This multi-centric cohort study [1] from four countries followed a group of infants with the HLA haplotype DR3–DQ2 or DR4–DQ8, from birth through the first few years of life; seeking the appearance of antibodies to tissue transglutaminase (tTG) (labeled as celiac disease autoimmunity), and development of celiac disease. This was part of a larger study evaluating the development of type 1 diabetes in a cohort of infants with genetic susceptibility (based on carrying the HLA haplotype DR3-DQ2 or DR4-DQ8) [2]. Over a median follow-up duration of nearly five years, the investigators reported 12% prevalence of celiac disease autoimmunity and 3% prevalence of celiac disease. They also identified that the respective risks of these two outcomes varied by the HLA genotype: 26% and 11% with homozygosity for DR3-DQ2 haplotype; 11% and 3% with DR3-DQ2/DR4-DQ8 haplotype; 8% and 3% with DR4-DQ8 homozygosity; and 3% and <1% among those with DR4-DQ8/DR8-DQ4 haplotype. There was statistically significant higher risk of celiac disease autoimmunity and celiac disease in infants from Europe (highest risk in Sweden), female gender, and those with family history of celiac disease.

Relevance: Data from developed countries suggest celiac disease prevalence of 1% [3] in the general pediatric population. A similar prevalence has been reported in India [4]. There is significantly higher prevalence among those with type I diabetes (3-10%) and family members of those with celiac disease (5-20%) [5-8]. HLA DQ2 and DQ8 are considered the most important genetic risk factor for celiac disease; these haplotypes are present in nearly all patients with celiac disease [9,10]. The absence of these haplotypes has very high negative predictive value for celiac disease [11,12]. However, HLA-DQ2 is present in about one-fourth to one-third of the unaffected Caucasian population; hence it alone has low positive predictive value [13], although combination with anti tTG IgA and endomysial antibody (triple test) improves it to nearly 100% [14]. Against this background, the cohort study [1] can be regarded as highly relevant. In the Indian context, there is limited information that celiac disease is associated with multiple DR3-DQ2 haplotypes [15,16] although population-based studies are unavailable.

Critical appraisal: The Critical Appraisal Skills Programme (CASP tool) [17] was used to evaluate this study (Table I).

TABLE I  Critical Appraisal of the Study Using the CASP Tool 

Extendibility: Some authors have sought to identify whether the HLA loci associated with celiac disease in Caucasian (European) population is similar in other ethnic groups. In a cohort of North Indian people, three of the several loci seen in Europeans, were observed to show strong association [18]. However other authors have suggested that there are three sets of HLA-DR3 haplotypes associated with development of type 1 diabetes [19]. Although the B8-DR3 haplotype confers the highest risk, there is considerable diversity of this haplotype in Indians unlike the single fixed haplotype observed in Caucasians [20]. These data suggest that while the results of the multi-centric cohort study may be relevant to India, the data cannot be directly extrapolated.

Conclusions: This cohort study suggests that infants with certain HLA haplotypes (especially DR3-DQ2 homozygosity) have higher risk for development of celiac disease autoimmunity and biopsy proven disease (compared to other haplotypes also associated with celiac disease). There are ethnic and gender variations, and family history confers increased risk.

Celiac disease (CD) results from a dysregulated immune response to dietary wheat and related cereal proteins. This disease has gained importance from its HLA-linked genetic basis as evident by ethnic and racial clustering, familial aggregation, twin concordance, strong association with certain syndromes, and autoimmunity. Many environmental factors, non-HLA linked genetics (innate, adaptive and mucosal barrier) and recently early childhood intestinal microbiota dysbiosis have also been implicated in the development of this disease [21]. However HLA-linked genetics and its "gene-dose effect" continue to be the stronghold of the etiopathogenesis and thus its utility as an aid in screening and diagnosis of CD [22].

Virtually all CD patients express HLA-DQ2 or DQ8 class II molecules that bind and present gluten peptides derived from exogenous protein antigens. These antigens are endocytosed by HLA class II positive antigen-presenting cells and degraded in an intracellular endosomal/ lysosomal compartment forming HLA class II-peptide complexes that are recognized by peptide-specific T cells [23]. What follows thereafter is a cascade of inflammation, villous atrophy and triggering of autoimmunity that makes the disease clinically evident.

Over the years, management of CD focused on withdrawal of gluten in diagnosed patients. However recently there has been a paradigm shift in expanding the scope of diagnosing CD early by screening high-risk groups. These high risk groups are: a) non-autoimmune first-degree relatives, selective IgA deficiency, Down syndrome, William’s syndrome and Turner syndrome); and b) autoimmune (type 1 diabetes mellitus, autoimmune thyroiditis and Sjorgen syndrome). The scope has been further enhanced by screening subjects with non- gastrointestinal conditions like dermatitis herpetiformis, delayed puberty, short stature, iron deficiency anemia, osteoporosis, arthritis, ataxia, polyneuropathy, dental enamel hypoplasia and recurrent abortions.

The study being analyzed here is a prospective longitudinal study that sheds light on relationship of HLA typing (both homozygous and heterozygous) and celiac autoimmunity (development of serum tissue transglutaminase antibodies twice at least 3 months apart) and CD (serology plus small bowel biopsy features) [1]. Of 786 subjects with CD autoimmunity, small bowel biopsy was possible in 350 and CD was confirmed in 83% (n=291) children. The major caveat in this robust study is that biopsy could not be performed in 436 subjects with tTG positivity. This may have resulted in underestimation of CD or overestimation as 21 out of 436 subjects were clubbed and presumed to be CD based only on high levels (>100 U) of tTG. It was additionally found that 1% of CD autoimmunity and 2% of diseased individuals developed Type 1 diabetes mellitus on follow-up where the community prevalence of diabetes mellitus is 0.3% [1].

The researchers also found that amongst the 4 countries, Sweden had the highest risk of CD or CD autoimmunity, almost double that of USA. The authors have attributed this phenomenon to non-genetic, dietary factors. The interplay between breastfeeding and early introduction of wheat during supplementation has been projected as crucial and complex. In an unpublished study the authors cited, the Swedish children were exclusively breastfed for a longer duration (median 4 weeks) as compared to those in USA (median 1 week). However, Swedish children were given gluten-containing cereals at the earlier age as compared to USA [1]. Possibly we could interpret that breastfeeding may be protective. It is prudent to breastfeed children longer and gluten-containing diet may be introduced later during infancy (our view point). Thus the study highlights role of environmental factors besides genetic susceptibility in causation of CD.

In high-risk groups as stated above, screening with HLA may be considered. Those who are HLA predisposed (especially HLA DQ2 homozygous) must be followed up with tTG serology screening as HLA-positivity does not mean CD. Negative tTG must be coupled with serum IgA to rule out IgA deficiency (level <5 mg/dL). Those who are tTG positive or IgA deficient must undergo small bowel biopsy. In Indian situations, cost is a major consideration for HLA typing as a first line screening among high risk groups. Therefore the option of either HLA first followed by serial tTG strategy or only serial serology testing on follow-up are the alternatives available [24]. Small bowel biopsy even today for diagnosing CD seems to be mandatory in India (authors’ viewpoint). Breastfeeding needs to be re-emphasized and awareness further strengthened by various campaigns.

Nevertheless, this study by Liu, et al. has knocked the door that opens into a whole new frontier of CD. It is time we take lessons, ponder upon and surge ahead!

Majority of patients with Celiac disease (CD) worldwide possess HLA-DR3-DQ2 haplotype while a few carry DR4-DQ8 or others. This paper by Liu, et al. [1], describes a well-planned prospective cohort study on children CD that has re-established the importance of HLA-DQ2 as a crucial risk factor both for development of CD autoimmunity as well as phenotypic manifestation of disease at an early age. This study has also affirmed that presence of double copies of HLA-DQ2 offers a greater risk for both the above outcomes than a single copy of DQ2 and is also associated with the earliest onset.

HLA-DQ2 association has a high negative predictive value and its absence makes the possibility of CD highly unlikely. Among various HLA-DQ2 heterodimers, DQ2.5 (DQA1*05:01+DQB1*02:01) confers the greatest risk, particularly in homozygosity; followed by DQ2.2 (DQA1*02:01 + DQB1*02:02) whether in heterozygous condition with DQ2.5 or in homozygosity. The authors studied presence of such HLA class II haplotypes amongst children of various origins (USA, Finland, Germany and Sweden) and shown that cumulative risk of CDA and CD was highest among children with DR3-DQ2/ DR3-DQ2 homozygotes followed by DR3-DQ2/DR4-DQ8 and DR4-DQ8/DR8-DQ4 haplotypes in descending order. The risk of CD autoimmunity was higher among children who had first degree relations with CD than with type 1 diabetes. The female predominance of CD autoimmunity was also evident in this study. Further, it is reported that risk of CD autoimmunity in Sweden was almost twice that of USA and that it developed at an earlier age despite haplotype matching. The authors attribute this difference to multiple environmental factors including interplay between breastfeeding and gluten exposure, age at which gluten was introduced in child’s diet and the probable type of infections, all of which are important in development of CD. Based on the above, the authors recommend initiation of screening for CD in at-risk children for an early diagnosis and the need to further investigate the intricate relationships between genetic ad environmental factors that may modulate development of CD in early childhood.

The DR3-DQ2 genomic segment of MHC is most commonly found as an integral component of the ancestral MHC haplotype AH8.1 i.e. the classical HLA-A1-B8-DR3 haplotype amongst the Caucasians. The extended DR3-DQ2 haplotypes in the Indian population are in contrast multiple and very different from the classical Caucasian AH8.1. The Indian DR3-DQ2 haplotypes are unique at numerous intermittent loci, are associated with CD and other autoimmune diseases. The significance of evolutionary divergence of such autoimmunogenic DR3-DQ2 haplotypes in the Indian population is not well understood. The possibility of association of different forms of DR3-DQ2 haplotypes with different disease variants of CD or T1D also remains to be determined. It is still an open question whether such haplotypes could be differently correlated with spectral phenotypes of CD i.e. GI symptomatic vs atypical symptoms vs no symptoms (Silent CD) or with no duodenal pathology (Latent CD) forms. Incidentally, HLA-DQ2 is the same genetic marker that is considered to be an important marker of un-responsiveness to HBV vaccine among the Caucasians but there is paucity of data on Indian population and it is not known if HBV response is compromised among Indian CD patients. Hence, MHC haplotypic associations need further elucidation particularly in conjunction with different disease forms in the Indian population and in comparison with analogous counterparts in other populations.

1. Liu E, Lee H, Aronsson CA, Hagopian WA, Koletzko S, Rewers MJ, et al. Risk of pediatric celiac disease according to HLA haplotype and country. New Engl J Med. 2014; 371:42-49.

2. Hagopian WA, Erlich H, Lernmark A. The Environmental Determinants of Diabetes in the Young (TEDY): Genetic criteria and international diabetes risk screening of 421000 infants. Pediatr Diabetes. 2011;12:733-43.

3. Murch S, Jenkins H, Auth M, Bremner R, Butt A, France S, et al. Joint BSPGHAN and Coeliac UK guidelines for the diagnosis and management of coeliac disease in children. Arch Dis Child. 2013;98:806-11.

4. Makharia GK, Verma AK, Amarchand R, Bhatnagar S, Das P, Goswami A, et al. Prevalence of celiac disease in the northern part of India: A community based study. J Gastroenterol Hepatol 2011;26:894-900.

5. Dube C, Rostom A, Sy R, Cranney A, Saloojee N, Garritty C, et al. The prevalence of celiac disease in average-risk and at-risk Western European populations: A systematic review. Gastroenterology 2005;128:S57-67.

6. Holmes GK. Screening for coeliac disease in type 1 diabetes. Arch Dis Child. 2002;87:495-8.

7. Atherton R, Ross A, Jessop F, Williams R, Heuschkel R, Zilbauer M. Coeliac disease in children with type 1 diabetes - Are current guidelines proving difficult to implement in practice? J Pediatr Gastroenterol Nutr 2014 Jul 23. [Epub ahead of print].

8. Rubio-Tapia A, Van Dyke CT, Lahr BD, Zinsmeister AR, El-Youssef M, Moore SB, et al. Predictors of family risk for celiac disease: A population based study. Clin Gastroenterol Hepatol. 2008; 6: 983-7.

9. Sollid LM, Thorsby E. The primary association of celiac disease to a given HLA-DQ alpha/beta heterodimer explains the divergent HLA-DR associations observed in various Caucasian populations. Tissue Antigens. 1990;36:136-7.

10. Sollid LM, Markussen G, Ek J, Gjerde H, Vartdal F, Thorsby E. Evidence for a primary association of celiac disease to a particular HLA-DQ alpha/beta heterodimer. J Exp Med. 1989;169:345-50.

11. Hadithi M, von Blomberg BM, Crusius JB, Bloemena E, Kostense PJ, Meijer JW, et al. Accuracy of serologic tests and HLA-DQ typing for diagnosing celiac disease. Ann Intern Med. 2007;147:294-302.

12. Husby S, Koletzko S, Korponay-Szabó IR, Mearin ML, Phillips A, Shamir R, et al. European Society for Pediatric Gastroenterology, Hepatology, and Nutrition guidelines for the diagnosis of coeliac disease. J Pediatr Gastroenterol Nutr. 2012;54:136-60.

13. Rubio-Tapia A, Hill ID, Kelly CP, Calderwood AH, Murray JA; American College of Gastroenterology. ACG clinical guidelines: Diagnosis and management of celiac disease. Am J Gastroenterol. 2013;108:656-76.

14. Klapp G, Masip E, Bolonio M, Donat E, Polo B, Ramos D, et al. Celiac disease: The new proposed ESPGHAN diagnostic criteria do work well in a selected population. J Pediatr Gastroenterol Nutr. 2013;56:251-6.

15. Yachha SK, Poddar U. Celiac disease in India. Indian J Gastroenterol. 2007;26:230-7.

16. Kaur G, Sarkar N, Bhatnagar S, Kumar S, Rapthap CC, Bhan MK, et al. Pediatric celiac disease in India is associated with multiple DR3-DQ2 haplotypes. Hum Immunol. 2002;63:677-82.

17. No authors listed. Critical Appraisal Skills Programme. 12 questions to help you make sense of a cohort study. Available from: http://www.caspinternational.org/mod_ product/uploads/CASP_Cohort_Studies_Checklist_ 14.10.10.pdf. Accessed August 16, 2014.

18. Senapati S, Gutierrez-Achury J, Sood A, Midha V, Szperl A, Romanos J, et al. Evaluation of European coeliac disease risk variants in a north Indian population. Eur J Hum Genet. 2014 Jul 23. doi: 10.1038/ejhg.2014.137. [Epub ahead of print]

19. Kumar N, Kaur G, Tandon N, Kanga U, Mehra NK. Genomic evaluation of HLA-DR3+ haplotypes associated with type 1 diabetes. Ann N Y Acad Sci. 2013;1283:91-6.

20. Kaur G, Kumar N, Szilagyi A, Blasko B, Fust G, Rajczy K, et al. Autoimmune-associated HLA-B8-DR3 haplotypes in Asian Indians are unique in C4 complement gene copy numbers and HSP-2 1267A/G. Hum Immunol. 2008;69:580-7.

21. Olivares M, Neef A, Castillejo G, Palma GD, Varea V, Capilla A, et al. The HLA-DQ2 genotype selects for early intestinal microbiota composition in infants at high risk of developing coeliac disease. Gut. 2014;Jun 17:doi: 10.1136/gutjnl-2014-306931. [Epub ahead of print]

22. Murray JA, Moore, SB, Van Dyke CT, Lahr BD, Dierkhising RA, Zinsmeister AR, et al. HLA DQ gene dosage and risk and severity of celiac disease. Clin Gastroenterol Hepatol. 2007;5:1406-12.

23. Van de Wal Y, Kooy Y, Van Veelen P, Vader W, Koning F, Peña S. Coeliac disease: it takes three to tango! Gut. 2000;46:734-7.

24. Srivastava A, Yachha SK, Mathias A, Parveen F, Poddar U, Agrawal S. Prevalence, human leukocyte antigen typing and strategy for screening among Asian first-degree relatives of children with celiac disease. J Gastroenterol Hepatol. 2010;25:319-24.