CVI Accepts, published online ahead of print on 3 July 2013 Clin. Vaccine Immunol. doi:10.1128/cvi.00290-13 Copyright 2013, American Society for Microbiology. All Rights Reserved. 1 2 3 Effect of prolonged incubation time on the results of the QuantiFERON TB Gold In-Tube assay for the diagnosis of latent tuberculosis infection 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 Joo-Won Min, MD 1.; Ha-Youn Lee, MD 2.; Ji Sun Lee 2 ; Jinwoo Lee 2 ; Jae Ho Chung, MD 1.; Sung Koo Han, MD 2.; Jae-JoonYim, MD 2. Division of Pulmonary and Critical Care Medicine, Department of Internal Medicine, Kwandong University College of Medicine 1, Division of Pulmonology and Critical Care Medicine, Department of Internal Medicine and Lung Institute, Seoul National University College of Medicine 2 Correspondence to: Jae-Joon Yim, MD. E mail: yimjj@snu.ac.kr, Fax: +82-2-762-9662, Phone: +82-2-2072-2059 Department of Internal Medicine, Seoul National University College of Medicine, 28 Yongon-Dong, Jongno-Gu, Seoul, 110-774, Republic of Korea. Running Title: Effect of longer Incubation on results of QFT-GIT Disclosure of Interest We declare that we do not have any affiliation with or financial interest in a commercial organization that poses a conflict of interest. 1
22 ABSTRACT 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 Background: Previous reports have shown that the sensitivity of the 6-day lymphocyte stimulation test is much higher than that of commercially available interferon-γ-release assays (IGRAs). The aim of this study was to elucidate the effect of prolonged incubation on the results of the QuantiFERON TB Gold In-Tube (QFT-GIT) assay. Methods: Patients aged > 20 years with suspected tuberculosis (TB) were recruited prospectively from May 1, 2009 to December 31, 2010. In addition, healthy volunteers with no history of TB treatment were included as controls. For each participant, three sets of the QFT-GIT assay were performed using 24-, 48-, and 72-h incubations, and the results were compared. Results: Thirty-seven patients with suspected pulmonary TB and 33 healthy controls were enrolled in this study. Of the 37 TB suspects, the QFT-GIT results were positive in 28 (75.7%) after a 24-h incubation. After prolonged incubation, the results differed in four of the 37 TB suspects (10.8%). Among 27 patients with culture-confirmed TB, sensitivity of QFT- GIT after 24-, 48-, and 72-h incubation were 85.2%, 81.5%, and 81.5%, respectively. Among the 33 healthy controls, the QFT-GIT results were positive in two (6.1%) after a 24-h incubation. The results changed in two of the 33 healthy controls (6.1%) after prolonged incubation. Specificity of QFT-GIT after 24-, 48-, and 72-h incubation were 93.9%, 87.9%, and 90.9%, respectively. 41 42 43 Conclusion: Use of the QFT-GIT assay with a prolonged incubation time did not increase sensitivity. The recommended incubation time of 16 24 h should be respected because prolonged incubation can cause indeterminate or false-positive results. 2
44 KEYWORDS: Interferon-gamma release assay, latent tuberculosis infection, incubation. 45 INTRODUCTION 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 Recently, interferon-γ-release assays (IGRAs) have been introduced into clinical practice as an alternative to the traditional tuberculin skin test (TST) (1, 2). Two IGRAs are available commercially: the QuantiFERON TB Gold In-Tube (QFT-GIT) assay (Cellestis; Victoria, Australia) and the T-SPOT.TB assay (Oxford Immunotec; Oxford, UK). A meta-analysis suggested that the QFT-GIT assay may not be as sensitive as the T-SPOT.TB assay (70% for QFT-GIT and 90% for T-SPOT.TB) (3), but the QFT-GIT is more convenient because it does not require mononuclear cell separation. Since the adoption of the QFT-GIT for routine clinical practice, various procedural modifications to improve its sensitivity have been investigated. Incubation without delay (4, 5) and incubation at 39 C rather than 37 C (6) have been reported to increase its sensitivity. Given that the sensitivity of the 6-day lymphocyte stimulation test is much higher than that of IGRAs (7), we postulated that prolonged incubation would increase the sensitivity of QFT- GIT for the diagnosis of Mycobacterium tuberculosis infection. In addition, in routine clinical practice in South Korea, sampling for the QFT-GIT assay is only performed between Monday and Thursday, but not on Friday. Because the assay measures interferon-γ (IFN-γ) release from whole blood after a 16- to 24-h incubation, post-incubation process had to be performed on weekends if samples were drawn on Friday. 63 64 In this context, we examined the effect of prolonged incubation on the results of the QFT-GIT assay. 3
65 METHODS 66 Participants 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 Patients with ages > 20 years and with suspected tuberculosis (TB) were recruited prospectively from May 1, 2009 to December 31, 2010 at Seoul National University Hospital and Kwandong University Myongji Hospital. In addition, 33 healthy volunteers with neither past history of TB treatment nor contact with active TB patients were recruited for comparison. Written informed consent was obtained from all participants, and the study protocol was approved by the Ethics Review Committee and Institutional Review Board at each center. Determination of sample size We expected that the sensitivity of the QFT-GIT assay could be improved from 70% to 80% by prolonging the incubation period from 24h to 48 or 72 h. To show this change, a required minimum sample size of 27 was estimated based on a one-tailed test with a 95% confidence interval (CI), a type II error of 0.20, and a power of 0.80. QuantiFERON TB Gold In-Tube assay From each participants, 9 ml of whole blood were drawn, and put directly into three sets of QFT-GIT blood collection tubes: 1 ml each for the Nil control tube, TB antigen tube (ESAT- 6, CFP-10, or TB7.7), and mitogen control tube. All tubes were incubated within 6 h of sampling. The first, second, and third sets of tubes were processed and analyzed after 24-, 48-, and 72-h incubations, respectively. The IFN-γ levels were measured using a QFT-GIT ELISA kit. Samples from the same patient were measured in the same ELISA plate. Samples with a TB antigen value minus Nil value 0.35 IU/mL and 25% of the Nil value, as determined by 4
87 the manufacturer s software, were considered positive. 88 Statistical analysis 89 90 91 92 93 The change in the IFN-γ levels with prolonged incubation was analyzed using a repeated measured nonparametric test. A P-value <0.05 was considered to indicate statistical significance. All statistical analyses were performed using SPSS 17.0 software (SPSS, Chicago, IL, USA). Downloaded from http://cvi.asm.org/ on July 26, 2018 by guest 5
94 RESULTS 95 Participants 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 Thirty-seven patients with suspected TB were enrolled in this study [median age, 66 years; age range, 27 90 years; 21 (56.8%) males]. Bacillus Calmette Guérin (BCG) scars were found in 12 patients (32.4%), and seven (18.9%) had a history of TB treatment. In 27 (73.0%) of the 37 patients, pulmonary TB was confirmed by culture of Mycobacterium tuberculosis. Thirty-three healthy volunteers were recruited for comparison [median age, 28 years; age range, 23 42 years; 19 (57.6%) males]. BCG scars were found in 25 (75.6%) participants. No volunteers had past history of TB treatment. Baseline results of the QFT-GIT assay After 24-h incubation, QFT-GIT assay results were positive in 28 (75.7%) of the 37 patients. Three patients showed indeterminate results. Among the 27 culture-confirmed TB patients, 23 (85.2%) yielded positive assay results. The median IFN-γ level (value from stimulation of TB antigen minus value from Nil stimulation) was 2.19 IU/mL (range, 0.01 10.00 IU/mL; Fig. 1). Two of the 33 healthy volunteers had positive QFT-GIT assay results, and 31 had negative results. The median IFN-γ level was 0.00 IU/mL (range, 0.13 to10.00 IU/mL; Fig. 1). 111 Change in QFT-GIT assay results after prolonged incubation 112 113 114 Patients with suspected TB The changes in QFT-GIT assay results according to incubation duration are shown in the Table 1. The results in four patients (10.8%) differed between 24-h and longer incubation 6
115 116 117 118 119 times. In two patients with negative results after 24-h incubation (IFN-γ levels of 0.01 and 0.11 IU/mL, respectively), the results were indeterminate after 48- and 72-h incubations because of the differences in the IFN-γ levels between mitogen and Nil stimulation (0.05 and 0.28, respectively, at 48 h; 0.23 and 0.46, respectively, at 72 h). 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 In one patient with an initial positive result (IFN-γ of 0.49 IU/mL), the result was still positive after a 48-h incubation (IFN-γ of 0.54 IU/mL), but was negative after a 72-h incubation (IFN-γ of 0.30 IU/mL). In another patient with an initial positive result (IFN-γ level of 0.53 IU/mL), the result was indeterminate after a 48-h incubation because the IFN-γ level between mitogen and Nil stimulation was 0.27 IU/mL. After 72-h incubation, the result was positive again (IFN-γ of 0.51 IU/mL). The IFN-γ level did not change between the 24-h (median, 2.19 IU/mL) and 48-h incubations (median, 2.60 IU/mL), although the IFN-γ level decreased between 48-h and 72-h incubations (median, 1.88 IU/mL; P = 0.003; Fig. 1). The difference in the IFN-γ level between mitogen and Nil stimulation did not change from prolonged incubation (P = 0.140). Healthy volunteers The results of two of the 33 healthy volunteers differed with prolonged incubation (Table 1). The initial results of both subjects were negative, with IFN-γ (TB antigen value minus Nil value) levels of 0.30 and 0.28 IU/mL, respectively. In the first volunteer, the result was positive after both the 48-h (IFN-γ of 0.67 IU/mL) and 72-h incubations (IFN-γ of 0.46 IU/mL). In the other volunteer, the result was positive after a 48-h incubation (IFN-γ of 1.47 IU/mL), and negative after a 72-h incubation (IFN-γ of 0.45 IU/mL). 137 The IFN-γ level increased between the 24-h (median, 0.00 IU/mL) and 48-h incubations 7
138 139 140 (median, 0.00 IU/mL; P = 0.034), but the IFN-γ level did not change between the 48-h and 72-h incubations (median, 0.00 IU/mL; P = 0.715; Fig. 1). The difference in the IFN-γ level between mitogen and Nil stimulation did not change after prolonged incubation (P = 0.317). 141 Sensitivity and Specificity of QFT-GIT assay after prolonged incubation 142 143 144 145 146 147 148 Among the 27 culture-confirmed TB patients, 23, 22, and 22 yielded positive assay results after 24-, 48-, and 72-h incubation, respectively. As a result, sensitivity of QFT-GIT after 24-, 48-, and 72-h incubation were 85.2%, 81.5%, and 81.5%, respectively. Among 33 healthy volunteers 31, 29, and 30 yielded negative assay results after 24-, 48-, and 72-h incubation, respectively. As a result, specificity of QFT-GIT after 24-, 48-, and 72-h incubation were 93.9%, 87.9%, and 90.9%, respectively. Downloaded from http://cvi.asm.org/ on July 26, 2018 by guest 8
149 DISCUSSION 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 Since the clinical adoption of the QFT-GIT assay, the impact of various procedural modifications on the assay results was reported. First, the importance of immediately incubating the tubes at 37 C has been underscored. According to two reports, immediate incubation reduces the incidence of indeterminate results (4, 5). In a study conducted in Zambia, a 24-h delay in the start of incubation produced indeterminate results in five of 109 TB patients (8). Modifications of the incubation process have been evaluated to improve the sensitivity of assays based on IFN-γ measurements. First, because fever augments the proinflammatory immune response in vivo, incubation was conducted at 39 C rather than 37 C (6). In that study, incubation at 39 C increased the IFN-γ level and IFN-γ-inducible protein 10 responsiveness to TB antigens in participants with a low initial response. Another study has reported much higher sensitivity for the 6-day lymphocyte stimulation test compared with the two commercially available IGRAs (the QFT-GIT and T-SPOT.TB assays) (7). Meanwhile, it would be very helpful if prolonged incubation would, at least, not affect IGRA assay results. Prolonged incubation would allow sampling on Fridays, and processing on following Monday. In addition, samples could be incubated for several days during transport to a central lab. 167 168 169 170 171 In the present study, we used QFT-GIT assay using incubation times (48 and 72 h) that are longer than the recommended time and compared the results with those of the standard incubation time (16 24 h). With prolonged incubation, the negative or indeterminate baseline results in four of the TB patients did not change to positive results. Instead, changes to indeterminate results occurred in several patients because of the decreased difference in IFN- 9
172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 γ levels between mitogen and Nil stimulation. As a result, sensitivity of QFT-GIT decreased slightly after prolonged incubation. Conversely, prolonged incubation changed the results from negative to positive in two healthy volunteers. Because these two volunteers had no evidence of past history of TB treatment or contact with TB, these results could be interpreted as false-positive results caused by prolonged incubation. Consequently, specificity of QFT-GIT also decreased slightly after prolonged incubation. To fully appreciate our results, the limitations of this study should be acknowledged. First of all, sensitivity from prolonged incubation in this study was drawn from patients with active TB. In a previous study recruiting participants with mostly remote latent TB infection, the explanation for the increased sensitivity was that prolonged incubation allowed memory T cells to respond, while only effector cells would be able to mount an IFN-γ response in the short term assay (7). Because the effector cells were still prevalent among patients with active TB, the lack of effect of prolonged incubation periods in our study could not be generalized. Secondary, although the healthy volunteers who participated in our study did not have history of past TB treatment or contact with active TB patients, they could have been exposed to tuberculous bacilli, because the annual incidence of TB in South Korea is 100 out of 100,000 (9). Consequently, specificities presented in this study could be underestimated. In conclusion, the use of prolonged incubation times did not increase the sensitivity of the QFT-GIT assay. The recommended incubation time of 16 24 h should be respected because prolonged incubation can cause indeterminate or false-positive results. 193 10
194 REFERENCES 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 1. Pai M, O'Brien R. 2008. New diagnostics for latent and active tuberculosis: state of the art and future prospects. Semin Respir Crit Care Med. 29:560-568. 2. Kang YA, Lee HW, Yoon HI, Cho B, Han SK, Shim YS, Yim JJ. 2005. Discrepancy between the tuberculin skin test and the whole-blood interferon gamma assay for the diagnosis of latent tuberculosis infection in an intermediate tuberculosisburden country. Jama. 293:2756-2761. 3. Pai M, Zwerling A, Menzies D. 2008. Systematic review: T-cell-based assays for the diagnosis of latent tuberculosis infection: an update. Ann Intern Med. 149:177-184. 4. Doberne D, Gaur RL, Banaei N. 2011. Preanalytical delay reduces sensitivity of QuantiFERON-TB gold in-tube assay for detection of latent tuberculosis infection. J Clin Microbiol. 49:3061-3064. 5. Herrera V, Yeh E, Murphy K, Parsonnet J, Banaei N. 2010. Immediate incubation reduces indeterminate results for QuantiFERON-TB Gold in-tube assay. J Clin Microbiol. 48:2672-2676. 6. Aabye MG, Ravn P, Johansen IS, Eugen-Olsen J, Ruhwald M. 2011. Incubation of whole blood at 39 degrees C augments gamma interferon (IFN-gamma)-induced protein 10 and IFN-gamma responses to Mycobacterium tuberculosis antigens. Clin Vaccine Immunol. 18:1150-1156. 7. Leyten EM, Arend SM, Prins C, Cobelens FG, Ottenhoff TH, van Dissel JT. 2007. Discrepancy between Mycobacterium tuberculosis-specific gamma interferon release assays using short and prolonged in vitro incubation. Clin Vaccine Immunol. 14:880-885. 8. Shanaube K, De Haas P, Schaap A, Moyo M, Kosloff B, Devendra A, Raby E, Godfrey-Faussett P, Ayles H. 2010. Intra-assay reliability and robustness of 11
219 220 221 222 QuantiFERON(R)-TB Gold In-Tube test in Zambia. Int J Tuberc Lung Dis. 14:828-833. 9. World Health Organization. Global tuberculosis control: WHO report 2011. Geneva: World Health Organization. 223 224 225 Downloaded from http://cvi.asm.org/ on July 26, 2018 by guest 12
226 227 228 229 Table 1. Results of the QuantiFERON TB Gold In-Tube assay according to incubation period. Patient QFT-GIT QFT-GIT QFT-GIT Final diagnosis Number (24-h incubation) (48-h incubation) (72-h incubation) 37 patients with suspected TB 33 patients with consistent results 1-21 TB Positive Positive Positive 22, 23 TB Indeterminate Indeterminate Indeterminate 24-28 TB, not confirmed Positive Positive Positive 29-32 TB, not confirmed Negative Negative Negative 33 TB, not confirmed Indeterminate Indeterminate Indeterminate 4 patients with inconsistent results 34 TB Positive Positive Negative 35 TB Positive Indeterminate Positive 36 TB Negative Indeterminate Indeterminate 37 TB Negative Indeterminate Indeterminate 33 healthy volunteers 31 people with consistent results 1-29 - Negative Negative Negative 30-31 - Positive Positive Positive 2 people with inconsistent results 32 - Negative Positive Negative 33 - Negative Positive Positive Abbreviations: QFT-GIT, QuantiFERON TB Gold In-Tube; TB, tuberculosis. 13
230 231 232 233 234 235 236 237 FIGURE 1. QFT-GIT assay results of TB suspects and healthy volunteers tested using prolonged incubation periods. Results that differed between the 48- and 72-h tests are indicated by gray dashed lines. Repeated measured nonparametric test was performed, and Wilcoxon Signed Ranked test was used to compare differences in means. A, TB antigen values minus Nil values decreased in prolonged incubation (p=0.016) in TB suspects. B, TB antigen values minus Nil values increased in prolonged incubation (p=0.045) in healthy voluteers. Downloaded from http://cvi.asm.org/ on July 26, 2018 by guest 14
Figure 1 Downloaded from http://cvi.asm.org/ on July 26, 2018 by guest