目的

中国是结核病高负担国家，准确诊断活动性结核病(Active Tuberculosis，ATB)是消除结核病的前提。部分肺结核和肺外结核标本无法检出结核分枝杆菌(Mycobacterium tuberculosis，MTB)，免疫学诊断方法可能辅助结核病诊断。当前， γ-干扰素释放试验尚不能准确鉴别ATB与潜伏结核感染(Latent Tuberculosis Infection，LTBI)。本研究中，通过ATB与LTBI细胞因子分泌特征差异，剔除干扰肽段，筛选结核毒力抗原ESAT-6和CFP-10的优势肽段重新组合成抗原1；筛选结核潜伏相关抗原Rv1733c SLP和Rv2028c SLP的优势肽段重新组合成抗原2，提高结核免疫学诊断方法区分不同结核感染状态的能力，同时降低成本。应用荧光免疫斑点法(Fluorescence-linked Immunospot Assay, FluoroSpot)，进一步在临床验证，评价抗原1和抗原2特异性FluoroSpot诊断结核感染、ATB和鉴别诊断ATB和LTBI的准确性。进一步的，探索结核特异性抗原刺激后，不同时点下多种细胞因子的分泌特征，以期缩短结核细胞免疫诊断结果获得时间。

方法

第一部分 结核特异性毒力抗原与潜伏相关抗原优势肽段库筛选、建立及其免疫原性验证

纳入病原学确诊的ATB患者和LTBI医务工作者，将ESAT-6、CFP-10、Rv1733c SLP和Rv2028c SLP肽段各分为4~6个肽段池，分别使用各肽段池刺激患者外周血单个核细胞，使用FluoroSpot方法检测分泌细胞因子IFN-γ和IL-2 的结核特异性T细胞，计算分泌IFN-γ⁺IL-2^-、IFN-γ^-IL-2⁺和IFN-γ⁺IL-2⁺特异性T细胞的频数和比例，根据细胞因子分泌特征，筛选出优势肽段组合成新的抗原1和抗原2；

纳入结核感染者，使用Elispot和FluoroSpot方法，分别评价抗原1和ESAT-6、CFP-10全肽，抗原2和Rv1733c SLP、Rv2028c SLP全肽刺激后，特异性T细胞频数差异及相关性；

纳入病原学确诊ATB患者和LTBI，与ESAT-6、CFP-10、Rv1733c SLP和Rv2028c SLP全肽比较，初步评价抗原1、抗原2特异性FluoroSpot方法对ATB与LTBI的鉴别诊断准确性。

第二部分 结核特异性抗原优势肽段库联合FluoroSpot区分不同结核感染状态的研究

纳入临床疑诊结核患者，进行T-SPOT.TB检测，使用抗原1和抗原2作为结核特异性抗原，进行FluoroSpot检测；

在病原学确诊的ATB患者中比较T-SPOT.TB和抗原1-FluoroSpot、抗原2-FluoroSpot检测结核感染的阳性率。

比较抗原1-FluoroSpot和T-SPOT.TB诊断ATB的准确性；

评价抗原1-FluoroSpot鉴别诊断ATB与LTBI的准确性；评价在抗原1-FluoroSpot的基础上联合抗原2对ATB和LTBI鉴别诊断准确性的提升作用；

在接受免疫抑制治疗的患者中，评价抗原1-FluoroSpot鉴别诊断ATB与LTBI的诊断准确性，评价在抗原1-FluoroSpot的基础上联合抗原2对ATB和LTBI鉴别诊断准确性的提升作用。

第三部分 结核特异性抗原刺激后不同时点细胞因子分泌特征的探索

纳入结核感染者，使用ESAT-6和CFP-10全肽作为结核特异性抗原。

使用数字ELISA技术中的单分子阵列技术(Single Molecule Array，Simoa)检测结核特异性抗原刺激2h、4h和16h 时IFN-γ的分泌量，报告抗原刺激孔与阴性对照孔的IFN-γ分泌情况，使用重复测量检验各时点结核特异性IFN-γ分泌量是否存在差异。

使用Luminex技术检测结核特异性抗原刺激2h、4h、6h、8h、10h、12h、14h和16h时IFN-γ、CCL4、CXCL1、IP-10、IL-1Ra、IL-2、IL-6和IL-17的分泌量，探索结核特异性细胞因子分泌随刺激时间的变化趋势。

结果

第一部分 结核特异性毒力抗原与潜伏相关抗原优势肽段库筛选、建立及其免疫原性验证

纳入病原学确诊的ATB患者 13例，LTBI医务工作者 10例，综合比较ATB与LTBI两组分泌IFN-γ和IL-2的结核特异性T细胞频数、分泌IFN-γ⁺IL-2^-和IFN-γ^-IL-2^＋T细胞频数和比例，筛选出ESAT-6优势肽段池1、2、5和CFP-10优势肽段池1、3、5，组合成抗原1；筛选出Rv1733c SLP优势肽段池2、3、5和Rv2028c SLP优势肽段池1、2、3，组合成抗原2。

纳入72例结核感染者进行免疫原性验证。Elispot结果显示抗原1刺激后，与ESAT-6和CFP-10全肽相比，分泌IFN-γ的特异性T细胞频数无显著差异(P=0.076)，相关性系数(R)为0.877(P＜0.001)。FluoroSpot结果显示，抗原1与ESAT-6和CFP-10全肽在分泌IFN-γ特异性T细胞频数上差异显著(P＜0.001)，R为0.829(P＜0.001)，在分泌IL-2特异性T细胞频数上无显著差异(P=0.156)，R为0.826(P＜0.001)，在分泌IFN-γ^＋IL-2^＋特异性T细胞频数上无差异(P=0.083)，R为0.821(P＜0.001)。FluoroSpot结果显示，抗原2与Rv1733c SLP和Rv2028c SLP全肽在分泌IFN-γ特异性T细胞频数上无差异(P＞0.999)，其R为0.769(P＜0.001)；在分泌IL-2特异性T细胞频数上无差异(P=0.092)，R为0.870(P＜0.001)；在分泌IFN-γ^＋IL-2^＋特异性T细胞频数上无差异(P=0.381)，R为0.627(P＜0.001)。

纳入病原学确诊ATB 患者11例，临床需与ATB鉴别的LTBI 46例进行初步临床验证。抗原1对ATB和LTBI的鉴别诊断AUC为0.907(95%CI: 0.825-0.989)，显著高于毒力抗原全肽（AUC=0.768，95%CI 0.633-0.904，P=0.043）；抗原2对ATB和LTBI的鉴别诊断AUC为0.906(95%CI: 0.779-0.973)，显著高于潜伏相关抗原全肽（AUC=0.795，95%CI 0.647-0.902，P=0.048）。

第二部分 结核特异性抗原优势肽段库联合FluoroSpot-IFN-γ/IL-2区分不同结核感染状态的研究

共纳入257例研究对象，其中ATB患者 55例，包括病原学确诊ATB 34例；LTBI 151例；Non-TBi 51例。其中接受免疫抑制治疗者110例，包括ATB 15例，LTBI 95例。

抗原1刺激后，ATB组分泌IFN-γ⁺IL-2^-特异性T细胞频数(P=0.007)和比例(P＜0.001)均显著高于LTBI组，LTBI组分泌IFN-γ⁺IL-2⁺特异性T细胞比例(P＜0.001)和IFN-γ^-IL-2^＋特异性T细胞比例(P=0.004)显著高于ATB组；抗原2刺激后，LTBI组分泌IFN-γ^-IL-2⁺的特异性T细胞频数(P=0.023)和比例(P=0.006)显著高于ATB组。

在34例病原学确诊ATB中，T-SPOT.TB检测结核感染的阳性率为88.2%，总分泌IFN-γ/IL-2特异性T细胞频数以5 SFC/2.5×10⁵ PBMC作为Cutoff值，抗原1-FluoroSpot诊断结核感染的阳性率为94.1%；总分泌IFN-γ/IL-2特异性T细胞频数以16 SFC/2.5×10⁵ PBMC作为Cutoff值，抗原2-FluoroSpot诊断结核感染的阳性率为47.1%；联合抗原1和抗原2-FluoroSpot诊断结核感染的阳性率为97.1%。

抗原1-FluoroSpot同时检测IFN-γ⁺IL-2^- T细胞频数和比例，诊断ATB的AUC为0.883(95%CI 0.835-0.931)，显著高于T-SPOT.TB(AUC=0.695，95%CI 0.620-0.769，P＜0.001)。以拟合指数0.131为cutoff值，抗原1-FluoroSpot诊断ATB的灵敏度、特异度、阳性预测值、阴性预测值、阳性似然比、阴性似然比分别为0.883(95%CI: 0.835-0.931)，92.7%(95%CI: 81.6%-97.6%)，71.3%(95%CI: 64.4%-77.3%)，46.8%(95%CI: 37.3%-56.6%)，97.3%(95%CI: 92.8%-99.1%)，3.23(95%CI: 2.57-4.06)，0.10(95%CI: 0.04-0.26).

抗原1-FluoroSpot同时检测IFN-γ⁺IL-2^-T细胞频数和比例，鉴别诊断ATB和LTBI的AUC为0.875(95%CI 0.821-0.928)，以拟合指数0.169为cutoff值，抗原1-FluoroSpot鉴别诊断ATB与LTBI的灵敏度、特异度、阳性预测值、阴性预测值、阳性似然比、阴性似然比分别为0.875(95%CI：0.821-0.928)，92.7%(95%CI：81.6%-97.6%)，70.9%(95%CI：62.8%-77.8%)，53.7%(95%CI：43.2%-63.9%)，96.4%(95%CI：90.5%-98.8%)，3.18(95%CI：2.45-4.13)，0.10(95%CI: 0.04-0.27)。抗原2联合抗原1-FluoroSpot可进一步将鉴别诊断ATB和LTBI的AUC提升至0.904(95%CI 0.863-0.945)。以拟合指数0.242为cutoff值，可将特异度提升至81.8%(95%CI：68.7% - 90.5%)，阴性似然比提升至0.22(95%CI：0.13-0.39)。

在接受免疫抑制治疗的患者中，抗原1-FluoroSpot同时检测IFN-γ⁺IL-2^-特异性T细胞频数和比例，鉴别诊断ATB和LTBI的AUC为0.887(95%CI 0.813-0.957)。抗原2联合抗原1-FluoroSpot可进一步将鉴别诊断ATB和LTBI的AUC提升至0.900(95%CI 0.826-0.974)。

第三部分 结核特异性抗原刺激后不同时点细胞因子分泌特征的探索

共纳入6位结核感染者进行Simoa检测，设置空白对照孔、阳性对照孔、结核特异性抗原孔。刺激2h 时，所有样本的抗原刺激孔均可检出IFN-γ的分泌，刺激4h时，抗原刺激孔IFN-γ分泌量较阴性对照孔明显升高。

共纳入5名结核感染者进行Luminex检测。首次检出IFN-γ分泌时点分别在2h(3人)和6h(2人)，检出量为1.8-9.2pg/ml，全时段IFN-γ分泌量为0-730.9 pg/ml；首次检出IP-10时点均在2h，分泌量为1.1-3.9pg/ml，全时段分泌量为3.26-8098.35pg/ml；首次检出IL-2时点均在2h，分泌量为0.68-4.15pg/ml，全时段分泌量为0.68-37.44pg/ml；首次检出CCL4时点均为2h，分泌量为33.52-188.97pg/ml，全时段分泌量为0-4013.07 pg/ml。IFN-γ、IP-10、IL-2、CCL4的分泌量有随结核特异性抗原刺激时间延长而升高的趋势。IL-1Ra、CXCL1和IL-6各时点分泌量与空白对照孔差异较小；IL-17分泌量各时点均较低。

结论

本研究筛选构建的抗原1和抗原2较好地保留了全肽的免疫原性，与全肽相比提升了鉴别诊断ATB和LTBI的能力。已申请中国发明专利。

抗原1 联合抗原2-FluoroSpot检测分泌IFN-γ/IL-2的特异性T细胞可提高结核感染的检出率。

抗原1-FluoroSpot检测分泌IFN-γ/IL-2的特异性T细胞对诊断ATB的准确性显著优于T-SPOT.TB。

在抗原1-FluoroSpot检测分泌IFN-γ/IL-2的特异性T细胞的基础上，联合抗原2可进一步提升鉴别诊断ATB与LTBI的准确性。在接受免疫抑制治疗的患者中，依然保持较好的鉴别诊断准确性。

IFN-γ、IP-10、IL-2、CCL4的分泌量存在随结核特异性抗原刺激时间延长而升高的趋势。具体可缩短的检测时点，尚需扩大样本量进一步验证。

Objective

China is a country with high tuberculosis burden. Accurate diagnosis of active tuberculosis (ATB) is a prerequisite for the elimination of tuberculosis. It is difficult to obtain specimens containing Mycobacterium tuberculosis (MTB) for some pulmonary tuberculosis and extrapulmonary tuberculosis. Tuberculosis immunodiagnosis methods may assist in the diagnosis of tuberculosis. Currently, tuberculosis immunodiagnosis methods lack sufficient accuracy in differentiating ATB from latent TB infection (LTBI). This study innovatively addresses this limitation through cytokine secretion profiling between ATB and LTBI to construct dominant antigen epitope libraries by screening virulent antigens ESAT-6 and CFP-10 (antigen 1), and latency-associated antigens Rv1733c SLP and Rv2028c SLP (antigen 2), with rigorous exclusion of cross-reactive peptides, enhance diagnostic specificity for discriminating different MTB infection states, particularly ATB vs LTBI, and optimize cost-effectiveness through epitope rationalization. Furthermore, we characterize temporal secretion patterns of MTB-specific multicytokine profiles to potentially reduce diagnostic turnaround time.

Methods

Part 1: Screening, Establishment and Immunogenicity Validation of Dominant Peptide Libraries of Tuberculosis-Specific Virulence antigens and Latent-Associated antigens

Pathogenetically diagnosed ATB patients and LTBI Healthcare workers were enrolled. The peptides of ESAT-6, CFP-10, Rv1733c SLP and Rv2028c SLP were divided into 4 to 6 peptide pools. Each peptide pool was used to stimulate the PBMCs of the patients respectively. FluoroSpot was used to detect the frequencies of IFN-γ⁺, IL-2⁺ and IFN-γ⁺IL-2⁺. The frequencies and proportions of IFN-γ⁺IL-2^-, IFN-γ^-IL-2⁺ and IFN-γ⁺IL-2⁺ were calculated. According to the secretion characteristics of cytokine subsets in the ATB and LTBI groups, the interfering peptides were eliminated, the dominant peptide libraries were screened out, and combined to antigen 1 and antigen 2.

Included individuals with tuberculosis infection were evaluated using Elispot and FluoroSpot methods to assess the differences and correlations in the frequency of specific T cells after stimulation with full peptides of antigen 1 and ESAT-6/CFP-10, as well as full peptides of antigen 2 and Rv1733c SLP/Rv2028c SLP.

Patients with ATB and LTBI were included. FluoroSpot was used to evaluate the differences in the diagnostic efficacy between antigen 1, antigen 2 and the full peptides of ESAT-6, CFP-10, Rv1733c SLP and Rv2028c SLP for differentiating ATB from LTBI respectively.

Part 2: Study on Differentiating Different Tuberculosis Infection States by the Dominant Peptide Libraries of Tuberculosis-Specific antigens Combined with FluoroSpot-IFN-γ/IL-2

Patients with suspected tuberculosis were included. T-SPOT.TB test was performed. Using antigen 1 and antigen 2 screened out in the first part as specific antigens, FluoroSpot test was performed.

Among the pathogenetically diagnosed ATB cases, the diagnostic accuracies of T-SPOT.TB, antigen 1-FluoroSpot, and antigen 2-FluoroSpot for tuberculosis infection were compared.

The diagnostic accuracies of antigen 1-FluoroSpot and T-SPOT.TB for diagnosing ATB were compared.

The diagnostic accuracies of antigen 1-FluoroSpot in the differential diagnosis between ATB and LTBI were evaluated. The improvement effect of combining antigen 2 on the basis of antigen 1-FluoroSpot on the differential diagnostic accuracy was evaluated.

In patients receiving immunosuppressive therapy, the diagnostic accuracies of antigen 1-FluoroSpot in the differential diagnosis between ATB and LTBI were evaluated. The improvement effect of combining antigen 2 on the basis of antigen 1-FluoroSpot on the differential diagnostic accuracy was evaluated.

Part 3 Exploration of the temporal secretion Patterns in TB-Specific Cytokines

Tuberculosis infected individuals were included, and full peptides of ESAT-6 and CFP-10 were used as tuberculosis-specific antigens.

The Single Molecule Array (Simoa) technique in Digital ELISA technology was used to detect the secretion of IFN-γ at 2 hours, 4 hours, and 16 hours after stimulation with tuberculosis-specific antigens. The secretion of IFN-γ in the antigen-stimulated wells and the negative control wells were reported. Repeated measurement was used to test whether there were differences in the secretion amounts of tuberculosis-specific IFN-γ at each time point.

The Luminex technique was used to detect the secretion of IFN-γ, CCL4, CXCL1, IP-10, IL-1Ra, IL-2, IL-6, and IL-17 at 2 hours, 4 hours, 6 hours, 8 hours, 10 hours, 12 hours, 14 hours, and 16 hours after stimulation with tuberculosis-specific antigens. The secretion of each cytokine in the antigen-stimulated wells, negative control wells, and positive control wells at each time point were reported.

Results

Part 1: Screening, Establishment and Immunogenicity Validation of Dominant Peptide Libraries of Tuberculosis-Specific Virulence antigens and Latent-Associated antigens

Thirteen cases of pathogenetically diagnosed ATB and 10 medical workers with LTBI were included. The frequencies and proportions of tuberculosis-specific IFN-γ⁺IL-2^-T cells and IFN-γ^-IL-2⁺T cells in the ATB group and the LTBI group were comprehensively compared. The dominant peptide pools 1, 2, and 5 of ESAT-6 and the dominant peptide pools 1, 3, and 5 of CFP-10 were screened out and combined to antigen 1. The dominant peptide pools 2, 3, and 5 of Rv1733c SLP and the dominant peptide pools 1, 2, and 3 of Rv2028c SLP were screened out and combined to antigen 2.

2.A total of 72 tuberculosis-infected individuals were included for correlation analysis. The Elispot results showed that there was no significant difference in the frequency of specific T cells between antigen 1 and the full peptides of ESAT-6 and CFP-10 (P=0.076), and the correlation coefficient (R) was 0.877 (P<0.001). The FluoroSpot results showed that there was a significant difference in the frequency of IFN-γ⁺ specific T cells between antigen 1 and the full peptides (P<0.001), with R being 0.829 (P<0.001). There was no significant difference in the frequency of IL-2⁺ specific T cells (P=0.156), and R was 0.826 (P<0.001). There was no difference in the frequency of IFN-γ⁺IL-2⁺ specific T cells (P=0.083), and R was 0.821 (P<0.001). The FluoroSpot results showed that there was no difference in the frequency of IFN-γ⁺ specific T cells between antigen 2 and the full peptides of Rv1733c SLP和Rv2028c SLP (P>0.999), and R was 0.769 (P<0.001). There was no difference in the frequency of IL-2⁺ specific T cells (P=0.092), and R was 0.870 (P<0.001). There was no difference in the frequency of IFN-γ⁺IL-2⁺ specific T cells (P=0.381), and R was 0.627 (P<0.001).

3.Eleven cases of ATB and 46 cases of LTBI were included for preliminary clinical verification. The AUC of antigen 1 for the differential diagnosis between ATB and LTBI was 0.907 (95% CI: 0.825, 0.989), which was significantly higher than that of the full peptides of the virulence antigen (AUC=0.768, 95%CI: 0.633-0.904, P=0.043). The AUC of antigen 2 for the differential diagnosis between ATB and LTBI was 0.906 (95% CI: 0.779-0.973), which was significantly higher than that of the full peptides of the latency-associated antigen(AUC=0.795, 95%CI: 0.647-0.902, P=0.048). The combination of antigen 1 and antigen 2 could increase the AUC of antigen 1 for the differential diagnosis between ATB and LTBI from 0.907 (95% CI: 0.825-0.989) to 0.972 (95% CI: 0.890-0.998).

Part 2: Study on Differentiating Different Tuberculosis Infection States by the Dominant Peptide Libraries of Tuberculosis - Specific antigens Combined with FluoroSpot

A total of 257 participants were included in this study, among which there were 55 patients with ATB, including 34 patients with pathogenetically diagnosed ATB (20 cases without underlying diseases); 151 patients of LTBI; and 51 participants of Non-TBi, among which there were 37 healthy participants. There were 110 participants receiving immunosuppressive therapy, including 15 ATB patients and 95 LTBI participants.

Under the stimulation of antigen 1, the frequency (P=0.007) and proportion (P<0.001) of IFN-γ⁺IL-2^- specific T cells in the ATB group were significantly higher than those in the LTBI group. The proportion of IFN-γ⁺IL-2⁺ specific T cells (P<0.001) and the proportion of IFN-γ^-IL-2⁺ specific T cells (P=0.004) in the LTBI group were significantly higher than those in the ATB group. Under the stimulation of antigen 2, the frequency (P=0.023) and proportion (P=0.006) of IFN-γ^-IL-2⁺ specific T cells secreted in the LTBI group were significantly higher than those in the ATB group.

Among the 34 cases of pathogenetically diagnosed ATB, the positive rate of tuberculosis infection detected by T-SPOT.TB was 88.2%. Taking the total frequency of IFN-γ/IL-2 specific T cells 5 SFC/2.5×10⁵ PBMC as the cutoff value, the positive rate of diagnosing tuberculosis infection by antigen 1-FluoroSpot was 94.1%. Taking the total frequency of IFN-γ/IL-2 specific T cells 16 SFC/2.5×10⁵ PBMC as the cutoff value, the positive rate of diagnosing tuberculosis infection by antigen 2-FluoroSpot was 47.1%. The positive rate of diagnosing tuberculosis infection by combining antigen 1 and antigen 2-FluoroSpot was 97.1%.

The AUC of the antigen 1-FluoroSpot assay for the diagnosis of all ATB cases is 0.883(95%CI: 0.835-0.931), which is significantly higher than that of T-SPOT.TB(AUC=0.695, 95%CI: 0.620-0.769, P<0.001). For the pathogenetically diagnosed ATB, the AUC of the antigen 1-FluoroSpot assay is 0.880 (95% CI: 0.820- 0.940), which is significantly higher than that of T-SPOT.TB(AUC=0.639, 95% CI: 0.546-0.731, P<0.001). For the clinically diagnosed ATB, the AUC of the antigen 1-FluoroSpot assay is 0.893(95%CI: 0.845-0.931).The sensitivity, specificity, positive predictive value, negative predictive value, positive likelihood ratio, and negative likelihood ratio of antigen 1-FluoroSpot for diagnosing ATB were 0.883 (95% CI: 0.835-0.931), 92.7% (95% CI: 81.6%-97.6%), 71.3% (95% CI: 64.4%-77.3%), 46.8% (95% CI: 37.3%-56.6%), 97.3% (95% CI: 92.8%-99.1%), 3.23 (95% CI: 2.57-4.06), and 0.10 (95% CI: 0.04-0.26), respectively.

The AUC of the antigen 1-FluoroSpot assay for detecting the frequency and proportion of IFN-γ⁺IL-2^-T cells in the differential diagnosis between ATB and LTBI is 0.875 (95%CI: 0.821-0.928). The sensitivity, specificity, positive predictive value, negative predictive value, positive likelihood ratio, and negative likelihood ratio of antigen 1-FluoroSpot for differential diagnosis between ATB and LTBI were 0.875 (95% CI: 0.821–0.928), 92.7% (95% CI: 81.6%–97.6%), 70.9% (95% CI: 62.8%–77.8%), 53.7% (95% CI: 43.2%–63.9%), 96.4% (95% CI: 90.5%–98.8%), 3.18 (95% CI: 2.45–4.13), and 0.10 (95% CI: 0.04–0.27), respectively. Combining antigen 2 with antigen 1-FluoroSpot further increased the AUC for differential diagnosis between ATB and LTBI to 0.904 (95% CI: 0.863–0.945). Using a fit index of 0.242 as the cutoff value, the specificity could be increased to 81.8% (95% CI: 68.7%–90.5%), and the negative likelihood ratio could be increased to 0.22 (95% CI: 0.13–0.39).

Among 110 patients receiving immunosuppressive therapy, there were 15 cases of ATB and 95 cases of LTBI. The AUC of the frequency and proportion of IFN-γ⁺IL-2^-specific T cells stimulated by antigen 1-FluoroSpot in the differential diagnosis between ATB and LTBI is 0.887(95%CI:0.813-0.957). Combining antigen 2 with antigen 1-FluoroSpot can increase the AUC of antigen 1-FluoroSpot for the differential diagnosis between ATB and LTBI from 0.887 (95% CI: 0.813-0.957) to 0.900 (95% CI: 0.826-0.974).

Part 3 Exploration of the temporal secretion Patterns in TB-Specific Cytokines

A total of 6 individuals with tuberculosis infection were included in the Simoa assay, with blank control wells, positive control wells, and tuberculosis-specific antigen wells set up. At 2 hours of stimulation, secretion of IFN-γ was detectable in the antigen-stimulated wells of all samples.

Five MTB-infected individuals were enrolled for Luminex detection. The time points of first detection of IFN-γ secretion were at 2 hours (in 3 patients) and 6 hours (in 2 patients), and the detected amount was 1.8 - 9.2 pg/ml. The secretion amount of IFN-γ throughout the whole period was 0 - 730.9 pg/ml. The time points of first detection of IP-10 were all at 2 hours, and the secretion amount was 1.1-3.9 pg/ml. The secretion amount throughout the whole period was 3.26 - 8098.35 pg/ml. The time points of first detection of IL-2 were all at 2 hours, and the secretion amount was 0.68 - 4.15 pg/ml. The secretion amount throughout the whole period was 0.68 - 37.44 pg/ml. The time points of first detection of CCL4 were all at 2 hours, and the secretion amount was 33.52 - 188.97 pg/ml. The secretion amount throughout the whole period was 0 - 4013.07 pg/ml. The secretion of IFN-γ, IP-10, IL-2 and CCL4 show an upward trend with the prolongation of tuberculosis-specific antigen stimulation. The secretion amounts of IL-1Ra, CXCL1 and IL-6 at each time point had little difference compared with those in the negative control wells, and the secretion amount of IL-17 was low at each time point.

Conclusion

The antigen 1 and antigen 2 screened in this study have well retained the immunogenicity of the full peptides and improved the ability to differentially diagnose ATB and LTBI. The screened antigens have applied for a Chinese invention patent.

Detecting specific T cells secreting IFN-γ/IL-2 by antigen 1 combined with antigen 2-FluoroSpot can improve the detection rate of tuberculosis infection.

The accuracy of antigen 1-FluoroSpot in diagnosing tuberculosis infection and ATB is significantly better than that of T-SPOT.TB.

On the basis of detecting IFN-γ/IL-2-secreting specific T cells by antigen 1-FluoroSpot, the combination of antigen 2 can further improve the accuracy of differentiating ATB from LTBI. This method still maintains good diagnostic accuracy for differentiation in patients receiving immunosuppressive therapy.

The secretion levels of IFN-γ, IP-10, IL-2 and CCL4 show a trend of increasing with the prolongation of tuberculosis-specific antigen stimulation time.The specific detection time points that can be shortened still need to be further verified by expanding the sample size.