Introduction

Compared to the general population, patients with celiac disease carry an increased burden of disease [1]. In addition to autoimmune diseases [2] and endocrine disorders [3], patients diagnosed with celiac disease have an increased risk of a variety of malignancies that include both T-cell and B-cell non-Hodgkin lymphoma, squamous carcinoma of the esophagus and oropharynx, and small intestinal adenocarcinoma [4–6]. In this report, we document a significantly increased risk of papillary carcinoma of the thyroid in patients with celiac disease.

Methods

We have maintained a database of patients diagnosed with celiac disease who were seen in our Celiac Disease Center since 1981. Data, including age at diagnosis, mode of presentation, adherence to the gluten-free diet, and other diagnoses, was prospectively entered into the database that was anonymized to protect privacy. Among these patients we identified 606, seen between July 1981 and April 2004, who had biopsy-proven celiac disease.

The prevalence of papillary thyroid cancer in our database was compared to data in the Surveillance, Epidemiology and End Point Results (SEER) program of the National Cancer Institute, Washington, D.C. [7]. Information was accessed using Surveillance Research Program, National Cancer Institute SEER*Stat software(available: www.seer.cancer.gov/seerstat) version 6.1.

Patient-years at risk were calculated from the date of diagnosis of celiac disease to either the date of diagnosis of cancer or the date of the most recent follow-up visit, whichever came first. We calculated the standardized morbidity ratio (SMR), ratio of observed to expected and 95% confidence intervals (CI), assuming that the observed number of cancers had a Poisson distribution. Histopathology of thyroid cancer was reviewed. The study was approved by the Institutional Review Board of Columbia University.

Results

The database consisted of 606 patients; 68% were female and the age at diagnosis of celiac disease was 41.5 ± 17.6 years. The prevalence of thyroid disease was 18.4%. The classical diarrhea-predominant presentation of celiac disease was noted in 43%. Three of these patients had papillary thyroid cancer, whereas only 0.133 patients were expected, resulting in a SMR of 22.52 (95% CI 14.90–34.04; P < .001; Table 1). The prevalence for papillary thyroid cancer was 494 per 100,000 compared to 22 per 100,000 of the general population.

Table 1 Results
Full size table

Of the patients with papillary thyroid cancer (Table 2), 2 were female and 1 male; the age at diagnosis of cancer was 44.3 years (range, 38–48 years), and mean duration since diagnosis of celiac disease 7.7 years (range, 1–19.4 years). Each had originally presented with diarrhea-predominant, symptomatic celiac disease and responded to a gluten-free diet. Adherence to the diet was strict as confirmed by an examination of the dietary history by an experienced dietician and negative celiac serologies. None of the patients had a family history of thyroid cancer or a history of exposure to radiation. Each patient presented with a thyroid nodule or mass and was euthyroid, although one had positive thyroid antibodies.

Table 2 Patients With Celiac Disease and Papillary Thyroid Cancer
Full size table

Two patients underwent total thyroidectomy and one hemithyroidectomy. One of the patients had a follicular variant of papillary thyroid carcinoma, 2 patients had multifocal and bilateral tumors, and 2 patients had lymph node involvement. Hashimoto's thyroiditis (as defined histologically because of the presence of diffuse lymphocytic and plasma cell infiltration with formation of lymphoid follicles, damage to the follicular basement membrane, and Hurthle cell change of the thyroid follicular epithelium) was identified in the resected nonmalignant thyroid tissue in only 1 patient, who had the positive thyroid antibodies.

Discussion

We observed a significantly increased risk of papillary thyroid cancer in our cohort of patients with celiac disease compared to United States national SEER data. This occurred despite adherence to a gluten-free diet that is considered protective against the development of malignancies [4]. One of the patients was on the diet for only 1 year.

Thyroid disease, especially Hashimoto's thyroiditis, is common in patients with celiac disease [3]. Papillary thyroid cancer occurs frequently in the setting of Hashimoto's thyroiditis [8, 9]. Because clinically significant papillary thyroid cancer is observed in fewer patients with Hashimoto's thyroiditis than those who harbor precursor lesions [9], it is likely that additional mutagenic events or immune mechanisms might play important roles in papillary thyroid cancer development. Because only 1 of our patients had Hashimoto's thyroiditis, additional mechanisms apart from autoimmune thyroiditis may be important for the development of papillary thyroid cancer in patients with celiac disease. Such mechanisms could include chronic genotoxic stress owing to chronic inflammation mediated by the lymphocytic infiltrates and/or alternatively an as-yet uncharacterized immune deregulation could lead to ineffective tumor surveillance and indirectly lead to an increased incidence of malignancies [10].

Thyroid cancer has not previously been reported as an associated malignancy in series of adult patients with celiac disease [5, 6, 11–15]. However, there were thyroid cancers among a European series of children with celiac disease and malignancies [16]; Freeman [17] has reported a thyroid lymphoma in a patient with celiac disease. Our patients were seen in a specialist referral setting that may have biased the results. In addition, we are not aware of any regional factor that could have influenced the increased occurrence of thyroid cancer in these patients. However, this study should alert clinicians to be aware of an increased risk of papillary thyroid cancer in patients with celiac disease. Examination of the thyroid should be performed as part of the regular physical examination, and thyroid nodules should be evaluated by ultrasonography and biopsy, to exclude thyroid cancer.