
84 AJTCCM VOL. 29 NO. 2 2023
ABSTRACTS
* Co-senior authors
Corresponding author: K Dheda (keertan.dheda@uct.ac.za)
Background. Extensively drug-resistant tuberculosis (XDR-TB) is an
increasing public health concern, as drug resistance is outpacing the
drug development pipeline. Alternative immunotherapeutic approaches
are needed.
Methods. Dendritic cells (DCs) were cultured from XDR-TB patient-
derived peripheral blood monocytes (N=30) by maturation with
Mycobacterium tuberculosis (MTB)-specic antigens, with/without a
maturation cocktail (interferon-γ, interferon-α, CD40L, interleukin
(IL)-1β, and TLR3, TLR7 and TLR8 agonists). Two peptide pools
were evaluated: (i) an ECAT peptide pool (ESAT6, CFP10, Ag85B
and TB10.4 peptides); and (ii) a PE/PPE peptide pool (PE and PPE
peptides). A sonicated lysate of HN878 served as an antigen control.
DCs were assessed for the expression of key maturation markers and
the secretion of 1-polarising cytokines. e ability of the DC-primed
peripheral blood mononuclear cells to restrict the growth of MTB-
infected monocyte-derived macrophages was evaluated using an in vitro
mycobacterial containment assay.
Results. In patients with XDR-TB, DCs matured with MTB antigen +
cocktail, compared with DCs matured with MTB antigen only, showed
significantly higher upregulation of key co-stimulatory molecules,
CD80, CD83, CD86, and CCR7 (p<0.001 for all comparisons), and
higher secreted levels of the IL-12p70 (0.67 v. 0.01 ng/mL/106 cells;
p<0.001). e matured DCs enhanced antigen-specic CD8+ T-cell
responses to ESAT6 (p=0.05) and Ag85B (p=0.03). Furthermore,
containment was signicantly higher with MTB antigen + cocktail v.
antigen alone (p=0.0002 for PE/PPE). PE/PPE + cocktail-matured DCs
achieved a higher magnitude of containment compared with ECAT +
cocktail-matured DCs (50% (interquartile range 39 - 75) v. 46% (15 -
62); p=0.02).
Conclusion. In patients with XDR-TB, an eector response primed
by the PE/PPE peptide pool and cocktail-matured DCs was capable of
restricting the growth of MTB in vitro. ese data support the generation
of a DC-based immunotherapeutic intervention for therapeutically
destitute patients with drug-resistant TB. Further mechanistic studies
and future phase 1 human clinical studies are warranted.
Cough aerosol sampling of tuberculosis
patients indicates that highly infectious
‘super-spreader’ individuals can be
accurately identied through predictive
modelling of GeneXpert cycle thresholds
S Meier, S Jaumdally, S Oelofse, A Esmail, D Willemse, A Pooran,
K Dheda
Centre for Lung Infection and Immunity, Division of Pulmonology,
Department of Medicine, Faculty of Health Sciences, University of
Cape Town and UCT Lung Institute, Department of Medicine, Faculty
of Health Sciences, University of Cape Town, South Africa
Corresponding author: S Meier (smeier@sun.ac.za)
Background. To control the spread of tuberculosis, particularly at
community level, it is critical to identify highly infectious individuals,
who should be isolated and their contacts traced. Given the current
resource limitations, it would be desirable to target the most infectious
individuals, especially in the context of screening (active case nding)
programmes. In a previous study, we used a cough aerosol sampling
system (CASS) to identify individuals who exhaled culturable and
presumably contagious Mycobacterium tuberculosis (MTB) in their
cough aerosols (particles <10 µM).
Objectives. To rapidly identify highly infectious super-spreader
individuals.
Methods. In the previous study, of 452 tuberculosis patients, 310 did
not emit any culturable cough aerosols (CASS negative), while 142 had
culturable MTB (CASS positive) in their respirable cough droplets.
Here, we performed a more granular analysis by dividing the CASS-
positive patients into two groups, those who emitted 1 - 9 (n=100) and
those who emitted ≥10 (the ‘super-spreaders’, n=42) colony-forming
units in their cough aerosols. We compared various demographic,
clinical, physiological and diagnostic characteristics between the three
groups and performed predictive linear modelling (with bootstrapping)
to evaluate whether super-spreaders could be accurately identied.
Results. Compared with CASS-negative patients, super-spreaders were
signicantly more likely (p<0.05 for all comparisons) to be male (79%
v. 55%; relative risk (RR) 2.6 (1.3 - 5.3)), to be HIV negative (79% v.
54%; RR 2.8 (1.4 - 5.6)), to be smokers (76% v. 60%; RR 1.97 (1 - 3.9)),
to have a persistent cough (86% v. 69%; RR 2.5 (1.1 - 5.7)), and to have
better lung function. In addition, super-spreaders had a signicantly
(p<0.0001 for all comparisons) shorter time to sputum culture positivity
(6.1 (0.5) v. 21.2 (1.1) days), a lower GeneXpert cycle threshold (CT)
average (CT 16.1 (0.5) v. 22.4 (0.3)), and higher average smear grade
(3.6 (0.2) v. 1.4 (0.1)). Predictive modelling using the Xpert CT alone
(preferred for its fast turnaround time) revealed that super-spreaders
could be distinguished from CASS-negative patients with a sensitivity
of 0.83 (0.78 - 0.86) and specicity of 0.73 (0.63 - 0.78).
Conclusion. Tuberculosis super-spreaders can be accurately identied
using readily available demographic, clinical and GeneXpert results,
showing great potential to inform isolation and contact tracing
protocols that could considerably limit the transmission of tuberculosis.
SARS-CoV-2 viral replication persists
in the human lung for several weeks
after symptom onset in mechanically
ventilated patients with severe COVID-19:
A prospective tissue-sampling study
M Tomasicchio, S Jaumdally, A Pooran, A Esmail, L Wilson,
A Kotze, L Semple, S Meier, K Pillay, R Roberts, R Kriel,
R Meldau, S Oelofse, C Mandviwala, J Burns, R Londt, M Davids,
C van der Merwe, A Roomaney, L Kühn, T Perumal, A J Scott, M J
Hale, V Baillie, S Mahtab, C Williamson, R Joseph, A Sigal,
I Joubert, J Piercy, D omson, D L Fredericks, M G A Miller,
M C Nunes, S A Madhi, K Dheda
Centre for Lung Infection and Immunity, Division of Pulmonology,
Department of Medicine, Faculty of Health Sciences, University of
Cape Town and UCT Lung Institute, Cape Town, South Africa
Corresponding author: K Dheda (keertan.dheda@uct.ac.za)
Background. The immunopathogenesis of severe COVID-19 is
incompletely understood. Remdesivir is not recommended in
mechanically ventilated (MV) patients. In the upper respiratory tract
(URT), replicating (culturable) SARS-CoV-2 is recoverable for ~ 4