Tuberculosis (TB) is one of the “big three” global health diseases, along with malaria and HIV/AIDS. You just can’t ignore the fact that TB is one of the top 10 causes of death in the world, causing 1.8 million deaths in 2015 in mostly low- and middle-income countries. Because TB is so common you’d think we have a simple POC (point-of-care) diagnostic test for it, like the rapid blood test or cheek swab we have for diagnosing HIV. But you’d be wrong.
Couple the lack of POC diagnostics with the fact that around 480 000 people developed multi-drug resistant TB (MDR-TB) in 2015 and you can start to see that we have a real problem here. Rapid TB diagnosis is vital to stopping the spread of TB. And that’s where the rats come in.
Current TB diagnostics in many resource-poor settings rely on microscopy and culturing the bacteria that cause TB. Both methods require sputum samples which are not the easiest to collect from patients. Additionally, many areas with a high TB burden also have a high HIV/AIDS burden and because folks with HIV have a weakened immune system they can develop TB with fewer actual TB bacteria present compared to someone with a healthy immune system. So now you have a co-infected person who doesn’t have very many TB bacteria in their body to begin with and you’re trying to find some of those bacteria on a microscope slide to diagnose the person with TB. Seems to me that you might misdiagnose quite a few of those patients as smear negative if you didn’t get lucky enough to find bacteria in their sputum.
What about the rats? One of the interesting things that we’re learning about diseases is that some emit volatile organic compounds (VOCs).You may have heard of dogs who are being trained to smell cancer. Now rats are being used to sniff out TB.
The African giant pouched rat is the rat of choice because it is native to sub-Saharan Africa and has a great nose. This process works because the TB bacteria emit a specific VOC that the rats can be trained to smell in the sputum samples. To test samples the rats are put into what looks similar to a biohazard laboratory hood with the door closed; that is, a glass and metal enclosed box, where the TB samples to be “tested” can be slide into and out of the box while the rats runs back and forth sniffing the samples. To indicate a sample is positive for TB the rat stops and scratches at the sample. To indicate a sample is negative, the rat simply keeps going to the next sample.
The rats sniff at least 100 samples per day. As a way to continue to measure accuracy and provide positive reinforcement, some of the samples are known positives where the rat is rewarded with a food treat for selecting the correct sample. Additionally, more than one rat sniffs the samples so there is built-in redundancy. The samples marked by any of the rats as positive then go on to receive a higher level, more sophisticated (read: more expensive) test.
While there are these more sophisticated diagnostic tests that provide an increased level of sensitivity and specificity to diagnose TB (compared to standard diagnostic procedures), they are very expensive compared to the rats. A rat costs about $6 500 to train and can do 100 samples in 20 mins. The GeneXpert machine, a highly-accurate DNA based test supported by the WHO, costs around $17 000 and takes 2 hours for one sample. Now, admittedly, the rats will probably never be as good as the DNA-based test; but the GeneXpert requires stable electricity, rats require food and cages. Which one of those options do you think is more feasible for resource-poor settings?
Using rats to diagnose TB is still in it’s early stages of development. More data are required before anyone can say for sure that rats are the new way to diagnose TB. But even if the rats never become the new way to diagnose TB and they only find a handful of new TB cases in resource-poor settings, those handful of cases are real people who have benefited from the work of these rats.