Some sporadic insights into academia.
Science is Fascinating.
Scientists are slightly peculiar.
Here are the views of one of them.

Tuesday, 4 August 2015

Buy one get one free?

PHiD-CV induces anti-Protein D antibodies but does not augment pulmonary clearance of nontypeable Haemophilus influenzae in mice

Synflorix – a highly effective vaccine for S. pneumoniae

In our recent study (Siggins et al) published in the journal Vaccine we were looking at a commercially available vaccine called Synflorix (technical name PHiD-CV). This is a vaccine made by GSK against the bacteria called Streptococcus pneumoniae. This bacteria causes pneumonia (unsurprisingly – the name is a bit of a give away) but more seriously it causes a range of conditions called invasive pneumococcal disease where the bacteria crosses from the airways into the rest of the body leading to diseases like meningitis and sepsis. One of the problems with preventing Streptococcus pneumoniae infections (or Strep pneumo) is that there are many different forms of the bug and so a vaccine needs to protect you against as many of them as possible. Synflorix is highly effective at providing protection against 10 of the Strep pnuemo strains.

Sweet on the outside, deadly in the middle

The aim of our study was slightly different. Bacteria are a bit like the blue liquorice allsorts, with a bobbly sugar coat hiding a disgusting middle. Our immune system only recognises the outside part of the bacteria which is made up of complex sugars (polysaccharides). These molecules, unlike proteins, are not particularly well recognised by the immune system (poorly immunogenic). One trick to improve the immunogenicity of the sugars is called conjugation. This is where the sugars are chemically bound to proteins, which are better recognised by the immune system, and in a biological sleight of hand you the white blood cells into improving the response. A number of different bacterially derived proteins are used as the carrier protein to boost the immune response including ones from Tetanus and Diphtheria. In the current study we were looking at the effect of a protein derived from non-typeable H. influenzae (NTHi) called protein D. The aim was to see whether the protein D part of the Synflorix vaccine could give broad cross protection against NTHi in addition to Strep pnuemo.

NTHi is the new HiB

I guess you are wondering why we would want protection against NTHi, or probably more honestly, what on earth is NTHi? Haemophilus influenzae (the Hi in NTHi) is a bug we all carry in our throats and noses, which sometimes causes illness and occasionally causes meningitis. However, Haemophilus is divided into different types based on its sugar coat. It is the B type or HiB that is highly invasive and there is a vaccine for this, included in the UK infant schedule at 2, 3 and 4 months, which has significantly reduced the incidence of meningitis. Haemophilus are grouped (or Typed) by the sugar they are covered in, some don’t have a sugar coat and are so non-typeable (the NT in NTHi). Whilst the non-coated strains rarely cause invasive/ meningitis disease, they do lead to a significant amount of community acquired pneumonia. NTHi lung infections are particularly common in patients who are long term smokers and have the ongoing lung damage from smoking (COPD). There is therefore an economic argument for reducing the burden of this disease, especially if it can be achieved as a side effect of a vaccine that is designed to prevent a different bacterial pathogen, which brings us back to the current study.

No cross protection

It is very complicated to perform the types of study needed to define whether Synflorix gives cross protection against NTHi in patients. This is because the majority of NTHi infections cause pneumonia and the causative agent of pneumonia is rarely defined when patients go to their GPs. To get around this we developed a mouse model of NTHi lung infection. When infected with NTHi mice got sick and we were able to detect the bacteria in the lungs and airways. This gave us a model to test the vaccine in. Mice were given Synflorix (the vaccine we wanted to test) or another pneumonia vaccine called Prevenar (which has the same Streptococcus components, but doesn’t use the NTHi protein). The vaccine was good at inducing an immune response and we were able to measure antibodies specific to NTHi Protein D. So far so good, unfortunately when we infected the vaccinated mice with NTHi, there was no difference between the vaccinated ones and the control/ unvaccinated mice, either in how sick they got or how much bacteria was in their lungs. From this we drew the conclusion that sadly in our hands and in our system, Synflorix does not give the extra protection against the second organism – NTHi. This doesn’t mean it isn’t a highly effective vaccine against S. pneumoniae, which it is.

The no’s have it.


This study was a piece of what we call negative data. Ben Goldacre (@Bengoldacre) goes into this much more eloquently than me, but negative data is really important. It may not be as exciting as finding something new and positive, but by publishing negative, you 1) stop someone else doing the same thing which 2) reduces the cost of repeated research, 3) reduces the numbers of people and animals that get used in the process. This work was kindly supported by the charity Sparks and whilst in this case did not change clinical practice, it is an important piece of basic research. It may inform policy about which vaccine to use and therefore free up money to be spent elsewhere. We are really grateful to the people who raise money for the charity and encourage you all to get on your running shoes and raise more money so we can perform the research to find out whether vaccines do (or just as importantly do not) work.