Bacteria-killing viruses have destroyed cancer cells in mice by redirecting the immunity they built up from vaccinations to attack tumours. Vaccinating mice against malaria and then using a harmless phage, a virus that infects bacteria, to direct that immune response towards cancer cells resulted in the tumours being eradicated in nearly half of the animals.

Immunotherapy, treatments that use our immune system to fight cancer, has transformed how we treat some forms of the condition, but many people still don’t benefit. One of the major challenges is helping the immune system to recognise tumours as a target.

Looking for a way around this, Amin Hajitou at Imperial College London and his colleagues focused on a phage that usually infects E. coli. After attaching to these bacteria and injecting its genetic material, the phage hijacks their genetic machinery to produce thousands of new phages, then destroys the bacteria.

The researchers genetically engineered this phage to recognise and bind to proteins called αvβ3 and αvβ5 integrins, which are abundant on many tumour cells and largely absent from healthy ones. They also modified its genetic cargo to deliver instructions for producing a malaria-specific antigen, a molecular signal the immune system recognises as foreign. “The phage acts like a targeted delivery vehicle,” says Hajitou.

The team then tested this approach in 60 mice with cancerous tumours just beneath their skin. Fifteen of them were given a malaria vaccine and two weeks later, were injected with the phages in their tails, receiving six injections in total over the next two weeks. The rest acted as controls, with 15 receiving no intervention, 15 being given the malaria vaccine alone and 15 getting only the engineered phages.

In 44 per cent of the treated mice, the tumours were eradicated and hadn’t returned a year later, when the study ended. Other treated mice lived longer than the control groups, which experienced no survival benefit.

“These modified viruses can be administered systemically into the mouse and they then specifically find out the tumour cells and infect these,” says David Withers at the University of Oxford. “This really improves upon current approaches to manipulate tumours, such as oncolytic viruses [which infect and destroy cancer cells], which require injections directly into the cancer, thus limiting where you can treat and, in the case of metastatic disease [when cancer spreads], [it’s] an impossible task trying to target all of the cancer sites.”

Adjusting the phage’s antigen-making instructions should mean that the approach also works when someone has been vaccinated against infections like seasonal flu or covid-19. “Other vaccines, stronger than malaria, should work even better,” says Hajitou. “The principle is to exploit pre-existing immune memory, [it’s] not something unique to malaria.”

The researchers are in talks with the Medicines and Healthcare Products Regulatory Agency (MHRA) in the UK about assessing the approach in an early-stage trial in people, which they hope to begin next year.