Thomas Nørrelykke Nissen 2016

Nissen, Thomas Nørrelykke. 2016. Immune modulating effects of Bacillus Calmette-Guérin vaccination at birth – Vaccine specific antibody responses and in vitro cytokine production.

Bacillus Calmette-Guérin (BCG) is the most widely used vaccine in medical history. It is solely used as tuberculosis prophylaxis, but mounting evidence suggests that BCG also has clinical non-specific effects (NSEs). Observational and randomised clinical trials in West Africa have shown that BCG has beneficial NSEs on childhood morbidity and mortality.
Immunological studies have also found altered cytokine response upon in vitro stimulation with unrelated antigens and changes in antibody levels to routine vaccines given later in life. The mechanism of these effects is not fully understood, but seems to involve induction of memory in the innate immune system.

A multicentre trial – ‘The Danish Calmette Study’ was initiated in 2012 to investigate the NSE of BCG in a high-income setting. At birth 4262 children were randomised to BCG or control and attended clinical follow-up at 3 and 13 months. The aim of this thesis was; a) to investigate the specific antibody responses from routine vaccinations in a sub-group of children assigned to BCG vaccination or control; b) to investigate cytokine responses when stimulating whole blood, 4 days, 3 and 13 months after randomisation, with a broad range of unrelated antigens and pathogens in a sub-group of children assigned to BCG vaccination or control; c) to assess the risk of adverse reaction in all children assigned to BCG and to evaluate the treatment and outcome of the prolonged and most severe adverse reactions.

In paper II we found no overall effect of BCG on antibody levels to routine vaccines. When testing for effect modification of ‘time of vaccination/randomisation’ we found that BCG was associated with higher antibody levels when randomised on day 2-7 whereas being randomised on day 0-1 was associated with lower antibody levels. This difference was statistical significant for anti-pertussis-toxin IgG and for anti-Pneumococcus Polysaccharide IgG serotypes 9v, 18c and 19f. For most antibodies girls had higher levels than boys.

In paper III we found no overall effect of BCG on cytokine responses to whole blood stimulation 4 days and 3 months after randomisation. At 13 months BCG vaccination induced lower levels of IL-10 causing BCG vaccinated children to have a higher TNF-α/IL-10 ratio.

Effect modification of ‘time of vaccination/randomisation’ was seen for all cytokines, except IFN-γ, to all stimulations in the 4 days samples. This tendency was in favour of the children randomised at day 2-7, but it did not reach statistical significance.

In paper IV we found that 10 of 2018 BCG vaccinated children developed suppurative lymphadenitis. This was almost five times more frequent than expected from the summary of product characteristics. The time from onset of symptoms to healing ranged from 63-328 days, but although a prolonged course of treatment all children responded well to conservative treatment alone. Significantly more parents of a child developing a suppurative lymphadenitis indicated that the vaccine had more adverse effects than parents of children not developing a suppurative lymphadenitis.

We were not able to show an effect of BCG on antibody levels to routine vaccinations or to cytokine response to whole blood stimulation. Analysis of effect modifications showed that age at BCG vaccination might be of importance when investigating heterologous BCG immunology. If BCG was to be re-introduced in Denmark a higher risk of suppurative lymphadenitis should be expected, but although prolonged time course, conservative treatment is successful.