Abstract
The Problem
M
And yet, this represents only the tip of the iceberg. If we use 6% as a conservative estimate of the prevalence of MPC in the United States, one would predict over 18 million patients with disease. Yet, the SEER database reports a U.S. prevalence of only half a million cases of all types and sizes of thyroid cancer (
Primum non nocere
These data force us to question whether the identification and eradication of all MPC is a worthwhile goal. Clearly all thyroid cancers are small when they first form. And clearly some thyroid cancers grow, invade, metastasize, and result in death. But despite the increasing identification of MPC, death rates from all thyroid cancers have remained relatively static and are reported by the SEER database to number approximately 1800 patients annually in the United States (
Discussion
In this issue of Thyroid, Ito and colleagues (16) report on their ongoing effort to define the natural history of MPC by observing it. They diagnosed MPC by ultrasound and fine-needle aspiration biopsy and excluded patients with regional lymph node metastases, distant metastases, or evidence of recurrent laryngeal nerve or tracheal invasion or tumors adjacent to the nerve or trachea, and patients whose cytopathology suggested a high-grade malignancy. They now have 1235 patients whose cancers have been observed with ultrasounds once or twice a year for at least 18 months, their average duration of observation is 5 years, with some patients having been observed for as long as 19 years. Among those observed for 10 years, only 8% had MPC that grew by 3 mm or more, and only 3.8% developed nodal metastases. No patient has developed distant metastases or died, and among the 191 patients having surgery after a period of observation, only one patient has had recurrent disease in remnant thyroid tissue after a partial thyroidectomy (and the recurrence is being observed). The authors suggest that “most low-risk papillary microcarcinoma lacking aggressive features are harmless, and immediate surgery for all of them is definitely an overtreatment.”
In a separate cohort of 230 Japanese MPC patients, Sugitani and colleagues (17) have also reported very similar findings demonstrating an increase in size of the primary tumor of 3 mm or more in only 7%, and newly identified lymph node metastases in only 1% during a mean of 5 years of observation.
Don't ask, don't tell
Confirmation of the appropriateness and safety of observation for MPC will be necessary before this approach is widely accepted. However, based on the available published data, the thyroid cancer disease management team at Memorial Sloan-Kettering Cancer Center in New York has been offering an observational approach as routine clinical management in properly selected intrathyroidal MPC for the last several years. Furthermore, it is evident that the authors of the 2009 American Thyroid Association Management Guidelines for Thyroid Nodules and Differentiated Thyroid Cancer (18) had considered some of the earlier publications of Ito and colleagues when they stated: “Routine FNA is not recommended for subcentimetric thyroid nodules.” Biopsy of very small nodules was only recommended in the presence of abnormal lymph nodes or in high-risk patients. Perhaps observation would be better accepted if we deferred from making the diagnosis of MPC by FNA unless the nodule first meets the exclusion criteria established by Ito et al. for observation (16), or unless the nodule grows or becomes associated with nodal metastases during observation.
Prophylactic central node dissection?
One can question the aggressive use of central node dissection for MPC reported by Ito and colleagues. Prophylactic central neck dissection identifies metastatic lymph nodes in up to two thirds of MPC patients (19 –22). Furthermore, small volume metastatic lymph nodes in the lateral neck can also be identified in up to 50% of MPC patients subjected to extensive neck dissections (19 –22). Despite the very high prevalence of microscopic occult lymph node metastases in MPC, loco-regional recurrence rates range from 2% to 6% whether or not prophylactic neck dissections are done, and whether or not radioactive iodine ablation is performed (23). Therefore, we must conclude that occult lymph node metastases occur early and often in the natural history of MPC, yet seldom develop into clinically significant disease.
It is important to note that prophylactic central neck dissection may be associated with significant operative risks. In experienced hands, it appears that prophylactic central neck dissection can be done with an acceptable risk of complications (24,25). And in experienced hands, the complications were similar following central node dissection done for recurrent disease versus prophylactic central node dissection (26). Unfortunately, approximately half of the thyroid operations in the United States are done by low-volume thyroid surgeons, who are very likely to have higher complication rates after a central node dissection than those reported by high-volume thyroid surgeons such as Ito and colleagues (27). In our opinion, it seems unnecessary to subject most MPC patients to a prophylactic central node dissection in an effort to minimize an already very low recurrence rate. A more rational approach would be to dissect the central compartments only in those very few patients who subsequently develop clinical recurrent disease.
Suppression
The current report by Ito and colleagues (16) also found that progression of the MPC is more common in younger than older patients. While they had an insufficient number of patients taking thyrotropin (TSH) suppressive therapy for accurate analysis, they did note that only one such patient demonstrated progression, and hypothesized that TSH suppression may be a “good alternative management” for young patients with MPC.
After decades of use, TSH suppressive therapy has become unpopular and the American Thyroid Association guidelines (18) specifically state: “routine suppressive therapy of benign thyroid nodules in iodine sufficient populations is not recommended.” While the efficacy of suppressive therapy for reducing nodule size is debated (28,29), TSH suppression appears to interfere with the process of goitrogenesis and prevent the formation of new nodules (30,31), and it is associated with reduced recurrence in stage 3 and 4 thyroid cancers (32). However, it is also associated with a threefold increased risk of atrial fibrillation (33) and reduced bone density in postmenopausal women (34). It is possible that “suppressing” TSH levels into the lower portion of the normal range may provide efficacy and ameliorate side effects (35,36). The finding that the incidence of papillary thyroid cancer was lower in Italy among centers that continue to use suppressive therapy is also provocative (37) and raises the question as to whether some of the recent increased prevalence of thyroid cancer could be related to the abandonment of TSH suppressive therapy over the last 20 years. Whether minimally suppressive doses of thyroid hormone have a role in young patients with MPC or patients with suspicious subcentimetric thyroid nodules are issues worthy of additional studies.
Conclusions
Both the series of Ito et al. (16) and Sugitani et al. (17) provide compelling evidence that properly selected patients with MPC can safely avoid immediate thyroid surgery and be followed with an active surveillance management approach. During an observation period of 5–10 years, an increase in size of the primary tumor is anticipated in approximately 5–10% with 2–4% developing clinically identifiable lymph node metastases. Importantly, appropriate therapy at the time of disease progression is associated with excellent clinical outcomes, suggesting that a “delay in treatment” is not associated with any clinically meaningful harm in the few patients that demonstrate disease progression while under observation. Since there appears to be little benefit to the routine early detection and treatment of MPC, we must critically reevaluate our current management paradigms to ensure that we are not causing more harm than good in our aggressive attempts to identify very small volume thyroid cancer. Furthermore, while we currently do not favor TSH suppression during observation or routine prophylactic central neck dissection in the management of MPC, additional studies are needed to address these important management issues. While widespread acceptance of an active surveillance management approach will require additional studies with longer follow-up, the currently available data do allow clinicians to discuss the option of observation rather than immediate surgery in properly selected MPC patients.