JCM Accepts, published online ahead of print on 15 May ... · 4 Davide Abate 1*, Alda Saldan 1*,...

32
- 1 - Comparison of CMV ELISPOT and CMV Quantiferon TM interferon-γ 1 releasing assays in assessing risk of CMV infection in kidney transplant 2 recipients 3 Davide Abate 1* , Alda Saldan 1* , Carlo Mengoli 1* , Marta Fiscon 1 , Cristina Silvestre 2 , 4 Loredana Fallico 1 , Marta Peracchi 1 , Lucrezia Furian 2 , Riccardo Cusinato 1 , Luciana 5 Bonfante 3 , Barbara Rossi 4 , Francesco Marchini 4 , Dino Sgarabotto 5 , Paolo Rigotti 2 , 6 Giorgio Palù 1 7 8 1 Department of Molecular Medicine - 2 Department of Surgery, Oncology and Gastroenterology, 9 3 Department of Medicine Nephrology I Unit, 4 Hemodialisis and Nephrology II 5 Transplant infectious 10 disease division, Padua General Hospital - University of Padua School of Medicine via Gabelli, 63 - 35121 11 Padova Italy. 12 * Authors share equal contribution. 13 14 Keywords: Human Cytomegalovirus, transplantation, T-cell immunity, ELISPOT, 15 Quantiferon 16 17 Running title: CMV ELISPOT and Quantiferon in kidney transplants 18 19 Astract: 250 words 20 Text: 2270 words 21 Figures: 3 22 Tables: 4 23 24 Corresponding author: 25 Dr. Davide Abate 26 Department of Molecular Medicine 27 University of Padova 28 Via Gabelli, 63 35121 29 Padova Italy. 30 Phone: +39 0498218940 31 Fax: +39 0498216379 32 E-mail: [email protected] 33 34 Copyright © 2013, American Society for Microbiology. All Rights Reserved. J. Clin. Microbiol. doi:10.1128/JCM.00563-13 JCM Accepts, published online ahead of print on 15 May 2013 on February 18, 2019 by guest http://jcm.asm.org/ Downloaded from on February 18, 2019 by guest http://jcm.asm.org/ Downloaded from on February 18, 2019 by guest http://jcm.asm.org/ Downloaded from

Transcript of JCM Accepts, published online ahead of print on 15 May ... · 4 Davide Abate 1*, Alda Saldan 1*,...

- 1 -

Comparison of CMV ELISPOT and CMV QuantiferonTM interferon-γ 1

releasing assays in assessing risk of CMV infection in kidney transplant 2

recipients 3

Davide Abate1*, Alda Saldan1*, Carlo Mengoli1*, Marta Fiscon1, Cristina Silvestre2, 4

Loredana Fallico1, Marta Peracchi1, Lucrezia Furian2, Riccardo Cusinato1, Luciana 5

Bonfante3, Barbara Rossi4, Francesco Marchini4, Dino Sgarabotto5, Paolo Rigotti2, 6

Giorgio Palù1 7

8 1Department of Molecular Medicine - 2Department of Surgery, Oncology and Gastroenterology, 9

3Department of Medicine Nephrology I Unit, 4Hemodialisis and Nephrology II 5Transplant infectious 10

disease division, Padua General Hospital - University of Padua School of Medicine via Gabelli, 63 - 35121 11

Padova Italy. 12

* Authors share equal contribution. 13

14 Keywords: Human Cytomegalovirus, transplantation, T-cell immunity, ELISPOT, 15

Quantiferon 16

17 Running title: CMV ELISPOT and Quantiferon in kidney transplants 18 19 Astract: 250 words 20 Text: 2270 words 21 Figures: 3 22 Tables: 4 23 24 Corresponding author: 25

Dr. Davide Abate 26 Department of Molecular Medicine 27 University of Padova 28 Via Gabelli, 63 35121 29 Padova Italy. 30 Phone: +39 0498218940 31 Fax: +39 0498216379 32 E-mail: [email protected] 33 34

Copyright © 2013, American Society for Microbiology. All Rights Reserved.J. Clin. Microbiol. doi:10.1128/JCM.00563-13 JCM Accepts, published online ahead of print on 15 May 2013

on February 18, 2019 by guest

http://jcm.asm

.org/D

ownloaded from

on F

ebruary 18, 2019 by guesthttp://jcm

.asm.org/

Dow

nloaded from

on February 18, 2019 by guest

http://jcm.asm

.org/D

ownloaded from

- 2 -

Abstract 35

Background. Assessing the CMV specific cell-mediated immunity (CMI) represents an 36

appealing strategy to detect transplant recipients at risk of infection. In this study we 37

compared two interferon gamma-releasing assays (IGRAs), CMV Quantiferon and CMV 38

ELISPOT, in predicting protective CMV specific T-cell responses. 39

Methods. 221 Quantiferon and ELISPOT tests were conducted on 120 adult KTR 40

(including 100 R+ and 20 D+/R-). As control cohort 39 adult healthy subjects (including 41

33 CMV seropositive and 6 CMV seronegative) were enrolled. CMV IgG serology was 42

used as reference for both tests. 43

Results. In CMV seropositive individuals, ELISPOT and Quantiferon provided 46% 44

concordance with serology, 12% discordance, 18% disagreement between ELISPOT or 45

Quantiferon with serology and 24% grey areas when one or both tests resulted weak 46

positive. None of CMV seronegative subjects showed detectable responses in ELISPOT 47

and Quantiferon. In transplant recipients, both ELISPOT and Quantiferon positively 48

correlated with each other and negatively correlated to CMV DNAemia in a significant 49

way (p-value <0.05). During the antiviral prophylaxis, all 20 D+/R- KTRs examined 50

displayed undetectable Quantiferon and ELISPOT and there was no evidence of CMV 51

seroconversion. The ROC statistical analysis revealed similar specificity and sensitivity 52

in predicting detectable viremia (AUC 0.66 and 0.62 for Quantiferon and ELISPOT 53

respectively). ELISPOT and Quantiferon values above 150 spots/200000 PBMCs and >1-54

6 IU IFN-g were associated with protection from CMV infection (OR 5 and 8.75 55

respectively). 56

on February 18, 2019 by guest

http://jcm.asm

.org/D

ownloaded from

- 3 -

Conclusion. In transplants both tests displayed similar ability in predicting CMV 57

infection. Both ELISPOT and Quantiferon require several ameliorations to avoid false 58

nagative results. 59

Abbreviations: KTR, kidney transplant recipients; R+, CMV seropositive transplant recipient; D+/R-, CMV 60

serogenative transplant recipient of a CMV positive donor; ROC, receiving operator characteristics; AUC, area under 61

the curve. PBMCs, peripheral blood mononuclear cells; OR, odd ratio. 62

63

Introduction 64

Human cytomegalovirus (CMV) represents one of the primary opportunistic pathogen 65

and a leading cause of morbidity and mortality in transplant subjects. The pre-emptive or 66

prophylaxis antiviral therapy has effectively reduced the impact and incidence of 67

symptomatic CMV infection, however the drug-related toxicity and the potential 68

emergence of drug resistant CMV strains discourage a prolonged antiviral regimen. 69

The cell mediated immunity (CMI), and in specific the CMV specific CD4+ and CD8+ T-70

cell response, is able to control viral replication and thus the onset of symptomatic 71

infections (9, 11, 14, 16, 24). In CMV D+/R- transplants, de novo CMI develops upon 72

primary CMV infection often originated from virus reactivation within the allograft, 73

while in CMV R+ transplants CMI recovers from previously existing immunity. The 74

assessment of CMV specific CMI has also been used to determine the individual risk of 75

infection and as a helpful indicator in making therapeutic decisions such as whether 76

initiate or terminate an antiviral treatment (1-4, 10, 13, 31). For this reason a diagnostic 77

test assessing the status of immune reconstitution has been recommended in the current 78

CMV management guidelines in transplant patients (17). 79

on February 18, 2019 by guest

http://jcm.asm

.org/D

ownloaded from

- 4 -

At the present, several methods are available to monitor the CMI in transplant recipients: 80

several methods rely on the functional analysis of T-cell secreted cytokines or T-cell 81

phenotype (ie IFN-γ, TNF-α, IL-2, CD107a, PD-1) upon antigen stimulation, while other 82

methods, such as tetramer assays are based on direct detection of antigen specific T-cells 83

or cell proliferation assays (5, 27, 28, 32). Moreover tetramer assays are limited to few 84

human leukocyte antigen (HLA) haplotypes and thus may not be applicable to all 85

patients. Most of these methods rely on advanced technologies and costly reagents or 86

require a long turnaround time-to-result that make the test of impractical use for 87

clinical/diagnostic purposes. These issues and the lack of standardization have limited 88

their use to highly specialized laboratories. In the recent years an increasing number of 89

reports focused on IGRAs as diagnostic standard for detecting CMI towards infectious 90

agents in humans (1-3, 7, 13, 18-20). IGRA tests provide a practical, standardized, rapid 91

and cost-effective tool to assess the pathogen specific CMI (reviewed in (12, 21, 25). 92

The most commonly used IGRAs, T-SPOT TB (ELISPOT) and TB-Gold (Quantiferon), 93

were developed for detecting Mycobacterium tuberculosis (TB) responses. Both T-SPOT 94

TB and TB-Gold are FDA approved tests for use in humans and display similar 95

characteristics in assessing tuberculosis infection in immunocompetent subjects (6, 30). 96

Some reports also support that T-SPOT TB may be useful in immunosuppressed 97

population since it has higher sensitivity compared to TB-Gold (23, 26). 98

In this study we compared CMV IFN-γ ELISPOT and Quantiferon tests to assess their 99

grade of agreement, correlation and ability to predict CMV infection. Both tests have 100

been used in experimental settings to detect CMV specific T-cell responses in transplants 101

(3, 8, 15, 18, 29, 34) 102

on February 18, 2019 by guest

http://jcm.asm

.org/D

ownloaded from

- 5 -

The main differences between CMV ELISPOT and Quantiferon include: 1) the stimulus 103

peptide composition designed to stimulate selectively CD8+ T cells (Quantiferon) or both 104

CD4+ and CD8+ T-cells (ELISPOT); 2) Quantiferon test evaluates the IFN-γ production 105

in a volume of 1ml of whole blood, while ELISPOT considers the IFN-γ production in a 106

given number of PBMCs isolated from blood; 3) Quantiferon quantitatively measures 107

IFN-γ as international units (IU), while ELISPOT quantifies the spot forming colonies 108

(SFC) produced by a given number of PBMCs; 109

110

Methods and Patients 111

Patients and definitions 112

221 ELISPOT and Quantiferon tests were performed on 125 KTRs enrolled in the study 113

from September 2009 to September 2012. As control group we enrolled also 39 adult 114

healthy subjects. The control group included 33 adult CMV IgG seropositive and 6 CMV 115

IgG seronegative healthy subjects with median age 52 (range 24-72 years old), 116

comprising 19 Caucasian male and 20 Caucasian female. 125 KTRs were voluntarily 117

recruited among the transplants subjects at 30, 60, 90, 180 and 360 days after transplant. 118

The main clinical characteristics of patients are shown in Table 1. Study exclusion 119

criteria included any condition of pre-existing or acquired immunodeficiency. The Padua 120

General hospital institutional review board and ethical committee approved all the 121

medical and diagnostic procedures. CMV seropositive recipients (R+) were treated 122

according to pre-emptive antiviral strategy once CMV DNAemia was detected >10.000 123

copies/ml of whole blood. CMV seronegative recipients of a CMV seropositive allograft 124

(D+/R-) were treated according to antiviral prophylaxis regimen for 180 days after 125

on February 18, 2019 by guest

http://jcm.asm

.org/D

ownloaded from

- 6 -

transplant. Standard antiviral therapy comprised the administration of (val)-ganciclovir 126

(Valcyte, Roche) at standard dose (900mg per day orally) corrected for renal 127

functionality. No cases of CMV drug resistant strains occurred during the study. CMV 128

disease is defined as fever, malaise, and /or gastrointestinal symptoms with concurrent 129

CMV DNAemia and absence of other ongoing infections. 130

131

Detection of CMV viremia (DNAemia) 132

In all cases shown CMV viremia was evaluated using real time polymerase chain reaction 133

(PCR) with an Abi Prism 7900 HT (Applied Biosystems). PCR primers, probes and PCR 134

conditions were described (22). The lowest detection limit is defined as <1000 copies/ml 135

of whole blood. CMV infection is defined as two sequential episodes of CMV DNAemia 136

>1000 copies. Routine surveillance for viral reactivation or infection included weekly 137

determination of CMV DNAemia during the first 100 days after transplant and continued 138

thereafter if clinically indicated. 139

140

CMV serology 141

CMV serostatus was assessed using IgG ELISA assay (Enzygnost; Dade Behring). IgM 142

ELISA (Enzygnost; Dade Behring) was used to detect primary CMV infection. 143

144

Quantiferon and ELISPOT tests 145

Both Quantiferon and ELISPOT were performed on freshly isolated blood: at the same 146

time peripheral blood was collected in 3 (3x1ml) Quantiferon (Cellestis) tubes (Positive 147

control, negative control-nil- and CMV stimulus) and 10 mls of peripheral blood were 148

on February 18, 2019 by guest

http://jcm.asm

.org/D

ownloaded from

- 7 -

collected in sodium citrate tubes for ELISPOT testing. Quantiferon blood tubes were 149

incubated overnight at 37°C and further processed according to manufacturer’s 150

instructions. Quantiferon data were acquired using Personal lab workstation (Adaltis). 151

For ELISPOT testing, PBMCs were extracted using Ficoll-plaque Plus gradient (GE 152

healthcare) and 200000 PBMCs/well were seeded in a 96 well ELISPOT plate (AID 153

diagnostic) and stimulated as described (3). Quantiferon and ELISPOT tests were 154

performed independently in double blind fashion. As indicated by manufacturer 155

datasheet, ELISPOT was considered positive >20 spots, while Quantiferon >0.2 IU. The 156

reported values refer to presence of CMI, not to protection from infection. 157

158

Statistical analysis 159

Stata software (StataCorp) was used to analyze the data. The correlation of ELISPOT to 160

Quantiferon and CMV DNAemia was obtained by negative binomial regression where 161

ELISPOT was expressed as number of spots, Quantiferon as IFN-γ cytokine 162

concentration, and CMV DNAemia was a binary (0, 1) variable. 163

ELISPOT and Quantiferon were evaluated by receiving operator curve (ROC) analysis, 164

using as endpoint (reference variable) the protection against the emergence of an episode 165

of detectable CMV viremia. Sensitivity and specificity were obtained for every possible 166

cutoff of the ELISPOT or Quantiferon. The overall odds ratio (OR) was calculated as: 167

sensitivity*specificity/[(1-sensitivity)*(1-specificity)]. 168

We considered ELISPOT and Quantiferon levels protective if no detectable events of 169

CMV DNAemia occurred within 60 days after ELISPOT and Quantiferon determination. 170

on February 18, 2019 by guest

http://jcm.asm

.org/D

ownloaded from

- 8 -

Given the low number of D+/R- patients analyzed, the statistical analysis to assess the 171

Quantiferon and ELISPOT ability to predict viremia was not possible. 172

173

174

Results 175

Quantiferon and ELISPOT in respect to CMV IgG serology in healthy adult individuals 176

We compared the results of Quantiferon and ELISPOT in a cohort of 39 controls 177

including 33 CMV IgG positive and 6 IgG negative subjects. None of these subjects 178

resulted positive for CMV IgM. In 6/6 CMV IgG seronegative controls both Quantiferon 179

and ELISPOT displayed undetectable values, in accordance with negative serology. 180

Of the 33 IgG seropositive healthy subjects analyzed, 4/33 (12%) displayed undetectable 181

CMI both with Quantiferon and ELISPOT, 15/33 (46%) resulted with positive CMI both 182

for Quantiferon and ELISPOT (>40 spot and >0.3 IU IFN-γ), while in 6/33 (18%) 183

ELISPOT and Quantiferon were discordant (4 ELISPOT +/ Quantiferon - and 2 184

ELISPOT - / Quantiferon +) and in 8/33 (24%) ELISPOT or Quantiferon displayed 185

borderline weak positive value (table 2). The 4/33 (12%) subjects displaying undetectable 186

CMI in Quantiferon and ELISPOT in disagreement with positive CMV IgG serology 187

produced detectable high responses in ELISPOT when stimulated with whole CMV 188

virion lysate (data not shown). 189

Quantiferon and ELISPOT in kidney transplants 190

A total 221 Quantiferon and ELISPOT tests were performed in a cohort of 120 kidney 191

transplants including adult 100 CMV R+ and 20 CMV D+/R-. ELISPOT and Quantiferon 192

analysis was performed at 30, 60, 90, 180 and 360 days after transplant. 193

on February 18, 2019 by guest

http://jcm.asm

.org/D

ownloaded from

- 9 -

In the group of 20 D+/R- all subjects analyzed displayed undetectable Quantiferon and 194

ELISPOT and resulted with negative CMV IgG and IgM serology throughout the 195

antiviral prophylaxis regimen. 196

In the group of 100 CMV R+ the time course of ELISPOT and Quantiferon revealed a 197

consistent increase over time that peaked at 180 days after transplant, followed by a slight 198

decrease at 360 days (figure 1). 199

Correlation of Quantiferon and ELISPOT and development of CMV DNAemia 200

In order to assess the correlation of Quantiferon and ELISPOT in transplants and the 201

relationship of Quantiferon and ELISPOT with CMV DNAemia we employed a negative 202

binomial regression statistical test. The results show that there is a statistical significant 203

correlation between Quantiferon and ELISPOT (p <0.037) and a statistical significant 204

inverse correlation between Quantiferon and ELISPOT and development of CMV 205

DNAemia (p<0.017) (table 3 and figure 2). Since the exponentiated coefficient of 206

Quantiferon is 100.0740 = 1.076781, a one-unit increase of Quantiferon predicts a 7.7% 207

increase of ELISPOT. 208

We also attempted to correlate ELISPOT and Quantiferon with the magnitude (peak of 209

viremia) and duration (days) of viremia using a linear regression approach but no 210

significant difference was found (data not shown). The statistical analysis showed that the 211

low number of cases might have caused the lack of significance. 212

Sensibility and specificity of Quantiferon and ELISPOT 213

In order to assess the sensitivity and specificity of Quantiferon and ELISPOT in 214

preventing the onset of CMV DNAemia we employed the ROC statistical analysis (figure 215

3A-B). The calculated OR for ELISPOT was 2.12. The maximum OR value observed for 216

on February 18, 2019 by guest

http://jcm.asm

.org/D

ownloaded from

- 10 -

ELISPOT was 5.01, at cutoff values ≥ 147. However, the OR mostly exceeded 4 in the 217

cut-off area ≥ 119 - ≥ 165. The Quantiferon calculated OR was 2.52. The maximum 218

value observed for OR was 8.75, at cutoff ≥ 6.1. However, this cutoff is very high, and 219

presumably not really useful, implying a low sensitivity of 19.13%. The ROC area, 220

standard error and 95% confidence interval for Quantiferon and ELISPOT are reported in 221

table 4. We also tested if the combining both ELISPOT and Quantiferon tests increases 222

the ROC area. As shown in figure 3C, the combination of both tests results in a modest 223

increase on the ROC area (0.67). 224

225

Discussion 226

Predicting the risk of infection in transplant recipients represents an innovative and 227

promising strategy to improve the clinical management of transplant recipients. 228

In this study we present the results of a comparative analysis of two IGRA tests, 229

ELISPOT and Quantiferon, widely used to assess the CMI in transplants. 230

In healthy CMV seronegative adult subjects both Quantiferon and ELISPOT were in 231

agreement with the negative IgG serology, while in CMV IgG seropositive adults 12% of 232

subjects resulted negative for both tests. This finding was unexpectedly high given 233

previous reports (34) showing a 97% agreement with serology: the negative results of 234

Quantiferon and ELISPOT opposing the positive CMV serology in healthy individuals 235

may probably depend on previously shown inability of certain individuals in recognizing 236

pp65 (ppUL83) stimulus peptide (33) or probably to atypical or non-common HLA 237

haplotypes of these subjects (Abate et. al. submitted). Indeed CMV IgG+ individuals with 238

negative CMI in Quantiferon and ELISPOT produced positive detectable high responses 239

on February 18, 2019 by guest

http://jcm.asm

.org/D

ownloaded from

- 11 -

when stimulated with not-HLA restricted whole CMV virion lysate (data not shown), 240

suggesting a limitation of the currently used stimuli in being recognized from 241

heterogeneous HLA types. The consistency of the results from healthy subjects should be 242

taken in account when immune monitoring is performed in transplants, in particular when 243

a negative result is obtained. In transplant subjects we have found that none of the D+/R- 244

receiving antiviral prophylaxis therapy resulted positive for Quantiferon, ELISPOT and 245

CMV IgG or IgM. This finding suggests that the current antiviral prophylaxis scheme is 246

highly effective in suppressing CMV reactivation from the allograft. This finding is also 247

consistent with previously published data on D+/R- kidney transplants being unable to 248

mount virus specific immune responses during the antiviral prophylaxis regimen (3). In 249

transplants, Quantiferon and ELISPOT displayed a positive statistical significant 250

correlation between the each other. Quantiferon and ELISPOT also showed a consistent 251

increase over time peaking at 180 days after transplant, followed by a decrease at 360 252

days: this may suggest either a general slow process of immune reconstitution boosted by 253

early post transplant episodes of CMV viremias, succeeding a steady state level of 254

antiviral immunity. 255

In transplant subjects both Quantiferon and ELISPOT displayed similar robustness 256

sensitivity and specificity and inverse correlation with development of CMV viremia 257

suggesting that values >150 spots/200000 PBMCs for ELISPOT and >1-6 IU IFN-γ for 258

Quantiferon may have good predictive value for protection from CMV viremia. The 259

proposed cutoffs refer to protection from CMV viremia and are different from the cutoff 260

of presence/absence of CMI proposed from the Quantiferon and ELISPOT 261

manufacturers. This study to our knowledge is the first comparison of Quantiferon and 262

on February 18, 2019 by guest

http://jcm.asm

.org/D

ownloaded from

- 12 -

ELISPOT in transplants and the results from the present study may better aid clinicians to 263

understand the limitations and the advantages of the two IGRA tests analyzed. This study 264

has also elucidated some critical aspects that may be improved for both tests: there is an 265

urgent need to overcome the unexpected high number of false negative results due to 266

HLA type inability to recognize the CMV stimulus composition. 267

268

Acknowledgements 269

We are grateful to Padua General Hospital - Azienda Ospedaliera di Padova for providing 270

us the IFN-γ ELISPOT plates and Cellestis and A.D.A, Italy, for providing Quantiferon 271

and control tubes and reagents. We thank Daniel Tinto, Alice Bianchin, Angela Bozza, 272

Simona Cofano and Valentina Fornasiero for the technical guidance. We also want to 273

thank Elisa Sefora Pierobon and Margherita Moro for medical advice. 274

275

Authors’ contribution 276

DA, CM, DS, PR, GP analyze the statistical data and wrote the manuscript. AS, LF, MP, 277

MF performed the tests and collected the experimental data. DA, LF, CS, PR, RC, LB, 278

RB, MF, DS, CM, GP supervised the clinical study. 279

280

Conflict of interests 281

The authors declare no conflict of interest. 282

283

284

285

on February 18, 2019 by guest

http://jcm.asm

.org/D

ownloaded from

- 13 -

286

287

288

Figure and figure legend 289

Figure 1 290 A 291

292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 B 307 308

0

5

10

15

20

30 60 90 180 360

days

309 310 311

0

200

400

600

800

1000

30 60 90 180 360

days on February 18, 2019 by guest

http://jcm.asm

.org/D

ownloaded from

- 14 -

Figure 1. ELISPOT (A) and Quantiferon (B) regression over time after transplant in 312 CMV R+ subjects. Hallow circles represent single observations. A “Lowess smoother” 313 (a locally weighted regression line, dashed line) was added for clarity. 314

315

on February 18, 2019 by guest

http://jcm.asm

.org/D

ownloaded from

- 15 -

Figure 2 316

0

100

200

300

400E

LIS

PO

T

0 2 4 6 8 10

QUANTIFERON

317

Figure 2. The fitted values of ELISPOT score (predictions of the negative binomial model) are indicated as 318 two lines, assuming an undetectable CMV DNAemia (dashed line) or a detectable CMV DNAemia (solid 319 line). Hollow diamonds indicate the individual observations. 95% confidence intervals for the predictions 320 are also indicated. ELISPOT values were limited to the scale 0–400, Quantiferon to 0-10. 321

322

323

324

325

326

327

328

329

330

331

332

333

on February 18, 2019 by guest

http://jcm.asm

.org/D

ownloaded from

- 16 -

Figure 3 334

A 335

B

7.06.0

3.5 3.0

1.6

.98.81 .79

.60.47 .43

.26

.19 .17 .16.12

.08.06 .05

.04.03

.02

.01

0

.2

.4

.6

.8

1

Sen

sitiv

ity

0 .2 .4 .6 .8 1

False positive rate

Quantiferon

336

355318301

147

119116 11395

8779

68 56 51 47

36

26 20

14 11

8 76 5 4

3 21

0

0

.2

.4

.6

.8

1

Sen

sitiv

ity

0 .2 .4 .6 .8 1

False positive rate

ELISPOT

337

338

339

340

on February 18, 2019 by guest

http://jcm.asm

.org/D

ownloaded from

- 17 -

C 341

0

.2

.4

.6

.8

1S

ensi

tivity

0 .2 .4 .6 .8 1

False positive rate

342

343 Figure 3. ROC curve analysis using only (A) Quantiferon (black solid dots, AUC = 0.6604) or (B) 344 ELISPOT (gray hollow dots, AUC = 0.6203) or (C) both tests in combination (AUC = 0.6731). The 345 endpoint (reference variable) was the protection against the emergence of an episode of CMV viremia. The 346 Quantiferon or ELISPOT scores were the classifying variables. Numbers on the curve represent the 347 absolute Quantiferon or ELISPOT values. 348 349 350

351

352

353

354

355

356

357

358

on February 18, 2019 by guest

http://jcm.asm

.org/D

ownloaded from

- 18 -

Table 1. Transplants characteristics 359

Patient’s characteristics KTRs (%)Total n. 120 Gender Male (%) 82 (71) Female (%) Age (median, range)

38 (29) 53 (21-82)

CMV serostatus D+/R+ and D-/R+ (%) 100 (83) D+/R- (%) 20 (17) Immunosuppressive regimen CNI, MMF, and steroids (%) 120 (100) Acute rejection episodes (%) 26 (22) Patients who experienced post-transplant CMV DNAemia All (%) 53 (44) R+ (%) 47 (89) R- (%) 6 (11) Patients with CMV disease All (%) 2 (2) R+ (%) 1 (50) R- (%) 1 (50) Patients who received treatment for CMV infection All (%) 32 (27) R+ (%) 27 (84) R- (%) 5 (16)

Abbreviations: CNI, calcineurin inhibitors; MMF, mycophenolate mofetil; mTOR, mammalian target or 360 rapamycin. 361 362

363

364

365

366

367

368

369

370

371

on February 18, 2019 by guest

http://jcm.asm

.org/D

ownloaded from

- 19 -

Table 2. ELISPOT and Quantiferon test in CMV IgG+ individuals 372

HS (%) ELISPOT Quantiferon15 (46%) Pos Pos 4 (12%) Neg Neg 4 (12%) Pos Neg 2 (6%) Neg Pos 4 (12%) Weak Pos Pos 1 (3%) Weak Pos Weak Pos 3 (9%) Pos Weak Pos

HS: healthy subjects; Positive values: ELISPOT >40, Quantiferon >0.3; Negative 373 values: ELISPOT <20, Quantiferon <0.2; Weak positives: ELISPOT >20-40<; 374 Quantiferon >0.2-0.3<. ELISPOT is expressed as number of spots/200.000 PBMCs; 375 Quantiferon as IU IFN-g 376 377

378

379

on February 18, 2019 by guest

http://jcm.asm

.org/D

ownloaded from

- 20 -

Table 3. Negative binomial regression of ELISPOT versus Quantiferon and CMV 380 DNAemia in kidney transplants. 381

382

Log likelihood = -1168.0423. LR chi2(2) = 11.00. Prob > chi2 = 0.0041 383 384 385 386 387 Table 4. ELISPOT and Quantiferon ROC 388

ROC Asymptotic Normal

Area Std. Err 95% Conf. Interval

ELISPOT 0.6203 0.0458 0.53048 0.7101

Quantiferon 0.6604 0.0447 0.57285 0.74797

ELISPOT and Quantiferon score as a predictor of CMV viremia episode avoidance (protection), ROC 389 curve method. The area under the curve (AUC) is reported, along with the standard error and the 95% 390 confidence interval. 391

392

393

ELISPOT Coefficient Std. Err. z P 95% Conf. Interval

QUANTIFERON 0.0740 0.0355 2.09 0.037 0.0045 0.1435

CMV DNAemia -0.7020 0.2946 -2.38 0.017 -1.2794 -0.1245

incept 4.6125 0.1448 31.85 0.000 4.3287 4.8963

Alpha 2.674709 0.2339 2.253408 3.174776

on February 18, 2019 by guest

http://jcm.asm

.org/D

ownloaded from

- 21 -

References 394

1. Abate, D., S. Cesaro, S. Cofano, M. Fiscon, A. Saldan, S. Varotto, C. Mengoli, 395 M. Pillon, E. Calore, M. A. Biasolo, R. Cusinato, L. Barzon, C. Messina, M. 396 Carli, and G. Palu. 2012. Diagnostic utility of human cytomegalovirus-specific 397 T-cell response monitoring in predicting viremia in pediatric allogeneic stem-cell 398 transplant patients. Transplantation 93:536-542. 399

2. Abate, D., M. Fiscon, A. Saldan, S. Cofano, C. Mengoli, D. Sgarabotto, C. 400 d'Agostino, L. Barzon, R. Cusinato, G. Toscano, G. Feltrin, A. Gambino, G. 401 Gerosa, and G. Palu. 2012. Human cytomegalovirus-specific T-cell immune 402 reconstitution in preemptively treated heart transplant recipients identifies 403 subjects at critical risk for infection. J Clin Microbiol 50:1974-1980. 404

3. Abate, D., A. Saldan, M. Fiscon, S. Cofano, A. Paciolla, L. Furian, B. Ekser, 405 M. A. Biasolo, R. Cusinato, C. Mengoli, L. Bonfante, B. Rossi, P. Rigotti, D. 406 Sgarabotto, L. Barzon, and G. Palu. 2010. Evaluation of cytomegalovirus 407 (CMV)-specific T cell immune reconstitution revealed that baseline antiviral 408 immunity, prophylaxis, or preemptive therapy but not antithymocyte globulin 409 treatment contribute to CMV-specific T cell reconstitution in kidney transplant 410 recipients. J Infect Dis 202:585-594. 411

4. Baldanti, F., D. Lilleri, and G. Gerna. 2008. Monitoring human 412 cytomegalovirus infection in transplant recipients. J Clin Virol 41:237-241. 413

5. Borchers, S., M. Bremm, T. Lehrnbecher, E. Dammann, B. Pabst, B. Wolk, 414 R. Esser, M. Yildiz, M. Eder, M. Stadler, P. Bader, H. Martin, A. Jarisch, G. 415 Schneider, T. Klingebiel, A. Ganser, E. M. Weissinger, and U. Koehl. 2012. 416 Sequential anti-cytomegalovirus response monitoring may allow prediction of 417 cytomegalovirus reactivation after allogeneic stem cell transplantation. PLoS One 418 7:e50248. 419

6. Chee, C. B., S. H. Gan, K. W. Khinmar, T. M. Barkham, C. K. Koh, S. 420 Liang, and Y. T. Wang. 2008. Comparison of sensitivities of two commercial 421 gamma interferon release assays for pulmonary tuberculosis. J Clin Microbiol 422 46:1935-1940. 423

7. Clari, M. A., B. Munoz-Cobo, C. Solano, I. Benet, E. Costa, M. J. Remigia, D. 424 Bravo, P. Amat, and D. Navarro. 2012. Performance of the QuantiFERON-425 cytomegalovirus (CMV) assay for detection and estimation of the magnitude and 426 functionality of the CMV-specific gamma interferon-producing CD8(+) T-cell 427

on February 18, 2019 by guest

http://jcm.asm

.org/D

ownloaded from

- 22 -

response in allogeneic stem cell transplant recipients. Clin Vaccine Immunol 428 19:791-796. 429

8. Costa, C., A. Saldan, F. Sinesi, F. Sidoti, C. Balloco, S. Simeone, A. 430 Piceghello, S. Mantovani, A. Di Nauta, P. Solidoro, and R. Cavallo. 2012. The 431 lack of cytomegalovirus-specific cellular immune response may contribute to the 432 onset of organ infection and disease in lung transplant recipients. Int J 433 Immunopathol Pharmacol 25:1003-1009. 434

9. Crough, T., C. Fazou, J. Weiss, S. Campbell, M. P. Davenport, S. C. Bell, A. 435 Galbraith, K. McNeil, and R. Khanna. 2007. Symptomatic and asymptomatic 436 viral recrudescence in solid-organ transplant recipients and its relationship with 437 the antigen-specific CD8(+) T-cell response. J Virol 81:11538-11542. 438

10. Cummins, N. W., P. J. Deziel, R. S. Abraham, and R. R. Razonable. 2009. 439 Deficiency of cytomegalovirus (CMV)-specific CD8+ T cells in patients 440 presenting with late-onset CMV disease several years after transplantation. 441 Transpl Infect Dis 11:20-27. 442

11. Danziger-Isakov, L., and P. S. Heeger. 2009. Clinical utility of measuring T-cell 443 immunity to CMV in transplant recipients. Am J Transplant 9:987-988. 444

12. Diel, R., D. Goletti, C. Lange, and D. Manissero. 2011. Interferon-gamma 445 release assays for the diagnosis of latent Mycobacterium tuberculosis infection. 446 Eur Respir J 38:1238-1239. 447

13. Egli, A., A. Humar, and D. Kumar. 2012. State-of-the-art monitoring of 448 cytomegalovirus-specific cell-mediated immunity after organ transplant: a primer 449 for the clinician. Clin Infect Dis 55:1678-1689. 450

14. Gerna, G., and D. Lilleri. 2006. Monitoring transplant patients for human 451 cytomegalovirus: Diagnostic update. Herpes 13:4-11. 452

15. Gerna, G., D. Lilleri, C. Fornara, G. Comolli, L. Lozza, C. Campana, C. 453 Pellegrini, F. Meloni, and T. Rampino. 2006. Monitoring of human 454 cytomegalovirus-specific CD4 and CD8 T-cell immunity in patients receiving 455 solid organ transplantation. Am J Transplant 6:2356-2364. 456

16. Kotton, C. N. 2013. CMV: Prevention, Diagnosis and Therapy. Am J Transplant 457 13 Suppl 3:24-40. 458

on February 18, 2019 by guest

http://jcm.asm

.org/D

ownloaded from

- 23 -

17. Kotton, C. N., D. Kumar, A. M. Caliendo, A. Asberg, S. Chou, D. R. 459 Snydman, U. Allen, and A. Humar. 2010. International consensus guidelines on 460 the management of cytomegalovirus in solid organ transplantation. 461 Transplantation 89:779-795. 462

18. Kumar, D., S. Chernenko, G. Moussa, I. Cobos, O. Manuel, J. Preiksaitis, S. 463 Venkataraman, and A. Humar. 2009. Cell-mediated immunity to predict 464 cytomegalovirus disease in high-risk solid organ transplant recipients. Am J 465 Transplant 9:1214-1222. 466

19. Lisboa, L. F., D. Kumar, L. E. Wilson, and A. Humar. 2012. Clinical utility of 467 cytomegalovirus cell-mediated immunity in transplant recipients with 468 cytomegalovirus viremia. Transplantation 93:195-200. 469

20. Manuel, O., S. Husain, D. Kumar, C. Zayas, S. Mawhorter, M. E. Levi, J. 470 Kalpoe, L. Lisboa, L. Ely, D. R. Kaul, B. S. Schwartz, M. I. Morris, M. G. 471 Ison, B. Yen-Lieberman, A. Sebastian, M. Assi, and A. Humar. 2013. 472 Assessment of Cytomegalovirus-Specific Cell-Mediated Immunity for the 473 Prediction of Cytomegalovirus Disease in High-Risk Solid-Organ Transplant 474 Recipients: A Multicenter Cohort Study. Clin Infect Dis. 475

21. Manuel, O., and D. Kumar. 2008. QuantiFERON-TB Gold assay for the 476 diagnosis of latent tuberculosis infection. Expert Rev Mol Diagn 8:247-256. 477

22. Mengoli, C., R. Cusinato, M. A. Biasolo, S. Cesaro, C. Parolin, and G. Palu. 478 2004. Assessment of CMV load in solid organ transplant recipients by pp65 479 antigenemia and real-time quantitative DNA PCR assay: correlation with pp67 480 RNA detection. J Med Virol 74:78-84. 481

23. Morris, M. I., J. S. Daly, E. Blumberg, D. Kumar, M. Sester, N. Schluger, S. 482 H. Kim, B. S. Schwartz, M. G. Ison, A. Humar, N. Singh, M. Michaels, J. P. 483 Orlowski, F. Delmonico, T. Pruett, G. T. John, and C. N. Kotton. 2012. 484 Diagnosis and management of tuberculosis in transplant donors: a donor-derived 485 infections consensus conference report. Am J Transplant 12:2288-2300. 486

24. Nebbia, G., F. M. Mattes, C. Smith, E. Hainsworth, J. Kopycinski, A. 487 Burroughs, P. D. Griffiths, P. Klenerman, and V. C. Emery. 2008. 488 Polyfunctional cytomegalovirus-specific CD4+ and pp65 CD8+ T cells protect 489 against high-level replication after liver transplantation. Am J Transplant 8:2590-490 2599. 491

on February 18, 2019 by guest

http://jcm.asm

.org/D

ownloaded from

- 24 -

25. Pai, M., L. W. Riley, and J. M. Colford, Jr. 2004. Interferon-gamma assays in 492 the immunodiagnosis of tuberculosis: a systematic review. Lancet Infect Dis 493 4:761-776. 494

26. Pai, M., A. Zwerling, and D. Menzies. 2008. Systematic review: T-cell-based 495 assays for the diagnosis of latent tuberculosis infection: an update. Ann Intern 496 Med 149:177-184. 497

27. Ravkov, E. V., I. Y. Pavlov, K. E. Hanson, and J. C. Delgado. 2012. Validation 498 of cytomegalovirus immune competence assays for the characterization of 499 CD8(+) T cell responses posttransplant. Clin Dev Immunol 2012:451059. 500

28. Ritter, M., T. Schmidt, J. Dirks, P. Hennes, I. Juhasz-Boss, E. F. Solomayer, 501 L. Gortner, B. Gartner, T. Rohrer, U. Sester, and M. Sester. 2012. 502 Cytomegalovirus-specific T cells are detectable in early childhood and allow 503 assignment of the infection status in children with passive maternal antibodies. 504 Eur J Immunol. 505

29. Schmidt, T., M. Ritter, J. Dirks, B. C. Gartner, U. Sester, and M. Sester. 506 2012. Cytomegalovirus-specific T-cell immunity to assign the infection status in 507 individuals with passive immunity: a proof of principle. J Clin Virol 54:272-275. 508

30. Sester, M., G. Sotgiu, C. Lange, C. Giehl, E. Girardi, G. B. Migliori, A. 509 Bossink, K. Dheda, R. Diel, J. Dominguez, M. Lipman, J. Nemeth, P. Ravn, 510 S. Winkler, E. Huitric, A. Sandgren, and D. Manissero. 2011. Interferon-511 gamma release assays for the diagnosis of active tuberculosis: a systematic review 512 and meta-analysis. Eur Respir J 37:100-111. 513

31. Sester, U., B. C. Gartner, H. Wilkens, B. Schwaab, R. Wossner, I. 514 Kindermann, M. Girndt, A. Meyerhans, N. Mueller-Lantzsch, H. J. Schafers, 515 G. W. Sybrecht, H. Kohler, and M. Sester. 2005. Differences in CMV-specific 516 T-cell levels and long-term susceptibility to CMV infection after kidney, heart 517 and lung transplantation. Am J Transplant 5:1483-1489. 518

32. Sester, U., D. Presser, J. Dirks, B. C. Gartner, H. Kohler, and M. Sester. 519 2008. PD-1 expression and IL-2 loss of cytomegalovirus- specific T cells 520 correlates with viremia and reversible functional anergy. Am J Transplant 8:1486-521 1497. 522

33. Sylwester, A. W., B. L. Mitchell, J. B. Edgar, C. Taormina, C. Pelte, F. 523 Ruchti, P. R. Sleath, K. H. Grabstein, N. A. Hosken, F. Kern, J. A. Nelson, 524

on February 18, 2019 by guest

http://jcm.asm

.org/D

ownloaded from

- 25 -

and L. J. Picker. 2005. Broadly targeted human cytomegalovirus-specific CD4+ 525 and CD8+ T cells dominate the memory compartments of exposed subjects. J Exp 526 Med 202:673-685. 527

34. Walker, S., C. Fazou, T. Crough, R. Holdsworth, P. Kiely, M. Veale, S. Bell, 528 A. Gailbraith, K. McNeil, S. Jones, and R. Khanna. 2007. Ex vivo monitoring 529 of human cytomegalovirus-specific CD8+ T-cell responses using QuantiFERON-530 CMV. Transpl Infect Dis 9:165-170. 531

532 533

on February 18, 2019 by guest

http://jcm.asm

.org/D

ownloaded from

on February 18, 2019 by guest

http://jcm.asm

.org/D

ownloaded from

on February 18, 2019 by guest

http://jcm.asm

.org/D

ownloaded from

on February 18, 2019 by guest

http://jcm.asm

.org/D

ownloaded from

on February 18, 2019 by guest

http://jcm.asm

.org/D

ownloaded from

on February 18, 2019 by guest

http://jcm.asm

.org/D

ownloaded from

on February 18, 2019 by guest

http://jcm.asm

.org/D

ownloaded from

Comparison of Cytomegalovirus (CMV) Enzyme-LinkedImmunosorbent Spot and CMV Quantiferon Gamma Interferon-Releasing Assays in Assessing Risk of CMV Infection in KidneyTransplant Recipients

Davide Abate,a Alda Saldan,a Carlo Mengoli,a Marta Fiscon,a Cristina Silvestre,b Loredana Fallico,a Marta Peracchi,a Lucrezia Furian,b

Riccardo Cusinato,a Luciana Bonfante,c Barbara Rossi,d Francesco Marchini,d Dino Sgarabotto,e Paolo Rigotti,b Giorgio Palùa

Department of Molecular Medicine,a Department of Surgery, Oncology and Gastroenterology,b Department of Medicine Nephrology I Unit,c Hemodialysis andNephrology II Unit,d and Transplant Infectious Disease Division,e Padua General Hospital, University of Padua School of Medicine, Padua, Italy

Volume 51, no. 8, p. 2501–2507, 2013. Page 2502, Materials and Methods, “Patients and definitions” section, line 2: “125 kidneytransplant recipients” should read “120 kidney transplant recipients.”

Page 2502, Materials and Methods, “Patients and definitions” section, line 7: “One hundred twenty-five KTRs” should read “Onehundred twenty KTRs.”

Page 2502, Materials and Methods, “Patients and definitions” section: Information about the transplant immunosuppression inductionprotocol used was inadvertently omitted. The following statement should be inserted at the end of line 9: “The following inductionimmunosuppressive treatments were employed: antithymocyte globulin (ATG) for 67/120 patients (56%) (median age of cohort, 60years), basiliximab for 47/120 patients (39%) (median age of cohort, 48 years), both ATG and basiliximab for 3/120 patients (2.5%)(median age of cohort, 34 years), and alefacept for 3/120 patients (2.5%) (median age of cohort, 33 years).” The overall results andconclusions of the study are unchanged.

Copyright © 2014, American Society for Microbiology. All Rights Reserved.

doi:10.1128/JCM.00110-14

AUTHOR CORRECTION

March 2014 Volume 52 Number 3 Journal of Clinical Microbiology p. 1025 jcm.asm.org 1025