical problem, since CAR-T therapy works better in some cancers than in others. You can learn simulation, parameter sweeps, data analysis, and how scientists think about translational design rules.

Research Questions

• How does tumor hypoxia change CAR-T cell infiltration and tumor killing in the model?
• What is the effect of CAR affinity on CAR-T effectiveness under low-oxygen conditions?
• Does increasing simulated T-cell exhaustion reduce tumor control more than lowering CAR affinity?
• To what extent does the spatial pattern of hypoxia alter the number of surviving tumor cells?
• Which combination of CAR affinity and exhaustion parameters gives the best balance between killing and persistence?
• How does the model response differ between a uniform tumor and one with an oxygen gradient?

Basic Materials

• Laptop or desktop computer with internet access.
• Google account for Colab access.
• PhysiCell model files or a public tutorial notebook.
• Spreadsheet software such as Google Sheets or Excel.
• Free plotting tool such as Python in Colab or Google Sheets charts.
• Notebook for tracking parameter settings and outcomes.
• Background reading on CAR-T therapy and solid tumors from PubMed or NIH.

Advanced Materials

• Laptop or desktop computer with strong memory and stable internet access.
• Google Colab Pro not required, but optional if notebook runtime limits interfere.
• Python environment with matplotlib, pandas, and numpy.
• PhysiCell source code or modified model files.
• Version control tool such as GitHub for saving code changes.
• ImageJ if you export and quantify spatial images from the simulation.
• Access to review articles on CAR-T, hypoxia, and tumor microenvironment from PubMed.

Software & Tools

• Google Colab: Runs the PhysiCell notebook in the browser without local installation.
• Python: Lets you change parameters, organize outputs, and make plots.
• pandas: Helps you clean simulation results and compare runs.
• matplotlib: Creates graphs of tumor size, CAR-T counts, and oxygen-related trends.
• ImageJ: Measures spatial patterns if you export model images for analysis.

Experiment Steps

1. Define the biological question you want the model to answer, such as whether hypoxia or exhaustion matters more for CAR-T failure.
2. Choose the model outputs you will measure, such as tumor cell count, CAR-T infiltration, or persistence over time.
3. Set up a baseline simulation and confirm that it produces stable, interpretable behavior before changing anything.
4. Plan a parameter sweep for CAR affinity, exhaustion, and oxygen gradient strength so you can compare many conditions fairly.
5. Design controls that separate the effect of each variable from the effect of random simulation noise.
6. Decide how you will turn raw outputs into a design rule, such as a threshold, ranking, or best-fit parameter region.

Common Pitfalls

• Changing too many model parameters at once, which makes it impossible to tell why the output changed.
• Comparing runs before checking that the baseline simulation behaves consistently, which can hide code or setup errors.
• Treating one noisy simulation as a real trend instead of repeating runs and comparing averages.
• Ignoring the oxygen gradient shape, which can make hypoxia look weaker or stronger than it really is.
• Measuring only final tumor size and missing the time course, which can hide early CAR-T failure or delayed recovery.

What Makes This Competitive

A strong version of this project goes beyond a simple parameter sweep. You can test whether the same design rule holds across different hypoxia patterns, tumor sizes, or exhaustion assumptions. You can also use sensitivity analysis or another tough statistical approach to show which variable really drives the outcome. That kind of careful modeling looks much closer to real translational research.

Project Variations

• Compare CAR-T performance in uniform oxygen versus a steep hypoxia gradient to see how spatial stress changes killing.
• Test whether different exhaustion decay rates change the best CAR affinity threshold for tumor control.
• Swap the tumor geometry or density pattern and see whether the same CAR-T design rule still holds.

Learn More

• PhysiCell documentation: Search for the official PhysiCell site and tutorial materials that explain the open-source cell simulator and example notebooks.
• NIH PubMed: Search review articles on CAR-T therapy, solid tumors, hypoxia, and T-cell exhaustion.
• NCI Cancer.gov: Read patient and research pages on CAR-T therapy and the tumor microenvironment.
• MIT OpenCourseWare: Look for systems biology, cancer biology, or computational biology course materials with modeling examples.
• Frontiers in Immunology: Search for open-access review and research articles on CAR-T cells and solid-tumor barriers.
• PLOS Computational Biology: Search for open-access papers on agent-based modeling, tumor microenvironments, and immune-cell simulations.