3.5.1 Pathogenesis of nontoxic goitre

‘Goitre’ is a clinical term defined by a thyroid enlargement above the gender- and age-specific reference range (Table 3.5.1.1). Goitre may arise from very different pathological conditions (Table 3.5.1.2) and may present with euthyroid, hyperthyroid, or hypothyroid function. On morphological grounds, a goitre may be diffuse or nodular. This chapter will focus on the pathogenesis of nontoxic goitre, also called simple or dysplastic goitre in the older literature.

Nodular goitre can be divided into solitary nodular and multinodular thyroid disease and constitutes a complex thyroid disorder with heterogeneous morphological functional and pathogenetic properties (1). Histologically, thyroid nodules are distinguished by morphological criteria according to the World Health Organization classification (2). On functional grounds, nodules are classified as either ‘cold’, ‘normal’, or ‘hot’ depending on whether they show decreased, normal, or increased uptake on scintiscan. In contrast to solitary nodular thyroid disease, which has a more uniform clinical, pathological, and molecular picture, multinodular goitre (MNG) usually comprises a mixed group of nodular entities, i.e. one usually finds a combination of hyperfunctional, hypofunctional, or normally functioning thyroid lesions within the same thyroid gland. The overall balance of functional properties of individual thyroid nodules within an MNG ultimately determines the functional status in the individual patient, which may be euthyroidism, subclinical hyperthyroidism, or overt hyperthyroidism. On the molecular level, thyroid nodules within a nodular goitre may represent polyclonal lesions or true monoclonal thyroid neoplasia.

The development of nodular goitre is influenced by extrinsic factors interacting with constitutional parameters of gender and age (1–6). The most important trigger for nodular (and diffuse) goitre is iodine deficiency (3). There is a direct correlation between goitre prevalence and iodine deficiency and vice versa between correction of iodine deficiency and regression of goitre incidence. For instance, iodine deficiency was common in Germany until the early 1990s, when iodized salt was introduced into food industries leading to a marked improvement in nutritional iodine supply as reflected in increased urinary iodine excretion (median 72 μg iodine/l urine in 1994 to 125 μg iodine/l urine in 2003). The use of iodized salt is currently estimated to be above 80% in private households, 70–80% in restaurants, and 35–40% in food industries. In 1994, the prevalence of diffuse goitre was 21% in the age group 18–30 years and 33% in the age group 46–65 years, while in 2002, an impressive reduction in goitre frequency was found with a goitre prevalence of 6% in the 18- to 30-year-olds and of 26% in the 46- to 65-year-olds, in the Papillon study, in which 96 000 German employees were investigated (4). Another recent epidemiological study (SHIP) has underscored this decrease in overall goitre prevalence due to improved iodine supply; thyroid nodules now tend to occur in normal-sized rather than enlarged thyroid glands (5). This may be explained by the thyroid’s inherent disposition to develop focal hyperplasia, discussed below.

Various other goitrogenic factors are known and are relevant to thyroid disease in situations with co-existing iodine deficiency. First, metabolites of various nutrients (e.g. cabbage, cauliflower, and broccoli) may interfere with iodine uptake. Second, industrial pollutants, including resorcinol and phthalic acid, are known to be goitrogenic. Third, deficiencies of selenium, iron, and vitamin A may exacerbate the pathogenic effects of iodine deficiency (3).

Other risk factors for nodular goitre have been suggested, but their putative impact on the prevalence of thyroid nodules occurring in a normal-sized or enlarged thyroid gland is less clear (1, 6). Smoking has been proposed as a risk factor for goitre, and nodules were also found with higher prevalence in goitres of smokers compared with nonsmokers. The impact of smoking on thyroid disease is most likely due to increased thiocyanate levels in smokers exerting a competitive inhibitory effect on iodide uptake. In line with this, the association is more pronounced in areas with iodine deficiency. Radiation is another environmental risk factor not only for thyroid malignancy but also for benign nodular thyroid disease. An increased prevalence of thyroid nodule disease has been associated with exposure to radionuclear fallouts and therapeutic external radiation.

Nodular thyroid disease and goitre are more frequent (2.5-fold to sevenfold) in women but the reasons for this still remain to be clarified. A growth-promoting effect of oestrogens has been described in vitro and oestradiol has been suggested to amplify growth factor-dependent signalling in normal thyroid cells and thyroid tumours. However, pregnancy-related thyroid enlargement appears to be mostly related to iodine deficiency, and in one German study increased MNG prevalence with parity was only observed in women who had not taken iodine supplementation during an earlier pregnancy. Several studies suggest that thyroid volume is also significantly correlated with body mass index. In agreement with this, a recent study has shown that in obese women, weight loss of more than 10% may result in a significant decrease in thyroid volume.

Lastly, because of the cumulative impact of external risk factors on the thyroid gland, the prevalence of thyroid nodular disease increases with age. For example, in a borderline iodine deficiency area, multinodular goitre was present in 23% of the studied population of 2656 Danish people aged 41–71 years and increased with age in women (from 20% to 46%) as well as men (from 7% to 23%) (1).

Thyroid nodules (and goitre) also occur in individuals without exposure to iodine deficiency, and not all individuals in an iodine-deficient region develop goitre. A familial clustering for nodular goitre is well documented and family and twin pair studies in endemic and nonendemic goitre regions have underscored a genetic predisposition for goitre development (1, 5). For example, twin studies show a concordance rate of 80% for monozygotic twins and of 42% for dizygotic twins in endemic regions and of 40–50% and 13% in nonendemic regions, respectively, strongly suggesting interplay between genetic and environmental factors. On the basis of twin studies, the contribution of genetic susceptibility to goitre development has been calculated to be 39% in endemic regions and 82% in a nonendemic area (7).

Genetic defects in enzymes involved in thyroid hormone synthesis (e.g. thyroglobulin (TG), thyroperoxidase (TPO), sodium-iodide symporter (NIS)) typically result in hypothyroid goitres but in some rare cases genetic variations in the TG, TPO, and NIS genes have also been reported in association with a (diffuse or) nodular euthyroid goitre. Furthermore, alterations in the pendrin gene account for the syndromic occurrence of euthyroid nodular goitre and congenital sensorineural hearing loss.

Since these monogenetic defects are exceptionally rare, linkage studies have been performed to identify susceptibility loci for nontoxic goitre on a broader scale (8). A locus on chromosome 14 (termed MNG1 locus) has been identified in a Canadian and a German study and was found to cosegregate with familial nontoxic goitre. In an Italian pedigree with euthyroid goitre, an X-linked autosomal pattern of inheritance with a putative genetic defect in the Xp22 region was suggested. Moreover, in a study by the European Thyroid Association working group on the ‘Genetics of euthyroid goiter’, 18 extended Danish, German, and Slovakian families were analysed in a genome-wide scan. Further putative candidate loci for nontoxic goitre were identified on chromosomes 3p, 2q, 7q, and 8p emphasizing the genetic heterogeneity of euthyroid goitre. However, no germline mutation that cosegregates with goitre in the affected families has been identified to date. Thus, for the majority of euthyroid goitres, a complex multifactorial pathogenesis including interactions between various environmental factors, gender-specific components, and the genetic background has to be assumed.

Development of nodular goitre most likely proceeds in two phases that involve global activation of thyroid epithelial cell proliferation (e.g. as the result of iodine deficiency or other goitrogenic stimuli) leading to hyperplasia and a focal increase of thyroid epithelial cell proliferation causing thyroid nodules. So far, the most common stimulus for focal proliferation is a somatic mutation.

Two driving pathogenetic events have to be considered (Fig. 3.5.1.1): first, iodine deficiency causing an increase in thyroid cell numbers (true hyperplasia), as observed in animal models, and second, H₂O₂ production and free radical formation, which occurs physiologically during thyroid hormone synthesis and may damage genomic DNA. Thus in a mouse model, the spontaneous mutation rate in the naïve thyroid gland has been found to be almost 10 times higher than in other organs (9, 10).

Both processes provide a mutagenic milieu, in which the likelihood of somatic mutations is increased. Whether these somatic mutations lead to thyroid nodular disease critically depends on the affected gene and most likely the environmental selection factors (e.g. iodine deficiency; Fig. 3.5.1.2). A proof of principle for this concept is the evolution of a toxic adenoma from a somatic TSH-receptor mutation (1, see Chapter 3.3.5). Other examples include the origin of papillary thyroid cancer based on BRAF mutations or RET/PTC rearrangements. These somatic mutations have been found already in microscopic lesions of thyroid autonomy and papillary microcarcinoma, respectively. Besides the driving mutation, increased growth factor production and auto- and paracrine action of secreted growth factors (e.g. IGF-1) has been found in monoclonal thyroid tumours and may further propel nodule development. The development of polyclonal thyroid lesions in a nontoxic goitre is less clear and putatively is linked to exogenous factors, e.g. intrathyroidal production of growth factors such as IGF-1, which act on the naturally functional and morphological heterogeneous thyroid follicles (11).

From the epidemiological data discussed above, one might expect an inherent progressive course of nodular thyroid disease. Studies aimed at accurate assessment of the nodules by ultrasonography differ in terms of follow-up period, definition of growth, type of thyroid lesion, and the background in which they are conducted. Moreover, the interobserver variability of long-term studies of nodule volumes is not known. With these caveats in mind, the following observations have been reported (1, 6). In iodine-sufficient areas, nodule ‘growth’ has been reported in 35% of US patients over a follow-up period of 4.9–5.6 years. On long-term follow-up over 15 years in an area of iodine sufficiency, only one-third of benign nodules showed growth as assessed by palpation and ultrasonography, compared with the majority of nodules which remained unchanged or even showed a decrease in size. In Germany, a mean 3-year follow-up of 109 consecutive patients showed a steady and significant (30% volume) increase in nodular size in 50% of patients. In a Danish study, only four (8%) of 45 cold nodules in an area of borderline iodine deficiency showed a change in size (5 mm in diameter), of which only one nodule actually increased and three nodules shrank over a follow-up period of 2 years. Thus, in iodine-deficient and iodine-sufficient settings a varying proportion of nodules will grow, and the speed of growth is highly heterogeneous.