The number of genetically modified animals being bred for experiments in the UK is to be reviewed by the Government after official figures showed they accounted for more than half of live animal tests last year .
Lib Dem Home Office minister Norman Baker said he was concerned about "potential excessive breeding" after GM creatures drove total live tests in 2013 to 4.12 million, an increase of 11,600 (0.3%) over 2012.
Scientists backed the rise saying the animals allowed them to study the early stages of diseases like cancer and dementia, b ut animal welfare groups criticised the increase.
Mr Baker said: "The overall 2013 procedure figures are similar to the previous year. However, there has been a decrease two years running in the number of animals used for experiments other than breeding.
"I am concerned by potential excessive breeding of genetically altered animals - there has been an increase in this, particularly with zebra fish.
"I have commissioned a review on what steps can be taken to drive these figures down."
The total number of procedures on animals has risen every year since the coalition Government pledged in 2010 to reduce them, although the number of tests that do not involve genetic manipulation or creation of "harmful mutations" fell by 111,600 (5%), the third year in a row it has gone down.
The number of procedures involving genetically modified animals has more than doubled since 1995, according to Annual Statistics of Scientific Procedures on Living Animals Great Britain 2013, released by the Home Office's Animals in Science Regulation Unit (ASRU).
The report also showed a 7% increase (216 tests) in the use of "non-human primates" - monkeys - in tests compared to 2012 although rodents like mice and rats accounted for 82% of test subjects.
Prof Roger Morris, acting head of chemistry at King's College London and a former head of its school of biomedical sciences, told a press conference at the Science Media Centre in London that mapping genomes could now be done for less than £10,000 in under a week, so work of greater accuracy and complexity was being carried out.
He said: "As scientists...I suspect on the average about 90% of our work involves molecules, cells in culture, modelling, computer analysis or studies from man where we are taking clinical material and seeing what is going wrong.
"In the middle sits the mice or the (other) experimental animals and the reason they are there is, only in experimental animals can we look at the early stages of disease.
"We don't need to model the late stages of cancer, the late stages of dementia, we have patients galore and they will all volunteer for any trials we like.
"But what we need to know is how to get at the beginning of the disease before it has eroded the brain, before cancer has invaded all of the body, and that we can only do with experiments in animals."
Professor Robin Lovell-Badge, head of stem cell biology and developmental genetics at the National Institute for Medical Research (NIMR) added that while stem cell research had made great strides, live testing was needed because scientists "need to understand how these stem cells respond to physiological signals within the body".
But animal groups questioned the usefulness of animal research to help solve human conditions.
The RSPCA's chief scientific officer, Dr Maggy Jennings, said: "It is depressing to see no sign of an end to this upward trend in the number of scientific procedures being carried out using animals, at a time when the validity of animal experiments and the quality of the science is being criticised as never before."
Michelle Thew, chief executive of the British Union for the Abolition of Vivisection, said: "The UK should be leading the way in reducing animal testing, yet we remain one of the world's largest users of animals in experiments."