Researchers have demonstrated "in theory" that a protein antigen from a childhood vaccine can be delivered into the cells of a malignant tumour to refocus the body's immune system against cancer, effectively halting it and preventing its recurrence.

The bacteria-based intracellular delivering (ID) system uses a non-toxic form of Salmonella that releases a drug, in this case, a vaccine antigen, after it's inside a solid-tumour cancer cell.

"As an off-the-shelf immunotherapy, this bacterial system has the potential to be effective in a broad range of cancer patients," said Neil Forbes, professor of chemical engineering, at the University of Massachusetts-Amherst, in a paper published in the journal Frontiers in Immunology.

The research offers promise toward tackling difficult-to-treat cancers, including liver, metastatic breast and pancreatic tumours.

The team has filed for a patent and plans to seek FDA approval in an effort to start clinical trials within a few years.

"The idea is that everybody is vaccinated with a whole bunch of things, and if you could take that immunisation and target it towards a cancer, you could use it to eliminate the cancer," Forbes explained.

To test their theory that this immune treatment could work, Forbes and team genetically engineered ID Salmonella to deliver ovalbumin (chicken egg protein) into the pancreatic tumour cells of an experimental model that had been immunised with the ovalbumin 'vaccine.' 

The researchers showed that the ovalbumin disperses throughout the cytoplasm of cells in both culture and tumours.

The ovalbumin then triggered an antigen-specific T-cell response in the cytoplasm that attacked the cancer cells. The therapy cleared 43 per cent of established pancreatic tumours, increased survival and prevented tumour re-implantation, the paper said.

"We had complete cure in three out of seven of the pancreatic experimental models," Forbes said "We're really excited about that; it dramatically extended survival."

The team then attempted to re-introduce pancreatic tumours in the immunised experimental model. The results were exceedingly positive. 

"None of the tumours grew, meaning that the experimental model had developed an immunity, not just to the ovalbumin but to the cancer itself," Forbes said.

 "The immune system has learned that the tumour is immunogenic. I'm doing further work to figure out how that's actually happening."