Skip to content

A Vaccine for Yash

by the RNA Foundry

PopVax +

In January 2025, 32-year-old Yashdeep Bindal was diagnosed with stage IV pancreatic cancer. He had just returned to work after the birth of his daughter, Maya.

At the RNA Foundry, we are developing a personalized cancer vaccine for Yash. We’re sharing his tumour sequencing data and our vaccine design so that other researchers can help refine it.

Our first mRNA-LNP vaccine, developed at the Foundry for an infectious disease, entered a clinical trial in Australia earlier this year. Once the phase 1 trial concludes, we will open-source the design and manufacturing information so others can build on our work.

We’re working with Yash and his family to develop a personalized vaccine for him in the open, too. Yash has authorized a public release of his data for this – if you work on cancer vaccines, antigen discovery or immunotherapy, download the package and tell us what you think. Which antigens would you include? What signaling elements would you use? What might we have missed?

Clinical history

Diagnosis

  1. Weight loss, fatigue, and abdominal and back pain began.

  2. Admitted to Stanford; diagnosed with metastatic pancreatic cancer. Liver biopsy on 15 January; FOLFIRINOX started on 17 January.

Chemotherapy

  1. Anaemia required frequent blood transfusions during the first chemotherapy cycle.

  2. Caris liver-tissue profiling identified the CDH1::NRG1 fusion.

  3. CT: pancreatic tumour mildly smaller; liver and lung lesions stable to slightly smaller. Suspected peritoneal disease and small-volume ascites.

NRG1 fusion targeting

  1. After the identification of a rare CDH1::NRG1 fusion, he was switched to zenocutuzumab (Bizengri), an antibody targeting this fusion’s signalling pathway. He had a strong initial response, with minimal side effects and CA19-9 falling from ~45,000 at diagnosis to 18.

  2. Rising CA19-9, growing lung nodules and a new peritoneal nodule. Liver and pancreatic disease remained controlled on zenocutuzumab.

  3. Peritoneal biopsy yielded insufficient tissue; liquid biopsy still detected the NRG1 fusion. Zenocutuzumab continued through October.

Treatment change

  1. Started gemcitabine/nab-paclitaxel (chemotherapy) plus afatinib (a tyrosine kinase inhibitor) in India. Rash limited afatinib use. Also receiving high-dose vitamin C infusions outside standard treatment.

Data package

We’re sharing our lead vaccine design for Yash in a GitHub repository, along with the code and sequencing data used to develop it. You can reproduce our work and explore alternatives, including MHC class II targets, immunogenicity prediction, and linker optimization for processing and presentation.

Have a vaccine design for Yash? Fork the repository, add your design in its own directory and open a pull request. Include:

  • Your selected antigens and the rationale for choosing them.
  • Annotated protein or mRNA sequences with unspecified components labelled.
  • Code, data sources, dependencies and commands to reproduce your work.

See the contribution guide, or email yash@rnafoundry.com with your ideas.

Explore the data and design on GitHub →

Lead vaccine design

One mRNA sequence encodes 24 selected antigen regions — read the methods used to design it, and suggest changes.

This lead design uses matched-normal white-blood-cell DNA alongside Yash’s tumour DNA and RNA. All 12 mutation regions meeting our eligibility criteria are included.

The RNA, from 5′ to 3′

724-amino-acid protein

5′ → 3′ · schematic widths · follow the numbered path
24 antigen regions + linkersRNA 142–2,070 nt

Each selected region contributes 25 amino acids. The order of the regions and the joining sequences were chosen jointly, taking the SEC and MITD boundaries into account. Explore the regions below.

The encoded regions

In their actual order · select to explore
DNA mutation 12Splice event 5CTA / self protein 5Gene fusion 2Short linkers

CTA / self protein · region 1 of 24

SPATA31A6 CTA / self protein

A reference protein region from a cancer–testis antigen (CTA) gene with positive supplied expression.

Encoded 25-aa regionProtein 27–51
SLSASQPPEPSLPLEHPSPEPPALF

Reference sequence; no mutationSelected HLA-binding peptide

Predicted HLA-I binding

33.6 nM at best

Best eligible short peptide for each allele. Lower values indicate stronger binding. Select a row to see the peptide in context.

ASQPPEPSL9 aa · HLA-C*01:02

Bars show affinity on a logarithmic scale. All six allele values are in nM.

Gene expression

13.61 transformed TPM
How expression is used in selection

Supplied gene expression.

The supplied gene value is used directly as the expression proxy.

Region score = affinity contribution sum × √expression proxy.

Reports

Pathology, tumour profiling and liquid biopsy reports.

Reports are included in the GitHub data package.