Tuberculosis 95 (2015) 639e650

Contents lists available at ScienceDirect

Tuberculosis journal homepage: http://intl.elsevierhealth.com/journals/tube

REVIEW

Interferon gamma release assays for monitoring the response to treatment for tuberculosis: A systematic review Vanessa Clifford 1, Yu He 1, Christel Zufferey, Tom Connell, Nigel Curtis* Department of Paediatrics, The University of Melbourne, Murdoch Children's Research Institute, Royal Children's Hospital Melbourne, Parkville, VIC, Australia

a r t i c l e i n f o

s u m m a r y

Article history: Received 16 March 2015 Accepted 4 July 2015

Introduction: The ability to monitor the response to therapy for tuberculosis (TB) and confirm adequate treatment would be a major advance. The utility of interferon gamma assays (IGRA) for this purpose remains uncertain. Methods: A systematic search of all studies investigating commercial IGRA to monitor anti-tuberculous treatment was done. Studies were included if they included an IGRA before the start of, and at least once during, treatment for active or latent TB. Results: We identified 30 studies, of which 24 used QuantiFERON-TB (QFT), three used T-SPOT.TB and three used both QFT and T-SPOT.TB. Most studies were done in low TB incidence countries. No uniform pattern was seen in IGRA conversion and reversion rates at the end of treatment for active or latent TB. In most studies, the majority of IGRA results remained positive at the end of treatment. In many studies, the quantitative levels of IFN-g decreased during treatment, particularly in active TB. There was significant heterogeneity in the included studies. Conclusion: While quantitative IGRA responses generally fall during treatment for TB, the large degree of variation in results between participants in each study means that IGRAs are unlikely to be useful for monitoring anti-tuberculous treatment in clinical practice for any individual patient. © 2015 Elsevier Ltd. All rights reserved.

Keywords: Mycobacterium tuberculosis Interferon gamma release assay QuantiFERON T-SPOT.TB Monitoring Antibiotics Diagnosis Systematic review

Contents 1. 2.

3.

4.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 640 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 640 2.1. Search strategy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 640 2.2. Study selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 640 2.3. Exclusion criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 640 2.4. Data extraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 640 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 640 3.1. Qualitative (categorical) changes in IGRA results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 641 3.2. Quantitative changes in IGRA results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 641 3.2.1. Active TB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 641 3.2.2. LTBI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 641 3.2.3. Children . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 641 3.2.4. Immunocompromised . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 641 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 644 Funding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 649 Competing interests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 649

* Corresponding author. Department of Paediatrics, The University of Melbourne, Royal Children's Hospital Melbourne, Flemington Road, Parkville, VIC 3052, Australia. Tel.: þ61 3 9345 6366; fax: þ61 3 9345 4751. E-mail address: [email protected] (N. Curtis). 1 The first two authors have contributed equally and wish to be regarded as joint first authors. http://dx.doi.org/10.1016/j.tube.2015.07.002 1472-9792/© 2015 Elsevier Ltd. All rights reserved.

640

V. Clifford et al. / Tuberculosis 95 (2015) 639e650

Ethical approval . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 649 Authors' contributions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 649 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 649

1. Introduction

2.4. Data extraction

A biomarker to indicate successful tuberculosis (TB) treatment would be a major advance for the management and control of TB globally. In addition to monitoring treatment in individual patients, it would enable the assessment of shorter course regimens for the treatment of both latent TB infection (LTBI) and active TB; currently the absence of a reliable surrogate marker of cure hampers trials of new TB therapies. Biomarkers to confirm adequate TB treatment would also help reduce transmission from subsequent reactivation following failed treatment. Recent research has focused on the use of interferon gamma (IFN-g) release assays (IGRAs) as a biomarker of treatment success. Animal and human studies have shown a relationship between the Mycobacterium tuberculosis (MTB) bacillary load and the magnitude of IFN-g responses to MTB antigens [1,2]. It has therefore been postulated that a decrease in the magnitude of IFN-g responses to MTB-specific peptides measured by IGRA can be used as a biomarker of cure [3]. To assess the utility of IGRAs for this purpose, a number of studies have investigated the kinetics of IGRA responses during the treatment of TB. We did a systematic review of the use of IGRAs in monitoring the response to treatment of LTBI or active TB.

A standardised table was used by two authors independently for data extraction. The following data were recorded where available: country of study; age of participants; TB status of participants (active TB or LTBI); immune status of participants (including HIV status); method of TB diagnosis; IGRA type; the number and proportion of positive IGRA results at baseline and at each study follow-up point; the number and proportion of participants with IGRA conversion or reversion from the start to end of treatment; quantitative level of IFN-g measured at start and end of treatment and the proportion of participants who had an IGRA at the end of treatment (‘completion rate’). Active TB was categorised as either ‘confirmed’ (culture or PCR positive) or ‘probable’ (clinical or histological diagnosis of TB with appropriate response to treatment). Where data were missing, authors of relevant papers were contacted by email for further information.

2. Methods 2.1. Search strategy Our review was done in accordance with the ‘preferred reporting items for systematic reviews and meta-analyses’ (PRISMA) statement. Original articles, letters to the editor and published abstracts were identified by searching MEDLINE (1995 to November 2014), EMBASE (to November 2014) and the Cochrane Central Register of Controlled Trials (CENTRALL). The following search terms were used: (tuberculosis/or latent tuberculosis/or lung tuberculosis/) and (gamma interferon/or enzyme linked immunospot assay/or immunoassay/) and (tuberculostatic agent/or ethambutol/or isoniazid/or pyrazinamide/or rifampicin/or treatment outcome/or patient monitoring/or treatment response/or drug monitoring/or kinetics/).

2.2. Study selection Studies were included if they had an IGRA before the start of, and at least once during, treatment for active or latent TB. No restrictions were placed on sample size or method of data collection.

2.3. Exclusion criteria Studies were excluded if participants did not receive treatment or results did not differentiate between participants who did and did not receive treatment. Studies without original data, case reports, animal studies, cross sectional studies, studies not published in English, and studies that used ‘in-house’ immunoasssays were also excluded.

3. Results A total of 1508 unique references were identified by the initial search. After review of title and abstracts, 75 full text articles were assessed for eligibility. Forty-five articles were excluded after review of the full texts. Of these 45 articles, 14 used an in-house ELISPOT [1,2,4e15], five were excluded because they did not use ESAT-6/CFP-10-specific stimulation [16e20], 4 were cross-sectional [21e24], two did not include original data [25,26], 15 did not provide longitudinal follow-up treatment-specific data [27e41], two were published in a language other than English [42,43] and three were conference abstracts [44e46]. A total of 30 articles were eligible for inclusion in the review (Figure 1, Tables 1 and 2). Studies were from countries with both high and low TB incidence; the majority were done in low incidence countries (21/30). Of the 30 included studies, 24 used QuantiFERON-TB (QFT) [47e70] for longitudinal assessment of IFN-g levels in participants, three used T-SPOT.TB [71e73] and three used both QFT and TSPOT.TB [74e76]. Twenty studies included only adults [47,48,50,53,55e59,61,62,64e66,68e71,73,76], three studies were done in children [54,63,75] and seven studies had a mixed population [49,51,52,60,67,72,74] (usually including only a small number of children and/or adolescents). Fifteen studies investigated active TB patients [49e51,56e59,61,64,67,69,71,73,74,76], 12 investigated LTBI patients [47,48,52,53,55,60,62,65,66,68,70,72] and three investigated both LTBI and active TB patients [54,63,75]. Most studies (23/30) had completion rates greater than 70%. One study included a significant proportion of HIV positive patients (54%) [56], with a further 3 studies including a small proportion (