Risk of Morbidity in Contemporary Celiac Disease

Nina R Lewis; Geoffrey KT Holmes


Expert Rev Gastroenterol Hepatol. 2010;4(6):767-780. 

In This Article


Diagnosed celiac disease has been traditionally linked with greatly increased risks of lymphoma and other malignancies,[2–4,7,51,53,66–68] although as alluded previously, some earlier studies have been limited in their ability to provide precise estimates of the malignancy risks experienced by people with celiac disease in comparison with the general population with their small sample size, selected nature from specialist celiac centers and ability to adjust for potential confounders. As larger, population-based studies have appeared, more modest although still increased risks are suggested, with the risk of lymphoma increased at least fivefold (HR: 4.8 [95% CI: 2.7–8.5])[7] with a 30% increased risk of any malignancy (HR: 1.3 [95% CI: 1.1–1.6][69]) in people with celiac disease in comparison with the general population.[2,3,7,66,70–72] These larger studies have also supported an increased risk of small intestinal adenocarcinoma, esophageal cancer and oropharyngeal cancer in diagnosed celiac disease. Although there is evidence that a gluten-free diet will reduce the lymphoma risk,[73] for these malignancies there is no evidence at present that treatment with a gluten-free diet will reduce their occurrence. Reasons for the apparent reduced risk of breast and lung cancers in celiac disease are unclear.[2–7] A population-based survey of 7416 women with celiac disease suggests that there are potentially adverse as well as favorable breast cancer risk profile features in celiac disease in comparison with the general population.[74] With BMI and height similar to the general population, this study[74] suggests that anthropometric exposures may not explain the apparent reduced risk of breast cancer in women with celiac disease.[75]

Most of the malignancies occurred shortly before or after celiac disease was detected, and a reduced malignancy risk was observed after excluding malignant events within the first few years of diagnosis.[3,7] This suggests that the excess risks of malignancy is probably, in part, due to the more detailed investigation (and thus increased likelihood of detection of occult or overt malignancy) of symptoms at presentation of celiac disease. Conversely, celiac disease is more likely to be detected during the investigation for cancer. It is important to note that lymphoproliferative malignancy is an infrequent complication of celiac disease when considering absolute risks. For example, the risk beyond 1 year after a diagnosis of celiac disease of any lymphoproliferative malignancy is approximately one in 1200 person-years.[7] Alternatively, approximately 20 small bowel lymphomas, 12 small bowel adenocarcinomas and 12 esophageal carcinomas would be encountered each year in the whole celiac population of the UK.[76] This information can be used to reassure patients that there is a very low chance they will develop cancer.

Malignancy in Undetected Celiac Disease

Concerns have remained as to a possible increased risk of malignancy in people with undetected celiac disease, which has been used to argue in favor of screening for celiac disease. Screening for undetected celiac disease is dependent on the performance of the serological testing used,[9] with poor specificity acting as a greater threat to validity in epidemiological screening studies, since a test with 98% specificity will have a 2% false-positive rate that would equal the number of true positives if disease prevalence was 2% and sensitivity was 100%. Three European case–control studies of similar study design have attempted to assess the risk of non-Hodgkin's lymphoma (NHL) in undetected celiac disease.[66,77,78] A total of 2397 patients with unselected NHL were included as cases with the presence of celiac disease in cases and controls determined by endomysial testing. While the smallest of the three investigations was unable to find an increased risk,[77] the largest observed a threefold increased risk of NHL confined to celiac disease diagnosed clinically before the study but not in silent celiac disease detected by screening.[78] In the other study the risk was also threefold, but the diagnosis of celiac disease preceded the onset of NHL in most cases (four out of six).[66] In a recent population-based Finnish cohort study of 8000 people gathered between 1978 and 1980 and observed for 20 years with reliance on the national cancer registry for detection of cancers (99% complete for incident cancers), no increase in risk of malignancy was observed in tissue transglutaminase-positive (n = 202; relative risk: 0.91 [95% CI: 0.60–1.37]) nor endomysial antibody-positive (n = 73; relative risk: 0.67 [95% CI: 0.28–1.61]) study participants compared with serology-negative controls.[79] These studies suggest that as the risk of malignancy in undetected celiac disease is not markedly increased over that of the general population, we should not use the fear of malignancy as an argument for early detection or screening for celiac disease.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: