Neoantigen-Targeted Cell Therapy
The strategy of targeting tumor-specific mutations (neoantigens) with engineered T cells — the leading approach for bringing cell therapy to solid epithelial cancers. Championed by Steven Rosenberg (NCI) at Session V.
The Problem
- Solid epithelial cancers cause ~90% of cancer deaths
- No antibody recognizes a surface molecule unique to a major solid epithelial cancer (despite 50+ years of monoclonal antibodies)
- Targeting shared antigens → fatal on-target/off-tumor toxicity
- Whole TIL (as grown for melanoma) did not work in epithelial cancers
The Approach
- Whole-exome and transcriptome sequencing identifies every mutation (~200 in epithelial cancers)
- Mutated 25-mer peptides loaded onto patient’s APCs and screened against expanded TIL
- ~1 in 70 mutated neoepitopes is genuinely immunogenic
- ~75% of patients have TIL recognizing at least one neoantigen
- Almost every neoantigen is unique to the individual patient
Three Strategies
| Strategy | Method | Status |
|---|---|---|
| Selected TIL | Isolate only neoantigen-reactive TIL subset | Nature Medicine 2025 |
| TCR transduction | Clone reactive TCR, transduce into patient’s normal lymphocytes | Nature Medicine 2024 |
| Shared-mutation TCRs | Libraries against KRAS and p53 hotspot mutations | Most exciting current work |
Key Data
- Selected TIL in chemorefractory pancreatic/colorectal/cholangiocarcinoma: ~24% RECIST response rate (median 4 prior regimens)
- Proof-of-principle patient (2014): cholangiocarcinoma, disease-free 12 years
- KRAS G12V–targeted TCR: pancreatic cancer regression
- p53 R175H–targeted TCR: colorectal liver/lung mets shrinking
- Visceral disease responds — lung, liver, bone metastases
Why KRAS and p53 Are the Unlock
- KRAS hotspot mutations in ~30% of all cancers
- p53 hotspot mutations in ~50% of all cancers
- Converts a hyper-personalized therapy into a partially shared one — the prerequisite for scaling
Cross-References
- Steven Rosenberg — pioneer
- CAR-T therapy — the broader modality
- David Baker — designed binders to disordered mesothelin stubs for CAR-T (Session XII)
Comments