TSH Suppression in Differentiated Thyroid Cancer Patients. Still More Questions than Answers after 30 Years

Treatment and follow-up of patients affected by differentiated thyroid cancer (DTC) has evolved over the last years. The initial single treatment approach for all patients ^1,2 has been replaced by an individualized approach ^3,4 with the intention of avoiding overtreatment and extensive follow-up in the great majority of patients with a good prognosis.

International guidelines have stratified the risk of recurrence of DTC by considering the clinical and pathological findings, as well as immediate postoperative data. ⁴ An initial risk stratification is now considered necessary to direct the extent of surgery and the decision on radioiodine remnant ablation. Furthermore, thyrotropin (TSH) suppression has been tailored to balance risks/benefits of potential DTC progression with adverse effects of exogenous subclinical hyperthyroidism, characterized by low-undetectable serum TSH and FT4 at the higher limit of its normal range during L-thyroxine (LT4) therapy. ^4,5

Although TSH plays a regulatory role in the progression of thyroid cancer through various signaling pathways and interaction with other factors related to progression, the complexity of this mechanism in humans is still unclear and requires further clarification. ⁶ Moreover, the concept of TSH suppression has changed over time because of the increased sensitivity of TSH measurements. Different definitions of TSH suppression contribute to difficulty in interpretating of the results from the numerous studies assessing the postoperative outcomes of DTC. Initial findings from the multicenter National Thyroid Cancer Treatment Cooperative Study Group (NTCCSG) Registry (NTCTCS) showed that in high-risk patients, but not in low-risk patients, LT4 suppressive treatment with undetectable serum TSH was associated with decreased rates of disease recurrence and cancer-related mortality in differentiated thyroid cancer. ⁷ However, when radioiodine (RAI) therapy and the age of the patient were considered in the multivariate analysis, the impact of TSH suppression in high-risk patients was not statistically significant. ⁷ Subnormal serum TSH levels were found to be associated with overall survival (OS) and disease-specific survival (DFS) in patients in Stage 2 compared with those with normal or elevated serum TSH levels in the second prospective study from the NTCCSG Registry on 2,936 patients affected by DTC. ⁸ Furthermore, subnormal to undetectable serum TSH levels were reported to be associated with improved survival in patients with Stage 3 and 4. ⁸ Even when distant metastatic disease was discovered during follow-up, the third NTCTCS Registry analysis found that moderate TSH suppression was associated with significantly improved 1–3 years OS and DFS. ⁹ More aggressive therapy with undetectable serum TSH did not produce any additional improvement at any stage of the disease. ⁹ TSH suppression was not linked to better OS or progression-free survival (PFS) across a mean follow-up of 7.2 years in a subsequent multicenter retrospective cohort study of intermediate and high-risk DTC patients. ¹⁰ These results, however, were limited by the low number of deaths. ¹⁰ Only two trials randomized DTC patients to receive TSH suppressive or replacement LT4 therapy. The first randomized controlled trials (RCT) was designed as a noninferiority study to test the hypothesis that DFS in patients without TSH suppression therapy would not be inferior to that in patients with TSH suppression therapy. ¹¹ A total of 441 patients were randomly assigned to group A (n = 221) or group B (n = 220). Disease-free—5-year survival, disease-specific 5-year survival, overall recurrence rates, and sites of recurrence were not significantly different between groups A and B. The confidence interval [CI] hazard ratio (HR) for recurrence ranged from 0.85 to 1.27 according to Cox proportional hazard model and was within the margin of 2.12 required to declare 10% noninferiority. Results from this study demonstrated the equivalency of not performing TSH suppression therapy compared with performing the therapy. The subsequent randomized controlled study included a small study group (76 patients were randomly assigned to receive postoperative TSH suppression therapy, and 72 patients were assigned to receive non suppression therapy. ¹² No significant difference for DFS (p = 0.09) and OS (p = 0.17) was detected between the two groups. However, for high-risk patients, the change was statistically significant (p = 0.04).

The 2016 American Thyroid Association (ATA) guidelines ⁴ recommended TSH suppression below 0.1 mIU/L only in patients with a higher risk of recurrence after primary treatment. ⁴ There is limited evidence on the benefit of TSH suppression in intermediate risk patients, such that moderate TSH suppression (0.1–0.5 mIU/L) may be used. ⁴ On the contrary, TSH suppression may be avoided in low-risk DTC patients; the lack of benefit of this treatment supports the necessity to protect these patients from the long-term adverse effects of iatrogenic subclinical hyperthyroidism, especially in the context of advanced age and the presence of comorbidities. ⁴ Moreover, the ATA guidelines ⁴ elaborated a dynamic risk strategy based on follow-up data to re-stratify the initial risk of recurrence following surgery. The response to treatment was defined as good, indeterminate, biochemical incomplete, and structural incomplete. ⁴ These guidelines recommend that: patients with excellent response be shifted to a low-normal serum TSH range (0.5–2 mIU/L) during follow-up; patients with structural response continue TSH suppression therapy; and patients with an indeterminate response receive a LT4 dose with a TSH goal of 0.1–0.5 mIU/L. The TSH target during LT4 therapy remains a controversial issue in DTC patients with biochemical incomplete response, in which thyroglobulin (Tg) trends, doubling time Tg levels, and the evaluation of cardiovascular risk factors may help clinicians determine if a more aggressive LT4 treatment might be beneficial.

ATA guideline recommendations on treatment of DTC patients at high and intermediate risk are based respectively on moderate- and low-quality evidence. ⁴ Therefore, Gubbi et al. ¹³ performed a comprehensive meta-analysis to analyze recent data examining the association between TSH suppression and survival in DTC patients defined as intermediate- and high-risk. Based on the 2016 ATA guidelines, ⁴ these patients were treated with total thyroidectomy and radioiodine from 1996 to 2019. ¹³ Nine trials including adult patients (18 years of age and older) with a 5-year follow-up period were chosen for the final assessment, relapse-free survival (RLFS) from four studies, DFS from four studies, and a composite of PFS form one study was considered. Five moderate-quality studies did not favor TSH suppression in intermediate- and high-risk patients, while three low-quality studies did favor TSH suppression. Based on seven studies, the composite outcome of PFS, DFS, and RLFS was not significantly different between the TSH-suppression and non-suppression groups (HR: 0.75; [CI: 0.48–43 1.17]). Similarly, a composite outcome evaluation of OS and DSS based on four trials did not support TSH suppression (HR: 0.69; [CI: 0.31–1.52]). The results did not support TSH suppression, even after a separate analysis of 5-year data from two excellent studies. ¹³

With regards to a secondary composite outcome of cardiac or skeletal adverse events, results were obtained from two retrospective studies ^14,15 and supported a significantly higher risk of developing cardiovascular and skeletal complications in the TSH-suppressed groups (HR: 1.82; [CI: 1.30–2.55]). Significant heterogeneity was observed.

In 2021, a meta-analysis by Lee et al. ¹⁶ included studies of adult patients with DTC and exogenous subclinical hyperthyroidism; these patients were reported to experience increased heart rate, left ventricular mass index, and interventricular septal thickness with impaired diastolic filling, compared with healthy controls. However, some limitations of the studies included in this analysis were retrospective study designs and small sample sizes (<50 patients per group).

Some studies reported a significantly higher risk of atrial fibrillation (AF) in DTC patients, compared with cancer-free controls or with the general population, after adjustment for risk factors. The risk of AF was not associated with TSH levels in some studies. However, in a retrospective study this risk was higher in DTC patients receiving higher doses of LT4. ¹⁷ Moreover, a higher cumulative dose of RAI was correlated with a higher risk of AF, independently of TSH levels. ¹⁸ The meta-analysis by Quiang et al. ¹⁹ found limited evidence of the increased risk of cardiovascular (CV) death and AF after adjustment for confounders. The risk of AF, adjusted for CV risk factors, was 1.66 [CI 1.22–2.27] in DTC survivors compared with the general population, after the evaluation of three studies including 4,428 patients. The risk of ischemic heart disease, stroke, and heart failure was not significantly different between DTC survivors and controls. However, significant heterogeneity was found among the study results. ¹⁹

There are some important limitations of the well performed and updated meta-analysis by Gubbi et al. A small number of studies were available for the evaluation, the bulk of which had a retrospective design. The use of different TSH assays among studies did not allow the authors to evaluate the risks of varying degrees of TSH suppression in younger adults versus older patients.

Future prospective randomized studies are necessary to clarify the utility of TSH suppression therapy according to age, histology of thyroid tumor, and tumor genotype. Such data are needed to clarify in which patients this treatment should be considered. In the meantime, the recommendations on LT4 therapy in DTC patients should not significantly change from those provided by the ATA in 2016. ⁴ The level of suppression should be weighed against the risk of potential side effects. ^{3 –5} Conclusive results on the exact duration of TSH suppression for this treatment to be beneficial, symptoms, quality of life, and adverse effects are needed.

Authors Disclosure Statement

B.B has no relevant conflicts of interest to declare.