| Summary: | <p><strong>Background</strong></p>
<p>Salmonella Typhi (S. Typhi), the aetiological cause of typhoid fever, is responsible for up to 20.6 million infections every year, the majority of which occur in children under 15 years of age. A Vi-tetanus toxoid conjugate vaccine (Vi-TT) has proven to be highly protective against the development of typhoid fever and vaccine implementation is expected to have a significant effect on rates of disease in endemic settings.</p>
<p><strong>Research question 1</strong></p>
<p>Cellular immunity is likely to play a substantial role in protection from typhoid fever. However, early progression of disease remains poorly understood in humans as these immunological responses can rarely be examined prior to symptom onset. Understanding of host responses is further impeded by lack of appropriate animal models of S. Typhi infection. Human challenge studies offer an opportunity to circumvent these issues and offer valuable insights.</p>
<p><strong>Research question 2</strong></p>
<p>Clinical presentation of typhoid fever is highly heterogeneous with severe complications often requiring hospitalisation. Identification of individuals at risk of worsened outcome is critical for proper treatment and improved prognosis. C-reactive protein (CRP) and white blood cell count (WBC) are commonly used as prognostic indicators of patient condition, however, measures such as CRP are not always available in resource limited settings. In addition, few previous studies have sought to assess whether these indicators are optimally suited to this purpose in the context of typhoid fever. Whether better correlates of disease severity can be identified to serve this purpose is an unanswered question.</p>
<p>To address these identified gaps in knowledge, a longitudinal study of cellular immunity was conducted following S. Typhi infection in a human challenge model. This work aimed to assess:</p>
<p>1. The progression of immune responses in peripheral blood after infection and the effect of Vi-vaccination.</p>
<p>2. The utility of blood count ratios as biomarkers of disease severity.</p>
<p>3. The impact of vaccination on severity biomarkers and symptom presentation.</p>
<p><strong>Methods</strong></p>
Peripheral blood samples (PBMCs) and full blood count were collected from individuals enrolled in a Phase IIb clinical trial and human challenge study evaluating the efficacy of two typhoid fever vaccines, Vi-PS and Vi-TT. Enrolled individuals were UK residents with no previous history of typhoid fever. Participants were randomised to receive either a single dose of Vi-TT (Bharat Biotech), Vi-PS (Sanofi Pasteur) or a control vaccine, MenACWY (GlaxoSmithKline). Approximately 1 month after vaccination, participants were challenged with 1-5 x 104 CFU of S. Typhi. PBMCs and full blood count were collected at baseline, following vaccination and for several days after infection until either diagnosis or for up to two weeks after challenge. Following diagnosis or completion of the 2-week observation period, participants were treated with antibiotics.</p>
<p>PBMCs where phenotyped by mass cytometry measuring 37 markers in total (n = 57 participants, protected: n = 28, diagnosed: n = 29). Cellular subsets were identified by clustering analysis and differential abundance of populations after infection were assessed by negative binomial mixture models. Signatures of cellular immunity were identified by partial least squares-discriminant analysis (PLS-DA) and their association with infection outcome was evaluated.
During challenge, participants completed an online diary self-assessing development of symptoms and relative severity. Participants were asked to report on the following with a grade between 0 (symptom not present) to 3 (severe): headache, feeling generally unwell, abdominal pain, constipation, cough, diarrhoea, joint pain, loss of appetite, muscle pain and nausea. Subpopulations of participants with differing symptom profiles at diagnosis were identified by latent profile analysis (LPA).</p>
<p><strong>Results</p>
Early infection:</strong> Within 24 hours of infection, PBMC composition was significantly perturbed by S. Typhi infection in all participants regardless of whether they were protected or developed acute disease (defined as either positive blood culture or persistent fever). PLS-DA identified 3 profiles of cellular immunity at the 24-hour timepoint (referred to henceforward as A, B and C) that were differentially associated with the development of typhoid fever. Profile A had the highest proportion of protected participants (80%, n = 15). These individuals exhibited a significant expansion of CD16+ monocytes after challenge in comparison to other profiles, as well as having higher frequencies of naïve B-cells at baseline. Conversely, participants in profile C had the lowest proportion of protected participants (8.3%, n = 12) and were found to have high baseline frequencies of terminally differentiated (TEMRA) T-cells and significant recirculation of CD62L+ memory IgA+ and IgG+ B-cells after infection. Remaining individuals developed an intermediatory profile (Profile B, n = 25), characterized by absence of both changes in monocytes and memory B-cells. Participants in all profiles showed significant expansion in CD62L+ CD56dim NK cells. Vaccination status was modestly associated with profile of cellular immunity although the proportion of vaccinees in each profile was not significantly different.</p>
<p><strong>Acute disease:</strong> Two symptom profiles, profile A (n = 34) and profile B (n = 12), were identifiable in individuals that developed acute symptoms after challenge. Participants in profile B reported significantly greater numbers of symptoms (p = 1.02 x 10-6) and higher prevalence of fever (p = 8.97 x 10-4). On evaluation of full blood count data, neutrophil-to-lymphocyte ratio (NLR) and eosinophil-to-lymphocyte ratio (ELR) were both significantly perturbed in all individuals with typhoid fever (p = 3.42 x 10-11 and p = 1.93 x 10-11 respectively), however perturbation of NLR and ELR was statistically greater in participants in profile B. This suggests these biomarkers could successfully identify more severe clinical presentation of disease. Additionally, NLR and ELR were found to be more sensitive identifiers of profile B participants than either CRP or WBC. Vi-TT vaccination, but not Vi-PS, significantly lowered NLR and number of self-reported symptoms. In vaccinees, NLR was significantly negatively correlated with Vi-IgG titres but not Vi-IgA or Vi-IgM suggesting Vi-IgG titres may play a significant role in reducing disease severity.</p>
<p><strong>Conclusions</strong>
In conclusion, signatures of immunity are identifiable in peripheral blood as early as 24 hours after oral challenge. These profiles have significant associations with outcome of infection but are only modestly associated with vaccination status.
This analysis suggests Vi-vaccination has a modulatory effect on symptom presentation in the event of acute disease. Vi-TT, but not Vi-PS vaccinees, developed significantly fewer symptoms than unvaccinated participants. NLR and ELR were significant, independent correlates of disease severity that may have prognostic potential in the management of typhoid fever.</p>
|
|---|