The Cellestis Story. Part Two.

I am assuming that you have already read The Cellestis Story. Part One. (You may want to go back and re-read as I have made a few minor changes).


In summary, TB is a very nasty bacterium found in every county in the world that results in around 2 million deaths per year. It can usually be treated with a cocktail of antibiotics (did I mention that it is not a pleasant treatment for the patient?). Up to 30% of the world population is infected with TB in it's latent form which can become active and transmissible at any time. Ultimately, the only way to reduce (and hopefully, one day, eliminate) this scourge is to find and treat the huge reservoir of latent TB. This is where Cellestis comes in.


Diagnosing TB.
Bearing in mind that TB can exist in two states (active and latent), we need to look at the diagnosis of TB in two parts.


Diagnosing Active TB.
Generally, diagnosing Active TB relies upon physically locating the bacterium itself. When the TB is pulmonary (ie in the lungs - the most common place) the most widely used method of diagnosis is the sputum culture. In simple terms this means the patient coughing up some sputum which is then grown in the laboratory and examined for the TB bacterium. Whilst this process can take up to 12 weeks (TB is a slow growing bacterium) there are other processes to achieve a speedier preliminary diagnosis including PCR which can achieve results in 12 hours.


X-rays are also commonly used to diagnose pulmonary TB. However, even TB disease diagnosed by X-ray requires confirmation. Of course none of this works if the TB is not in the lungs and, of course, none of these tests will diagnose latent TB. 


Diagnosing Latent TB.
As difficult as it is to diagnose active TB, it is even more difficult to diagnose latent TB. Remembering that latent TB is TB that is in a dormant, or inactive state, there is actually nothing to look for. There is no TB in the sputum, the blood or any other bodily fluid. Because it is dormant and not spreading, it cannot be seen by X-ray or CT-Scan.


So, how to diagnose a patient with latent TB?


You use the immune system of the body itself. In simple terms, if you have TB (latent or active) then the body's immune system (hopefully) knows how to fight the TB bacteria and is doing so. Therefore, if you can demonstrate that a person's immune system knows how to fight TB then you might presume that the person has TB. This was realized around 120 years (!) ago and led to the development of the skin test for TB. In it's various incarnations it is known as skin test, mantoux test, TST, heaf test or tine test. All of these tests are based on the same principle and I shall refer to them collectively as TST.


TST (Tuberculin Skin Test).
The TST involves injecting a small amount of tuberculin (a grab bag of proteins that look a lot like TB) under the skin (usually on the forearm). If the body's immune system knows how to fight TB then it should recognise these proteins and attack. That immune response will create a lump (induration) after two or three days at the site of the injection and the size of the lump is measured to determine whether the patient has TB. Whilst the TST has been very useful and extremely widely used, it has an enormous number of deficiencies. Some of the major ones are:

• - Because the immune system has a long memory, the positive reaction from the TST can equally be due to a past infection of TB, rather than a current one. Generally, it is considered that once a person has had TB, the TST is no longer a viable diagnostic for that person.
• - Because many of the proteins in the PPD are the same as those in the BCG vaccination, anybody that has been vaccinated for TB will show as a positive to the TST, regardless whether they have TB or not. Often patients do not know whether they have had a BCG vaccination or not. Some countries (Japan, for example) have a population that is close to 100% vaccinated - the TST is worthless in that situation.
• - The TST is a two step process. Injection of the PPD followed by reading of the lump two or three days later. Depending upon the environment, it is not unusual for 30% of patients to not return for their reading.
• - The TST is very subjective. Despite the extensive training of TB practitioners, there is a large "inter operator variability". That is, a lump that one practitioner sees as positive, another may see as negative.
• - If the patient has a weakened immune system (eg HIV) then the TST is likely to show as negative, even if the patient has TB. Even more importantly, the test does not indicate this situation. A patient with TB can thereby be sent on their way with a "clean bill of health" to perhaps develop active TB and spread the TB further.
• - If the time of the Doctors and Nurses administering and reading the TST are taken into account, the TST is an expensive test. If you add the lost work time of the patient, even more so.

Having said all of the above, the TST has played a very important role in controlling the spread of TB. If only there was something better ....

QuantiFERON-TB Gold 

The QuantiFERON (QFT) diagnostic uses a similar concept as the TST (using the body's own immune system) but approaches it in a different way that does away with the problems of the TST. Generically, QFT is an IGRA (Interferon Gamma Release Assay).

It is a bit complex and, in truth, the fine details are beyond me. The following description of the processes involved is therefore necessarily "broad-brush". Whilst it may not be scientifically exact, it is essentially valid.

The battle warriors of the immune system are T-cells. An immune response to an infection involves two different types of T-cells, Effector T-cells and Memory T-cells. When Effector T-cells recognize an enemy that they are equipped to fight they attack the infection immediately (within a few hours) and rally more troops to the battle by generating Interferon-gamma. Once the battle is over, the effector T-cells will eventually die. On the other hand, the memory T-cells remain in the blood (essentially) forever with the memory of how to fight a particular infection. When they recognize an enemy that they are familiar with they do not engage in battle themselves but instead generate Interferon-gamma that will eventually result in the production of the appropriate Effector T-cells. The Memory T-cells take some time to get their act together - this is an important distinction. 

The QFT approach to diagnosis, therefore, is to measure that production of Interferon-gamma as an indicator of infection. The process is quite straightforward. A small sample of blood is taken from the patient and it is added to some proteins that are very specific to the TB bacterium. If the patient is currently fighting a TB infection then their blood will contain effector T-cells that will recognize those proteins as TB and "do their thing". Part of which is to generate Interferon-gamma. The QFT diagnostic measures the Interferon-gamma as an indication of TB infection. Note that because the diagnostic looks at the Interferon-gamma produced in the first few hours only, it is not confounded by the production of Interferon-gamma that is produced by the memory T-cells some days later.

So, what does this mean when we compare QFT with TST?

• - The QFT only recognizes a current infection of TB. A past infection of TB will not produce a false positive reading (as would happen with TST). 
• - Because the specific proteins used in the QFT are present in TB but not in the BCG vaccination, the fact that a patient has previously been vaccinated for TB will not confound the QFT result.
• - QFT is a one step process as far as the patient is concerned. Once the blood sample has been taken the patient does not need to return for a "reading". Therefore, no resources are wasted on non-returns as with TST.
• - The QFT is performed in a laboratory, using commonly available instrumentation. There is no operator subjectivity involved (as with TST). The result is a pure metric.
• - If the patient has a weakened immune system then (as for the TST) there may be no readable immune response to the QFT proteins. However, the QFT method will recognise this and report an "indeterminate" result. This is extremely important as that patient will not be incorrectly diagnosed as negative (as would happen with TST). Instead the Doctor has the information to launch a further investigation of the TB status of the patient.
• - The total cost of a QFT test is less than that of a TST when all factors are taken into account. Cost Comparison - TST vs QFT.

It is worth noting that neither TST nor QFT can distinguish between active and latent TB. Once a patient has been diagnosed as having TB by either of these tests, further diagnosis is required to make the distinction. This may involve the diagnostics mentioned earlier and, importantly, symptomatic diagnosis.

Of course, it is easily seen that QFT can alert to active TB disease that cannot easily be diagnosed with the traditional tests for active TB (eg non-pulmonary TB).

I should mention that there is another IGRA diagnostic commercially available for TB (T-Spot). I will deal with that later but for the moment just take my word for it that it has some significant disadvantages when compared to QFT and presents no commercial competition.

Sensitivity and Specificity.

Sensitivity and Specificity are two measure that are used to define the accuracy of a diagnostic.

Sensitivity is the percentage of people in a group that have a disease that will be correctly identified as such by the diagnostic. Specificity is the percentage of people in a group that do not have the disease that will be correctly identified as such by the diagnostic.

The FDA quoted figures for QFT are a Sensitivity of ~89% and a Specificity of > 99%. 

Because the TST has never been subjected to the rigours of the FDA approval process there are no reliable sensitivity and specificity figures available. The available figures for TST vary enormously from environment to environment. In an ideal situation, it's sensitivity might approach that of QFT but it's specificity can be as low as 30% and even in a highly controlled  environment is unlikely to exceed 80%. Think about that - every person incorrectly identified as having TB is likely to be subjected to around six months of treatment with some quite nasty antibiotics (and alcohol is contraindicated during treatment!).

Next: The Cellestis Story. Part Three.