Re: “High Thyroid Cancer Mortality Rate in Japan: A Result of Nonaggressive Treatment Strategy,or Just an Aging Population?” By Magner ( Thyroid 2018;28:818–819)

Dear Editor:

We read with interest the letter from Dr. Magner who made several observations about thyroid cancer management and mortality in the United States compared to Japan (1).

Thyroid Cancer Incidence and Mortality in Japan Compared to the United States

Dr. Magner observed that the age-adjusted mortality due to thyroid cancer in the United States is 0.5/100,000 and the Japanese rate is 1.4/100,000. One reason for the difference is that the rates he quoted are age adjusted using each country's own population. Age-adjusted rates are the sum of the age-group mortality rate/100,000 × the proportion of the population in that age group. If the proportion in each age group differs across the countries being compared, the age-adjusted rates cannot be validly compared.

As shown in Table 1, people aged ≥55 years make up just 27% of the U.S. population, while in Japan, they make up 40%. A more appropriate comparison is found in the study by LaVecchia et al. published in 2015 (2). They compared thyroid cancer mortality rates around the world, adjusting each country's rates to the world standard population. The differences between Japan and the United States were still present but were much smaller (men's mortality per 100,000: 0.31 in Japan and 0.25 in the United States; women's mortality per 100,000: 0.39 in Japan and 0.24 in the United States) (2).

Examining mortality rate by age group, Dr. Magner correctly pointed out that mortality due to thyroid cancer in Japan is higher in the ≥75-year age group than it is in the United States. It is also slightly higher in each of the other age groups, with the gap widening with advancing age. To understand why this might be, it is important to know that mortality is a measure of deaths due to a disease in an entire population, not just deaths among those affected with the disease. This means that competing risks of death from other causes can affect mortality rates of the disease of interest.

In the United States, thyroid cancer ranks 29th as a cause of death, far down the list after leading causes such as heart disease, stroke, dementia, pulmonary diseases, and diabetes, among others (3). The average life expectancy in the United States is shorter than it is in Japan, and if some people who might have otherwise died of thyroid cancer die instead of these other illnesses, this will have a tendency to lower the thyroid cancer mortality rate. In Japan, thyroid cancer ranks 20th as a cause of death, and although people live longer, a smaller proportion of Japanese people die from the top causes of death in the United States (3). This might be a reason why mortality due to thyroid cancer is slightly higher in Japan than it is in the United States.

In summary, the differences in the world standard age-adjusted mortality between Japan and the United States are very small: <0.15/100,000. If the difference is real, we think competing causes of death is one potential reason, and this is a more likely explanation than differences in treatment patterns based on the reasons outlined below.

Thyroid Cancer Treatment in Japan Compared to the United States

Dr. Magner wondered whether “less aggressive treatment of middle aged and Japanese patients may be putting older cancer survivors at greater risk of death.” We do not think this is the case. Both the Japanese guidelines and the American Thyroid Association guidelines suggest total thyroidectomy for higher-risk cancers. The Japanese guidelines also recommend prophylactic central neck dissection, which is more aggressive than common U.S. practice (4,5). Additionally, publications from our institutions show papillary thyroid cancer survival rates that are comparable to U.S. data at both the population and the institutional level (6 –11).

Regarding the clinical management differences between Japan and the United States, it is correct that radioactive iodine is used less often in Japan than it is in the United States, and this is because disposal of radioactive materials is very difficult. However, radioactive iodine treatment affects recurrence but has not been shown to improve survival. It is also true that levothyroxine is prescribed less often for patients who have undergone hemithyroidectomy. One of us performed a randomized trial and found that for low-risk cancers, at seven years, no suppression was not inferior to levothyroxine suppression for disease-free survival (12). This study is still ongoing. Currently, in Japan, people with higher-risk cancers are prescribed suppressive doses of levothyroxine.

Treatment Decisions and the “Number Needed to Treat”

Lastly, Dr. Magner suggested there should be a broader discussion of the concept of “number needed to treat” when considering whether a person with a 2.2 cm cancer and two positive lymph nodes should undergo total thyroidectomy and receive radioactive iodine. We agree that treatment harms and benefits as they relate to aggressiveness are worth discussing. In Japan, these discussions are needed to respond to the unprecedented rapid pace of the aging of the population. We think it is important to clarify the benefit of the Japanese conservative treatment strategy for low-risk PTC, and it is crucial to consider not only the rate of mortality, recurrence, or surgical complications, but also the patient-reported outcomes of thyroid-conserving surgery and non-use of radioactive iodine and levothyroxine.

Dr. Magner used a colon cancer screening analogy to express his concerns. We find it confusing to compare screening programs to treatment decisions. Population screening programs are undertaken after consideration of population disease prevalence and disease lethality (among other things), generally with the goal of decreasing population mortality rates. Treatment decisions are made for an individual person based on the balance of a particular treatment's efficacy and harms, in the context of the individual's prognosis, values, and treatment preferences.

The balance between what makes sense at a population level and what should be done at the individual level is an important topic for the development of healthcare policy and clinical guidelines. We hope to see more discussions like this about thyroid cancer treatment aggressiveness, supported by high-quality research data as it emerges.