AI-Assisted mRNA Cancer Vaccine Shrinks Tumours in Dog, Raising Hopes for Human Oncology

An mRNA cancer vaccine developed with the help of artificial intelligence has reduced tumour size by half in a rescue dog named Rosie, prompting researchers to explore its potential applications in human oncology. Paul Conyngham, a Sydney-based technology entrepreneur, used ChatGPT and machine learning tools to design the treatment after his dog received a terminal mast cell cancer diagnosis in 2024.

How the mRNA Cancer Vaccine Was Developed

Conyngham paid $3,000 for DNA sequencing at the University of New South Wales (UNSW) Ramaciotti Centre for Genomics. He then processed the genomic data through multiple analytical pipelines, using the AI program AlphaFold to identify mutated proteins and match them to potential drug targets. The resulting half-page formula was handed to Pall Thordarson, director of the UNSW RNA Institute, who synthesised a bespoke mRNA nanoparticle vaccine.

Research Team and Institutional Involvement

Martin Smith, associate professor of computational biology and director of the Ramaciotti Centre for Genomics at UNSW, supported the sequencing effort. Thordarson, a specialist in nanomedicine, manufactured the vaccine. Rachel Allavena, a canine immunotherapy professor at the University of Queensland’s School of Veterinary Science, administered the experimental drug under existing ethics approval. Conyngham drove 10 hours to the Gatton laboratory in December 2025 for Rosie’s first injection, followed by a booster the following month and a third dose scheduled for the week of March 15, 2026.

“This is the first time a personalised cancer vaccine has been designed for a dog. Ultimately, we’re going to use this for helping humans. What Rosie is teaching us is that personalised medicine can be very effective, and done in a time-sensitive manner, with mRNA technology.”

Pall Thordarson, Director, UNSW RNA Institute

Clinical Observations and Tumour Response

Prior to treatment in November 2025, Rosie carried a tennis ball-sized tumour on her hock. By January 2026, seven weeks into the trial, that tumour had reduced by approximately half. Allavena noted improvements in the dog’s energy levels, coat condition, and overall comfort. However, one large tumour did not respond to the initial treatment, and Conyngham is now working with UNSW on a second round of DNA sequencing to investigate why.

“It raises the question, if we can do this for a dog, why aren’t we rolling this out to all humans with cancer? It gives hope to a lot of people.”

Martin Smith, Associate Professor of Computational Biology, UNSW

Implications for Human Health Technology

The mRNA technology used in Rosie’s vaccine is the same platform deployed in Covid-19 vaccines. Oncologists are currently testing it in dozens of human clinical trials, though it has not yet entered mainstream cancer treatment. David Thomas, inaugural director of the UNSW Centre for Molecular Oncology, said the case demonstrates the value of citizen science, where individuals with computing backgrounds contribute meaningfully to medical research. Conyngham, who co-founded Core Intelligence Technologies and previously served as a director of the Data Science and AI Association of Australia, has 17 years of experience in machine learning and health technology applications.

“The striking thing about this is the idea of citizen science, where a punter in the street, with a computer science background, can use their skills in the scientific process. That’s a very impressive thing.”

David Thomas, Director, UNSW Centre for Molecular Oncology

Next Steps and Future Outlook

Conyngham is developing a second vaccine targeting the tumour that did not respond to the first treatment. UNSW researchers began new genetic sequencing work during the week of March 15, 2026. He said the total cost of Rosie’s treatment has reached tens of thousands of dollars. Moreover, he expressed optimism that the approach could eventually convert some terminal cancer diagnoses into manageable conditions by anticipating tumour mutations. The AI-driven methodology used in this case has drawn interest from the Canine Cancer Alliance in the United States, whose founder Mari Maeda helped connect the research teams involved.