Serum TSH as a thyroid cancer marker

TOPIC: Thyroid nodules

Title: Higher serum thyroid stimulating hormone level in thyroid nodule patients is associated with greater risks of differentiated thyroid cancer and avanced tumor stage.

Authors: Haymart MR, Repplinger DJ, Leverson GE, Elson DF, Sippel RS, Jaume JC, & Chen H.

Reference: Journal of Clinical Endocrinology & Metabolism 93: 809-814, 2008



Thyroid nodules are very common, but malignancy is identified in only 5-10% of patients with thyroid nodules. Clinical, radiological, pathological & biochemical markers of malignancy are needed to distinguish benign from malignant thyroid nodules to avoid surgery in the majority of patients with benign thyroid nodules. Fortunately, fine needle aspiration (FNA) can reliably distinguish benign from malignant thyroid nodules in 70-90% of patients. Unfortunately, between 10-30% of FNA are inconclusive leading to potentially unnecessary surgeries and the search for complementary markers of malignancies. Moreover, an optimal presurgical test would not only help distinguish between benign and malignant thyroid nodules, but would provide insight into stage and aggressiveness of disease. Interestingly, we may have been using such a test since the 1970s without realizing its potential utility.


The primary aim of the study was to correlate a preoperative TSH level with the diagnosis of differentiated thyroid cancer. A secondary aim was to correlate the preoperative TSH with the stage of disease in patients with differentiated thyroid cancer (DTC).

Main Outcome Measures

Preoperative serum TSH was compared with final thyroid nodule diagnosis at histopathology and stage of disease in patients with DTC.


The likelihood of malignancy increased with increasing preoperative TSH (TSH <0.06 mU/L: 16% malignancy; TSH 0.40-1.39 mU/L: 25%; TSH 1.40-4.99 mU/L: 35%; and TSH >5.0 mU/L: 52%). The mean TSH was higher in patients with advanced stage (stage III/IV) disease at 4.9 mU/L, compared with lower stage (stage I/II) disease at 2.1 mU/L.


Preoperative serum TSH is independently associated with malignancy in patients with thyroid nodules as a continuous variable and divided into ranges in both univariate and multivariate analyses.


Fine-needle aspiration (FNA) is the cornerstone in the evaluation of patients with thyroid nodules, but it is not a perfect test. Addressing any clinical problem, we utilize multiple methods to determine diagnosis, prognosis and plan of management. In the evaluation of patients with thyroid nodules, we rely on clinical evaluation (family history, radiation history, symptoms, physical examination). Several recent studies have led to the recommendation that all patients with thyroid nodules should undergo ultrasonographic examination to determine if the palpable nodule is in fact a thyroid nodule, if there are other non-palpable nodules and to evaluate nodules for suspicious ultrasonographic features (shape, echo characteristics, borders, blood flow) (see Cooper et al., Thyroid 16:109, 2006).

Despite these preoperative clinical, radiological and cytological tools, we are left with the diagnosis of an indeterminate nodule in some patients. Many investigators have searched for biomarkers that would distinguish benign from malignant nodules. A biomarker that could further distinguish low risk from high risk thyroid cancer would even be better. Some of these molecular markers are currently available at certain institutions (BRAF and Ras mutations, galectin-3 immuno-staining) and others are still being developed. Current investigations are using proteomics and genomics to identify relevant serum markers to help distinguish benign from malignant thyroid nodules.

Amazingly, there may be a simple serum marker to add to the clinical, radiological and cytological evaluation of patients with thyroid nodules: serum TSH. The current study by Haymart & coll. sought to address this question in a retrospective cohort analysis of 843 patients who underwent surgery for nodular thyroid disease and had a preoperative serum TSH. The authors showed that a higher serum TSH is associated with malignancy and that this represents an independent risk factor in a multivariate analysis. These data also support previous studies (Kumar H et al. Thyroid 9:1105, 1999; Boelaert K et al. JCEM 91:4295, 2006), thus strengthening the argument that we should use a preoperative serum TSH together with ultrasonographic features of thyroid nodules and FNA cytology as part of our assessment of patients with thyroid nodules.

A more clinically relevant subgroup to study is patients with indeterminate FNA cytology. Unfortunately, the current study did not analyze this subgroup, but did examine a small subgroup of patients with suspicious biopsies and showed a trend of a higher TSH in those who had malignant nodules (TSH: 3.7 mU/L) versus those who ultimately had benign nodules (TSH: 1.4 mU/L). The differences were not significant, likely due to the small number of patients with suspicious FNA cytology (N=18). The earlier large study by Boelaert et al. did show a significant association between serum TSH and malignancy in a large subgroup (N=235) of patients with indeterminate FNA cytology.

The present study does extend previous studies by examining 2 important patient subgroups, not previously reported. Patients with thyroid nodules less than 1 cm (N=128) who had surgery were analyzed and the same relationship between higher TSH and risk of malignancy was again observed: TSH < 0.06 mU/L: 25% malignant, compared with TSH 0.4-1.39 mU/L: 50% malignant, and compared with TSH 2.5-4.99 mU/L: 61% malignant. One limitation of the present study was that 50% of the patients had a unilateral resection, so the entire gland was examined in only 50% of the patients. Finally, this study examined the relationship of serum TSH and tumor stage in a subset of 239 patients with DTC. There was a significantly increased mean TSH between patients with ultimately benign thyroid nodules (TSH 1.4 mU/L), those with stage I/II DTC (TSH 2.1 mU/L) and those with stage III/IV DTC (TSH 4.9 mU/L). Papillary thyroid cancer comprised 87% of those with thyroid cancer in this cohort. These data are interesting in light of studies showing an association of the outcome and TSH suppression in patients with stage III/IV (Cooper DS et al. Thyroid 8:737, 1998) and more recently in patients with stage II DTC (Jonklass J et al. Thyroid 16:1229, 2006). This study further supports the measurement of serum TSH in patients with thyroid nodules, which has been recommended and done for many years. What we are now learning is that higher TSH appears to be independently associated with increased risk of malignancy, even in patients with small nodules, and higher TSH correlates with advanced stage DTC. This new information can help guide consideration of FNA for patients with borderline nodule size (<1-1.5 cm) and recommendations for surgery in some patients with indeterminate FNA cytology. Summary and commentary prepared by Bryan Haugen (Related to Chapter 18 of TDM)