The Diagnostic Value Of Mycobacterium Tuberculosis RD1 Gene Encoding Antigen-specific Cellular Response To Active Tuberculosis And Latent Tuberculosis Infection And Related Basic Research

Background China is one of the 22 countries in the world with high-burden of tuberculosis, early and accurate diagnosis of tuberculosis is critical to control TB epidemic. Traditional methods for diagnosing tuberculosis have their shortcomings:relevance ratio of acid-fast stains is low, culture of M. tuberculosis takes long time, and histopathologic examination is usually invasive. The tuberculin skin test (TST) is widely used, but its accuracy of diagnosis is affected by immunosuppressive condition and BCG vaccination. T-SPOT. TB on PBMC (Oxford Immunotec, Abingdon, UK) can be used to diagnose tuberculous infection through detecting the frequencies of IFN-y producing T cells which respond to the M. tuberculosis specific antigens encoded by the RD1 region(ESAT-6 and CFP-10). Case-control studies have shown that T-SPOT. TB on PBMC has higher sensitivity and specificity than TST and is not influenced by immunosuppressive condition and BCG vaccination. Prospective cohort studies on evaluation of performance of T.SPOT.TB on PBMC in the diagnosis of tuberculosis are mostly conducted in countries with low and intermediate epidemics of tuberculosis, data from the 22 high burden countries is very limited. There is no published data from Mainland of China. Otherwise, it is difficult to achieve confirmed bacteriological diagnosis of tuberculous serositis/meningitis. In regions with high epidemic of tuberculosis, high infection rate of latent TB reduces the specificity of T-SPOT. TB on peripheral blood in diagnosis of active tuberculosis. While in patients with serous effusion, the sensitivity of T-SPOT.TB in diagnosis of active tuberculosis is also affected. Since antigen-specific effector T cells tend to accumulate to lesion sites, by detecting specific interferon-y release T cells in serous effusion/cerebrospinal fluid via stimulation of RD1-encoded M. tuberculosis antigen, it is expected to diagnose tuberculous serositis/meningitis more sensitively and more specifically. Healthcare workers have an increased risk of tuberculosis infection compared with the general population. Diagnosis of latent tuberculosis based merely on TST would lead to considerable false positive results in individuals with BCG vaccination, an accurate approach is necessary to evaluate the state of latent tuberculosis infection (LTBI) in healthcare workers. There is no published data on performance of T-SPOT. TB for detecting LTBI among healthcare workers in a general hospital in China. Cellular immune response is mainly induced in Mycobacterium tuberculosis infection, CD4+T cells subsets contribute to the immune response against M. tuberculosis.To explore the response induced by RD1-encoded specific antigens in different diseases status in M.tuberculosis infection can improved our understanding of the machanism of immune response. there is no published data on change of RD1-encoded antigen specific T cells in serousl effusion/cerebrospinal fluid along with the decrease of antigen load.Objective 1) To evaluate the utility value of T-SPOT.TB on PBMC for diagnosis of active tuberculosis in high epidemic region, and analyze the factors which influence the sensitivity and specificity of T-SPOT. TB on PBMC; to assess the significance of frequencies of tuberculosis-specific IFN-y producing T cells in diagnosis of active tuberculosis, and investigate the cut-off value of spot-forming cells which differentiates active tuberculosis from inactive tuberculosis LTBI and previoustuberculosis).2)To establish the method of detecting RD1-encoded M. tuberculosis antigen specific interferon-?release T cells in serous effusion (hydrothorax, hydroperitoneum, hydropericardium, joint fluid)/cerebrospinal fluid through T-SPOT. TB; to estimate the value of applying T-SPOT. TB on serous effusion and cerebrospinal fluid in diagnosis of active tuberculosis compared with T-SPOT. TB on peripheral blood; To explore the significance of spot forming cell frequency in diagnosis of active tuberculosis through comparing T-SPOT. TB on peripheral blood, serous effusion and cerebrospinal fluid, so as to improve the sensitivity and specificity of T-SPOT. TB.3) To compare the performance of T-SPOT.TB for detecting LTBI with tuberculin skin test among healthcare workers in a general hospital in Beijing, and evaluate diagnostic concordance and risk factors for LTBI.4) To explore association between RD1-encoded M. tuberculosis antigen-specific IFN-y secreting CD4+T cells and clinical disease state, change of RD1-encoded antigen specific T cells in serousl effusion/ cerebrospinal fluid along with effective treatment (decrease of antigen load).Methods 1)Five hundred and ninety one in-patients (HIV-negative, at ages of sixteen or above) of suspected active tuberculosis or of those that need to be differentiated from active tuberculosis were prospectively enrolled at Peking Union Medical College Hospital from December 2005 to March 2009. According to clinical manifestations and data of radiology, smear acid-fast stains, TB culture, histopathological examination as well as response to anti-TB therapy, the enrolled patients were finally assigned to four predefined groups, i.e. bacteriologically and/or histologically confirmed tuberculosis, clinically diagnosed tuberculosis, clinical indeterminate and. active tuberculosis excluded, respectively. Based on final diagnosis, the utility value of T-SPOT.TB on peripheral blood for diagnosis of active tuberculosis was evaluated.2) One hundred and seven in-patients (HIV-negative, at ages of sixteen or above) of suspected tuberculous meningitis or of suspected tuberculous serositis with serous effusion (hydrothorax, hydroperitoneum, hydropericardium, joint fluid) as main manifestations were prospectively enrolled at Peking Union Medical College Hospital from June 2005 to March 2010. T-SPOT. TB detecting of peripheral blood and serous effusion/cerebrospinal fluid was conducted for each patient at the same time. The patients were followed up and assigned to three groups as active tuberculosis, clinical indeterminate and active tuberculosis excluded respectively, according to criterions above. Based on final diagnosis, the applications of T-SPOT.TB on peripheral blood and serous effusion/cerebrospinal fluid as well as TST in diagnosis of tuberculous serositis/meningitis were compared.3) We conducted a cross-sectional study of 101 healthcare workers in PUMCH,a general tertiary hospital in Beijing. Subjects with radiological features suggestive of prior tuberculosis were excluded. The T-cell-based interferon?release assay, T-SPOT.TB, was compared with TST to identify LTBI.4) CD4+T cells were depleted from PBMC/PEMC by immunomagnetic microbeads, patients with tuberculous serositis/meningitis were followed up when available.Results 1)Five hundred and eighty three enrolled cases obtained complete clinical data. Among which, forty seven cases were bacteriologically and/or histologically diagnosed of tuberculosis, ninety five cases were clinically diagnosed of tuberculosis, thus a total of one hundred and thirty nine cases (23.9%) were diagnosed of active tuberculosis; forty six cases (7.9%) need further diagnosis, and three hundred and ninety eight cases (68.3%) were diagnosed of active tuberculosis excluded. Nine enrolled cases were lost to follow up. The sensitivity, specificity, NPV and PPV of T-SPOT.TB on PBMC for diagnosis of active tuberculosis were 83.0%,65.0%,91.6% and 45.3%, respectively. The sensitivity for bacteriologically and/or histologically diagnosed patients was significantly higher than that for clinically diagnosed patients (95.7% vs 76.4%, P=0.005).Logistic regression analysis showed that serous effusion was independent risk factor affecting the diagnosis sensitivity of T-SPOT.TB on PBMC (RR=0.19,95% CI,0.07-0.53). The sensitivity of T-SPOT.TB on PBMC in patients without serous effusion was significantly higher than in patients with serous effusion (90.4% vs 71.4%, P=0.004) History of previous tuberculosis or radiological features of old tuberculosis was independent risk factor affecting the diagnosis specificity of T-SPOT.TB on PBMC (RR=0.54,95%CI,0.34-0.86), patients with previous tuberculosis history or radiological changes of old tuberculosis showed a significant decrease in diagnosis specificity (71.0% vs 47.2%, P=0.001).In patients that diagnosed of active tuberculosis, the median of spot forming cell frequencies of T-SPOT.TB on PBMC was 442 SFCs/106PBMC, spot forming cell frequencies were significantly higher for bacteriologically/histologically confirmed patients than for clinically diagnosed patients (604 SFCs/106 PBMC vs 276 SFCs/106 PBMC,P=0.033). In patients that diagnosed of active tuberculosis excluded, the median of spot forming cell frequencies of T-SPOT.TB on PBMC was 124 SFCs/106 PBMC, significantly lower than that in active tuberculosis patients (P<0.001). If a cut-off value of spot forming cell frequencies of T-SPOT.TB on PBMC was set to 72 SFCs/106 PBMC, the sensitivity, specificity and AUROC for diagnosis of active tuberculosis were 71.9%,76.4% and 0.810, respectively (95%CI,0.769-0.852). If a cut-off value of spot forming cell frequencies of T-SPOT.TB on PBMC was set to 652 SFCs/106 PBMC, the specificity for diagnosis of active tuberculosis were 95%. 2) Thirty-two cases were with final diagnosis of tuberculous serositis (seven of which were confirmed bacteriologically or histologically), thirty-six cases were excluded from tuberculous serositis, twelve cases were diagnosed of tuberculous meningitis (three of which were confirmed bacteriologically), fourteen cases were excluded from tuberculous meningitis, eight cases were not clarified a diagnosis, and five cases were lost to follow up. In patients with tuberculous serositis, the sensitivity of T-SPOT.TB on serous effusion was 96.9%, which was significantly higher than that on peripheral blood (78.1%, P=0.021), both were significantly higher than that of TST (37.5%). The specificity of T-SPOT.TB on serous effusion was 88.9%, which was higher than that on peripheral blood (77.8%), but showed no significant difference (P=0.206). The PPV of T-SPOT.TB on serous effusion was 88.6%, which was significantly higher than that of TST(52.9%) but showed no significant difference with that on peripheral blood (75.6%, P=0.166).The NPV of T-SPOT.TB on serous effusion was 97.0%, which was significantly higher than that on peripheral blood (80.0%, P=0.030), and both were significantly higher than that of TST. The frequencies of IFN-y release specific T cells detected by T-SPOT.TB on serous effusion were significantly higher than that on peripheral blood (1276[160-3648] vs 456[244-744], P=0.001). Taking 60 SFCs/106 PEMC as the cut-off value, the sensitivity, specificity, PPV, NPV and AUROC of T-SOPT.TB on serous effusion in diagnosis of tuberculous serositis were 90.6%,94.4%,93.5%,91.9% and 0.974 (95%CI,0.934-1.013), respectively. In the patients with tuberculous meningitis, the sensitivities of T-SPOT.TB on peripheral blood and on cerebrospinal fluid were 97.1% and 58.3% respectively, both were higher than that of TST, and the specificity and PPV of T-SPOT.TB on cerebrospinal fluid were all 100%.3) In 101 healthcare workers, 29 (28.7%,95%CI 19.9%-37.5%) were positive by T-SPOT.TB compared to 53 (55.2%,95%CI 64.0%-81.8%) by TST. Agreement between two tests was low (57.3%,k=0.18,95%CI,0.01-0.52). In multivariate analysis, direct exposure to sputum-smear positive TB patients was a significant risk factor for T-SPOT. TB (OR=3.68,95%CI,1.44-9.40), however, no risk factors were independently associated with a positive TST result. Of the 51 participants that had direct contact with sputum-smear positive TB cases,21 (41.2%,95%CI 27.7%-54.7%) were T-SPOT.TB positive, which was significantly higher than the rate of positivity for the other 50 participants without direct contact (16%, 8/50,95%CI 5.8%-26.2%, P=0.005). Pooled frequency of ESAT-6 and CFP-10-specific IFN-y secreting T cell was 268 SFCs/million (IQR 90-522) for participants of having direct contact with sputum-smear positive TB patients, significantly higher than that (94, IQR 40-231) for participants without direct contact (P=0.045).4) The frequencies of RD1-encoded M. tuberculosis antigens, including ESAT-6, CFP-10, Rv3879c and recombinant ESAT-6 protein, specific IFN-y secreting T cells were higher for active tuberculosis patients than for LTBI; tuberculous antigen specific T cells accumulated at site of active disease and decreased during treatment in patients with tuberculous serositis/meningitis, especially at infectious site.Conclusion 1)The sensitivity of T-SPOT.TB on PBMC for diagnosis of active tuberculosis showed in this study is in consistence with those in foreign reports, and the specificity showed is lower than those in countries with low epidemic of tuberculosis. As independent risk factors of T-SPOT.TB on PBMC for diagnosis of active tuberculosis, serous effusion affected the sensitivity, and history of previous tuberculosis or radiological features of old tuberculosis affected the specificity. The high NPV (91.6%) suggested that negative result of T-SPOT.TB on PBMC was valuable for excluding active tuberculosis. The low PPV (45.3%) revealed that, in regions with high epidemic of tuberculosis, positive result of T-SPOT.TB on PBMC had limited value for diagnosing active tuberculosis with a cut-off value of 24SFc/106 PBMC. The spot forming cell frequencies were significantly higher for active tuberculosis patients than for patients with latent tuberculosis or with old tuberculosis infection, an elevated cut-off value of spot forming cell frequencies could raise the specificity and PPV of T-SPOT.TB for diagnosis of active tuberculosis.2) Compared with T-SPOT.TB on peripheral blood, T-SPOT.TB on serous effusion could significantly improve the sensitivity for diagnosis of tuberculous serositis, T-SPOT.TB on cerebrospinal fluid could significantly improve the specificity and PPV for diagnosis of tuberculous meningitis. When the cut-off point was raised to 60 SFCs/106 PEMC, the specificity and PPV for diagnosis of tuberculous serositis were significantly improved while the sensitivity and NPV were not affected. T-SPOT.TB on serous effusion/cerebrospinal fluid showed high value of application in diagnosis of tuberculous serositis/meningitis, which could promote the diagnosis of tuberculous serositis/meningitis diagnosis to an improved accuracy.3) T-SPOT.TB is more closely correlated to risk factors for LTBI than TST in healthcare worker in a general hospital in Beijing. Direct exposure to sputum-smear positive TB patients increased the positivity of the T-SPOT.TB assay in these subjects. The findings of our preliminary study indicate that the TST is probably overestimating the real prevalence of LTBI in healthcare workers. T-SPOT.TB is a more accurate test than TST for identifying LTBI in healthcare workers.4) CD4+T cells were thought to be the major cell type activating macrophage and producing IFN-y in protective immunity for tuberculosis. Compared with LTBI, immune response in patients with active tuberculosis was stronger and more extensive. Tuberculous antigen specific INF-y secreting T cells accumulated at site of active disease and decreased with antigen load during treatment in patients with tuberculous serositis/meningitis, especially at infectious site.