Is There a Role for Levothyroxine Therapy in Euthyroid Patients on Active Surveillance for Papillary Microcarcinoma?

Over the past 10 years, consensus has developed regarding proper patient selection, surveillance strategy, and indications that warrant consideration of transition to surgical intervention in patients with papillary microcarcinoma (PMC) on active surveillance (AS). ^1,2 However, indications for levothyroxine (LT4) therapy during AS are not well defined.

Yamamoto et al., in this issue of Thyroid, present very interesting data from the Kuma Hospital experience exploring the impact of LT4 therapy on clinical outcomes in an updated cohort of 2509 patients with PMC followed with AS. ³ Compared with patients not on LT4 therapy at diagnosis, patients on LT4 therapy at the time of AS initiation had a statistically significant lower mean time-weighted detailed thyrotropin (TSH) score, lower rate of conversion to surgery, and smaller tumor volume doubling rate. While showing trends suggestive of benefit, no statistically significant differences were seen regarding change in tumor size, or cumulative incidence of disease progression during AS between the two groups.

While the time-weighted detailed TSH score was statistically different between the two groups (2.98 ± 0.24 in patients on LT4 at the start of AS vs. 3.10 ± 0.19 for patients not on LT4 therapy at the initiation of AS, p < 0.01), the clinical significance of such a small difference is questionable as both groups have TSH scores that largely reflected serum TSH values in the lower half of the normal reference range.

Thus, it is reasonable to ask whether LT4 therapy may be a surrogate marker of other important clinical factors that could impact clinical outcomes during AS. For example, since underlying chronic lymphocytic thyroiditis has been shown in some studies to correlate with improved clinical outcomes in differentiated thyroid cancer, ⁴ it is possible that the trend toward better outcomes shown in patients taking LT4 during AS could be related to differences in the immune response or tumor microenvironment in addition to the effect of LT4 on TSH levels.

Conversely, the improved outcomes in the LT4 therapy group may have been positively impacted by the practice patterns of the experienced Kuma Hospital clinicians to initiate LT4 therapy in euthyroid patients with high normal TSH values at the time of AS initiation. In the absence of a randomized placebo control group, it is difficult to know whether the improved clinical outcomes were related to the LT4 therapy, time-weighted TSH values, proper selection of patients for initiation of LT4 therapy at the start of AS, or some other unmeasured concurrent risk factor.

In our original series, ⁵ patients were followed on AS without LT4 therapy as long as the TSH levels were within the normal reference range. But as additional data were published suggesting that TSH values consistently >3 mIU/L may be associated with increased tumor growth during AS, ^{6 –9} we modified our practice to initiate LT4 therapy as needed to maintain a TSH level ≤3 mIU/L while carefully avoiding TSH values below the normal reference range, ¹⁰ recognizing that the potential risks associated with subclinical hyperthyroidism could easily outweigh the benefits of a slightly lower TSH in patients with low-risk papillary thyroid cancer on AS. ^11,12

The risk of causing inadvertent subclinical hyperthyroidism is not to be minimized as previous studies have documented significant limitations in our ability to consistently use LT4 therapy to achieve specific TSH goals. ^13,14 It is also important to acknowledge that the TSH goals recommended for low-risk thyroid cancer demonstrating an excellent response after thyroid surgery may or may not be applicable to low-risk thyroid cancer being followed with AS, which may be more analogous to a structural incomplete response than to an excellent or indeterminant response to therapy. ¹⁵

We do need to be mindful that any therapeutic intervention (e.g., LT4 therapy, thermal ablation) proposed in patients being followed with AS represents a step away from conservative management and toward more aggressive management options. Since the outcome of patients with PMC is excellent with a traditional AS management strategy, we must exercise caution to assure that any well-meaning more aggressive therapeutic interventions do not cause more harm than good in these patients with very low-risk thyroid cancer.

Perhaps the most intriguing aspects of the Yamamoto article are the descriptions of the tumor volume and tumor volume kinetics after initiation of LT4 therapy at the time of tumor enlargement during AS. While not definitive or statistically significant, the data do suggest that LT4 therapy at the time of tumor progression may have decreased the tumor volume doubling rate and perhaps decreased the size of the nodule in some patients. ³ However, it is important to recognize that as part of the natural history of the disease, PMC tumor growth rates can change over time and a decrease in tumor volume may take years to become apparent. ¹⁰

Thus, in the absence of a randomized placebo control group, we cannot be certain that changes in growth rates and tumor volumes after initiation of LT4 at the time of disease progression represent a therapeutic effect or just the underlying natural history of the PMC in an individual patient. Up to this point, we have not programmatically offered LT4 therapy at the time of tumor progression for patients in our AS program. However, we are having ongoing discussions within our disease management team to determine whether we should reconsider that practice in light of the Yamamoto data. ³

In conclusion, while the data regarding LT4 therapy and TSH goals during observation are incomplete, there are compelling observations suggesting that LT4 may have a role both in primary prevention of disease progression (particularly if the TSH >3 mIU/L at the start of AS) and possibly in secondary prevention of disease progression (initiation of LT4 therapy at the time of disease progression).

However, we do need to keep in mind that changes in clinical practice that are widely adopted before establishment of a strong evidence base, even when based on very plausible mechanistic explanations, often are shown to be ineffective in subsequent randomized clinical trials. ¹⁶ Thus, well-controlled prospective randomized clinical trials are needed to better define the role of LT4 therapy in both the primary and secondary disease progression setting.