Unravelling the rise in thyroid cancer incidence and addressing overdiagnosis

Abstract

Since the 1980s, the global incidence of thyroid cancer has risen sharply while thyroid cancer mortality rates have remained relatively stable. Much of this increase can be attributed to the overdiagnosis of small indolent papillary thyroid cancers, largely driven by the widespread, and sometimes inappropriate, use of imaging studies, especially thyroid and neck ultrasonography. To a lesser extent, increases in the incidence rates of advanced stage thyroid cancer have also contributed to the observed trends. Studies investigating the roles of obesity and environmental exposures have yielded heterogeneous results that suggests there may be a role for managing these modifiable risk factors to reduce the risk of thyroid cancer. In this Review, we describe changes in incidence of thyroid cancer over the past forty years, outline known and potential risk factors, and explore strategies to address the public health issues of overdiagnosis and the subsequent overtreatment of thyroid cancer.

Introduction

Thyroid cancer, the most common endocrine malignancy, is responsible for more than 820,000 cases globally, ranking seventh in cancer incidence according to 2022 data from the International Agency for Research on Cancer GLOBOCAN.^1,2 Although thyroid cancer affects both men and women, the global incidence rate in women is three times higher than in men, with thyroid cancer representing one in every 25 cancers diagnosed among women in 2022 (Figure 1).^1,3 Thyroid cancer has also disproportionately impacted vulnerable patient populations and has been more prominent in some countries (e.g., China, South Korea, and the United States) than others.^2,4 In the United States (US), for example, thyroid cancer is one of the most common cancers among adolescents and young adults.^5,6

Figure 1. The global incidence of thyroid cancer.

Based on data from the International Agency for Research on Cancer GLOBOCAN, the incidence rate in women is three times higher than in men, with thyroid cancer representing one in every 25 cancers diagnosed among women.

Between 1980 and 2017, global thyroid cancer incidence rates increased while mortality rates remained relatively stable.^2,7,8 In some countries (e.g., Austria, Canada, France, Ireland, Israel, Italy, South Korea, and the US), the incidence of thyroid cancer exhibited an upward trend until the early 2010s, after which it began to decline.⁸ Studies have consistently attributed the rapid rise in incidence of thyroid cancer primarily to increased detection of small low-risk papillary thyroid cancers that was driven by the widespread use of thyroid and neck ultrasonography and other imaging modalities.^2,9–15 In a study utilizing the Cancer Incidence in Five Continents database to examine population-based thyroid cancer incidence rates for 26 countries, Li et al found that overdiagnosis accounted for 34% to 93% of all patients diagnosed with thyroid cancer between 2008 and 2012, with the high between-country variability likely reflective of local medical practices.¹⁶

While the overdiagnosis of small indolent thyroid cancers accounts for much of the global increase in thyroid cancer incidence, it is not solely responsible for the rising trends (Figure 2). Studies using large registry data have demonstrated that alongside the increase in low-risk thyroid cancers, there is a true increase in the incidence of larger tumors that could not be entirely explained by overdiagnosis. For instance, in a study of thyroid cancer cases ascertained from the US SEER cancer registry, Lim et al found that there was a significant, though less rapid, increase in both the incidence (3.5% per year since 1981) and mortality rates of advanced-stage papillary thyroid cancer (2.9% per year between 1994–2013).¹⁰ Similarly, Pandeya et al found that the age-standardized incidence rates of stage III/IV differentiated thyroid cancer in Queensland, Australia increased from 0.54 (1982–1986) to 1.18 (2002–2008).¹⁷ The authors also performed a sensitivity analysis with only clinical stage IV cancers to account for the possibility that the observed rise in stage III thyroid cancer was due to more frequent performance of prophylactic neck dissections in the latter years. This yielded similar trends of increased incidence of stage IV thyroid cancer between 1982 and 2006.¹⁷ Although larger tumors can occasionally be identified incidentally, most are not. In a retrospective analysis of 225 Italian patients who had undergone surgery for papillary thyroid cancer, Cosme et al found that only 11.5% (N=3 of 26) of the tumors measuring greater than 4.0 centimeter in size were discovered incidentally while the remaining cases were identified through evaluation of palpable or visible thyroid nodules or for compressive symptoms.¹⁸ Thus, concurrent with an overwhelming increase in small low-risk thyroid cancers is a real, albeit smaller, increase in the incidence of larger and more advanced thyroid cancers that cannot be explained by the widespread use of more sensitive imaging studies or more extensive thyroid surgeries. However, the exact cause of this increase in thyroid cancer incidence remains poorly understood, with exposure to ionizing radiation the only well-established modifiable risk factor. As discussed in the below section on “potential thyroid cancer risk factors”, recent studies suggest that obesity and environmental exposures may play a role.

Figure 2. Factors contributing to global trends in the incidence of thyroid cancer.

The overdiagnosis of small indolent thyroid cancers, largely driven by the widespread and sometimes inappropriate use of imaging studies, especially thyroid and neck ultrasonography, accounts for much of the global increase in thyroid cancer incidence. To a lesser extent, increases in the incidence rates of advanced stage thyroid cancer have also contributed to the observed trends. The exact cause of thyroid cancer remains poorly understood, with exposure to ionizing radiation the only well-established modifiable risk factor.

The Relationship Between Radiation Exposure and Risk of Thyroid Cancer

Radiation exposure is the only well-established modifiable risk factor for thyroid cancer. On the largest scale, radiation exposure from nuclear events have consistently been found to be associated with an increased risk of thyroid cancer. A notable example is the 1945 atomic bombings of Hiroshima and Nagasaki, which exposed more than 100,000 civilians to external whole-body radiation from high-energy gamma rays.¹⁹ In a study of 4,091 survivors, Imaizumi et al observed a significant linear dose-response relationship between radiation exposure and the prevalence of thyroid cancer (Table 1).²⁰ The increase in relative risk of developing thyroid cancer as a result of radiation exposure from the atomic bombings extended from childhood up to at least 50 years of age at the time of exposure.²¹ Another example comes from the 1986 explosion at the Chernobyl nuclear power plant, which exposed millions of individuals to massive amounts of radioactive isotopes of iodine. Since 1990, a significant increase in the incidence of papillary thyroid cancer was observed, particularly among radiation-exposed children, in the surrounding areas of Ukraine, Belarus, and the Russian Federation.^22–25 Researchers hypothesize that much of the increased incidence of thyroid cancer was due to radiation-induced deoxyribonucleic acid (DNA) damage with downstream effects on genes involved in thyroid cell growth and differentiation. In a study that examined the genetic profile of papillary thyroid cancers from 359 individuals exposed as children to the Chernobyl radioactive fallout, Morton et al demonstrated a radiation dose-dependent increase in clonal DNA mutations, with the most common mutated genes being BRAF, RAS, and RET.²² Moreover, it is well-documented that children and adolescents exposed to the Chernobyl radioactive fallout have a sizeable dose-related increase in thyroid cancer, even 10–15 years after exposure.^26,27 In contrast, studies examining the risk of thyroid cancer among exposed adults following the Chernobyl nuclear disaster have been more equivocal.²⁸ More recently, the Fukushima nuclear power plant meltdown in 2011 resulted in the release of significant quantities of radioactive iodine isotopes into the atmosphere. However, studies examining the relationship between the Fukushima radiation exposure and thyroid cancer have been less clear, with some researchers hypothesizing that the data is confounded by widespread screening of children and adolescents in Fukushima Prefecture with thyroid ultrasonography following the nuclear accident.^29,30

Table 1.

Characteristics of studies on the relationship between radiation exposure and risk of thyroid cancer.

Authors (year)	Study Participants	Cohort Size	Age of Participants	Main Findings
Exposure to the 1945 atomic bomb in Hiroshima and Nagasaki
Imaizumi M, et al. (2006)20	Hiroshima and Nagasaki atomic bomb survivors	4091 individuals	Mean age: 70 years +/− standard deviation 9 years	There was a significant linear radiation dose-response relationship for thyroid nodules, including malignant tumors. The authors estimated that about 37% of thyroid cancers were associated with radiation exposure.
Parker LN, et al. (1974)21	Individuals who were within 2000 meters from center of the atomic bomb in Hiroshima or Nagasaki, and who had experienced acute symptoms of radiation exposure	74 individuals with thyroid cancer were age- and sex-matched with a symptomfree proximal group, a more distally exposed group, and a group of individuals who were not in either city at the time of the bombing	Age range (in 1945: Less than 10 to older than 50 years old	Clinically detected thyroid cancer is more prevalent in individuals who were exposed to atomic radiation more than 25 years ago, particularly women and those aged <20 years at time of exposure, compared to the control groups. An increased risk of thyroid cancer was found up to at least 50 years of age at the time of radiation exposure.
Exposure to the 1986 nuclear power plant explosion in Chernobyl
Morton LM, et al. (2021)22	Individuals from Ukraine who were diagnosed with papillary thyroid cancer	440 individuals (N=359 with estimated childhood 131-I exposure from the Chernobyl accident, and N=81 unexposed children born after 1986)	Age at time of thyroid cancer diagnosis: mean of 28.0 years, range of 10.0–45.6 years	Post-Chernobyl papillary thyroid cancers demonstrate radiation dose-dependent increases in clonal DNA double-strand breaks, which likely represents an early carcinogenic event in thyroid tumorigenesis following radiation exposure.
Pacini F, et al. (1997)23	Individuals from Belarus who were diagnosed with thyoid cancer between May 1986 and December 1995	472 individuals with thyroid cancer from Belarus were compared with 369 age-matched controls with thyroid cancer from Italy or France	Aged <21 years at time of thyroid cancer diagnosis	Between 1986 and 1989, the annual incidence of thyroid cancer increased significantly (from 3–8 in 1986–1989 to >90 in 1993–1995), with children aged </=5 years at the time of the accident accounting for the majority of thyroid cancers. When compared with age-matched controls, the post-Chernobyl Belarus thyroid cancers affected younger children, were almost always papillary, and were more aggressive at presentation.
Shibata Y, et al. (2001)24	Children born between 1/01/1983 and 12/31/1989 who were living within a radius of 150 km from the Chernobyl nuclear power plant	21,601 children	Age at time of screening thyroid ultrasound: 8–17 years	Exposure to the fallout after Chernobyl was significantly associated with the risk of thyroid cancer in children. Compared to children in group I (DOB: 1/01/1987 – 12/31/1989), children in group III (DOB: 1/01/1983 – 4/26/1986) and group II (DOB: 4/27/1986 – 12/31/1986) had a 121 and 11 times greater odds of developing thyroid cancer, respectively.
Kazakov VS, et al. (1992)25	Children residing in Belarus and Minsk City who were diagnosed with thyroid cancer between 1986 and the end of the first half of 1992	131 children	Information not reported	The overall incidence of thyroid cancer in children increased within a few years of exposure to radioactive iodine after the Chernobyl accident. The incidence increased from an average of just two cases per year in 1986 to 55 in 1991.
Zablotska LB, et al. (2011)26	Individuals in Belarus aged </= 18 years at the time of the Chernobyl accident	11,611 individuals (N=87 cases of thyroid cancer and N=11,524 non-cases)	Age at screening: 10–34 years	10–15 years after the Chernobyl accident, the risk of thyroid cancer significantly increased among individuals exposed to the fallout as children or adolescents. There was a significant dose-response relationship between 131-I thyroid dose and thyroid cancer, which was linear-exponential across the full range of doses, but consistent with linearity below 5 gray.
Exposure to the 2011 Fukushima nuclear power plant meltdown
Suzuki S, et al. (2016)29	Fukushima Prefecture residents (at time of nuclear power plant accident) aged ≤18 years (as of 1/01/2011) who underwent thyroid ultrasound screening	300,476 individuals	Average age at time of thyroid ultrasound screening: 10.7 years (standard deviation 5.0 years)	The authors reported a high overall thyroid cancer prevalence of 37.3 per 100,000 in children and adolescents, with no significant differences between evacuated and non-evacuated areas
Tsuda T, et al. (2016)30	Fukushima Prefecture residents aged ≤18 years who underwent thyroid ultrasound screening	298,577 (81% of residents) underwent the first round screening by the end of December 2014	Age </= 18 years on 3/11/2011	The authors identified an excess of thyroid cancers detected by ultrasound among children and adolescents within 4 years of the nuclear power plant accident. The highest incidence rate ratio (605 cases of thyroid cancer per one million) was observed in an area where residents were not even evacuated - this was 2.6 higher that that in Fukushima Prefecture.
Exposure to radiation from medical therapy
Bhatti P, et al. (2010)31	Five-year survivors of a childhood cancer diagnosed between 1970–1986	12,547 individuals	Age at first cancer diagnosis: <5 to >/= 15 years	The risk of thyroid cancer increased linearly with radiation doses up to about 20 gray, at which point the relative risk peaked at 14.6-fold. Sex, age at exposure and time since exposure were significant modifiers of the radiation-related risk of thyroid cancer.
Schneider AB, et al. (1993)32	Patients at a single institution who were treated with radiation therapy for benign conditions of the head and neck (1939–1963)	4,296 individuals (including 2,643 individuals with follow-up information available)	All patient were treated before their 16th birthday. Mean age at first radiation treatment: 4.4 years	Thyroid cancer was rare in the first 5–10 years after radiation exposure, with the majority of cases occurring in the between 20–40 years after exposure. Age at radiation exposure was a significant factor for developing thyroid cancer, with younger age associated with higher risk.
Exposure to radiation from diagnostic imaging
Han MA and Kim JH. (2018)37	Meta-analysis of 9 studies	5,647,109 individuals (including 1,790 cases of thyroid cancer)	Information not reported for overall study cohort	Exposure to diagnostic radiation was associated with a 1.5 times increased risk of thyroid cancer. Specifically, exposure to computed tomography scans and dental x-rays were associated with a 1.46 and 1.69 increased risk of thyroid cancer, respectively.
Kitahara CM, et al. (2018)38	Adult members of the United States radiologic technologists (USRT)	89,897 individuals, including 476 with a diagnosis of thyroid cancer	Median age at time of thyroid cancer diagnosis: 51 years for women and 54 years for men	The authors reported no evidence of an association between cumulative occupational radiation dose to the thyroid and risk of thyroid cancer.
Smith-Bindman R, et al. (2025)40	University of California San Francisco International CT Dose Registry	61,510,000 patients who collectively underwent a total of 93 million computed tomography (CT) exams in 2023	Age range: 0 – 99 years	The authors estimated 103,000 radiation-induced new cancers diagnoses, including 7,000 new thyroid cancers diagnoses, to results from these 93 million CT exams in 2023. The estimated radiation-induced cancer risks were higher in children and adolescents. The most frequently projected cancers in children was thyroid cancer.

On a smaller scale, radiation from medical therapy have been found to increase the risk of thyroid cancer. In a study of 12,547 five-year survivors of childhood cancer who had undergone radiation therapy, Bhatti et al found that the risk of thyroid cancer increased linearly with radiation doses up to about 20 gray, at which point the relative risk peaked at 14.6-fold (Table 1). Additionally, younger age at the time of radiation exposure was associated with a higher radiation-induced risk.³¹ Similar findings were reported by Schneider et al in their study of 2,634 patients who received radiation therapy to the head and neck before their 16^th birthday and were followed for a mean of 33 years. The authors also found that while the increased relative risk of thyroid cancer peaked 25–29 years after radiation exposure, the tumorigenic effects of radiation persisted for at least 40 years.³²

The contribution, if any, of radiation exposure from diagnostic imaging studies on risk of thyroid cancer is still subject to debate. While a single diagnostic exam with CT imaging, for example, exposes patients to relatively low doses of radiation, many individuals undergo multiple scans over the course of their lifetime. Over the past few decades, the use of diagnostic imaging studies has substantially increased in many countries.^33,34 For instance, according to the US National Council on Radiation Protection and Measurements, Americans were exposed to more than seven times as much ionizing radiation from medical procedures in 2006 compared to the early 1980s, largely due to higher utilization of CT and nuclear medicine scans.³⁵ However, the radiation dose affecting the thyroid gland per diagnostic imaging study has markedly decreased from the 1960s to the 1990s and then remained relatively unchanged between 2000 and 2010.³⁶ Given the variability in radiation doses across diagnostic imaging studies and over time, it is not surprising that research examining the relationship between diagnostic radiation exposure and thyroid cancer risk have yielded heterogeneous results (Table 1).^37–39 A 2025 study by Smith-Bindman et al analyzed data from 62 million patients who collectively underwent a total of 93 million CT examinations in 2023 to project the number of future incidence of CT-associated cancers. The authors estimated that, if current radiation dosing and utilization practices continue, CT-associated exposures could account for 102,700 new cancer cases annually in the US, including an estimated 7,000 new cases of thyroid cancer.⁴⁰

Potential Thyroid Cancer Risk Factors

Apart from radiation exposure, thyroid cancer has few known modifiable risk factors. Given that iodine is essential for thyroid hormone production and the prevalence of thyroid goiter and thyroid dysfunction is higher in iodine-deficient populations,⁴¹ several studies have explored the potential role of iodine intake to thyroid cancer development. However, the findings from these studies have been inconsistent.⁴² While studies have been conducted across diverse populations (e.g., in China, Croatia, Denmark, Germany, Italy, Sweden, Thailand, and the US), the ecological design of many of these studies, coupled with the simultaneous rise in use of thyroid and neck ultrasonography, complicates efforts to determine the true impact, if any, of national salt iodination programs on thyroid cancer incidence (Table 2).^43–47 Moreover, there is debate on the length of the latency period required, if any, to observe the effect of iodine supplementation on thyroid cancer incidence.⁴³

Table 2.

Characteristics of studies on potential thyroid cancer risk factors.

Authors (year)	Data Source (Years)	Cohort Size	Main Findings
Role of iodine status and risk of thyroid cancer
Mitro SD, et al. (2016)44	The Songkhla Registry, the Chiang Mai Registry, and the Khon Kaen Registry (1990 – 2009)	2,749 individuals with thyroid cancer	Between 1990 and 2009, the pooled incidence of papillary thyroid cancers (PTC) significantly increased while the incidence of follicular thyroid cancers (FTC) significantly decreased. The largest magnitude of change occurred in historically iodine-deficient regions. The authors interpreted the heterogeneity in thyroid cancer incidence trends by region and histology to be suggestive of an iodination effect.
Blomberg M, et al. (2012)45	Nationwide Danish Cancer Registry (1943 – 2008)	6,629 individuals with thyroid cancer	Between 1943 and 2008, the age-standardized incidence rate of thyroid increased from 0.41 to 1.57 per 100,000 in men and from 0.90 to 4.11 per 100,000 in women. The increase occurred primarily in the last 16 years of the study period, and due almost exclusively to the diagnosis of papillary thyroid cancer.
Pettersson B, et al. (1996)46	Swedish cancer registry (1958 – 1981)	5,838 individuals with thyroid cancer	In iodine-deficient areas, the relative risk of developing thyroid cancer was 0.92 for all histologic types combined, 0.80 for PTC and 0.87 for anaplastic thyroid carcinoma. Residence in iodine-deficient regions was associated with a 1.7 to 2-fold increased risk of FTC. The incidence of both PTC and FTC increased during the study period, with similar trends in iodine-deficient and iodine-sufficient areas.
Poljak NK, et al. (2011)47	Croatian National Cancer Registry (1997 – 2006)	1,149 individuals with thyroid cancer	The mean age-standardized incidence rate of thyroid cancer in Dalmatia was significantly greater than that in Slavonia (9.32 vs 6.2 per 100,000 inhabitants, respectively, P<0.001). PTC accounted for 80.0% of all thyroid cancers in Dalmatia and 63.7% in Slavonia, whereas FTC accounted for 20.9% and 12.4%, respectively.
Relationship between higher body mass index (BMI) and risk of thyroid cancer
Kitahara CM, et al. (2020)50	National Institutes of Health-American Association of Retired Persons (NIH-AARP) Diet and Health Study cohort data (1995 – 2015)	604 incident cases of papillary thyroid cancer, and 456,507 non-cases	Overweight and obesity were associated with a 1.26-fold and 1.30-fold increased risks of PTC, respectively, and nearly 3-fold and more than 5-fold increased risks of large (>4 cm) PTC vs normal weight. The authors estimated that one of every six PTC diagnosed among adults aged >/=60 years in 2015 were attributable to overweight and obesity.
Kwon H, et al. (2019)51	Korean National Health Screening database (2009 – 2012)	11,323,006 adults, including 50,464 individuals with thyroid cancer.	Higher BMI and larger waist circumference were associated with increased risk of thyroid cancer (both P < 0.001). Weight gain in lean subjects was associated with increased risk of thyroid cancer (hazard ratio [HR] 1.15; 95% confidence interval [CI] 1.11–1.19), whereas weight reduction in obese subjects was associated with decreased risk of thyroid cancer (HR 0.89, 95% CI 0.86–0.93).
Fussey JM, et al. (2020)52	UK Biobank (2006 – 2010)	379,708 participants of European ancestry, including 425 with thyroid cancer	The authors did not observe any significant associations between thyroid cancer and higher BMI or higher waist-hip ratio. The Mendelian randomization analysis did not support a causal link for obesity with thyroid cancer.
Kitahara CM, et al. (2016)53	Pooled analysis of 22 prospective cohort studies	2,996 cases of thyroid cancers and 2,091,051 non-cases	Increased risk of thyroid cancer was observed for greater values (per 5 cm or 5 kg/m2) of height (HR 1.07; 95% CI 1.04–1.10), baseline BMI (HR 1.06; 95% CI 1.02–1.10), and waist circumference (HR 1.03; 95% CI 1.01–1.05).
Youssef MR, et al. (2021)54	Meta-analysis of 31 studies	24,489,477 individuals	Normal and underweight were associated with decreased risk of thyroid cancer (Relative risk [RR] 0.68, 95% CI 0.65–0.71 and RR 0.92, 95%CI 0.91–0.93, respectively). In contrast, overweight and obese cohorts were more likely to develop thyroid cancer (RR 1.26, 95% CI 1.24–1.28 and RR 1.50, 95% CI 1.45–1.55, respectively). Weight gain increased the risk of developing thyroid cancer (RR 1.18, 95% CI 1.14–1.22), while weight loss decreased the risk (RR 0.89, 95% CI 0.85–0.93).
Role of environmental exposures on risk of thyroid cancer
Hoffman K, et al. (2017)58	Patients with PTC (diagnosed 4/2014 – 1/2016 at single US institution)	70 patients with PTC and 70 age- and gender-matched controls (from same institution or surrounding hospitals)	Cases of PTC were significantly more likely to have high concentrations of tris(2-chloroethyl) phosphate (TCEP) and decabromodiphenyl ether (BDE-209) in their house dust. Associations between flame retardants and PTC varied by the presence of BRAF V600E mutation, with high exposure generally more strongly related to BRAF V600E(−) tumors.
Omidakhsh N, et al. (2022)59	Thyroid cancer cases: California Cancer Registry (1999 – 2012). Controls: Parkinson's Disease Environment and Gene (PEG) case control study (2001 – 2011)	2,067 cases of thyroid cancer and 1,003 controls	The risk of thyroid cancer increased in an exposure-response fashion with the number of pesticides that subjects were exposed to in the 20-year cumulative exposure window. Larger numbers of pesticides contributed to increasing risk of thyroid cancer (1–9 pesticides: odds ratio [OR] 1.29, 95% CI 1.03–1.61; ≥10 pesticides: OR 1.34, 95% CI 1.08–1.68).

The rising incidence of thyroid cancer has paralleled the global increase in rates of obesity.⁴⁸ Excess body mass index (BMI), which is a measure of general adiposity, has been associated with an increased risk of several non-thyroid cancers.⁴⁹ In this context, numerous studies have examined the relationship between higher BMI and thyroid cancer risk (Table 2).^50–52 In a pooled analysis of 22 cohort studies, Kitahara et al observed a positive association between thyroid cancer incidence and several anthropometric factors, including higher baseline BMI (Hazard ratio [HR] 1.06, 95% confidence interval [CI] 1.02–1.10), greater BMI increases during adulthood (HR 1.07, 95% CI 1.00–1.15), and greater waist circumference (HR 1.03, 95% CI 1.01–1.05).⁵³ Similarly, a meta-analysis that pooled data from a cohort of 24 million individuals found a dose-response relationship between increasing body weight and thyroid cancer risk. Specifically, overweight and obese patients had an significantly higher risk of thyroid cancer (relative risk [RR] 1.26–1.50), whereas underweight and normal weight patients had a significantly lower risk (RR 0.68–0.92).⁵⁴ While these findings support a potential contributory role of obesity in the development of thyroid cancer, several limitations need to be considered. First, while retrospective studies have identified an association between obesity and thyroid cancer, none have established causality. Moreover, the underlying molecular mechanisms linking obesity and thyroid cancer remain poorly understood.⁵⁵ Second, at the population level, overweight and obese individuals are more likely to utilize healthcare resources, which can lead to a higher incidence of thyroid cancer due to overdiagnosis rather than a true increase.⁵⁶

A limited number of studies have explored the role of environmental exposures, including flame retardants and pesticides, in the development of thyroid cancer. While promising, these patient cohorts have often been relatively small and homogeneous (e.g., from a single institution or a specific region). To meet fire safety standards for furniture and electronics, the use of flame retardants has increased over the past several decades. In 2008, the US Environmental Protection Agency (EPA) estimated that more than 80% of the US population’s exposure to polybrominated diphenyl ether (PBDEs), a flame retardant chemical that is structurally similar to thyroid hormones, comes from household dust.⁵⁷ In a study of 70 patients with papillary thyroid cancer, Hoffman et al found that higher levels of some flame retardant components in household dust were associated with increased odds of developing thyroid cancer (Table 2).⁵⁸ In regards to the impact of pesticide use, Omidakhsh et al performed a case-control study of thyroid cancer cases identified though the California Cancer Registry (1999–2012) and found that the risk of thyroid cancer increased proportionately with the total number of pesticides that participants were exposed to over the 20-year period.⁵⁹

The Impact of Screening Ultrasonography on Thyroid Cancer Incidence

The diagnosis of thyroid cancer often begins with the identification of a thyroid nodule, which can be detected through palpation or imaging (Figure 3). While palpation of the thyroid gland is a common clinical practice, it has limited diagnostic accuracy in detecting thyroid nodules, with sensitivity as low as 20% for solitary thyroid nodules.⁶⁰ In comparison, imaging studies, particularly high-resolution ultrasonography, are much more sensitive in detecting thyroid nodules. For example, in a study of 635 German patients with no history of thyroid disease who underwent screening with high-frequency (13 MHz) thyroid ultrasound, Guth et al found thyroid nodules to be present in 68% of the cohort, of which 74% of the thyroid nodules measured less than 1.0 centimeter in size.⁶¹

Figure 3. The diagnostic cascade: the role of thyroid ultrasound in driving overdiagnosis and overtreatment.

The diagnosis of thyroid cancer often begins with the identification of a thyroid nodule, which can be detected through palpation or imaging.

The overuse of thyroid and neck ultrasonography has been a major driver of the overdiagnosis and subsequent overtreatment of thyroid cancer worldwide.^11,62–64 Ultrasonography is occasionally utilized as an initial imaging test to screen for thyroid cancer. A striking example of this phenomenon is seen in South Korea after the government launched a free national cancer screening program in 1999. While thyroid cancer screening was not included in the program, many hospitals offered patients the option to add on a screening thyroid ultrasonography for a nominal fee. Soon thereafter, the incidence of thyroid cancer increased - slowly during the 1990s then rapidly after 2000 - while the thyroid cancer mortality rate remained relatively stable.¹² By 2009, the age-standardized incidence rate of thyroid cancer per 100,000 had increased more than seven-fold, from 6.3 in 1999 to 47.5 in 2009.⁶⁵ A study of ten major hospital in South Korea found that the annual number of thyroid ultrasounds performed nearly doubled between 2001 and 2004 while the number of thyroid nodule FNA performed almost quadrupled during the same period.⁶⁶ In a study that analyzed data from the Cancer Incidence in Five Continents database, Li et al determined that about 93% (N=140,000) of the thyroid cancer cases diagnosed in South Korean women between 2008–2012 were attributable to overdiagnosis.¹⁶ With nearly all patients diagnosed with thyroid cancer undergoing surgical management, the median size of the thyroid tumors decreased from 1.8 cm in 1999 to 0.8 cm in 2008.^12,66 This trend was also observed at the hospital level. At one surgical center in South Korea, the proportion of patients undergoing thyroid surgery for sub-centimeter tumors increased from 14% in 1995 to 56% in 2005.¹²

Similarly, in clinical practice in the US, thyroid ultrasonography has been ordered for clinically unsupported reasons such as patient request, abnormal thyroid function test results, and positive thyroid antibody test results.^67–70 Consequently, this has contributed to the overdiagnosis of small non-palpable indolent thyroid cancers (Figure 3).^13,15,71,72

The Impact of Incidentally Detected Thyroid Nodules on Thyroid Cancer Incidence

Thyroid ultrasonography is standard of care to further evaluate and characterize thyroid nodules that are palpated on physical exam or incidentally discovered on imaging for non-thyroid indications (e.g., chest computed tomography, carotid ultrasonography). Studies examining the rates of incidental thyroid nodules identified on non-thyroid imaging studies have reported figures ranging from 0.7% to 16.8%.^73–75 This variability is partly due to differences in imaging modality examined and variability in reporting practices among radiologists.^76,77

With the increasing use of diagnostic imaging studies, including computed tomography (CT) scans, in medical practice in many countries,³³ the identification of incidental thyroid nodules has become a major contributor to the use of thyroid ultrasonography. In a retrospective review of patients who underwent thyroid ultrasounds between 2015 and 2017 at a single US institution, Jacome et al found that 31.6% of the ultrasonography orders were triggered by incidental findings on a prior imaging study (e.g., neck or chest CT scan).⁷⁰ Furthermore, in a systematic review of 18 studies, conducted between 1991 and 2018, that enrolled a total of 4,668 patients with thyroid cancer, Lincango-Naranjo et al concluded that 49% of patients had their thyroid cancer discovered incidentally. Incidental pathways included the scenario in which the thyroid nodule harboring the thyroid cancer was incidentally discovered on imaging, and the scenario in which the thyroid cancer was found incidentally on histological examination after surgery for a benign thyroid condition (e.g., Graves’ disease).⁶³ Among the patients with incidental thyroid cancers, 76% had tumors measuring less than 1.0 centimeter in size and 35% were detected through imaging, of which ultrasonography was the most common imaging modality (27%).⁶³ Given the increased prevalence of thyroid nodules with age,⁶¹ its impact on older individuals is substantial. In a study using US SEER-Medicare data, which includes a cohort of individuals aged 65 and older, Haymart et al demonstrated that greater thyroid ultrasonography use was associated with increased diagnosis of low-risk thyroid cancer.¹⁵ Moreover, the authors determined that the use of thyroid ultrasonography per 100,000 people enrolled in Medicare increased at a rate of 20.9% per year, which led to at least 6,594 patients aged 65 and older being diagnosed with thyroid cancer between 2002 and 2013.¹⁵

The increasing use of thyroid and neck ultrasonography has been paralleled by an increase in the number of thyroid nodule fine needle aspirations (FNA) and thyroid surgeries performed.⁷⁸ In a US study using private and public insurance claims data, Sosa et al found a 59% increase in the incidence of thyroid cancer between 2006 and 2011. During this five-year study period, the number of thyroid nodule FNA more than doubled (annual percentage change of 21.5%) and thyroid surgeries performed for thyroid nodule(s) increased by 31%, with total thyroidectomies rising 12% annually.⁷⁹ Likewise, an analysis using the Taiwan National Health Insurance Research Database claims data from 2004 to 2010 revealed a more than three-fold increase in the age-standardized rates of thyroid nodule FNA (from 15.0 in 2004 to 67.4 in 2010 per 100,000 person-year), corresponding to a 94.8% increase in the age-standardized annual incidence rate of thyroid cancer (from 4.8 to 9.3 per 100,000 person-year).⁸⁰ Similar trends have been reported in other countries including France and South Korea.^81,82

Potential Magnitude of Thyroid Cancer Overdiagnosis

In a meta-analysis of 35 studies conducted over six decades (1949–2007), Furuya-Kanamori et al found the prevalence of differentiated thyroid cancer in autopsy studies to be 11.2% (the prevalence reported by each of the 35 studies ranged from 1.5% to 35.6%).^83,84 This suggests that many individuals who die of other causes have indolent thyroid cancers. Therefore, if one were to screen all adults, thyroid cancer may be identified in up to one-third of adults.

While research into biological causes for sex difference in thyroid cancer incidence have yielded mixed results,^85–87 numerous studies have identified gender-based differences in health care utilization, including greater use of thyroid and neck ultrasonography in women, which likely contributes to the higher incidence rates of thyroid cancer observed in women.^15,78,88 Using US SEER data (1975–2017) and review of prior autopsy studies, LeClair et al found that the higher rates of thyroid cancer among women compared to men was mostly limited to the detection of small (</=2.0 centimeter) subclinical papillary thyroid cancers. In contrast, the incidence ratio by gender approached 1:1 as the thyroid cancer type lethality increased, and for subclinical papillary thyroid cancers at autopsy.⁸⁹ Furthermore, in a meta-analysis of 2,302 autopsies from 12 pooled study populations, LeClair et al reported a similar prevalence of subclinical papillary thyroid cancer in women (14.0%) and men (10.8%).⁸⁹

The harm in identifying and subsequently evaluating small clinically insignificant thyroid nodules lies, in part, in the risk of diagnosing indolent thyroid cancers, which may trigger a cascade of medical interventions that each carry its own risks.^90–92

Strategies to Address the Overdiagnosis and Overtreatment of Thyroid Cancer

In response to the widespread use of thyroid and neck ultrasonography, many professional societies have developed guidelines and recommendations aimed at minimizing its misuse and the downstream effects of overdiagnosing small indolent thyroid cancers. The US Preventive Services Task Force recommends against screening for thyroid cancer in asymptomatic adults.^92,93 In addition, the Choosing Wisely campaign in the US and in Australia advocated against routine thyroid ultrasonography in patients with abnormal thyroid function tests and no palpable thyroid abnormality.^94,95 While these guidance is helpful, it has not been adequate to curb the inappropriate use of thyroid and neck ultrasonography in current clinical practice. In addition to guidelines advocating against the misuse of thyroid and neck ultrasonography, it is necessary to have clear evidence-based guidelines on the appropriate indications for thyroid and neck ultrasonography.⁹⁶ Furthermore, patient education on the harms of inappropriate thyroid and neck ultrasonography is also important. In a survey of US physicians involved in thyroid cancer care, Chen et al found that 32.7% reported ordering thyroid ultrasound in response to patient request.⁶⁷ In South Korea, in response to the observed thyroid cancer epidemic, a multilevel intervention was implemented that recognized patients as key participants in promoting the use of thyroid and neck ultrasonography. Specifically, there was increased education to the general public on the relationship between use of thyroid ultrasonography and thyroid cancer overdiagnosis. For example, major newspapers ran headlines such as “What caused jump in thyroid cancer cases?”, television broadcasters ran hours-long investigative reports on the thyroid cancer “epidemic”, and medical professionals uploaded videos about the relationship between thyroid ultrasonography use and thyroid cancer to YouTube, which is a popular video platform in South Korea.^97,98 In combination, such efforts contributed to a 35% decrease in the number of thyroid surgeries performed and an estimated 30% reduction in the incidence of thyroid cancer in South Korea.⁹⁷

To address thyroid cancer overdiagnosis, the 2015 American Thyroid Association (ATA) clinical practice guidelines recommend against biopsy of sub-centimeter thyroid nodules.⁹⁹ Similarly, the 2016 revised clinical guidelines in South Korea increased the size of thyroid nodules subject to FNA to be at least 1.0 centimeter in size.¹⁰⁰ In addition, the four major international thyroid nodule risk stratification systems (ATA, American College of Radiology Thyroid Imaging and Reporting And Data System (ACR TI-RADS), European Thyroid Imaging and Reporting Data System (EU-TIRADS), and revised Korean Thyroid Imaging Reporting and Data System (K-TIRADS)) all recommend a ‘default’ lower size limit of 1.0 centimeter for FNA of the sonographically highest risk nodules.^99,101–103 The rationale for this lower size limit was to minimize overdiagnosis of small, likely indolent papillary microcarcinomas.

To address the overtreatment of thyroid cancer, interventions that target patient and physician education, and healthcare payment structures are important. In addition, for patients with low-risk differentiated thyroid cancer, interventions may be necessary to address patients’ perceptions of the thyroid “cancer” label. In a survey of 1,068 US residents, Dixon et al found that disease label (cancer, tumor, or nodule) played a role in patients’ preference, independent of treatment risks or prognosis.¹⁰⁴ Similarly, in an Australian web-based survey of 2,054 participants with no history of thyroid cancer, Nickel et al found that participants were more willing to accept potential harms associated with treatment options when the condition was described as a “cancer” compared with a “lesion”.¹⁰⁵

For select patients with sub-centimeter biopsy-proven thyroid cancer, active surveillance in which patients undergo serial neck ultrasounds to identify disease progression that would then prompt consideration for surgical management, may be a consideration. Active surveillance for low-risk papillary thyroid microcarcinoma was adopted as a treatment option in the 2010 Japanese clinical guidelines for the treatment of thyroid nodules.¹⁰⁶ Subsequently, the 2015 ATA and 2019 European Society for Medical Oncology (ESMO) clinical practice guidelines on management of thyroid cancer considered active surveillance to be a safe and effective alternative to immediate surgery in select patients including those with cytologically-confirmed very low-risk papillary thyroid microcarcinoma.^99,107 Informed by more than two decades of experience with active surveillance for low-risk thyroid cancer in the clinical trial setting, the Japan Association of Endocrine Surgery (JAES) Task Force published consensus statements on the use of active surveillance for low-risk papillary thyroid cancer in 2021.¹⁰⁸ Alongside the growing enthusiasm for active surveillance within the medical community, there is increasing interest among patients.^109,110 In a study of 100 Canadian patients with small low-risk papillary thyroid cancer who were offered the choice of active surveillance or surgery, 71% elected to pursue active surveillance.¹¹¹ Using qualitative methods, the authors then identified the following factors as strongly influential of the patients’ disease management choices: personal perceptions about cancer or thyroidectomy, contextual factors, family considerations, and trust in health care providers.¹¹¹ While the data on active surveillance for low-risk papillary thyroid microcarcinoma have been favorable in select patient populations, there remains limitations in generalizing study findings beyond the clinical trial setting and in clinical practice outside of Japan where the studies with the largest patient cohorts were conducted.¹¹²

For select patients with low-risk differentiated thyroid cancer, clinical practice guidelines have recommended thyroid lobectomy and total thyroidectomy as appropriate surgical treatment options.^99,107,113 Potential advantages of thyroid lobectomy include lower risk of post-operative complications, including recurrent laryngeal nerve injury, transient and permanent hypoparathyroidism, and hemorrhage/hematoma, as compared to that for total thyroidectomy.¹¹⁴ In addition, the overall risk of post-operative hypothyroidism after hemithyroidectomy is about 22% whereas total thyroidectomy necessitates thyroid hormone replacement.¹¹⁵ Despite clinical practice guideline recommendations encouraging the consideration of thyroid lobectomy for select patients with low-risk thyroid cancer, the incidence of total thyroidectomy have remained high in many countries, including Italy, South Korea, and the US.^116–118 To increase uptake of thyroid lobectomy in select patients with low-risk thyroid cancer, multi-level interventions are needed that directly educate patients about their treatment options, address surgeon attitudes,¹¹⁹ facilitate shared decision-making between patients and surgeons, and change insurance payment structures to encourage thyroid lobectomy when medically appropriate.

Summary

In recent decades, there has been a worldwide increase in the incidence of thyroid cancer with relatively stable thyroid cancer mortality rates. The observed epidemiological trend has primarily been driven by overdiagnosis of small indolent papillary thyroid cancers due to the increased use of thyroid and neck ultrasonography. There also appears to be a true increase, albeit smaller, in the incidence of thyroid cancer that is attributed to modifiable risk factors, some of which are known and some of which are currently being studied. Despite the low-risk nature of many of the thyroid cancer diagnoses, and clinical practice guidelines recommending consideration of less-intensive treatment options, many patients with thyroid cancer continue to be overtreated. Multi-level interventions are necessary to address the overdiagnosis and overtreatment of low-risk thyroid cancer.

Key Points.

• Between 1980 and 2017, global thyroid cancer incidence rates increased while mortality rates remained relatively stable

• Radiation exposure is the only well-established modifiable risk factor for thyroid cancer

• Recent studies suggest that obesity and environmental exposures may play a role in the development of thyroid cancer

• The overdiagnosis of small indolent thyroid cancers, largely driven by increased utilization of thyroid ultrasonography, accounts for much of the global increase in thyroid cancer incidence

• Multi-level interventions are necessary to address the public health issue of thyroid cancer overdiagnosis and subsequent overtreatment