| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Editorial |
Thyroid Unit Massachusetts General Hospital and Harvard Medical School Boston, Massachusetts 02114
Address all correspondence and requests for reprints to: Douglas S. Ross, Thyroid Unit ACC-730S, Massachusetts General Hospital, Boston, Massachusetts 02114. E-mail: dross{at}partners.org.
Thyroid Nodules and Thyroid Cancer: An Epidemic
There is an epidemic of thyroid nodules and thyroid cancer. This epidemic is probably due to improvements in medical technology and an increased ascertainment of nodular thyroids (1). Thyroid nodules may be detected as an incidental finding on neck and chest computed tomography scans, magnetic resonance images, and carotid Doppler studies. High-resolution ultrasonography detects nodules in more than 50% of women (2). Fine-needle aspiration cytology is the primary and frequently initial tool for assessing the risk of malignancy in thyroid nodules and selecting patients for thyroid surgery (3). At Massachusetts General Hospital, the number of thyroid aspirates processed by the cytology laboratory increased over 3-fold between 1995 and 2005. Although some endocrinologists would like to see the volume of thyroid nodule aspirates equal that of cervical pap smears, it is critical that the endocrine community establish guidelines that select and limit the number of invasive procedures.
This has become a formidable challenge because aggressive evaluation of nodular thyroids has resulted in a 2.4-fold increase in thyroid cancer incidence over the past three decades (1). One could argue that the increase in thyroid cancer detection validates an aggressive approach toward evaluating thyroid nodules, but the biology of thyroid cancer is complex. Micropapillary thyroid cancers (cancers under 10 mm) have been reported as incidental findings in 0.5–13% of autopsies in the United States (4) and up to 35% of autopsies in some European countries, and these cancers are detected in all age groups. If one assumes 6% as the prevalence of micropapillary thyroid cancer in the United States, one calculates that over 17.5 million people in the United States have thyroid cancer. However, the Surveillance, Epidemiology and End Results (SEER) program of The National Cancer Institute reports a prevalence of thyroid cancer in the United States of only 350,000 individuals (5). Thus, despite a 2.4-fold increase in incidence of thyroid cancer, we have still detected only 2% of the cancers present in the population. And half of these excess thyroid cancers are micropapillary cancers (1). Considering the anxiety, costs, and complications suffered by many of these patients, one can reasonably question the benefits of increased cancer detection, and it is clear that we not only need guidelines that select nodules for further evaluation but also need tools that allow us to predict which thyroid cancers will remain indolent, and which thyroid cancers will become aggressive.
Predictors of Thyroid Malignancy
There are a number of well-established predictors of malignancy in thyroid nodules that include hard and fixed lesions on physical examination, rapid growth of nodules, large size, associated hoarseness, dysphagia, or lymphadenopathy. A prior history of low-dose irradiation to the head or neck region in childhood (6), age less than 20 or greater than 70 yr (7), and male sex (7) also increase the risk of malignancy. In many, but not all studies, clinically solitary nodules are more likely to be malignant than nodules in multinodular goiters (7, 8); in a recent study, the risk of malignancy was similar in a patient with a single or multiple nodules, but the risk of malignancy was higher in a solitary nodule than in a nonsolitary nodule (9). Recent studies have concentrated on the ultrasonographic characteristics of thyroid nodules. High-risk nodules are hypoechoic and solid, and in addition have microcalcifications, irregular borders, central blood flow on Doppler imaging, and a high anteroposterior to transverse ratio (10, 11). Using these characteristics to select which nodules require further intervention, one can potentially reduce the number of nodules requiring fine-needle aspiration to less than one third.
In the current issue of JCEM, Boelaert et al. (12) have identified a surprisingly ordinary, yet novel predictor of malignancy in thyroid nodules—the serum TSH concentration. All of their patients had normal serum free T4 and T3 concentrations, but some had subclinical hyperthyroidism or subclinical hypothyroidism. The prevalence of malignancy was 2.8% for patients with serum TSH less than 0.4 mU/liter, 3.7% for those with serum TSH 0.4 to 0.9 mU/liter, 8.3% for patients with serum TSH 1.0 to 1.7 mU/liter, 12.3% for those with serum TSH levels of 1.8 to 5.5 mU/liter, and 29.6% for those whose serum TSH concentrations exceeded 5.5 mU/liter. A positive titer of thyroid peroxidase antibodies was associated with a higher risk of malignancy, but was not identified as an independent predictor of malignancy using binary logistic regression analysis.
Why Should Serum TSH Be a Risk Factor for Malignancy?
Serum TSH is a well-established growth factor for thyroid nodules, and suppression of serum TSH concentrations by administering exogenous thyroid hormone may interfere with the growth of established nodules as well as the formation of new thyroid nodules (13). Animal data demonstrate that TSH suppression in rats exposed to radioiodine prevents the formation of thyroid cancers (14), however there are no data in humans that suggest a protective effect of TSH suppression on oncogenesis. Suppression of serum TSH concentrations by administering exogenous thyroid hormone is associated with reduced recurrence and mortality in patients with thyroid cancer (15). It is therefore possible that the higher rate of malignancy with increasing serum TSH concentrations reflects a trophic effect of TSH on thyroid tissue that promotes neoplasia and carcinogenesis.
Alternatively, lower serum TSH concentrations may be associated with autonomous function in nodules, which has been long recognized as indicative of benign disease (3, 16). Previously, thyroid scintigraphy was used to distinguish autonomous ("hot") from nonfunctioning ("cold") thyroid nodules, to select those nodules that required a biopsy. Scintigraphy has been relegated as a second line test, because it is not as cost-effective as fine-needle aspiration biopsy (90–95% of nodules do not appear to be autonomous) (17). However, using routine scintigraphy, autonomous function can be appreciated only in larger nodules associated with subnormal serum TSH values, which are associated with suppressed iodine uptake in the surrounding normal thyroid tissue. Subtle autonomous function in nodules can be appreciated if one performs thyroid scintigraphy while the patient is taking exogenous thyroid hormone in a dose that results in a subnormal serum TSH concentration (suppression scan) (3). This exercise identifies more nodules with autonomous function than routine scintigraphy and could certainly be used to test the hypothesis that reduced malignancy rates in nodules associated with lower serum TSH concentrations reflect autonomy. Although suppression scans are not widely used, they have been recommended for further assessment of nodules with aspirates found to have a microfollicular pattern (indeterminate cytology or follicular neoplasm) (3). In the study of Boelaert et al. (12), within the subgroup of patients who had nodules with indeterminate cytology, nodules were less likely to be malignant if the serum TSH concentration was low.
Potential Uses of Serum TSH for Predicting Malignancy in Clinical Practice
Boelaert et al. (12) have provided a formula to predict malignancy in thyroid nodules that uses serum TSH, age and gender, and type of goiter. It is notable that their study included patients between 1984 and 2002, ultrasonography was not routinely used in their practice during most of this period, and the studies regarding the use of ultrasonographic criteria to predict malignancy were not yet available. Presumably, most of their patients had palpable thyroid abnormalities, and it must be determined whether serum TSH remains a risk factor for malignancy in a typical contemporary North American endocrine practice where most of the thyroid nodules are not palpable. It is unlikely that a patient with a palpable thyroid nodule over 15 mm will escape fine-needle aspiration regardless of their TSH value. However, serum TSH concentration may prove to be a useful addition to the proposed ultrasonographic criteria to limit the number of nodules under 15 mm that require further evaluation. For example, Papini et al. (10) found for 8- to 15-mm nodules that if they biopsied only solid and hypoechoic nodules that had microcalcifications, irregular borders, or central blood flow, they could detect 87% of the thyroid cancers and biopsy only 31% of the nodules. Cappelli et al. (11) found for 8- to 15-mm nodules that if they biopsied only nodules with an anteroposterior to transverse ratio greater than 1, and additionally had microcalcifications, irregular borders, central blood flow, or were hypoechoic, they could detect 84% of the cancers and biopsy only 16% of the nodules. The current study by Boelaert et al. (12) suggests the possibility that a combination of ultrasonographic criteria and clinical criteria, including the serum TSH concentration, may provide better prediction of malignancy. Additional studies are needed.
Serum TSH concentrations may also be useful for management of some patients with indeterminate cytology. As suggested above, the serum TSH concentration may be indicative of autonomy within a nodule, and might therefore predict the results of a suppression scan. If this were the case, suppression scans might be avoided, especially in elderly patients and other patients at risk for cardiac arrhythmia from serum TSH suppression. Although the majority of patients with predominately microfollicular cytology (follicular neoplasm) require further evaluation with scintigraphy or surgical excision, there is a group of patients with mixed macrofollicular and microfollicular lesions (possible follicular neoplasm) in whom clinical criteria, as well as patient and physician anxiety, help determine whether the nodule is excised or observed. Until further studies that directly address this issue are done, the data of Boelaert et al. indirectly suggest that patients with borderline cytology and higher serum TSH concentrations be treated more aggressively than those with lower serum TSH concentrations.
Received August 14, 2006.
Accepted August 29, 2006.
References
This article has been cited by other articles:
 |
|
 |
|
  J. J. Cronan Thyroid Nodules: Is It Time to Turn Off the US Machines? Radiology, June 1, 2008; 247(3): 602 - 604. [Full Text] [PDF]
|
|
|
|
 |
|
 R. L. Sedjo, T. Byers, E. Barrera Jr., C. Cohen, E. T. H. Fontham, L. A. Newman, C. D. Runowicz, A. G. Thorson, M. J. Thun, E. Ward, et al. A Midpoint Assessment of the American Cancer Society Challenge Goal to Decrease Cancer Incidence by 25% Between 1992 and 2015 CA Cancer J Clin, November 1, 2007; 57(6): 326 - 340. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R Jones, R Spendiff, S Fareedi, and P S Richards The role of ultrasound in the management of nodular thyroid disease Imaging, March 1, 2007; 19(1): 28 - 38. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
