Abstract:
Identifying a blood circulating cellular biomarker that can be used to assess severity
of disease and predict the time to culture conversion (TCC) in patients with
multidrug resistant tuberculosis (MDR-TB) would facilitate monitoring response to
treatment and may be of value in the design of future drug trials. We report on
the frequency of blood Ki67+HLA-DR− CD4+ T regulatory (Treg) cells in predicting
microbiological outcome before initiating second-line treatment for MDR-TB. Fifty-one
patients with MDR-TB were enrolled and followed over 18 months; a subset of
patients was sputum culture (SC) negative at baseline (n = 9). SC positive patients
were divided into two groups, based on median TCC: rapid responders (≤71 days
TCC; n = 21) and slow responders (>71 days TCC; n = 21). Whole blood at
baseline, months 2 and 6 was stimulated with M tuberculosis (Mtb) antigens and
Treg cells were then identified as CD3+CD4+CD25hiFoxP3+CD127−CD69− and further
delineated as Ki67+HLA-DR− Treg. The frequency of these cells was significantly
enlarged at baseline in SC positive relative to SC negative and smear positive
relative to smear negative patients and in those with lung cavitation. This difference
was further supported by unsupervised hierarchical clustering showing a significant
grouping at baseline of total and early differentiated memory Treg cells in slow
responders. Conversely, there was a clustering of a lower proportion of Treg
cells and activated IFNg-expressing T cells at baseline in the rapid responders.
Examining changes over time revealed a more gradual reduction of Treg cells in
slow responders relative to rapid responders to treatment. Receiver operating curve analysis showed that baseline Mtb-stimulated Ki67+HLA-DR− Treg cells could predict the TCC of MDR-TB treatment response with 81.2%sensitivity and 85%specificity (AUC
of 0.87, p < 0.0001), but this was not the case after 2months of treatment. In conclusion,
our data show that the frequency of a highly defined Mtb-stimulated blood Treg cell
population at baseline can discriminate MDR-TB disease severity and predict time to
culture clearance.
