Antiviral immune response against HTLV-1 invalidates T-SPOT.TB® results in patients with HTLV-1-positive rheumatic diseases

Abstract

Background: T-SPOT.TB^®, one of the screening tests for latent tuberculosis infection (LTBI), yields invalid results in human T-cell leukemia virus type 1 (HTLV-1)-positive patients with rheumatoid arthritis. However, the detailed mechanisms behind this invalidation are unclear. Additionally, it remains unclear whether T-SPOT.TB^® or QuantiFERON-TB (QFT) is more useful in HTLV-1-positive patients with rheumatic disease (RD).

Method: Among all of the HTLV-1-positive RD patients who visited our department between August 2012 and December 2022, 44 patients who were screened using T-SPOT.TB® were included in the analysis. QFT testing was performed in 33 of the 44 patients, and the results were compared with that of T-SPOT.TB®. Furthermore, we performed a culture experiment mimicking T-SPOT.TB® using peripheral blood mononuclear cells (PBMCs) obtained from HTLV-1-positive patients with RD. Additionally, T-cell subsets with autonomous product IFN-γ were analyzed using a flow cytometer.

Results: Of the included patients, 13 (29.5%) were invalid for T-SPOT.TB® because of the increased number of negative control spots. The median HTLV-1 proviral load in the invalid group was higher than that in the valid group (2.45 vs. 0.49 copies/100 PBMCs, respectively, p = 0.002). QFT was performed in all 33 patients, including 13 patients who were invalid in T-SPOT.TB®. The main source of IFN-γ production was CD8+ T-cells in the T-SPOT.TB® mimic experiment. Furthermore, Tax-expressing CD4+ T-cells and Tax-specific cytotoxic CD8+ T-cells were more frequently observed in patients with invalid results than in patients with valid results. CD4+ T-cell depletion in the T-SPOT.TB® mimic experiment reduced the population of IFN-γ producing CD8+ T cells.

Conclusion: T-SPOT.TB® may be invalidated by the interaction between Tax-expressing CD4+ T-cells and cytotoxic CD8+ T-cells. Moreover, HTLV-1-associated immune reactions due to contact between these cells may be unlikely to occur in QFT using whole blood. Therefore, our results reveal the superiority of QFT over T-SPOT.TB® as a screening test for LTBI in HTLV-1-positive patients with RD.

Keywords: HTLV-1; IFN-γ-releasing assay; cytotoxic T lymphocytes; latent tuberculosis infection; rheumatic disease.

Figure 1

Number of participants with rheumatic disease (RD) who underwent T-SPOT.TB^® in HTLV-1 RD Miyazaki Registry. This registry enrolled 47 patients. Of the 47 patients, 44 underwent T-SPOT.TB^® assay during their clinical course. These 44 patients were categorized into two groups according to the T-SPOT.TB^® assay results as valid and invalid groups.

Figure 2

Comparison of the population of HTLV-1-infected cells in participants with rheumatic disease (RD) who demonstrated valid or invalid T-SPOT.TB^® assay results using HAS-Flow. The population of HTLV-1-infected cells was determined using HAS-Flow. (A) Representative data of CADM1 versus CD7 plot in CD4+ T-cells in participants with RD who exhibited valid or invalid T-SPOT.TB^® assay results. Gating of the CD4+ T-cell population in peripheral blood. CADM1 versus CD7 plot of CD4+ T-cell population. The upper-right and lower-right regions were named D and N, respectively, following the previous study. (B) The population of HTLV-1-infected cells in D, N, and D+N regions was compared between the valid (n = 29) and invalid (n = 13) groups. The bold horizontal line indicates the median values. *P < 0.05 by Mann–Whitney U test.

Figure 3

Analysis of IFN-γ-producing cells in peripheral blood mononuclear cells (PBMCs) in participants with rheumatic disease (RD) who demonstrated valid and invalid T-SPOT.TB^® results. PBMCs were obtained from participants with RD cultured without stimulation. PBMCs were collected after 24 h, and the population of IFN-γ-producing cells was analyzed using flow cytometry. (A) Representative data of CD8 versus IFN-γ plot of CD3+ T-cell population in participants with RD who exhibited valid or invalid T-SPOT.TB^® assay results. (B) Representative data of CD4 versus IFN-γ plot of CD3+ T-cell population in participants with RD who demonstrated valid or invalid T-SPOT.TB^® assay results. (C) Population of IFN-γ-producing CD8+ T cells and (D) IFN-γ-producing CD4+ T cells in participants with RD who showed valid (n = 5) or invalid (n = 5) T-SPOT.TB^® assay results. The bold horizontal line indicates the median values. **P < 0.01 by Mann–Whitney U test.

Figure 4

Population of Tax-specific cytotoxic lymphocytes (CTLs) and Tax-expressed HTLV-1-infected cells in participants with rheumatic diseases (RD) who demonstrated valid or invalid T-SPOT.TB^® assay results. Peripheral blood mononuclear cells (PBMCs) were obtained from participants with RD, and the population of Tax-specific CTLs was determined using flow cytometry. (A) Representative data of CD8 versus Tax-tetramer plot in participants with RD who exhibited valid or invalid T-SPOT.TB^® assay results. (B) Population of Tax-specific CTLs in participants with RD who showed valid (n = 5) or invalid (n = 5) T-SPOT.TB^® assay results. PBMCs were obtained from participants with RD cultured without stimulation. PBMCs were collected after 24 h, and the population of Tax-expressed CD4+ T cells or IFN-γ-producing CD8+ T cells was determined using flow cytometry. (C) Representative data of CD4 versus Tax plot in whole PBMCs and CD8+ T-cell-depleted PBMCs including Tax-specific CTLs of RD participants who demonstrated valid or invalid T-SPOT.TB^® assay results. (D) Population of Tax-expressed CD4+ T cells in whole PBMCs in participants with RD who showed valid (n = 5) or invalid (n = 5) T-SPOT.TB^® assay results. (E) Population of Tax-expressed CD4+ T cells in PBMC-depleted CD8+ T cells, including Tax-specific CTLs, in participants with RD who showed valid (n = 5) or invalid (n = 5) T-SPOT.TB^® assay results. (F) Representative data of CD8 versus IFN-γ plot of CD3+ T-cell population plot in whole PBMCs and CD4+ T-cell-depleted PBMCs in participants with RD who demonstrated invalid T-SPOT.TB^® assay results. (G) IFN-γ producing CD8+ T cells in whole PBMCs and CD4+ T-cell-depleted PBMCs in participants with RD who exhibited invalid (n = 5) T-SPOT.TB^® assay results. The bold horizontal line indicates the median values. **P < 0.01 by Mann–Whitney U test.

Figure 5

IFN-γ-ELISPOT assay mimicking control panels of T-SPOT.TB^® after depletion of CD4, CD8, and Tax-specific cytotoxic lymphocytes (CTLs). An ELISPOT assay mimicking the control panels of T-SPOT.TB^® was performed using whole peripheral blood mononuclear cells (PBMCs) or CD4, CD8, and Tax-specific CTL-depleted PBMCs. (A) Representative data of the IFN-γ-ELISPOT assay. After 24 h of culture, 354 spots were observed in culture conditions without stimulation. No spots were observed in the brefeldin A treatment condition. Numerous spots were observed in phorbol myristate acetate (PMA)/ionomycin (Iono) treatment conditions as a positive control. Decreased spots were confirmed in each culture condition of CD4, CD8, and Tax-specific CTL-depleted PBMCs, respectively. (B) Number of IFN-γ spots in each culture condition. The IFN-γ ELISPOT assay was repeated thrice. The bar presents mean ± SD.

Figure 6

Mechanism of the invalid result of T-SPOT.TB^® in patients with positive HTLV-1. (A) Tax-specific cytotoxic CD8+ T cells recognize the Tax peptide/MHC-I complex antigen presented on CD4+ T cells and produce IFN-γ in patients with a high HTLV-1 proviral load. Hence, the spot counts of negative control in T-SPOT.TB^® may be high. (B) Less contact is observed between CD4+-infected cells and CTLs in QFT using whole blood samples and culturing them in tubes, and IFN-γ production may be reduced.