Rethinking the natural history of infection with M. tuberculosis

Research output: Contribution to journalArticlepeer-review


The tuberculin skin test is associated with exposure to TB, and conversion of a TST is associated with the risk of disease. A consistent number of TB cases occur over the years in persons with positive TSTs. The question is: How much of this is reactivation vs. re-exposure in an endemic setting? The actual case rate is really quite low, even among those with positive TSTs, so either only a minority of those infected with M. tuberculosis develop disease, i.e., the M. tuberculosis organism really does a poor job, or it is possible that there are exposures that do not reflect infection. In either case, we need to do a better job of identifying those at risk for developing TB. IFN-γ is associated with TB exposure, and it is quantitative, with natural variations over time. Dr Lewin sohn offered some food for thought. He expects to see rising amounts of IFN-γ with exposure, and this rise will indicate either a recent exposure to M. tuberculosis or a reactivation. Another prediction is that a negative IFN-γ may not distinguish remote exposure from no exposure. Some will be transiently exposed, and will develop TEM/TDC (like tetanus). The duration of these responses will depend upon the environment. T-cells reflect immunologic history, and his prediction is that some will vet a memory response just like tetanus, but he would predict these would be at low risk. What about understanding T-cell memory and persistent infection? In guinea pigs, conversion of the TST is associated with exposure to M. tuberculosis, but many of the animals will have skin test reversions in which no bacteria can be detected either pathologically or microbiologically. These data challenge the notion of 'once infected, always infected'. In humans, we do not yet have reliable tools to distinguish exposure from infection. However, post-mortem series of those with a positive TST often fail to reveal mycobacteria, and if present it is often found in areas of normal lung. CD8 T-cells are present at high frequency in both active and latent TB infection. CD8 T-cells preferentially recognize heavily infected cells, and may be a surrogate for bacterial burden and/or disease progression. D8 cells recognize MHCII-negative cells and may play a unique role in chronic/ persistent phases of infection. They may be a good surrogate for intercellular infection. CD8 T-cell responses to ESAT-6 and CFP-10 were measured in young children with pulmonary TB, or in the setting of household exposure. Here, CD8 responses were associated with pulmonary TB, but not residence in a household. In contrast, the CD4 response was present in both groups. These data were consistent with the hypothesis that exposure to M. tuberculosis and infection could be immunologically differentiated. Dr Lewinsohn's prediction is that a CD8 response will identify those at risk for progression but additional antigens will be needed to improve sensitivity. We know what we do not know, and what we need is a way to find out who are at greatest risk. Who are the acutely infected? We need to think next about what kind of studies need to be done and where, and whom we study.

Original languageEnglish (US)
Pages (from-to)S7-S10
JournalInternational Journal of Tuberculosis and Lung Disease
Issue number6 SUPPL. 1
StatePublished - Jun 2010

ASJC Scopus subject areas

  • Pulmonary and Respiratory Medicine
  • Infectious Diseases


Dive into the research topics of 'Rethinking the natural history of infection with M. tuberculosis'. Together they form a unique fingerprint.

Cite this