Risk of Celiac Disease in Children With Type 1 Diabetes Is Modified by Positivity for HLA-DQB1*02-DQA1*05 and TNF 308A ZDENEK SUMNIK, MD, PHD 1 ONDREJ CINEK, MD, PHD 1 NINA BRATANIC, MD 2 OLGA KORDONOURI, PHD 3 MICHAL KULICH, PHD 1 BARNABAS ROSZAI, MD 4 ANDRAS ARATO, MD, PHD 5 JAN LEBL, MD, PHD 6 GYULA SOLTESZ, MD, PHD 4 THOMAS DANNE, MD 7 TADEJ BATTELINO, MD, PHD 2 EDIT SCHOBER, MD 8 OBJECTIVE — The overlap between genetic susceptibility to celiac disease (CD) and to type 1 diabetes is incomplete; therefore, some genetic polymorphisms may significantly modify the risk of CD in subjects with type 1 diabetes. This study aimed to investigate whether the suscep- tibility to CD in diabetic children is modified by positivity for HLA-DQB1*02-DQA1*05 and DQB1*0302-DQA1*03 and by alleles of single nucleotide polymorphisms within the genes encoding CTLA4, transforming growth factor (TGF)-, tumor necrosis factor (TNF)-, inter- feron (IFN)-, interleukin (IL)-1, IL-2, IL-6, and IL-10. RESEARCH DESIGN AND METHODS — Genotypic data were compared between 130 case subjects (children with type 1 diabetes and CD diagnosed using endomysium antibodies) and 245 control subjects (children with type 1 diabetes only, optimally two per case, matched for center, age at type 1 diabetes onset, and type 1 diabetes duration). The subjects were recruited from 10 major European pediatric diabetes centers performing regular screening for CD. The polymorphisms were determined using PCR with sequence-specific primers, and the risk was assessed by building a step-up conditional logistic regression model using variables that were significantly associated with CD in the univariate analysis. RESULTS — The best-fitted model showed that risk of CD is increased by presence of HLA- DQB1*02-DQA1*05 (odds ratio 4.5 [95% CI 1.8 –11], for homozygosity, and 2.0 [1.1–3.7], for a single dose) and also independently by TNF -308A (1.9 [1.1–3.2], for phenotypic positivity), whereas IL1--889T showed a weak negative association (0.6 [0.4 – 0.9]). CONCLUSIONS — The results indicate that the risk of CD in children with type 1 diabetes is significantly modified both by the presence of HLA-DQB1*02-DQA1*05 and by a variant of another gene within the major histocompatibility complex, the TNF -308A. Diabetes Care 29:858 – 863, 2006 C eliac disease (CD) is the second most prevalent autoimmune condi- tion accompanying type 1 diabetes, after autoimmune thyroid disease. In Eu- rope, CD is 10 times more frequent in patients with type 1 diabetes than in the respective general populations (1). The reason for the association has not yet been fully elucidated. The environment of the ongoing diabetic autoimmunity may be a stimulant to the development of CD, a disease that possesses autoimmune fea- tures. Also, the genetic susceptibility to both conditions substantially overlaps: a certain degree of susceptibility to both diseases is associated with the HLA mol- ecules DQB1*0302-DQA1*03 and DQB1*02-DQA1*05, encoded either in cis or trans. The CTLA4 polymorphisms have also been implicated, as well as several polymorphisms within the genes encod- ing for cytokines. However, the overlap of the genetic susceptibility is incomplete. This is well demonstrated in the HLA-DQ region: the main determinant of CD risk is HLA- DQB1*02-DQA1*05, whereas for type 1 diabetes in populations of central and northern Europe, it is HLA-DQB1*0302- DQA1*03. There are also strong indica- tions that certain variants of the tumor necrosis factor (TNF)- gene, another gene residing within the major histocom- patibility complex, are associated with CD independently of HLA-DQ (2–5), whereas in type 1 diabetes, the TNF asso- ciation is secondary to that of HLA (6,7), although this opinion is not universally accepted (8,9). CTLA4 +49A/G is a marker associated with both diseases: type 1 diabetes (as well as several other autoimmune diseases) is associated with the G allele (rev. in 10), whereas for CD, the studies on the association have yielded conflicting results. Thus, CD and type 1 diabetes differ in the strength or direction of genetic asso- ciation with particular alleles of the same genes. Such a difference might be useful for determining the level of CD risk among patients with type 1 diabetes. This idea is appealing because genetic testing ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● From the 1 Motol University Hospital, the Second Medical School, Charles University, Prague, The Czech Republic; the 2 University Children’s Hospital, Department of Pediatric and Adolescent Endocrinology, Ljubljana, Slovenia; the 3 Clinic of General Pediatrics, Otto-Heubner-Centrum, Charite, Campus Virchow- Klinikum, Humboldt University, Berlin, Germany; the 4 Department of Pediatrics, University of Pecs, Pecs, Hungary; the 5 First Department of Pediatrics, Semmelweis University, Budapest, Hungary; the 6 Department of Pediatrics, The Third Medical School, Charles University, Prague, The Czech Republic; 7 Kinderkranken- haus auf der Bult, Hannover, Germany; and the 8 University Children’s Hospital, University of Vienna, Vienna, Austria. Address correspondence and reprint requests to Ondrej Cinek, MD, PhD, Department of Pediatrics, Motol University Hospital, Charles University Prague, V Uvalu 84, CZ-150 06, Prague, The Czech Republic. E-mail: ondrej.cinek@lfmotol.cuni.cz. Received for publication 9 October 2005 and accepted in revised form 15 January 2006. Z.S. and O.C. contributed equally to this work. Abbreviations: CD, celiac disease; IHWC, 13th International Histocompatibility Workshop and Con- ference; IL, interleukin; PCR-SSP, PCR with sequence-specific primers; SNP, single nucleotide polymor- phism; TNF, tumor necrosis factor. A table elsewhere in this issue shows conventional and Syste `me International (SI) units and conversion factors for many substances. © 2006 by the American Diabetes Association. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Pathophysiology/Complications O R I G I N A L A R T I C L E 858 DIABETES CARE, VOLUME 29, NUMBER 4, APRIL 2006