The best treatment for tuberculosis (TB) could depend on which version of a particular gene the patient has. Researchers from the UK, USA and Vietnam combined studies in zebrafish with clinical work to identify a gene that controls the inflammatory response in TB. It is one of the first applications of personalised medicine outside of cancer treatment.

'It's like a 'Goldilocks' gene. Depending on what versions of the LTA4H gene you have inherited, you could see an inflammatory response to TB that is 'too much', 'too little', or 'just right'', explains Dr Sarah Dunstan, head of human genetics at Oxford University's Vietnam unit and study co-author. 'You are likely to benefit most from a treatment tailored to your own genes'.

Variations in the LTA4H gene alter different biological pathways, leading to either too much or too little inflammation after infection by Mycobacterium tuberculosis. In zebrafish, both extremes allowed the bacteria to thrive and multiply. The study went on to show that blocking the appropriate biological pathway with drugs could restore just the right level of inflammatory response in the fish.

Dr Guy Thwaites of King's College London, who led the clinical study in Vietnam says: 'This is a fundamental discovery. It is now possible to think about the use of simple but rapid genetic tests to determine how people will respond to tuberculosis infection and whether they would benefit from steroids'.

The most severe form, tuberculosis meningitis, affects the brain and is seen in around one or two in every 100 cases. The team found that only those with LTA4H genes that cause too much inflammation benefited from the standard treatment, a combination of antibiotics to kill the bacteria and the steroid dexamethasone to reduce inflammation.

There was also the suggestion that steroids may have an adverse effect on patients with the version of LTA4H that causes a reduced inflammatory response. However, this result is not statistically significant.

'The findings could apply much more widely than just in tuberculous meningitis, or other forms of TB', said Dr Thwaites. 'Since the inflammation pathways governed by the LTA4H gene are central to many infections, there could be implications for many diseases'.

There were an estimated 9.4 million cases of TB and 1.7 million deaths worldwide in 2009. It is estimated that one third of the world's population is infected with latent TB.