Does DNA hold the key to improving drug-resistant TB treatment?

Tuberculosis (TB) is one of the most deadly infectious diseases in the world. It is caused by a bacteria that infects and gradually destroys the lungs, and is transmitted when someone with TB coughs or spits or sneezes – and someone else inhale the germs. More people die from TB than from HIV/AIDS. Despite this, there is currently no effective vaccine against TB, and most of the drugs that we use to treat it were developed more than 40 years ago.

The emergence of drug-resistant strains of TB now pose an additional major threat to public health worldwide.

Part of the problem is the way we test people for TB. The best way to treat patients with drug-resistant TB is to put them on the right drugs immediately after they are diagnosed.

But to test if a TB strain is resistant to a particular antibiotic is unfortunately a lengthy process. It includes isolating the bacteria from the cough sample, growing it and then sampling all the different drugs to see which can stop the bacteria from growing. Because the TB bacteria grows very slowly this process can take several weeks, and patients are often needlessly given drugs with severe side effects while waiting for their results.

There is only one rapid assessment widely available in hospitals in South Africa and it’s limited because it only tests for resistance to one drug.

We have just completed a new study that shows we could use genomics to cut the testing process down to a matter of days. By extracting and analysing DNA directly from the sputum of someone who has TB, in less than a week scientists will be able to tell which strain of TB the person has, whether it is resistant to drugs – and how it should be treated.

The problem of drug resistance

Drug resistance is one of the many challenges to controlling TB. TB treatment requires a combination of four drugs over six months. There are several drugs that can be used in this combination. These are known as first-line drugs. But due to treatment being lengthy, many TB patients don’t complete this course. As a result drug-resistant strains of the bacteria emerge in their bodies.

Treating drug-resistant TB is extremely difficult because doctors have to use a different set of drugs, known as second-line drugs. These are stronger but less effective. They are also more toxic. They have many side-effects, including deafness and severe depression. These drugs also have to be taken for at least 18 months – which is often the reason patients fail to complete their treatment correctly. The bacteria continues to mutate and the result is what we call extensively drug-resistant TB.

These strains are resistant to the best first- and second-line drugs and they can be transmitted to healthy individuals. There are few drugs left to treat patients with extensively drug-resistant TB and up to 75% of patients die if infected with this strain of TB.

What we found

We analysed the DNA sequences of 498 drug-resistant strains of TB to see if we could identify the mutations that cause resistance.

What we found was that for both the first- and second-line drugs we could predict pretty well whether the strain was resistant or not – just from the DNA sequence alone. We were also able to find new mutations that cause resistance to a particularly toxic drug called cycloserine, which can cause suicidal depression.

The ultimate aim of a rapid drug resistance test can only be achieved if we can identify all the mutations that cause resistance. This research takes us another step forward in this direction. The information can be used to develop the next generation of rapid molecular TB diagnostic tests, which will allow us to use toxic drugs like cycloserine in only those patients who will benefit from the drug.

Working around resistance

In one of our previous studies, we traced the emergence of drug-resistant TB back to the 1950s in South Africa’s KwaZulu-Natal province. The findings highlighted the inevitability of drug resistance, as the bacteria that causes TB seems to be able to develop resistance to every new drug we throw at it.

This is particularly important now as there are now two new drugs that have been developed to treat TB: bedaquiline and delamanid. If we don’t use these drugs extremely carefully resistance will develop rapidly and we will be back to square one.

Using advanced genome sequencing technologies will ultimately help us reach a much faster diagnosis, and ensure that treatment is optimal. Patients will be able to get a personalised set of drugs that will definitely work for them and avoid getting toxic drugs that don’t benefit them. In this way we’ll also hopefully guard against new resistant strains of TB developing.