Training someone's own white blood cells to fight cancer is an experimental approach worth trying, researchers say.

On Wednesday, investigators reported their early-stage findings of a new technique they tried in nine patients with advanced melanoma that had spread from the skin to other parts of the body. At this stage, average survival is less than a year.

Lifeguards in Sydney stand together to form the words 'slip slop slap' on the backs of their shirts during a sunscreen campaign in Australia to raise awareness of melanoma, a cancer that is the subject of a new experimental therapy. Will Burgess/Reuters

The subjects were given "adoptive immunotherapy," which involves removing specific T cells from their blood. The cells are manipulated in the lab, using genetic engineering techniques, to recognize and destroy tumour cells.

When the T cells are given back to the patient, the body's natural immune system is boosted.

Previously, the success of adoptive immunotherapy unravelled because the T cells were unable to remember to fight cancer in the long term without the use of other treatments that work but often have harsh side-effects.

"Our technique opens the way to therapies that produce less-toxic, long-lasting immune system attacks on cancer cells," said the study's lead author, Dr. Marcus Butler of the Dana-Farber Cancer Institute in Boston.

While the T cells survived for long periods, the treatment did not stop the disease from progressing in the majority of participants, the team reported in Science Translational Medicine, a sister publication of the journal Science that is devoted to biomedical science.

One patient did improve completely and now has no evidence of cancer 25 months after the therapy.

In the next part of the experiment, five patients whose disease had progressed despite T cell treatment were also given the drug ipilimumab, which boosts the cells' anti-tumour response.Dr. Marcus Butler and his team found a way to enable anti-tumour T cells to survive in the bloodstreams of melanoma patients for more than a year in some cases. Sam Ogden/Dana Farber Cancer Institute

Three of the patients experienced a long-term shrinkage of their tumours, and two others saw their disease stabilize.

"I think all of us in the medical oncology community are interested in increasing the number of patients who respond from this kind of therapy, and also when they do have a response that these responses are durable," Butler said in a journal statement.

The Phase I trial looks at the safety of the technique, which needs to be investigated in a larger number of patients.

The next step is to study how combining the technique with other therapies may boost the number and effectiveness of the memory T cells, said the study's senior author, Dr. Naoto Hirano of Dana-Farber and the Ontario Cancer Institute in Toronto.

The study was funded by grants from the U.S. National Institutes of Health, Immunotherapy Fund 1, the S. Craig Lindner Fund for Cancer Research, the Rudolf E. Rupert Foundation for Cancer Research, the Cancer Research Institute/Ludwig Institute for Cancer Research Cancer Vaccine Collaborative, Friends of the Dana-Farber Cancer Institute, the Dunkin' Donuts Rising Stars Program, and the American Society of Hematology.