Early BCG vaccination is unrelated to pulmonary immunity against Mycobacterium tuberculosis in adults
- Christian Herzmann1,2,
- Giovanni Sotgiu3,
- Tom Schaberg4,
- Martin Ernst5,
- Steffen Stenger6 and
- Christoph Lange1,7,8⇑
- for the German TB or Not TB Consortium9
- 1Division of Clinical Infectious Diseases, Research Center Borstel, Borstel, Germany
- 2Center for Clinical Studies, Research Center Borstel, Borstel, Germany
- 3Epidemiology and Medical Statistics Unit, Dept of Biomedical Sciences, University of Sassari, Research Medical Education and Professional Development Unit, AOU Sassari, Sassari, Italy
- 4Center of Pneumology, Agaplesion Deaconess Hospital Rotenburg, Rotenburg, Germany
- 5Division of Immune-Cell Analytics, Research Center Borstel, Borstel, Germany
- 6Institute for Medical Microbiology and Hygiene, University Hospital of Ulm, Ulm, Germany
- 7German Center for Infection Research (DZIF), Clinical Tuberculosis Unit, Borstel, Germany
- 8Dept of Internal Medicine, University of Namibia School of Medicine, Windhoek, Namibia
- 9For a list of the participating centres of the TB or Not TB network see the Acknowledgements section
- Christoph Lange, Division of Clinical Infectious Diseases, DZIF Clinical Tuberculosis Unit, Research Center Borstel, Parkallee 35, 23845 Borstel, Germany. E-mail: clange{at}fz-borstel.de
To the Editor:
Vaccination with Mycobacterium bovis bacille Calmette–Guérin (BCG) is performed for the prevention of tuberculosis. M. bovis BCG vaccination is among the most commonly applied of all vaccines worldwide [1]. M. bovis BCG vaccination efficiently reduces the morbidity and mortality of tuberculosis in children, especially miliary tuberculosis and meningitis [2].
Although recent investigations of Mycobacterium tuberculosis-specific immune responses by interferon-γ release assays (IGRAs) provide evidence on the effect of M. bovis BCG vaccination on the prevention of primary infection with M. tuberculosis [3–5], it has been suggested that this effect diminishes during adolescence [2, 6]. Consequently, adults are probably not protected from pulmonary tuberculosis by BCG vaccination.
To date, no study has investigated the impact of M. bovis BCG vaccination performed in childhood on pulmonary immune responses in adults. The objective of this study was to assess the effect of childhood M. bovis BCG vaccination on systemic and pulmonary immune responses to M. tuberculosis in healthy adult individuals exposed to patients with acid-fast bacilli (AFB)-positive sputum smear-positive tuberculosis in Germany.
An observational, cross-sectional, multicentre study was conducted by the German Ministry of Education and Research-funded research consortium on “Pulmonary Tuberculosis – Host and Pathogen Determinants of Resistance and Disease Progression (TB or Not TB)”. Healthcare workers (HCWs) with 1) ongoing professional contact with patients with AFB sputum smear-positive tuberculosis, 2) a cumulative professional exposure of at least 2 years, and 3) no clinical signs and/or symptoms of active tuberculosis were recruited at 18 German pulmonary medicine centres (centres are listed in the Acknowledgements section).
Furthermore, household contacts (HHCs) without evidence of active tuberculosis were enrolled at three urban municipal healthcare centres (i.e. Frankfurt, Hamburg and Hannover); their enrolment required 1) the absence of clinical signs and/or symptoms of active tuberculosis, and 2) cumulative exposure of >40 h to an AFB sputum smear-positive patient with culture-proven pulmonary tuberculosis. Individuals with a history of pulmonary tuberculosis who completed a standard course of tuberculosis treatment >6 months before enrolment and did not experience a relapse were recruited in a control group.
Epidemiological, clinical and demographic data, including BCG vaccination status, were captured using an ad hoc standardised questionnaire. An unblinded physician verified M. bovis BCG vaccination by clinical examination of a scar or through the subject’s vaccination passport. IGRAs were performed on cells collected from the peripheral blood of all subjects by the QuantiFERON Gold In-Tube (QFT; Cellestis Qiagen, Chadstone, Australia) or T-SPOT.TB (ELISPOT; Oxford Immunotec, Oxford, UK) at the attending physician’s discretion. The use of two different test systems impaired the overall comparability of the data but reflected clinical practice in the 18 centres involved. Tuberculin skin tests were not regularly performed according to German guidelines [7]. The laboratory staff was blinded to the clinical data, vaccination status and tuberculosis disease status of the participants.
All enrolled contacts were offered a bronchoscopy with bronchoalveolar lavage (BAL), unless bronchoscopy was contraindicated for medical reasons. If the subject agreed, flexible bronchoscopy was performed according to the current German guidelines [8]. The bronchoscope was wedged into a subsegmental bronchus of the middle lobe. BAL was performed with a total volume of 200–250 mL sterile normal saline.
A T-SPOT.TB was performed on BAL mononuclear cells (BALMCs). Interferon-γ secretion was evaluated in response to the following mycobacterial antigens: 6-kDa early secretory antigenic target (ESAT-6), 10-kDa culture filtrate protein (CFP-10) and purified protein derivative (PPD; Statens Serum Institutet, Copenhagen, Denmark). The assay was conducted according to the manufacturer’s instructions with the exception that 200 000, instead of 250 000, mononuclear cells were plated per well for improving the visibility of the spots.
Qualitative variables were expressed as percentages, whereas mean±sd) and median (interquartile range (IQR)) were used for parametric and nonparametric quantitative variables, respectively. The Shapiro–Wilk test, Chi-squared test, t-test and Mann–Whitney test were performed as applicable using Stata 11.0 (StataCorp LP, College Station, TX, USA).
Between June 2008 and July 2012, 512 eligible HCWs and HHCs with known BCG vaccination status were identified (263 HCWs, 166 HHCs and 83 with unclassified exposure). The exposure classification was assigned through a questionnaire, but this information was not captured in the initial phase of the study. Of the remaining 505 HIV-uninfected HCWs and HHCs, 322 (63.8%; 153 HCWs, 113 HHCs and 56 with unclassified exposure) were immunised with M. bovis BCG during childhood, while 183 (36.2%; 103 HCWs, 53 HHCs and 27 with unclassified exposure) were not. In 22 (4.4%) persons (12 HCWs and 10 with unclassified exposure), QFT results were unavailable. In 483 contacts with known blood IGRA status, 170 (35.2%; 92 HCWs, 55 HHCs and 23 with unclassified exposure) had a positive test result. To assess pulmonary immune responses, 88 (17.4%) out of 505 individuals (31 unvaccinated and 57 M. bovis BCG vaccinated) underwent bronchoscopy with subsequent BALMC IGRA. Of these, 26 (29.5%; 17 HCWs, four HHCs and five with unclassified exposure) had a positive test result, with 10 (32%) in the unvaccinated and 16 (28%) in the vaccinated group.
M. tuberculosis-specific IGRA responses were not associated with M. bovis BCG vaccination status (OR 1.09 (p=0.66) and 0.81 (p=0.68) for blood and BAL, respectively) (fig. 1). However, the frequency of PPD-responsive BALMCs was higher in M. bovis BCG vaccinated versus unvaccinated persons, with a median (IQR) of 22 (6–101) versus 8 (1–44) spot-forming cells per 200 000 BALMCs, respectively (marginal statistical significance, p=0.0468), indicating a possible enhanced immune response in vaccinated subjects. With respect to M. tuberculosis-specific immune responses, though, no vaccination-related differences were found after stimulation with the antigens CFP-10 and ESAT-6. For CFP-10, a median (IQR) of 3 (0–17) versus 5 (1–15) spot-forming cells per 200 000 BALMC was observed for vaccinated versus unvaccinated subjects, respectively; for ESAT-6, the median values (IQR) were 2 (0–19) versus 4 (0–11) spot-forming cells per 200 000 BALMCs for the same respective groups. No difference in systemic or pulmonary M. tuberculosis-specific immune responses were observed among subgroups of HCWs and HHCs (data not shown).
We investigated the impact of M. bovis BCG vaccination during childhood on pulmonary and systemic M. tuberculosis-specific immune responses in healthy adult HCWs and HHCs with recent contact with tuberculosis patients. These immune responses are considered to be correlates of infection with M. tuberculosis [9]. Using our methodology, the results from this study suggest that vaccination has no substantial effect on the adaptive immune system with regards to protection against M. tuberculosis infection in adulthood.
The interpretation of the results is based on two observational and experimental findings. First, we did not observe a difference in the frequency of latent tuberculosis infection, assessed by IGRA responses from peripheral blood cells, among recent adult close tuberculosis contacts with and without a history of M. bovis BCG-vaccination in childhood. Second, local pulmonary ESAT-6- and CFP-10-specific immune responses measured by BAL-IGRA did not differ between the groups, while PPD-specific pulmonary immune responses were augmented in M. bovis BCG-vaccinated individuals.
In summary, we present data from the largest cohort reported, to our knowledge, of healthy adult contacts of tuberculosis patients evaluated by bronchoscopy for pulmonary immune responses against M. tuberculosis and M. bovis BCG. In support of previous epidemiological data, we were unable to find a protective effect of M. bovis BCG vaccination against the development of latent infection with M. tuberculosis by adulthood [2]. The results underline the importance of the analysis of functional M. tuberculosis-specific immune responses from human lung cells for the assessment of correlates of immunity in future tuberculosis vaccine trials.
Acknowledgments
The participating collaborators centres of the TB or Not TB network in Germany are: M. Allewelt, D. Anacker (both Evanglische Lungenklinik, Berlin), K. Avsar (Asklepios Fachkliniken München-Gauting, Munich), S. Blaas (Klinik Donaustauf, Donaustauf), J. Doehling (Research Center Borstel, Borstel), M. Fischer (Klinikum Nuremberg Nord, Nuremberg), A. Glaewe (Research Center Borstel, Borstel), P. Hammerl (Lungenfachklinik, Immenhausen), S. Jaletzky (Deakoness Hospital, Rotenburg), S. Junghanß (Fachkrankenhaus Coswig, Coswig), F. Kunitz (Helios Klinikum Emil von Behring, Berlin), L. Krabbe (Research Center Borstel, Borstel), S. Kohr (Agaplesion Pneumologische Klinik Waldhof, Elgershausen, Greifenstein), B. Lehnigk (Lungenclinic Großhansdorf, Großhansdorf), V. Leucht (Fachkrankenhaus Coswig, Coswig), J. Meywirth (Krankenhaus Diekholzen, Diekholzen), M. Mowe (Lungenklinik Hemer, Hemer), R. Muetterlein (Fachkrankenhaus Parsberg, Parsberg), A. Neher (Asklepios Fachkliniken München-Gauting, Munich), M. Pletz (Hannover Medical School, Hannover), J. Paepke (Helios Klinikum Emil von Behring, Berlin), C. Prignitz (Krankenhaus Bethanien Solingen, Solingen), A. Quassem (Lungenfachklinik Lostau, Lostau), K.D. Schneider (Hannover Oststadtkrankenhaus, Hannover) and N. Vorreiter (Lungenfachklinik, Immenhausen).The authors are grateful for the technical assistance and support of Franziska Daduna, Jessica Hofmeister, Karoline Gaede, Stefan Niemann (all Research Center Borstel, Borstel, Germany), Daniel Mayer (Dept of Medical Microbiology and Hygiene, University of Ulm, Ulm, Germany), Jana Reppe, Lukas Tittmann (both Biobank Popgen, Kiel, Germany), and Steffi Jaletzki (Deaconess Hospital Rotenburg, Rotenburg, Germany).
Footnotes
Support statement: This study was funded by the German Ministry of Education and Research (Bundesministerium für Bildung und Wissenschaft reference 01KI0784; grant to Christoph Lange, Steffen Stenger and Tom Schaberg). Christoph Lange is supported by the DZIF.
Conflict of interest: Disclosures can be found alongside the online version of this article at erj.ersjournals.com
- Received March 4, 2014.
- Accepted May 14, 2014.
- ©ERS 2014