The Value of Nothing – Challenges of Developing Diagnostics for Endemic Diseases

How can you treat something if you don’t know what the problem is? This is the first challenge for medics and vets, one for which diagnostic testing holds the key. In the case of infectious diseases, diagnostics also play a huge role in disease prevention, enabling the detection of infected individuals so that measures may be put in place to limit further spread and contain potential cases. Control and treatment efficacy can however only be as good as your diagnostic test, which is why developing robust, reliable and sensitive tests are all important. But when the disease you want to test for is endemic, it can raise some additional challenges.

Developing a diagnostic test

There are a number of important considerations to bear in mind when looking to design a diagnostic test for an infectious disease, including:

• Do you want to detect the infectious agent directly (e.g., by culture or PCR/quantitative PCR) or do you want to detect the infection indirectly (e.g. sandwich ELISA for host antibodies or levels of biomarkers in the host)?
  
  
• At what point during infection can the marker (direct or indirect) be detected, is there a lag (as with the formation of an antibody response) before the positive individuals will become obvious?
  
  
• What is the best test medium from which to detect the desired target (blood, urine, nasal swab…etc.)
  
  
• What is the most appropriate test format (i.e. point of care or lab-based) and are there restrictive time pressures for requiring results (for some diseases every minute counts)?
  
  
• How do you determine what is positive and what is negative (i.e. do you need a cut-off value)?
  
  
• How much will it cost? Development and retail costs must be balanced with what the target market are willing to pay.

The importance and problem of negative controls

No matter which approach you choose however, it is not enough to be able to say “Look! All of these samples that are truly positive test positive with my test.” It is just as important to know that samples that are truly negative will test negative too otherwise the test is of little or no value. As part of the development process it is important to run the assay on a panel of known positive and negative samples that span the spectrum from negative to weak positive through to strong positive to see just how good your test is. From this data, the sensitivity and specificity of your test can then be calculated. But for all of this to be possible you need a bank of negative samples to prove the point.

For some diseases, such as anthrax, Ebola or cholera (in some countries at least), which are not common in the global population (thankfully!), obtaining negative samples is not a big problem. However, for diseases which may pass asymptomatically in some individuals (so a sample you believe to be negative may actually represent a positive) or which is endemic in the population (so finding any samples that are negative is a challenge), this may not be straight forward.

Magical negatives and where to find them - Assistance from an unexpected source

To many people say “Iceland” and they think “Game of Thrones”, “breath-taking scenery” or “that volcano that no-one can pronounce”. But Iceland is also home to a very special population of horses – the Icelandic horse. Prized for their range of gaits including the tolt, they are exported all around the world. However, Icelanders have not imported horses for over 1,000 years, leading to the development of a genetically isolated population, and that’s not all they’re isolated from. Being an island put’s Iceland at a big advantage when it comes to disease control and they lack many infectious diseases, such as equine influenza and equine viral arteritis in their equine population that plague the rest of the world. They also lack strangles, caused by infection with the bacterium Streptococcus equi (S. equi), which is endemic in the rest of the world and has the dubious glory of being crowned the most frequently diagnosed infectious disease of horses.

Being a naïve population to many infectious agents however means that were a disease like strangles to make its way into the population, they would be at high risk of infection as they lack prior immunity. So, when an outbreak of respiratory disease spread rapidly through the equine population in 2010, there was fear that it could be one such disease. Thankfully, the culprit turned out to be an infectious agent that was already present in the population, Streptococcus zooepidemicus, a relative S. equi, albeit a strain to which the population had little immunity, explaining the rapid spread.

In trying to discover what the cause of the disease actually was, however, many samples, including serum, were taken from the Icelandic equine population and stored. To assist the investigation, Sigríður Björnsdóttir, from MAST, the Icelandic Food and Veterinary Authority, Vilhjálmur Svansson from the Institute For Experimental Pathology, University of Iceland, Keldur and their colleagues joined forces with the Bacteriology team at the Animal Health Trust who were then developing a diagnostic ELISA test for strangles.

This relationship, also in association with the Wellcome Sanger Institute, turned out to be beneficial all round. Whilst the Icelandic team gained insights into what was causing the outbreak and how it was spreading from genome analysis of collected bacterial isolates, the Icelandic serum samples also provided ideal negative controls for the strangles ELISA test (once it had been confirmed that strangles wasn’t the cause of this outbreak – big sigh of relief all round).

Sigríður Björnsdóttir commented “Our main goal was to reduce and prevent the suffering of our animals, which was a major welfare concern, and to protect the sustainability of the breed. However, we have realized how important the protection of the population is also for providing negative controls for research and the production of tests for equine diseases worldwide. We're happy to be in this situation.”

Dr Carl Robinson, Senior Scientist at the Animal Health Trust, Newmarket, UK, and first author of the related publication of the strangles ELISA test commented “Whilst obtaining positive samples for evaluating the test was not a big challenge, negative samples were more problematic”. He continued “The test is based on measurements of equine antibody levels to S. equi, which may stay elevated for some time following infection - months even. Some animals can also form what is termed a “carrier state” where they appear perfectly healthy but harbor infection. Both of these factors mean that finding true negatives in any population where the infection is endemic is a real headache. The availability of the serum samples taken in Iceland, however, gave the perfect set of negative controls.” Ultimately this veritable negative gold mine lead to the robust validation of the test for strangles offered by and licensed from the Animal Health Trust which is now used by labs around the globe to help thousands of horses.