Articles

The epidemiological landscape of thyroid cancer worldwide:

GLOBOCAN estimates for incidence and mortality rates in
2020
Margherita Pizzato, Mengmeng Li, Jerome Vignat, Mathieu Laversanne, Deependra Singh, Carlo La Vecchia, Salvatore Vaccarella

Summary
Lancet Diabetes Endocrinol Background Thyroid cancer incidence rates have increased in many countries and settings; however, mortality rates
2022; 10: 264–72 have remained stable at lower rates. This epidemiological pattern has been largely attributed to an overdiagnosis
Published Online effect. Timely evidence for the global epidemiological status is necessary to identify the magnitude of this problem
March 7, 2022
and the areas mostly affected by it. We therefore aimed to provide an up-to-date assessment on the global distribution
https://doi.org/10.1016/
S2213-8587(22)00035-3 of thyroid cancer incidence and mortality rates in 2020.
See Comment page 235
Department of Clinical Sciences
Methods We extracted age-standardised incidence and mortality rates per 100 000 person-years of thyroid cancer as
and Community Health, defined by the International Classification of Diseases for Oncology 10th Revision (code C73), for 185 countries or
University of Milan, Milan, Italy territories by sex and 18 age groups (ie, 0–4, 5–9, … , 80–84, and ≥85 years) from the GLOBOCAN database. Both
(M Pizzato MD, incidence and mortality estimates were presented by country and aggregated across the 20 UN-defined world regions
Prof C La Vecchia MD);
International Agency for
and according to the UN’s four-tier Human Development Index (ie, low, medium, high, and very high) in 2020.
Research on Cancer, Lyon,
France (M Pizzato, J Vignat MSc, Findings Globally, in 2020, the age-standardised incidence rates of thyroid cancer were 10·1 per 100 000 women
M Laversanne MSc, D Singh PhD, and 3·1 per 100 000 men, and age-standardised mortality rates were 0·5 per 100 000 women and 0·3 per 100 000 men.
S Vaccarella PhD); Department
of Cancer Prevention, State Key
In both sexes, incidence rates were five times higher in high and very high Human Development Index countries
Laboratory of Oncology in than in low and medium Human Development Index countries, whereas mortality rates were relatively similar across
South China, Collaborative different settings. Incidence rates in women differed by more than 15 times across world regions, with the highest
Innovation Center for Cancer
incidence rates being in the Federated States of Micronesia and French Polynesia (18·5 per 100 000 women),
Medicine, Sun Yat-sen
University Cancer Center, North America (18·4 per 100 000), and east Asia (17·8 per 100 000, with South Korea reaching 45 per 100 000).
Guangzhou, China (M Li PhD) Mortality rates were less than one per 100 000 in most countries and in both sexes. South Korea had the highest
Correspondence to: incidence-to-mortality rate ratio in both sexes, followed by Cyprus and Canada.
Dr Margherita Pizzato,
Department of Clinical Sciences Interpretation The current thyroid cancer epidemiological landscape is strongly suggestive of a large effect of
and Community Health,
University of Milan, 20133 Milan,
overdiagnosis in many countries and settings worldwide, confirming the relevance of thyroid cancer overdiagnosis as
Italy a global public health problem.
[email protected]
Funding None.

Copyright © 2022 World Health Organization. Published by Elsevier Ltd. All rights reserved.

Introduction epidemic of thyroid cancer worldwide at any point in

Thyroid cancer—the most common malignancy of the recent times is characterised by a large variability in the
endocrine system—encompasses a variety of histotypes incidence rates between and within countries.
that are heterogeneous in frequency and clinical Conversely, mortality rates have remained stable or
behaviour, ranging from the most frequent and indolent declined, converging towards similarly lower rates in
papillary tumours (ie, accounting for the majority of all most countries and in both sexes.5,6 These
thyroid cancers in both sexes) to the progressively rarer epidemiological features of thyroid cancer have been
but highly aggressive follicular, medullary, and anaplastic primarily attributed to the increased detection of
tumours.1 Thyroid cancer shows a strong female indolent thyroid cancers,7 with overdiagnosis accounting
dominance, with most populations having an incidence for up to 60–90% of all patients diagnosed with thyroid
that is about three times higher in women than in men.2 cancer.8,9
Since the early 1980s, an increase in incidence rates, As overdiagnosis of thyroid cancer is now a public
principally concerning early papillary thyroid health problem of global relevance, monitoring and
carcinomas, has been observed initially in very high- provision of timely evidence for the evolution of the
income countries and subsequently in some current epidemiological patterns is important.10 Such
middle-income countries, such as China3 and India.4 evidence will be key to developing tailored prevention
These increases are still continuing in many places but strategies to limit overdiagnosis at the national, sub­
are not geographically homogeneous, so that the national, and regional level. In this Article, we aimed to

264 www.thelancet.com/diabetes-endocrinology Vol 10 April 2022

 Articles

Research in context
Evidence before this study cancer estimates projected to 2020 for 185 countries or
We searched PubMed for previously published studies territories, 20 UN-defined world regions, and by Human
addressing thyroid cancer patterns in incidence and mortality. Development Index, from the best available recorded data from
We particularly focused on studies exploring overdiagnosis national or subnational cancer registries and national vital
characteristics within and across countries. An example search registry systems. Incidence rates in women were about three
strategy used includes the following: “((((thyroid) AND (cancer times higher than in men. A large geographical variability of
OR neoplasm OR carcinoma OR neoplasms [MeSH Terms]) AND thyroid cancer incidence emerged, with high rates seen in many
(risk))) AND English [Language] AND (“1980”[Date - settings worldwide, including mostly high-income countries,
Publication]: “2021”[Date - Publication]))”. We also reviewed but also some low-income and middle-income countries.
references from retrieved articles to identify additional studies. Conversely, consistently low mortality rates due to thyroid
Incidence rates increased in many countries and settings, cancer were reported almost everywhere. The largest gap
whereas mortality rates remained stable at lower levels or between incidence and mortality rates in women was observed
declined; this epidemiological pattern has been largely in South Korea, whereas the smallest incidence-to-mortality
attributed to an overdiagnosis effect. The global relevance of ratio emerged in some Pacific Ocean islands (eg, Samoa,
overdiagnosis of thyroid cancer calls for continued and Vanuatu, and Solomon Islands).
updated monitoring of this effect, trends, and geographical
Implications of all the available evidence
expansion.
The evidence from this study highlights that overdiagnosis of
Added value of this study thyroid cancer has become a major public health issue in many
We provided the most up-to-date snapshot on the global settings worldwide. Monitoring epidemiological patterns is
distribution of thyroid cancer incidence and mortality rates crucial to develop tailored prevention strategies to limit
using data from the GLOBOCAN project. We included national overdiagnosis, with its clinical and financial implications.

provide, using the GLOBOCAN database,11 the most up-

Incidence Mortality
to-date assessment on the global distribution of thyroid
cancer incidence and mortality rates in 2020, with a Women Men Women Men
focus on the variability across and within regions and Cases ASR Cases ASR Death ASR Death ASR
according to different levels of the Human Development World 448 915* 10∙1 137 287* 3∙1 27 740* 0∙5 15 906* 0∙3
Index. Very high HDI 162 780 15∙5 49 863 4∙5 8166 0∙4 4951 0∙3
High HDI 247 066 13∙6 74 559 4∙1 12 465 0∙6 6469 0∙3
Methods Medium HDI 30 832 2∙7 10 393 0∙9 4728 0∙4 3640 0∙4
Data sources Low HDI 8068 2∙5 2440 0∙9 2375 0∙9 845 0∙4
The number of new cases of, and deaths from, thyroid ASR=age-standardised rate. HDI=Human Development Index. *The estimated numbers of cancer cases and cancer
cancers, as defined by the International Classification of deaths in a given region might not correspond to the sum of the estimated numbers of cancer cases and deaths in the
Diseases for Oncology 10th Revision (code C73), were individual countries of that region; the population of a region might include some small country populations for which
it was not possible to provide estimates.
extracted from the GLOBOCAN 2020 database for
185 countries or territories by sex and 18 age groups Table: Numbers and rates per 100 000 of incidence and mortality for thyroid cancer by sex, according to
(ie, 0–4, 5–9, 10–14, 15–19, 20–24, 25–29, 30–34, 35–39, the HDI 2020
40–44, 45–49, 50–54, 55–59, 60–64, 65–69, 70–74, 75–79,
80–84, and ≥85 years).12,13 Corresponding population data
for the year 2020 were extracted from the UN website.14 population-based cancer registry data including
This study did not involve any human contact, but only submissions to the Cancer Incidence in Five Continents
record linkage analysis of health-care databases. (volume 11) from 2008 to 2012, and more recent data
from the African Cancer Registry Network. Mortality
Procedures rate estimates were obtained using the most recent
The data sources and hierarchy of methods used in national vital registration data from WHO.15–17 More
compiling the cancer estimates have been described in details about the GLOBOCAN dataset and a panel For more on GLOBOCAN
detail elsewhere.12 In brief, the GLOBOCAN addressing country-specific details about estimation dataset see https://gco.iarc.fr/
today/data-sources-methods
2020 database includes national cancer estimates methods, data sources, as well as calendar years included
For more on country-specific
projected to 2020 derived from the best available are available online.
details see https://gco.iarc.fr/
recorded data available from national or subnational today/data/methods/
cancer registries and national vital registry systems in Statistical analysis globocan2020_annex_a.xlsx
185 countries or territories of the world. Incidence rate We calculated two summary measures (ie, incidence and
estimates are derived from national or subnational mortality rates) using direct standardisation—namely,

www.thelancet.com/diabetes-endocrinology Vol 10 April 2022 265

 Articles

the age-standardised rate—using the world population as More details about the age-standardisation is available.18
a reference, per 100 000 person-years. By denoting ai with The results are presented by country and aggregated
age-specific rates for the age group i, and wi with the across 20 UN-defined world regions14 and according to
corresponding weight from the standard population the UN’s four-tier Human Development Index in 2020.19
(where i=1,2, ... 18, so to include all age groups from By using the UN’s four-tier Human Development Index,
0-4 years to ≥85 years), the age-standardised rate is we could assess the cancer burden at varying levels of
calculated as follows: development (ie, low, medium, high, and very high
18 Human Development Index).
Σaiwi We did all statistical analyses using R (version 4.1).
i=1
Age-standardised rates =
18
Σwi Role of the funding source
i=1 There was no funding source for this study.

Age-standardised incidence rates

per 100 000 women
>16·6 2·1–4·3
11·3–16·6 1·3–2·1
8·3–11·3 <1·3
6·5–8·3 No data
4·3–6·5 Not applicable

Age-standardised mortality rates

per 100 000 women
>1·24 0·32–0·41
0·84–1·24 0·25–0·32
0·59–0·84 <0·25
0·51–0·59 No data
0·41–0·51 Not applicable

Figure 1: Age-standardised incidence rates (A) and mortality rates (B) of thyroid cancer per 100 000 person-years for 185 countries or territories among
women in 2020

266 www.thelancet.com/diabetes-endocrinology Vol 10 April 2022

 Articles

A
Regional rate
The Federated States of Country rate
Micronesia and French Polynesia Guam French Polynesia
North America
USA Canada
East Asia
Mongolia South Korea
South Europe
North Macedonia Cyprus
Australia and New Zealand
New Zealand Australia
South America
Guyana Brazil
West Asia
Yemen Turkey
West Europe
Netherlands France
Melanesia
Papua New Guinea New Caledonia
Central-eastern Europe
Bulgaria Hungary
Central America
El Salvador Costa Rica
North Europe
Estonia Latvia
Southeast Asia
Timor-Leste Singapore
Caribbean
Haiti Puerto Rico
North Africa
Sudan Morocco
Southern Africa
Lesotho South Africa
East Africa
Zambia Ethiopia
South-central Asia
Tajikistan Sri Lanka
West Africa
Cape Verde
Central Africa
CongoChad

0 5 10 15 20 25 30 35 40 45
Age-standardised incidence rate per 100 000 women

B
Regional rate
Country rate
Melanesia
Solomon Islands Vanuatu
The Federated States of
Micronesia and French Polynesia Guam Samoa
East Africa
Zambia Djibouti
West Asia
Bahrain United Arab Emirates
North Africa
Tunisia Morocco
Central America
Belize Honduras
Southeast Asia
Timor-Leste Laos
West Africa
Guinea Cape Verde
South America
French Guiana Ecuador
Middle Africa
Congo Chad
East Asia
Japan South Korea
Caribbean
Bahamas Jamaica
Central-eastern Europe
Belarus Poland
Southern Africa
Swaziland Namibia
South-Central Asia
Bhutan Sri Lanka
West Europe
Luxembourg Austria
North Europe
Sweden Latvia
South Europe
Slovenia Serbia
North America
Canada USA
Australia and New Zealand
New Zealand
Australia
0 1 2 3 4 5 6
Age-standardised mortality rate per 100 000 women

Figure 2: Age-standardised incidence rates (A) and mortality rates (B) of thyroid cancer by 20 UN-defined world regions among women in 2020
Rates are shown in descending order, and the lowest and the highest national rates of each world region are superimposed.

www.thelancet.com/diabetes-endocrinology Vol 10 April 2022 267

 Articles

Results southern Europe (Cyprus ranks second worldwide

Globally, in 2020 the estimated number of new cases of [30 per 100 000]), as well as Australia and New Zealand.
thyroid cancer is about 449 000 in women and 137 000 in Intermediate-to-high incidence rates in women are seen
men, corresponding to age-standardised incidence rates of in South America (Brazil has the highest rates in the
approximately 10·1 per 100 000 women and 3·1 per continent [18·5 per 100 000]), as well as west Asia and
100 000 men. Mortality rates from this disease west Europe (with rates widely ranging between five per
are 0·5 per 100 000 women and 0·3 per 100 000 men for a 100 000 in the Netherlands and 23 per 100 000 in France).
global total of about 44 000 deaths (ie, 28 000 in women and Thyroid cancer is less commonly diagnosed
16 000 in men). In each sex, incidence rates are five times in women from northeast Europe, central America, and
greater in high and very high Human Development Index especially southeast Asia and the Caribbean with
countries than in low and medium Human Development aggregate incidence rates being about six per
Index countries. Mortality rates are similar across different 100 000 women. Registered incidence rates remain low
settings, although slightly higher in low Human in Africa—ie, less than six per 100 000 women, except in
Development Index countries in women (table). Cape Verde, where it reaches a relatively high rate
A more than 15 times difference in incidence rates of 26 per 100 000 women. Similar patterns apply to the
across different regions of the world in women exist incidence rates in men, although with rates approximately
See Online for appendix (figures 1A, 2A). The Federated States of Micronesia and two-thirds lower than in women (appendix pp 2, 4).
French Polynesia have the highest incidence rates in Central Africa shows a female-to-male ratio of less than
women (18·5 per 100 000), with the French Polynesian two, whereas corresponding ratios are more than five in
surpassing 25 per 100 000 women. High incidence rates Melanesia and the Federated States of Micronesia and
in women are also seen in North America (18·4 per 100 000, French Polynesia. The absolute variation in incidence
in which Canada has the third highest incidence rate rates across regions is substantial in men but
worldwide [27 per 100 000 women]), east Asia (17·8 comparatively smaller than in women.
per 100 000 woman, in which South Korea has the highest Most regions worldwide have mortality rates in women
incidence rate worldwide [45 per 100 000]), between 0·20 per 100 000 and 0·80 per 100 000
(figures 1B, 2B). In Melanesian women, mortality rates
exceed those of any other country, reaching about
Incidence three per 100 000. The Federated States of Micronesia
448 915 new cases
and French Polynesia rank second in women with a
East Asia
198 695 (44·3%) mortality rate of about 1·5 per 100 000 (Samoa has the
North America highest mortality rate globally—ie, five per 100 000).
45 696 (10·2%)
South America East Africa and west Asia have mortality rates of about
38 952 (8·7%) one per 100 000 women, with Djibouti and
Central and east Europe the United Arab Emirates having the highest mortality
24 935 (5·6%)
Southeast Asia rates among their regions, with about
24 426 (5·4%) 2·5 per 100 000 women. Intermediate-to-high regional
South-central Asia
23 697 (5·3%) mortality rates (between 0·5 per 100 000 and
West Europe one per 100 000) are seen among women in the African
19 043 (4·2%)
Others
continent (except for southern Africa, whose mortality
73 471 (16·4%) rates are lower and aligned with European and American
mortality rates of <0·4 per 100 000), in Central
Mortality and South America (with Ecuador and Honduras having
27 740 deaths
the highest mortality rate in this region [about
East Asia
7813 (28·2%) 1·5 per 100 000 women]) and in southeast Asia as well as
South-central Asia east Asia—notably, South Korea with a mortality rate of
3622 (13·1%)
Southeast Asia
about 0·5 per 100 000. The regions with the lowest
3096 (11·2%) mortality rates were Europe (ranging between
South America east Europe [0·4 per 100 000 women] and south Europe
1939 (7·0%)
Central and east Europe [0·3 per 100 000]), North America (0·3 per 100 000), as
1726 (6·2%) well as Australia and New Zealand (0·2 per 100 000).
West Asia
1400 (5·0%)
Mortality rates in men are globally consistent with those
East Africa in women, ranging mostly between 0∙20 per 100 000 and
1334 (4·8%) 0·40 per 100 000 (appendix pp 3, 5). Higher mortality
Others
6810 (24·5%) rates are seen in west Asia and southeast Asia, whereas
lower mortality rates are seen in southern-to-northern
Figure 3: Pie charts showing the distribution of incidence cases and deaths of Europe and in southern Africa. West Africa has a large
thyroid cancer by UN-defined world regions in 2020 among women heterogeneous mortality rate, ranging from less than

268 www.thelancet.com/diabetes-endocrinology Vol 10 April 2022

 Articles

0·1 per 100 000 women in The Gambia to 1·6 per 100 000
in Togo, with Togo having the highest mortality rate for South Korea 44·6
0·4
men (appendix p 5). Cyprus 30·2
0·3
Figure 3 shows the distribution of incident cases and Canada 26·9
0·3
deaths of thyroid cancer according to world regions in 26·3
Cape Verde
women; these data for men are summarised in the 2·0
French Polynesia 26·3
appendix (p 6). In both sexes, almost a half of all incidences 1·7
24·8
and a quarter of deaths are estimated to occur in east Asia New Caledonia
1·9
in 2020; North and South Americas account for about 18% Italy 24·6
0·3
of the total incidence of thyroid cancers. The share of France 23·1
0·2
deaths due to thyroid cancers is consistent with the 22·4
Turkey
incidence portion among women in central-to- 0·8
Israel 20·9
eastern Europe (about 6%) and among men in west Europe 0·5
19·4
(about 5%). The share of cancer deaths in Costa Rica
0·6
south-central Asia (about 13% among women and about Croatia 19·3
0·3
21% among men) and southeast Asia (about 11% among Switzerland 18·8
0·4
women and about 9% among men) are higher than the 18·6
Portugal
corresponding incidence percentages for both regions 0·3
Brazil 18·5
(about 5% in women and 6% in men). East Africa 0·4
18·3
accounted for about 5% of the total thyroid cancer deaths Vanuatu
4·2
in women. Puerto Rico 18·2
0·2
Figure 4 shows the age-standardised incidence and China 17·5
0·5
mortality rates of thyroid cancer in countries with the 17·4
Austria
highest incidence rates in 2020 among women; these 0·3
USA 17·4
data for men are summarised in the appendix (p 7). 0·4
16·9
South Korea has the highest rate ratio gap between Latvia
0·6
incidence and mortality rates in both sexes, followed Samoa 16·8
5·0
by Cyprus and Canada. The lowest rate ratio gap is seen Solomon Islands 16·7
1·2
in Samoa and Vanuatu among women, and in Saudi 16·7
Hungary Incidence
Arabia and Turkey among men. 0·4
Australia 16·6 Mortality
0·2
Discussion 0 10 20 30 40 50
This up-to-date global analysis for 2020 shows a large Age-standardised rate per 100 000 women
geographical variability in the distribution of thyroid
Figure 4: Bar plots of age-standardised incidence and mortality rates of thyroid cancer per 100 000 person-
cancer incidence worldwide, with elevated incidence rates years in countries with the highest incidence rates in 2020 among women
in several regions and countries across all settings Incidence rates are shown in descending order, the corresponding values for the age-standardised mortality rates
worldwide. Conversely, systematically low mortality rates are superimposed.
were observed almost everywhere. The geographical
hetero­geneity in incidence rates is wider for women, symptoms or visible neck masses.20 Since the 1980–90s, the
whose rates are in each country around three times higher advent of the neck ultrasonography and of ultrasound-
than in men. Incidence rates in both sexes are five times guided fine-needle biopsy, together with the spread of new
greater in high and very high Human Development Index imaging technologies (eg, portable ultrasound machines
countries compared with low and medium Human and CT), enabled the diagnosis of previously undetectable
Development Index countries, whereas mortality rates are millimetric nodules.21,22 These techniques have led to the
rather similar across different settings. Disparities in detection of a number of indolent thyroid lesions,7,10 and to
incidence rates for women emerged within regions, with the consequent increase in the number of diagnoses and in
east Asia and south Europe being among the most hetero­ the incidence rates of thyroid cancer without affecting
genous areas, and even across regions, with a 15 times mortality rates, which remained at low rates or even
difference in incidence rates between the Federated States declined in many places. Besides the growing discrepancy
of Micronesia and French Polynesia and most of between incidence and mortality, other typical
the African continent. A large gap between incidence and epidemiological features of thyroid cancer overdiagnosis
mortality rates in women was observed in some countries, includes the following: the fact that the rising incidence is
such as South Korea, Cyprus, and Canada, whereas the almost exclusively due to an increased detection of papillary
smallest incidence-to-mortality ratio emerged in some thyroid cancer;23,24 the large geographical hetero­geneity of
Pacific Ocean Islands. the increases in incidence; the known existence of a large
Thyroid cancers were historically found by palpation, reservoir of subclinical tumours in the thyroid gland, as
frequently in patients already presenting with compression described in autopsy studies25,26 (of note, the prevalence of

www.thelancet.com/diabetes-endocrinology Vol 10 April 2022 269

 Articles

thyroid tumours in autopsies was found to be similar and increasing rates observed in those regions with a
between sexes, as opposed to diagnosed cases, which are better access to health care.4 Similar patterns and
three times more prevalent in women than in men27); and substantial geographical variability has emerged in China,
the gradual change of the age-specific curves of incidence, where higher incidence rates were observed in urban areas
which now peaks at around the age of 40 years or 50 years, with higher accessibility to the health care services than in
instead of increasing with age, similar to the period before rural areas.3
the introduction of modern diagnostic techniques and The intensified surveillance and scrutiny of the thyroid
intense surveillance. These epidemiological character­istics gland has led to an increased detection of small, early stage
are strongly suggestive of a large effect of overdiagnosis.28,29 tumours but it might have also lead to the discovery of a
The organisation of the health systems, including certain number of clinically silent, larger, and advanced-
regulation and access to diagnosis and care, as well as the stage cancers, as an increase of these advanced-stage
availability and facilities for thyroid cancer screening cancers was reported in some countries.10,47 It cannot be
practices are hence key determinants for the observed excluded, however, that this finding might be the result of a
geographical variability in incidence rates.30 The most true increase in the incidence of the disease due to the
striking example has been observed in South Korea,31 enhanced exposure to selected risk factors.48 Clinical or
where from the early 2000s a national screening subclinical iodine deficiency, for instance, is still present in
programme for other common malignancies contributed some European areas49,50 and, to a greater extent, in some
to the spread of opportunistic examinations for thyroid high mortality areas of central Asia and Africa.51 Ionising
cancer by fee-for-service providers.32,33 The proportion of radiation exposure for medical reasons, one of the major
thyroid cancer cases in South Korean women that were thyroid cancer risk factors, has increased in areas with a
attributable to overdiagnosis between 2008 and 2012 was broad access to health care, due to the increase of medical
about 90%.9 In 2020, thyroid cancer remains the most imaging procedures.52 Radiation exposure, particularly
common malignancy diagnosed in South Korea and the during childhood, could also contribute to the exceedingly
gap between incidence and mortality rates is still among high incidence rates observed in some Pacific Ocean Islands,
the widest worldwide. Overdiagnosis patterns also where atomic tests in the atmosphere were done between
appeared in some market-oriented areas where the access 1966 and 1996.53 In these regions, lifestyle and nutritional
to health care overrules regulatory controls (eg, the USA transition (ie, from a traditional diet based on fish and fruit
and Brazil34), as well as in some high-quality public health to a Western-style nutrition rich in saturated fats and added
systems with an easy and broad access to thyroid gland sugars54), along with still undefined genetic factors,53 could
diagnostic examinations (eg, Canada and some south have also partly contributed to comparatively high incidence
European countries).35 Incidence rates and temporal and mortality rates. These regions are characterised by an
changes among children and adolescents were found to be intense activity of volcanoes, whose effect has been related
strongly correlated with adult ones at the international to a certain excess in risk even in other volcanic areas—
level. This finding suggests an expanding role of eg, Hawaii, Iceland, and Italy.50,55 The exposure to certain
overdiagnosis also in younger age groups.36 The plateauing endocrine-disrupting chemicals, including some pesticides,
of thyroid cancer incidence rates observed over the past flame retardants, and food packaging materials, was
decade in South Korea, as well as in other high incidence associated to an increase in thyroid hormone axis disorders,
countries (eg, the USA37,38 and France39), might be linked to potentially resulting in thyroid cancers.56,57 In addition,
the increasing awareness of overdiagnosis threats, and excess adiposity might have contributed to the increase in
reflect the adoption of newer, more conservative diagnostic thyroid cancer incidence over the past decades, given that
guidelines.24,40 These guidelines included recommendations body-mass index is associated with thyroid cancer risk.58,59
against screening for thyroid cancer in asymptomatic adult Detection biases might have partly influenced this
populations free from specific risk factors, and watchful- association, as a more intense diagnostic scrutiny could
waiting approaches for low-risk lesions.41,42 In these occur among patients with underlying obesity-related
countries where access to specialistic diagnosis is guided chronic conditions. Hormone-related events in womens’
by strong regulatory rules (eg, the UK and Nordic lifetime (eg, contraception and pregnancy) seem to have a
countries), thyroid cancer incidence rates, especially weak to negligible role in thyroid carcinogenesis,60,61 while
papillary cancers, remained generally low.43 Geographical they contribute to expose women to more intensive medical
heterogeneity in incidence rates has been observed not surveillance.62 Overall, the possible contribution of these
only across countries, but also among regions within the factors on the large variability in incidence rates, if any, is
same country;44,45 in many low-income and middle-income likely to be small. Although the role of many of the above-
countries undergoing economic transition, low-regulated mentioned risk factors has not been clearly established,
scrutiny of the thyroid gland has risen in high health their exposure might be associated with an increase in
quality contexts, as a consequence of the increased access access to health-care services and consequently higher
to high sensitive diagnostic facilities.46 Namely, incidence incidental detection of subclinical tumours. Therefore,
rates in India, albeit ranking among the lowest worldwide, detection bias is difficult to detect and account for in a
are affected by a ten times regional variation, being high context of large overdiagnosis.63

270 www.thelancet.com/diabetes-endocrinology Vol 10 April 2022

 Articles

With respect to the database used for this analysis, the Data sharing
GLOBOCAN estimates for 2020 might be limited by the Data used in this study are retrievable from the web-based platform the
Global Cancer Observatory.
accuracy in certain contexts since 2020 projections are For more information on Global
Acknowledgments Cancer Observatory see https://
based on incidence and mortality trends from the past
We thank the cancer registries worldwide who compiled and submitted gco.iarc.fr/
years.12 However, when building these estimates, the their data for the GLOBOCAN project.
GLOBOCAN editors paid specific attention to the
Editorial note: the Lancet Group takes a neutral position with respect to
estimates of certain cancers, such as thyroid cancer; for territorial claims in published maps and institutional affiliations.
instance, for the estimates of incidence in South Korea
References
the most recent rates, rather than projections of past 1 Kitahara CM, Sosa JA. The changing incidence of thyroid cancer.
trends, were used. The COVID-19 pandemic was not Nat Rev Endocrinol 2016; 12: 646–53.
taken into consideration in this study, although it could 2 Rahbari R, Zhang L, Kebebew E. Thyroid cancer gender disparity.
Future Oncol 2010; 6: 1771–79.
have affected the registration process, particularly in low- 3 Li M, Zheng R, Maso LD, Zhang S, Wei W, Vaccarella S. Mapping
income and middle-income countries. Similarly, under overdiagnosis of thyroid cancer in China. Lancet Diabetes Endocrinol
regi­
stra­tion could have affected some countries 2021; 9: 330–32.
characterised by geopolitical issues in 2020. In general, 4 Panato C, Vaccarella S, Dal Maso L, et al. Thyroid cancer incidence
in India between 2006 and 2014 and impact of overdiagnosis.
the methods of incidence and mortality estimation are J Clin Endocrinol Metab 2020; 105: 2507–14.
undertaken at the national level and hence their validity 5 La Vecchia C, Malvezzi M, Bosetti C, et al. Thyroid cancer
depends upon the representativeness and quality of the mortality and incidence: a global overview. Int J Cancer 2015;
136: 2187–95.
source information from the country itself. Diagnostic 6 O’Grady TJ, Gates MA, Boscoe FP. Thyroid cancer incidence
and therapeutic improve­ments during the most recent attributable to overdiagnosis in the United States 1981–2011.
years might not be reflected in the provided estimates. Int J Cancer 2015; 137: 2664–73.
7 Brito JP, Al Nofal A, Montori VM, Hay ID, Morris JC. The impact of
The unavailability of information about thyroid cancer subclinical disease and mechanism of detection on the rise in
histopathological characteristics (eg, histotype and stage) thyroid cancer incidence: a population-based study in Olmsted
and on the pathways to diagnosis (eg, incidental or sympto­ County, Minnesota during 1935 through 2012. Thyroid 2015;
25: 999–1007.
matic detection) falls among study limitations. Thus, 8 Brito JP, Hay ID. Overdiagnosis of papillary carcinoma—who
caution is needed in the interpretation of the estimated benefits? Nat Rev Endocrinol 2017; 13: 131–32.
rates. 9 Li M, Maso LD, Vaccarella S. Global trends in thyroid cancer
Changes in histological classification guidelines, together incidence and the impact of overdiagnosis.
Lancet Diabetes Endocrinol 2020; 8: 468–70.
with difficulties in disease ascertaining and in certifying the 10 Lim H, Devesa SS, Sosa JA, Check D, Kitahara CM. Trends in
causes of death, might have affected the accuracy of thyroid cancer incidence and mortality in the United States,
incidence and mortality rates in some countries,64 likely to a 1974–2013. JAMA 2017; 317: 1338–48.
11 Sung H, Ferlay J, Siegel RL, et al. Global cancer statistics 2020:
greater extent in some low-income and middle-income GLOBOCAN estimates of incidence and mortality worldwide for
countries. In these countries, delayed, incomplete, and 36 cancers in 185 countries. CA Cancer J Clin 2021; 71: 209–49.
inaccurate reporting might have biased incidence and 12 Ferlay J, Colombet M, Soerjomataram I, et al. Cancer statistics for
the year 2020: an overview. Int J Cancer 2021; published online
mortality estimates, possibly resulting in an underestimation April 5. https://doi.org/10.1002/ijc.33588.
of both incidence and mortality rates. 13 WHO. Cancer today: data & methods. https://gco.iarc.fr/today/data-
In summary, this study provides a timely report of the sources-methods (accessed Sept 6, 2021).
global situation of thyroid cancer, using estimates 14 UN. Methodology. https://unstats.un.org/unsd/methodology/m49/
(accessed Sept 6, 2021).
projected to 2020 from the GLOBOCAN database. A large 15 Ferlay J, Colombet M, Soerjomataram I, et al. Estimating the global
geographical variability of thyroid cancer incidence exists, cancer incidence and mortality in 2018: GLOBOCAN sources and
with very high rates in many countries and settings methods. Int J Cancer 2019; 144: 1941–53.
worldwide, including in low-income and middle-income 16 Ferlay J, Soerjomataram I, Dikshit R, et al. Cancer incidence and
mortality worldwide: sources, methods and major patterns in
countries, and conversely consistently low mortality rates GLOBOCAN 2012. Int J Cancer 2015; 136: E359–86.
from the disease. These results suggest that overdiagnosis 17 Ferlay J, Shin H-R, Bray F, Forman D, Mathers C, Parkin DM.
of thyroid cancer is still expanding in many places world­ Estimates of worldwide burden of cancer in 2008:
GLOBOCAN 2008. Int J Cancer 2010; 127: 2893–917.
wide, confirming its relevance as a global public health 18 Boyle P, Parkin DM. Statistical methods for registries. In: Jensen
problem. Continued monitoring of the magnitude, trends, OM, Parkin DM, MacLennan R, et al, eds. Cancer registration:
and geographical expansion of this problem at the national principles and methods. Lyon: IARC Scientific Publications, 1991,
126–58.
and international level is therefore essential. 19 UN. Human development reports 2020. The next frontier. Human
Contributors development and the Anthropocene. http://hdr.undp.org/en/2020-
SV, MP, and MLi conceptualised and designed the study. JV and MLa report (accessed Sept 6, 2021).
accessed the data and did the formal analysis. MP and SV wrote the 20 Dal Maso L, Tavilla A, Pacini F, et al. Survival of 86 690 patients with
report. CLV and DS contributed to data interpretation. All authors had thyroid cancer: a population-based study in 29 European countries
full access to all the data in the study and had final responsibility for the from EUROCARE-5. Eur J Cancer 2017; 77: 140–52.
decision to submit for publication. 21 Brito JP, Morris JC, Montori VM. Thyroid cancer: zealous imaging
has increased detection and treatment of low risk tumours. BMJ
Declaration of interests 2013; 347: f4706.
We declare no competing interests.

www.thelancet.com/diabetes-endocrinology Vol 10 April 2022 271

 Articles

22 Vaccarella S, Dal Maso L, Laversanne M, Bray F, Plummer M, 44 Dal Maso L, Panato C, Franceschi S, et al. The impact of
Franceschi S. The impact of diagnostic changes on the rise in overdiagnosis on thyroid cancer epidemic in Italy,
thyroid cancer incidence: a population-based study in selected high- 1998–2012. Eur J Cancer 2018; 94: 6–15.
resource countries. Thyroid 2015; 25: 1127–36. 45 Salazar-Vega J, Ortiz-Prado E, Solis-Pazmino P, et al. Thyroid cancer
23 Davies L, Welch HG. Increasing incidence of thyroid cancer in in Ecuador, a 16 years population-based analysis (2001–2016).
the United States, 1973–2002. JAMA 2006; 295: 2164–67. BMC Cancer 2019; 19: 294.
24 Miranda-Filho A, Lortet-Tieulent J, Bray F, et al. Thyroid cancer 46 Veedu JS, Mathew A. Are we missing the elephant in the room?
incidence trends by histology in 25 countries: a population-based A case for thyroid cancer overdiagnosis as the etiology for its
study. Lancet Diabetes Endocrinol 2021; 9: 225–34. increasing incidence in India. J Glob Oncol 2018; 4: 1–3.
25 Ho AS, Davies L, Nixon IJ, et al. Increasing diagnosis of subclinical 47 Yan KL, Li S, Tseng C-H, et al. Rising incidence and incidence-
thyroid cancers leads to spurious improvements in survival rates. based mortality of thyroid cancer in California, 2000–2017.
Cancer 2015; 121: 1793–99. J Clin Endocrinol Metab 2020; 105: dgaa121.
26 Furuya-Kanamori L, Bell KJL, Clark J, Glasziou P, Doi SAR. 48 Chen AY, Jemal A, Ward EM. Increasing incidence of differentiated
Prevalence of differentiated thyroid cancer in autopsy studies over thyroid cancer in the United States, 1988–2005. Cancer 2009;
six decades: a meta-analysis. J Clin Oncol 2016; 34: 3672–79. 115: 3801–07.
27 Zamora-Ros R, Rinaldi S, Biessy C, et al. Reproductive and 49 Dal Maso L, Lise M, Zambon P, et al. Incidence of thyroid cancer
menstrual factors and risk of differentiated thyroid carcinoma: in Italy, 1991–2005: time trends and age–period–cohort effects.
the EPIC study. Int J Cancer 2015; 136: 1218–27. Ann Oncol 2011; 22: 957–63.
28 Morris LGT, Sikora AG, Tosteson TD, Davies L. The increasing 50 Kim J, Gosnell JE, Roman SA. Geographic influences in the global
incidence of thyroid cancer: the influence of access to care. Thyroid rise of thyroid cancer. Nat Rev Endocrinol 2020; 16: 17–29.
2013; 23: 885–91. 51 Dal Maso L, Bosetti C, La Vecchia C, Franceschi S. Risk factors for
29 Vaccarella S, Franceschi S, Bray F, Wild CP, Plummer M, thyroid cancer: an epidemiological review focused on nutritional
Dal Maso L. Worldwide thyroid-cancer epidemic? The increasing factors. Cancer Causes Control 2009; 20: 75–86.
impact of overdiagnosis. N Engl J Med 2016; 375: 614–17. 52 Enewold L, Zhu K, Ron E, et al. Rising thyroid cancer incidence in
30 Franceschi S, Vaccarella S. Thyroid cancer: an epidemic of disease the United States by demographic and tumor characteristics,
or an epidemic of diagnosis? Int J Cancer 2015; 136: 2738–39. 1980–2005. Cancer Epidemiol Biomarkers Prev 2009; 18: 784–91.
31 Choi YM, Kim WG, Kwon H, et al. Changes in standardized 53 Truong T, Rougier Y, Dubourdieu D, et al. Time trends and
mortality rates from thyroid cancer in Korea between 1985 and geographic variations for thyroid cancer in New Caledonia, a very
2015: analysis of Korean national data. Cancer 2017; 123: 4808–14. high incidence area (1985–1999). Eur J Cancer Prev 2007;
32 Ahn HS, Kim HJ, Welch HG. Korea’s thyroid-cancer ‘epidemic’— 16: 62–70.
screening and overdiagnosis. N Engl J Med 2014; 371: 1765–67. 54 Cléro E, Leux C, Brindel P, et al. Pooled analysis of two case-control
33 Ahn HS, Kim HJ, Kim KH, et al. Thyroid cancer screening in South studies in New Caledonia and French Polynesia of body mass index
Korea Increases detection of papillary cancers with no impact on and differentiated thyroid cancer: the importance of body surface
other subtypes or thyroid cancer mortality. Thyroid 2016; 26: 1535–40. area. Thyroid 2010; 20: 1285–93.
34 Veiga LHS, Neta G, Aschebrook-Kilfoy B, Ron E, Devesa SS. 55 Pellegriti G, De Vathaire F, Scollo C, et al. Papillary thyroid cancer
Thyroid cancer incidence patterns in Sao Paulo, Brazil, and the US. incidence in the volcanic area of Sicily. J Natl Cancer Inst 2009;
SEER Program, 1997–2008. Thyroid 2013; 23: 748–57. 101: 1575–83.
35 Guay B, Johnson-Obaseki S, McDonald JT, Connell C, Corsten M. 56 Leung AM, Korevaar TIM, Peeters RP, et al. Exposure to thyroid-
Incidence of differentiated thyroid cancer by socioeconomic status disrupting chemicals: a transatlantic call for action. Thyroid 2016;
and urban residence: Canada 1991–2006. Thyroid 2014; 24: 552–55. 26: 479–80.
36 Vaccarella S, Lortet-Tieulent J, Colombet M, et al. Global patterns 57 Alsen M, Sinclair C, Cooke P, Ziadkhanpour K, Genden E,
and trends in incidence and mortality of thyroid cancer in children van Gerwen M. Endocrine disrupting chemicals and thyroid cancer:
and adolescents: a population-based study. an overview. Toxics 2021; 9: 14.
Lancet Diabetes Endocrinol 2021; 9: 144–52. 58 Schmid D, Ricci C, Behrens G, Leitzmann MF. Adiposity and risk
37 Powers AE, Marcadis AR, Lee M, Morris LGT, Marti JL. Changes in of thyroid cancer: a systematic review and meta-analysis. Obes Rev
trends in thyroid cancer incidence in the United States, 2015; 16: 1042–54.
1992 to 2016. JAMA 2019; 322: 2440–41. 59 Kitahara CM, Pfeiffer RM, Sosa JA, Shiels MS. Impact of
38 Pereira M, Williams VL, Hallanger Johnson J, Valderrabano P. overweight and obesity on US papillary thyroid cancer incidence
Thyroid cancer incidence trends in the United States: association trends (1995–2015). J Natl Cancer Inst 2020; 112: 810–17.
with changes in professional guideline recommendations. Thyroid 60 Negri E, Maso LD, Ron E, et al. A pooled analysis of case-control
2020; 30: 1132–40. studies of thyroid cancer. II. Menstrual and reproductive factors.
39 Colonna M, Uhry Z, Guizard AV, et al. Recent trends in incidence, Cancer Causes Control 1999; 10: 143–55.
geographical distribution, and survival of papillary thyroid cancer 61 Caini S, Gibelli B, Palli D, Saieva C, Ruscica M, Gandini S.
in France. Cancer Epidemiol 2015; 39: 511–18. Menstrual and reproductive history and use of exogenous sex
40 Lubitz CC, Sosa JA. The changing landscape of papillary thyroid hormones and risk of thyroid cancer among women: a meta-
cancer: epidemiology, management, and the implications for analysis of prospective studies. Cancer Causes Control 2015;
patients. Cancer 2016; 122: 3754–59. 26: 511–18.
41 US Preventive Services Task Force. Screening for thyroid cancer: 62 Sierra MS, Soerjomataram I, Forman D. Thyroid cancer burden
US preventive services task force recommendation statement. in Central and South America. Cancer Epidemiology 2016;
JAMA 2017; 317: 1882–87. 44: S150–57.
42 Haugen BR, Alexander EK, Bible KC, et al. 2015 American thyroid 63 Vaccarella S, Dal Maso L. Challenges in investigating risk factors for
association management guidelines for adult patients with thyroid thyroid cancer. Lancet Diabetes Endocrinol 2021; 9: 57–59.
nodules and differentiated thyroid cancer: the American Thyroid 64 James BC, Mitchell JM, Jeon HD, Vasilottos N, Grogan RH,
Association Guidelines Task Force on thyroid nodules and Aschebrook-Kilfoy B. An update in international trends in incidence
differentiated thyroid cancer. Thyroid 2016; 26: 1–133. rates of thyroid cancer, 1973–2007. Cancer Causes Control 2018;
43 Lortet-Tieulent J, Franceschi S, Dal Maso L, Vaccarella S. Thyroid 29: 465–73.
cancer ‘epidemic’ also occurs in low- and middle-income countries.
Int J Cancer 2019; 144: 2082–87.

272 www.thelancet.com/diabetes-endocrinology Vol 10 April 2022