ORIGINAL STUDIES Interferon Gamma, Interferon-Gamma-Induced-Protein 10, and Tuberculin Responses of Children at High Risk of Tuberculosis Infection Roberta Petrucci, MD, MTropPaed,* Nabil Abu Amer, MD, DTMH, MTropMed,* Ricardo Queiroz Gurgel, MD, MSc, PhD,† Jeevan B. Sherchand, MSc TM, PhD,‡ Luiza Doria, MD, MSc, PhD,† Chamala Lama, BSc, MSc,‡ Pernille Ravn, MD, PhD,§ Morten Ruhwald, MD,§ Mohammed Yassin, MD, MTopMed, PhD,* Gregory Harper, BSc,* and Luis Eduardo Cuevas, MD, DTCH, MtopMed* Background: Children in contact with adults with pulmonary tu- berculosis (TB) are at risk for infection and disease progression, and chemoprophylaxis may reduce this risk. The identification of infec- tion is based on the tuberculin skin test (TST) and interferon- (INF-) release assays. Other biomarkers such as interferon-- induced-protein 10 (IP-10) may have potential for the diagnosis of latent TB infections. Objectives: To describe IP-10 concentrations and their association to TST and INF- responses in children recently exposed to adults with smear-positive TB in Brazil and Nepal. Methods: Two surveys using the same design were undertaken to describe TST, INF-, and IP-10 responses in 146 children in Nepal and 113 children in Brazil. Results: The concordance of TST and QuantiFERON-TB gold in-tube (QFT-IT) was high ( 0.73 in Brazil and 0.80 in Nepal). IP-10 responses were higher in children with both positive TST and positive QFT-IT (medians 1434 pg/mL in Brazil and 1402 pg/mL in Nepal) and lowest in children with both negative TST and negative QFT-IT (medians 206 pg/mL in Brazil and 81 pg/mL in Nepal). Children with negative TST and positive QFT-IT had higher IP-10 concentrations than children with positive TST but negative QFT-IT. Conclusions: IP-10 is a potential marker to identify latent TB infections that is expressed in large quantities and with good agree- ment with QFT-IT. The reasons for the discrepant results observed are discussed. Key Words: interferon-gamma, IP-10, latent tuberculosis, children (Pediatr Infect Dis J 2008;27: 1073–1077) C hildren in contact with adults with pulmonary tubercu- losis (TB) are at high risk of infection and disease progression. Up to half of infected infants and 15% of older children develop active TB, most within 2 years of infection. The timely detection of latent TB infections (LTBI) and the provision of chemoprophylaxis therefore are critical to reduce the risk of disease progression in this vulnerable population. For many years, the tuberculin skin test (TST) was the main diagnostic test used for the identification of LTBI. Although this method has reasonable performance in some settings, 1 often children receiving bacillus calmette-gue ´rin (BCG) after birth or infected with nontuberculous mycobacteria have false positive indurations 1 and children with intercurrent serious ailments such as measles, malnutrition, or immune suppres- sion have false negative reactions. 2,3 In vitro T-cell-based interferon gamma release assays (IGRAs) have recently be- come available as alternative or complementary tests to the TST. These tests use Mycobacterium tuberculosis complex specific antigens to elicit the release of interferon gamma (INF-) by T cells that have been sensitized by a previous TB infection. Although it is often suggested that the IGRAs detect LTBI more accurately than TST, 4 the lack of a stan- dard for LTBI has hampered the assessment of their sensi- tivity and specificity. IGRAs have used varying combinations of M. tuberculo- sis antigens since their initial development. 5 For example, earlier versions of the assays used single antigens (eg, Early Secretory Antigenic Target-6; ESAT-6) and more recent formulations have used a combination of antigens, with a reported increase in the sensitivity of the tests, suggesting that further antigen combinations may continue to optimize their performance. Current IGRAs are solely based on the detection of INF-, even though the complex immunologic interactions elicited after infection results in the expression of a large number of other pro- and anti-inflammatory cytokines. 6,7 Accepted for publication April 18, 2008. From the *Liverpool School of Tropical Medicine, Liverpool, UK; †Federal University of Sergipe, Aracaju, Brazil; ‡Microbiology and Parasitology and Health Research Laboratory, Tribhuvan University Institute of Med- icine, Nepal; §Department of Infectious Diseases, Copenhagen Univer- sity Hospital, Denmark; and Clinical Research Centre, Copenhagen University Hospital, Denmark. Copenhagen University Hospital (Ruhwald and Ravn) applied for a patent disclosing IP-10 as a marker for Mycobaterium tuberculosis infection. Pernille Ravn received a single payment of €2000 from Cellestis Ltd. (2006) to develop clinical guidelines for the test. The remaining authors do not have any conflict of interest to declare. Address for correspondence: Luis Eduardo Cuevas, MD, DTCH, MtopMed, Liverpool School of Tropical Medicine, Pembroke Place, Liverpool L3 5QA, UK. E-mail: lcuevas@liverpool.ac.uk. Copyright © 2008 by Lippincott Williams & Wilkins ISSN: 0891-3668/08/2712-1073 DOI: 10.1097/INF.0b013e31817d05a3 The Pediatric Infectious Disease Journal • Volume 27, Number 12, December 2008 1073