Thyroid gland (endocrine system)

Thyroid cancer

How surgery, radioactive iodine, genetic testing and targeted drugs shaped thyroid cancer care, and what overdiagnosis taught.

5-year relative survival · United States, SEER 8 registries

92.95%→98.43%

1975 (observed value; first year in the table) to 2018 (observed value; last observed year in the table). Read these with the caveats under “Then and now”.Source

24 milestones, by decade

The 1909 Nobel Prize in Physiology or Medicine was awarded to Theodor Kocher 'for his work on the physiology, pathology and surgery of the thyroid gland', according to the Nobel Prize API record (prize date 28 October 1909). The motivation covers the thyroid gland in general and does not single out cancer, so it is a marker for the history of thyroid surgery rather than of thyroid cancer treatment.
On the SEER Cancer Stat Facts page for thyroid cancer (both sexes, all races, age-adjusted), the rate of new cases in the SEER 8 series is 5.00 per 100,000 in 1975, 14.94 in 2014 (the highest observed value) and 14.86 in 2023, while the US death rate is 0.54 per 100,000 in 1975 and 0.50 in 2024. The page assigns the rise in incidence to no cause. Davies and Welch, analysing SEER 9 data for 1975 to 2009 (incidence 4.9 to 14.3 per 100,000 with mortality stable at about 0.5, and virtually the entire increase in papillary cancer), described an epidemic of diagnosis rather than of disease; that is their interpretation of an earlier period and is not a statement on the SEER page.
In SELECT, a randomised double-blind phase 3 trial in progressive radioiodine-refractory thyroid cancer, 261 people were assigned to lenvatinib and 131 to placebo, and placebo recipients could receive open-label lenvatinib at progression. Median progression-free survival was 18.3 months with lenvatinib and 3.6 months with placebo (hazard ratio 0.21; 99% confidence interval 0.14 to 0.31), and the response rate was 64.8% versus 1.5%. Median overall survival was not reached in either group, and 6 of the 20 deaths during the treatment period in the lenvatinib group were considered drug-related. The FDA approval letter for lenvatinib in this indication is dated 13 February 2015.

Evidence current to October 2026 · Checked against primary sources to the extent possible

A guided tour of 12 milestones. You can pause or step through it yourself.

Filter milestonesshowing all 24

Treatment type

Setting

Evidence laneTrack

Size shows significance; ring style shows setting:

Prevention / screening
Localised to perioperative
Locally advanced
Metastatic
Maintenance / relapsed

Time scale is compressed before 1800 and expanded after 1945; breaks are marked on the axis. Hold Ctrl or ⌘ and scroll to zoom, or use the buttons.

The age of rays

1909 · Origin · SurgeryKocher and thyroid surgery: the 1909 Nobel PrizeIn 1909 the Swiss surgeon Theodor Kocher won the Nobel Prize in Physiology or Medicine for his work on the thyroid gland. Histories describe thyroid surgery moving from an operation thought too dangerous to attempt to one of the safest. The prize honoured thyroid surgery in general, not cancer surgery specifically.Qualitative history. Nobel Prize in Physiology or Medicine 1909 to Theodor Kocher alone, for work on the physiology, pathology and surgery of the thyroid gland (awarded 28 October 1909). In a lecture on 4 April 1883 Kocher reviewed 102 operated cases; of 34 total excisions, 3 patients died of the operation and 1 had a cancerous goitre. He named the later condition cachexia strumipriva (Tröhler 2011) and afterwards took care to leave part of the gland intact (van Gijn 2012).

Chemotherapy

1946 · Treatment · RadiationRadioiodine therapy of functioning thyroid cancer metastasesIn the 1940s a physician at Montefiore Hospital in New York, Samuel Seidlin, gave radioactive iodine to people whose thyroid cancer had spread. A 1999 memoir dates the first such patient to 1943 and says some tumours took up the iodine only after the normal thyroid was removed. His group's report appeared in JAMA in 1946.Historical series; the 1946 JAMA report (Seidlin, Marinelli, Oshry) has no PubMed abstract and was not read, so no outcome figures are recorded. Per a 1999 memoir (Siegel): first patient 1943, previously thyroidectomised with functioning metastases of thyroid adenocarcinoma; Seidlin held that removing normal thyroid raises TSH and removes competition for iodine, inducing metastases to function; patients followed 1942 to 1955. Hertz and Roberts treated hyperthyroidism (31 March 1941).

Molecular origins

Targeted therapy

Immunotherapy

Precision & engineered

Then and now

5-year relative survival · United States, SEER 8 registries
- 92.95%
  1975 (observed value; first year in the table)
All races, both sexes; SEER Cancer Stat Facts category 'thyroid cancer' (papillary, follicular, medullary and anaplastic cancers are named on the page as the four main types, but the series is not split by type). SEER 8 five-year relative survival series, 1975–2018. The page states no age range for this series.
Modeled trend value for that year: 92.64%. The lowest observed value in the table is 90.97% in 1977. Relative survival excludes deaths from other causes. 'Observed' is the table's label for the yearly value, as opposed to the Joinpoint modeled-trend value in the adjacent column. The chart note reads 'SEER 8 5-Year Relative Survival Percent from 1975–2018, All Races, Both Sexes'; the table column is headed 'Year' and the page does not say it is the year of diagnosis. The page does not say what accounts for the change in this series and attributes it to no cause. Davies and Welch (PMID 24557566) interpreted a different period's incidence and mortality pattern as overdiagnosis, and the USPSTF recommendation page states that overdiagnosis and overtreatment are likely consequences of screening; neither document analyses this survival series. Read the series beside the incidence and death-rate values given with the mortality records. Compare only with other values from this same SEER 8 series, not with the SEER 21 values.
Source
5-year relative survival · United States, SEER 8 registries
- 98.87%
  2009 (observed value; highest observed value in the table)
All races, both sexes; SEER Cancer Stat Facts category 'thyroid cancer' (papillary, follicular, medullary and anaplastic cancers are named on the page as the four main types, but the series is not split by type). SEER 8 five-year relative survival series, 1975–2018. The page states no age range for this series.
Modeled trend value for that year: 98.37%. The table's modeled-trend column peaks at 98.48% in 2010. Relative survival excludes deaths from other causes. 'Observed' is the table's label for the yearly value, as opposed to the Joinpoint modeled-trend value in the adjacent column. The chart note reads 'SEER 8 5-Year Relative Survival Percent from 1975–2018, All Races, Both Sexes'; the table column is headed 'Year' and the page does not say it is the year of diagnosis. The page does not say what accounts for the change in this series and attributes it to no cause. Davies and Welch (PMID 24557566) interpreted a different period's incidence and mortality pattern as overdiagnosis, and the USPSTF recommendation page states that overdiagnosis and overtreatment are likely consequences of screening; neither document analyses this survival series. Read the series beside the incidence and death-rate values given with the mortality records. Compare only with other values from this same SEER 8 series, not with the SEER 21 values.
Source
5-year relative survival · United States, SEER 8 registries
- 98.43%
  2018 (observed value; last observed year in the table)
All races, both sexes; SEER Cancer Stat Facts category 'thyroid cancer' (papillary, follicular, medullary and anaplastic cancers are named on the page as the four main types, but the series is not split by type). SEER 8 five-year relative survival series, 1975–2018. The page states no age range for this series.
Modeled trend value for that year: 98.26%. Later years (2019–2023) show a modeled trend value only (98.23% in 2019, 98.11% in 2023). Relative survival excludes deaths from other causes. 'Observed' is the table's label for the yearly value, as opposed to the Joinpoint modeled-trend value in the adjacent column. The chart note reads 'SEER 8 5-Year Relative Survival Percent from 1975–2018, All Races, Both Sexes'; the table column is headed 'Year' and the page does not say it is the year of diagnosis. The page does not say what accounts for the change in this series and attributes it to no cause. Davies and Welch (PMID 24557566) interpreted a different period's incidence and mortality pattern as overdiagnosis, and the USPSTF recommendation page states that overdiagnosis and overtreatment are likely consequences of screening; neither document analyses this survival series. Read the series beside the incidence and death-rate values given with the mortality records. Compare only with other values from this same SEER 8 series, not with the SEER 21 values.
Source
5-year relative survival · United States, SEER 21 (excluding IL)
- 98.3%
  2016–2022
Thyroid cancer, all types together (the headline 5-year relative survival on the page). The page does not label the headline figure with an age range, race or sex; the stage table beneath it is labelled 'All Races, Both Sexes'.
Relative survival excludes deaths from other causes. SEER 21 (excluding IL), 2016–2022, is a different registry set and period from the SEER 8 series (1975–2018): do not join them into one line. By SEER Combined Summary Stage the page gives 99.9% (localized, 63% of cases), 98.1% (regional, 31%), 48.3% (distant, 3%) and 91.9% (unknown or unstaged, 3%); its prose says 63.5% are diagnosed at the localized stage and the table shows 63%. The page gives one figure for all thyroid cancers and does not show papillary, follicular, medullary and anaplastic cancers separately, so it cannot show how survival differs between types. The page makes no statement about detection or overdiagnosis. Compare only with other values from this SEER 21 series.
Source
5-year relative survival · United States, SEER 21 (excluding IL)
- 99.9%
  2016–2022; localized stage at diagnosis (63% of cases)
All races, both sexes; thyroid cancer, all types together; localized stage (confined to the primary site) at diagnosis.
Relative survival excludes deaths from other causes. SEER 21 (excluding IL), 2016–2022, is a different registry set and period from the SEER 8 series (1975–2018): do not join them into one line. By SEER Combined Summary Stage the page gives 99.9% (localized, 63% of cases), 98.1% (regional, 31%), 48.3% (distant, 3%) and 91.9% (unknown or unstaged, 3%); its prose says 63.5% are diagnosed at the localized stage and the table shows 63%. The page gives one figure for all thyroid cancers and does not show papillary, follicular, medullary and anaplastic cancers separately, so it cannot show how survival differs between types. The page makes no statement about detection or overdiagnosis. Compare only with other values from this SEER 21 series.
Source
5-year relative survival · United States, SEER 21 (excluding IL)
- 98.1%
  2016–2022; regional stage at diagnosis (31% of cases)
All races, both sexes; thyroid cancer, all types together; regional stage (spread to regional lymph nodes) at diagnosis.
Relative survival excludes deaths from other causes. SEER 21 (excluding IL), 2016–2022, is a different registry set and period from the SEER 8 series (1975–2018): do not join them into one line. By SEER Combined Summary Stage the page gives 99.9% (localized, 63% of cases), 98.1% (regional, 31%), 48.3% (distant, 3%) and 91.9% (unknown or unstaged, 3%); its prose says 63.5% are diagnosed at the localized stage and the table shows 63%. The page gives one figure for all thyroid cancers and does not show papillary, follicular, medullary and anaplastic cancers separately, so it cannot show how survival differs between types. The page makes no statement about detection or overdiagnosis. Compare only with other values from this SEER 21 series.
Source
5-year relative survival · United States, SEER 21 (excluding IL)
- 48.3%
  2016–2022; distant stage at diagnosis (3% of cases)
All races, both sexes; thyroid cancer, all types together; distant stage (cancer has metastasized) at diagnosis.
Relative survival excludes deaths from other causes. SEER 21 (excluding IL), 2016–2022, is a different registry set and period from the SEER 8 series (1975–2018): do not join them into one line. By SEER Combined Summary Stage the page gives 99.9% (localized, 63% of cases), 98.1% (regional, 31%), 48.3% (distant, 3%) and 91.9% (unknown or unstaged, 3%); its prose says 63.5% are diagnosed at the localized stage and the table shows 63%. The page gives one figure for all thyroid cancers and does not show papillary, follicular, medullary and anaplastic cancers separately, so it cannot show how survival differs between types. The page makes no statement about detection or overdiagnosis. Compare only with other values from this SEER 21 series.
Source
Age-standardised mortality · United States
- 0.54per 100,000
  1975 (observed value; first year in the table)
Both sexes, all races; thyroid cancer, all types together; age-adjusted death rate in the 'Death Rate — U.S.' column (the page gives U.S. Mortality as the data source and does not name the standard population). Counts deaths in the whole U.S. population, not deaths after a diagnosis.
Modeled trend value for that year: 0.56 per 100,000. The observed value is highest in 1977 at 0.57. The observed values run between 0.43 (1991 and 1994) and 0.57 (1977) per 100,000 over 1975–2024. For comparison, the page's rate of new cases (SEER 8 column, both sexes, all races, age-adjusted) is 5.00 per 100,000 in 1975, 14.94 in 2014 (the highest observed value), 13.74 in 2018 and 14.86 in 2023 (the 2020 value, 12.03, is displayed but not used in the trend fit); the page's text says incidence was stable over 2014–2023 and death rates stable over 2015–2024 in its statistical models. Incidence and death rate are different measures, and the page attributes the pattern to no cause. Davies and Welch (PMID 24557566; SEER 9 areas, 1975–2009, incidence 4.9 to 14.3 per 100,000, mortality about 0.5 per 100,000 throughout) interpreted a similar pattern, driven by papillary cancer, as overdiagnosis, an 'epidemic of diagnosis'; that interpretation is theirs and is not stated on the SEER page. Mortality and survival after diagnosis are different measures; the page does not link them (see the survival records).
Source
Age-standardised mortality · United States
- 0.43per 100,000
  1991 (observed value; lowest observed value in the table, equalled in 1994)
Both sexes, all races; thyroid cancer, all types together; age-adjusted death rate in the 'Death Rate — U.S.' column (the page gives U.S. Mortality as the data source and does not name the standard population). Counts deaths in the whole U.S. population, not deaths after a diagnosis.
Modeled trend value for that year: 0.44 per 100,000. The observed values run between 0.43 (1991 and 1994) and 0.57 (1977) per 100,000 over 1975–2024. For comparison, the page's rate of new cases (SEER 8 column, both sexes, all races, age-adjusted) is 5.00 per 100,000 in 1975, 14.94 in 2014 (the highest observed value), 13.74 in 2018 and 14.86 in 2023 (the 2020 value, 12.03, is displayed but not used in the trend fit); the page's text says incidence was stable over 2014–2023 and death rates stable over 2015–2024 in its statistical models. Incidence and death rate are different measures, and the page attributes the pattern to no cause. Davies and Welch (PMID 24557566; SEER 9 areas, 1975–2009, incidence 4.9 to 14.3 per 100,000, mortality about 0.5 per 100,000 throughout) interpreted a similar pattern, driven by papillary cancer, as overdiagnosis, an 'epidemic of diagnosis'; that interpretation is theirs and is not stated on the SEER page. Mortality and survival after diagnosis are different measures; the page does not link them (see the survival records).
Source
Age-standardised mortality · United States
- 0.5per 100,000
  2024 (observed value; last year in the table)
Both sexes, all races; thyroid cancer, all types together; age-adjusted death rate in the 'Death Rate — U.S.' column (the page gives U.S. Mortality as the data source and does not name the standard population). Counts deaths in the whole U.S. population, not deaths after a diagnosis.
Modeled trend value for that year: 0.50 per 100,000. The page's text gives 0.5 per 100,000 based on 2020–2024 deaths. The observed values run between 0.43 (1991 and 1994) and 0.57 (1977) per 100,000 over 1975–2024. For comparison, the page's rate of new cases (SEER 8 column, both sexes, all races, age-adjusted) is 5.00 per 100,000 in 1975, 14.94 in 2014 (the highest observed value), 13.74 in 2018 and 14.86 in 2023 (the 2020 value, 12.03, is displayed but not used in the trend fit); the page's text says incidence was stable over 2014–2023 and death rates stable over 2015–2024 in its statistical models. Incidence and death rate are different measures, and the page attributes the pattern to no cause. Davies and Welch (PMID 24557566; SEER 9 areas, 1975–2009, incidence 4.9 to 14.3 per 100,000, mortality about 0.5 per 100,000 throughout) interpreted a similar pattern, driven by papillary cancer, as overdiagnosis, an 'epidemic of diagnosis'; that interpretation is theirs and is not stated on the SEER page. Mortality and survival after diagnosis are different measures; the page does not link them (see the survival records).
Source

Survival figures are affected by earlier detection (lead time) and overdiagnosis, and by changes in who is diagnosed. For screened cancers, read them alongside mortality.

Evidence chains

A lane follows one defined population and setting, where each new trial tests against what was then standard. Compare only within a lane, and read every comparison as illustrative.

Radioiodine ablation after thyroidectomy: lower versus higher activity

Patients after total thyroidectomy for differentiated thyroid cancer at low risk (HiLo: stage T1 to T3, no distant metastasis, age 16 to 80 years; ESTIMABL1: age 18 years or older, defined pT1 or pT2 categories; per the abstracts) · Adjuvant · endpoint: Ablation success rate

Results are not comparable across trials.

This lane shows a landmark rate (ablation success at 6 to 9 months), not a time-to-event measure. Only HiLo is drawn: it compares the 1.1 GBq and 3.7 GBq groups with both stimulation methods pooled. ESTIMABL1 tested the same two activities in a similar 2-by-2 design, but the abstract gives only an overall ablation rate and no rate by arm, so its result is described in the milestone and not drawn. Eligibility, the definition of ablation success and the timing of assessment differ between the two trials (HiLo at 6 to 9 months; ESTIMABL1 at 8 months by neck ultrasonography and stimulated thyroglobulin). These trials tested a lower activity, not omission of radioiodine. Ablation success is a surrogate and says nothing here about recurrence or survival. Compare only within each trial's own arms.

Each pair shows one trial: its control arm and its experimental arm. Yesterday's experimental arm is often today's control. Populations, definitions and follow-up differ.

experimental arm
control arm
different endpoint from the lane's
step line through experimental values

Ablation success rate, rate, across 1 trial in this lane. HiLo (2012): experimental 85 % vs control 88.9 %.

Data table

Trial	Year	Endpoint	Control (%)	Experimental (%)
HiLo	2012	Ablation success rate	88.9 %	85 %

Low-risk differentiated thyroid cancer: radioiodine ablation versus none

Patients with low-risk differentiated thyroid cancer after total thyroidectomy (ESTIMABL2: eligibility as defined in the paper; IoN: complete R0 resection with pT1 to pT3 (TNM7) or pT3a (TNM8) and N0, Nx or N1a disease, per the abstract) · Adjuvant · endpoint: RFS

No comparable values to plot for this lane yet.

Radioiodine-refractory differentiated thyroid cancer: kinase inhibitor versus placebo

Locally advanced or metastatic radioiodine-refractory differentiated thyroid cancer with progressive disease. DECISION (sorafenib) and SELECT (lenvatinib) were run before COSMIC-311 (cabozantinib), which required prior VEGFR-targeted therapy · Metastatic · endpoint: PFS

Results are not comparable across trials.

Each drug is compared only with its own placebo arm. The three trials ran in different eras and populations: COSMIC-311 required prior VEGFR-targeted therapy and DECISION and SELECT did not, and they differ in progression window, geography and the proportion of patients crossing over to the active drug after progression. COSMIC-311's interim report gives a median only for the placebo arm (the cabozantinib median was not reached) and a hazard ratio, so it is listed in the data table but not drawn. SELECT's hazard ratio carries a 99% interval and COSMIC-311's a 96% interval, not 95%. Medians and hazard ratios from different trials are not pooled or ranked.

Each pair shows one trial: its control arm and its experimental arm. Yesterday's experimental arm is often today's control. Populations, definitions and follow-up differ.

experimental arm
control arm
different endpoint from the lane's
step line through experimental values

PFS, median, across 2 trials in this lane. DECISION (2013): experimental 10.8 months vs control 5.8 months; SELECT (2015): experimental 18.3 months vs control 3.6 months.

Data table

Trial	Year	Endpoint	Control (months)	Experimental (months)
DECISION	2013	PFS	5.8 months	10.8 months
SELECT	2015	PFS	3.6 months	18.3 months

Advanced medullary thyroid cancer: kinase inhibitor versus placebo

Unresectable locally advanced or metastatic medullary thyroid cancer. ZETA randomised 2:1 to vandetanib 300 mg/day or placebo; EXAM randomised 2:1 to cabozantinib 140 mg/day or placebo in patients with documented radiographic progression (per the abstracts) · Metastatic · endpoint: PFS

Results are not comparable across trials.

ZETA's abstract reports a hazard ratio for progression-free survival and no median for either arm, so ZETA is listed in the data table but not drawn; only EXAM's medians can be drawn. The trials enrolled different populations (ZETA locally advanced or metastatic disease; EXAM metastatic disease with documented radiographic progression) with different RET mutation and prior-therapy mix. Compare each drug only with its own placebo arm. EXAMINER (a phase 4 comparison of cabozantinib 60 mg tablets with 140 mg capsules) is a dose comparison and LIBRETTO-531 (selpercatinib versus cabozantinib or vandetanib) has an active control, so neither is placed beside the placebo-controlled trials. Single-arm studies in RET-altered thyroid cancer are not drawn. Overall survival did not differ significantly in EXAM and was immature in ZETA.

Each pair shows one trial: its control arm and its experimental arm. Yesterday's experimental arm is often today's control. Populations, definitions and follow-up differ.

experimental arm
control arm
different endpoint from the lane's
step line through experimental values

PFS, median, across 1 trial in this lane. EXAM (2012): experimental 11.2 months vs control 4 months.

Data table

Trial	Year	Endpoint	Control (months)	Experimental (months)
EXAM	2012	PFS	4 months	11.2 months

Standard of care, then and now

How the usual sequence of treatment changed for one setting. Choose a point in time; each step links to the milestone behind it.

Radioiodine-refractory differentiated thyroid cancer: systemic therapy

Progressive, locally advanced or metastatic differentiated thyroid cancer that no longer responds to radioactive iodine · Metastatic

2021. Based only on the trials cited in each step. Cabozantinib was approved for disease that progressed after VEGFR-targeted therapy; targeted options for RET, NTRK and BRAF alterations are described in other milestones.

Text version

1946: Radioiodine for functioning metastases
2013: Radioiodine for functioning metastases → Progressive, iodine-refractory: sorafenib
2015: Radioiodine for functioning metastases → Progressive, iodine-refractory: sorafenib → Progressive, iodine-refractory: lenvatinib
2021: Radioiodine for functioning metastases → Progressive, iodine-refractory: sorafenib → Progressive, iodine-refractory: lenvatinib → After VEGFR therapy: cabozantinib

Advanced medullary thyroid cancer: systemic therapy

Progressive, unresectable locally advanced or metastatic medullary thyroid cancer · Metastatic

2023. Based only on the studies cited in each step. A randomised trial compared selpercatinib with cabozantinib or vandetanib in kinase inhibitor-naive RET-mutant disease (interim analysis; open-label).

Text version

2011: Symptomatic or progressive: vandetanib
2012: Symptomatic or progressive: vandetanib → Progressive, metastatic: cabozantinib
2020: Symptomatic or progressive: vandetanib → Progressive, metastatic: cabozantinib → RET-mutant: selpercatinib
2023: Symptomatic or progressive: vandetanib → Progressive, metastatic: cabozantinib → RET-mutant, first line: selpercatinib

Reversals and dead ends

Approaches that were abandoned or withdrawn, and trials whose results were later questioned, are part of the story.

2023
NRG/RTOG 0912: pazopanib plus chemoradiotherapy in anaplastic cancer
The trial report describes anaplastic thyroid cancer as having no standard radiotherapy-based local treatment. A US randomised trial added pazopanib or placebo to paclitaxel plus radiotherapy. Survival was not significantly better with pazopanib (median 5.7 versus 7.3 months). The authors report the approach was feasible and that the data may generate hypotheses.
NRG/RTOG 0912 (double-blind phase 2; 71 eligible of 89 enrolled): paclitaxel plus IMRT 66 Gy with pazopanib (n=36) vs placebo (n=35). Median OS 5.7 months (95% CI 4.0 to 12.8) vs 7.3 (4.3 to 10.6); HR 0.86 (95% CI 0.52 to 1.43; one-sided log-rank p=0.28). 1-year OS 37.1% vs 29.0%. Grade 3-5 adverse events 88.9% vs 85.3% (p=0.73).
2023
Pralsetinib in RET thyroid cancer; MTC indication withdrawn 2023
Pralsetinib, a RET-targeted pill, shrank tumours in roughly 56 to 77 of every 100 people with RET-mutant medullary thyroid cancer in a single-arm study, and gained US accelerated approval in 2020. In 2023 that use was withdrawn at the sponsor's request after the confirmatory trial proved hard to run, not, the sponsor said, because of new safety or efficacy data.
ARROW thyroid cohorts (phase 1-2, single-arm, pralsetinib 400 mg daily; masked central review). Interim: ORR 71% (15 of 21) treatment-naive and 60% (33 of 55) cabozantinib/vandetanib-pretreated RET-mutant MTC; 89% (8 of 9) RET fusion-positive thyroid cancer. Update (cutoff 18 October 2021): ORR 77.4% (62 patients), 55.7% (61), 90.9% (22). Accelerated approval 1 December 2020; MTC indication removed 20 July 2023 at sponsor request; phase 3 AcceleRET-MTC withdrawn, no patients enrolled.

Today and tomorrow

Standard of care today

Screening: US Preventive Services Task Force recommendation page (final recommendation statement dated 9 May 2017; page text read on 20 October 2026): the Task Force recommends against screening for thyroid cancer in asymptomatic adults (grade D) and concludes with moderate certainty that screening results in harms that outweigh the benefits. The page says the recommendation applies to screening in asymptomatic adults and not to people with throat or neck symptoms or lumps, nor to people at increased risk (its examples include radiation exposure and inherited syndromes). Its rationale states that evidence is inadequate to estimate the accuracy of neck palpation or ultrasound as screening tests and that overdiagnosis and overtreatment are likely consequences of screening. Screening positions of other bodies were not read.; US Preventive Services Task Force 2017
Prevention: Hereditary medullary thyroid cancer (multiple endocrine neoplasia type 2A):
in a 2005 report from a genetic screening programme, 50 carriers of a RET mutation characteristic of MEN 2A, aged 19 years or younger, underwent total thyroidectomy
5 to 10 years later, basal and stimulated calcitonin levels were at or below the assay's detection limit in 44 of 50 (proportion 0.88
95% CI 0.76 to 0.95). The abstract describes a single group of 50 patients and no comparison group, and the authors wrote that a longer evaluation period would be needed to confirm that those patients were free of disease. The American Thyroid Association medullary thyroid carcinoma guideline (2015
PMID 25810047) is cited by identifier only
its text was not opened and no recommendation on timing of surgery is stated; Skinner MA, Moley JA, Dilley WG, Owzar K, Debenedetti MK, Wells SA Jr 2005 Guideline: Wells SA Jr, Asa SL, Dralle H, Elisei R, Evans DB, Gagel RF, et al 2015
Adjuvant: Low-risk differentiated thyroid cancer after thyroidectomy:
two randomised non-inferiority trials tested leaving out postoperative radioiodine. In ESTIMABL2 (730 patients evaluable at 3 years), the percentage without an event (a composite of functional, structural and biologic abnormalities) was 95.6% (95% CI 93.0 to 97.5) in the no-radioiodine group and 95.9% (95% CI 93.3 to 97.7) in the radioiodine group, a difference of -0.3 percentage points (two-sided 90% CI -2.7 to 2.2) against a non-inferiority margin of 5 percentage points. In IoN (UK
504 randomised), 5-year recurrence-free rates in the intention-to-treat analysis were 97.9% (95% CI 96.1 to 99.7) without ablation and 96.3% (95% CI 93.9 to 98.7) with ablation, an absolute difference of 0.5 percentage points (95% CI -2.2 to 3.2) against a margin of 5 percentage points
the authors' interpretation applies to pT1, pT2 and N0 or Nx tumours without adverse features, and recurrences were more frequent with pT3 or pT3a tumours and with N1a disease. Intermediate-risk disease is the subject of an open trial (INTERMEDIATE
see the horizon). The American Thyroid Association guidelines (2015, PMID 26462967
2025, PMID 40844370), the ESMO guideline (2019, PMID 31549998) and the NCCN Guidelines Insights, version 1.2025 (PMID 40639400) are cited by identifier only
their text was not opened and no recommendation from them is stated; Leboulleux S, Bournaud C, Chougnet CN, Zerdoud S, Al Ghuzlan A, Catargi B, et al 2022 Mallick U, Newbold K, Beasley M, Garcez K, Wadsley J, Johnson SJ, et al 2025 Haugen BR, Alexander EK, Bible KC, Doherty GM, Mandel SJ, Nikiforov YE, et al 2016 Ringel MD, Sosa JA, Baloch Z, Bischoff L, Bloom G, Brent GA, et al 2025 Guideline: Filetti S, Durante C, Hartl D, Leboulleux S, Locati LD, Newbold K, et al 2019 Guideline: Haddad RI, Bischoff L, Applewhite M, Bernet V, Blomain E, Brito M, et al 2025
Locally advanced: Anaplastic thyroid cancer:
on 4 May 2018 the FDA approved dabrafenib in combination with trametinib for locally advanced or metastatic anaplastic thyroid cancer with a BRAF V600E mutation and no satisfactory locoregional treatment options (supplement approval letters for both drugs
neither letter uses accelerated-approval wording, so the approval type is an inference). The supporting evidence is ROAR, an open-label, non-randomised phase 2 basket study: in the full cohort of 36 patients with unresectable or metastatic disease (data cut-off 14 September 2020
median follow-up 11.1 months), the investigator-assessed objective response rate was 56% (95% CI 38.1 to 72.1), including three complete responses, median progression-free survival was 6.7 months, median overall survival was 14.5 months and 12-month overall survival was 51.7%. In NRG/RTOG 0912, a randomised phase 2 trial in 71 eligible patients, adding pazopanib to paclitaxel and radiotherapy did not significantly improve overall survival: median 5.7 months with pazopanib and 7.3 months with placebo (hazard ratio 0.86
95% CI 0.52 to 1.43
one-sided log-rank p=0.28). The American Thyroid Association anaplastic thyroid cancer guidelines (2021
PMID 33728999) are cited by identifier only
their text was not opened and no recommendation is stated; U 2018 U 2018 Subbiah V, Kreitman RJ, Wainberg ZA, Cho JY, Schellens JHM, Soria JC, et al 2022 Sherman EJ, Harris J, Bible KC, Xia P, Ghossein RA, Chung CH, et al 2023 Guideline: Bible KC, Kebebew E, Brierley J, Brito JP, Cabanillas ME, Clark TJ Jr, et al 2021
Metastatic: Radioiodine-refractory differentiated thyroid cancer, FDA actions read in approval letters:
sorafenib for locally recurrent or metastatic, progressive differentiated thyroid carcinoma refractory to radioactive iodine (22 November 2013)
lenvatinib for locally recurrent or metastatic, progressive, radioactive iodine-refractory differentiated thyroid cancer (13 February 2015, the effective date stated in a replacement letter)
cabozantinib tablets for adults and children 12 years and older with locally advanced or metastatic differentiated thyroid cancer that has progressed after prior VEGFR-targeted therapy and is radioactive iodine-refractory or ineligible (17 September 2021, effective date stated in a corrected letter). These three letters carry no accelerated-approval wording, so their approval type is an inference. For RET fusion-positive thyroid cancer, selpercatinib received accelerated approval on 8 May 2020 and a letter of 12 June 2024 provides for conversion to traditional approval for patients 2 years and older
pralsetinib received accelerated approval on 1 December 2020 and the revised prescribing information dated 12/2025, issued with the 22 December 2025 supplement approval, still lists RET fusion-positive thyroid cancer as an accelerated approval. Trial results in the abstracts: in DECISION (sorafenib, 417 patients) median progression-free survival was 10.8 months versus 5.8 months with placebo (hazard ratio 0.59
95% CI 0.45 to 0.76)
in COSMIC-311 (cabozantinib after VEGFR-targeted therapy, 187 patients) the interim analysis gave median progression-free survival not reached versus 1.9 months with placebo (hazard ratio 0.22
96% CI 0.13 to 0.36). EMA authorisation pages read: Lenvima (marketing authorisation issued 28 May 2015
monotherapy for adults with progressive, locally advanced or metastatic differentiated thyroid carcinoma refractory to radioactive iodine)
Nexavar (marketing authorisation issued 19 July 2006
the same thyroid indication appears in the page's product information, and a news item dated 25 April 2014 says EMA recommended extending its use to differentiated thyroid cancer
the extension decision date was not shown)
Cabometyx (marketing authorisation issued 9 September 2016
adults with locally advanced or metastatic differentiated thyroid carcinoma refractory or not eligible for radioactive iodine who progressed during or after prior systemic therapy
the extension date was not shown)
Retsevmo (conditional authorisation issued 11 February 2021, later switched to standard authorisation without a date on the page
lists advanced RET fusion-positive thyroid cancer that is radioactive iodine-refractory). The EU authorisation for Gavreto (granted 18 November 2021 for lung cancer) was withdrawn on 24 October 2024 at the holder's request for commercial reasons. The ESMO systemic therapy update (2022
PMID 35491008) and the American Thyroid Association 2025 guideline are cited by identifier only
their text was not opened and no sequencing recommendation is stated; U 2013 U 2015 U 2021 US Food and Drug Administration 2020 US Food and Drug Administration 2024 US Food and Drug Administration 2020 US Food and Drug Administration 2025 Brose MS, Nutting CM, Jarzab B, Elisei R, Siena S, Bastholt L, et al 2014 Brose MS, Robinson B, Sherman SI, Krajewska J, Lin CC, Vaisman F, et al 2021 European Medicines Agency 2026 European Medicines Agency 2026 European Medicines Agency 2026 European Medicines Agency 2026 European Medicines Agency 2025 Guideline: Filetti S, Durante C, Hartl DM, Leboulleux S, Locati LD, Newbold K, et al 2022 Ringel MD, Sosa JA, Baloch Z, Bischoff L, Bloom G, Brent GA, et al 2025
Metastatic: Advanced medullary thyroid cancer, FDA actions read in approval letters:
vandetanib for symptomatic or progressive medullary thyroid cancer in patients with unresectable locally advanced or metastatic disease (6 April 2011
the letter says use in indolent, asymptomatic or slowly progressing disease should be carefully considered because of treatment-related risks)
cabozantinib capsules for progressive, metastatic medullary thyroid cancer (29 November 2012)
selpercatinib for RET-mutant medullary thyroid cancer (accelerated approval 8 May 2020
conversion to traditional approval on 27 September 2024, the letter carrying the heading 'Subpart H fulfilled' and listing the LIBRETTO-531 primary analysis among the requirements fulfilled). The pralsetinib indication for RET-mutant medullary thyroid cancer, approved under accelerated approval on 1 December 2020, was removed from the US prescribing information by a supplement approved 20 July 2023
the letter states that, after a meeting with FDA about the challenges of fulfilling the confirmatory study, the sponsor requested withdrawal and waived the expedited withdrawal procedures. In LIBRETTO-531 (291 randomised
selpercatinib versus physician's choice of cabozantinib or vandetanib
interim efficacy analysis, median follow-up 12 months), median progression-free survival by blinded independent central review was not reached with selpercatinib and was 16.8 months in the control group (hazard ratio 0.28
95% CI 0.16 to 0.48), and crossover to selpercatinib was permitted. EMA pages read: Caprelsa (marketing authorisation granted 17 February 2012 according to the page
the page's summary text says it was given 'conditional approval'
indicated for aggressive and symptomatic medullary thyroid cancer in patients aged 5 years and older)
Cometriq (conditional authorisation issued 21 March 2014)
Retsevmo (see above
lists advanced RET-mutant medullary thyroid cancer). The American Thyroid Association medullary thyroid carcinoma guideline (2015
PMID 25810047) and the ESMO systemic therapy update (2022
PMID 35491008) are cited by identifier only
their text was not opened and no recommendation is stated; US Food and Drug Administration 2011 US Food and Drug Administration 2012 US Food and Drug Administration 2020 US Food and Drug Administration 2024 US Food and Drug Administration 2020 US Food and Drug Administration 2023 Hadoux J, Elisei R, Brose MS, Hoff AO, Robinson BG, Gao M, et al 2023 European Medicines Agency 2026 European Medicines Agency 2026 European Medicines Agency 2026 Guideline: Wells SA Jr, Asa SL, Dralle H, Elisei R, Evans DB, Gagel RF, et al 2015 Guideline: Filetti S, Durante C, Hartl DM, Leboulleux S, Locati LD, Newbold K, et al 2022
Relapsed / refractory: Tumour-agnostic FDA actions read in approval letters that cover thyroid cancer only as one of many tumour types:
dabrafenib with trametinib for unresectable or metastatic solid tumours with a BRAF V600E mutation that have progressed after prior treatment and have no satisfactory alternative options, in patients 6 years and older, not for colorectal cancer (accelerated approval, 22 June 2022)
selpercatinib for solid tumours with a RET gene fusion (accelerated approval 21 September 2022
conversion to traditional approval 14 July 2026, letter headed 'Subpart H fulfilled')
larotrectinib (accelerated approval 26 November 2018
conversion to regular approval 9 April 2025, letter headed 'Subpart H fulfilled') and entrectinib (accelerated approval 15 August 2019) for solid tumours with an NTRK gene fusion. In a pooled analysis of three phase 1 and 2 larotrectinib trials in 29 patients with TRK fusion-positive thyroid carcinoma (data cut-off July 2020), the investigator-assessed objective response rate was 71% (95% CI 51 to 87) among 28 evaluable patients
it was 86% for papillary and follicular histologies and 29% for anaplastic histology
the abstract describes no comparison group. EMA: the Vitrakvi page states conditional marketing authorisation on 19 September 2019 for solid tumours with an NTRK gene fusion. Which of these labels apply to an individual person with thyroid cancer depends on the label wording and on testing, and no guideline text was read; U 2022 US Food and Drug Administration 2022 US Food and Drug Administration 2026 US Food and Drug Administration 2018 US Food and Drug Administration 2025 US Food and Drug Administration 2019 Waguespack SG, Drilon A, Lin JJ, Brose MS, McDermott R, Almubarak M, et al 2022 European Medicines Agency 2026

Trials to watch

Dabrafenib plus trametinib versus cabozantinib in BRAF V600E thyroid cancer (ECOG-ACRIN) NCT06475989
In BRAF V600E-mutant differentiated thyroid cancer that has progressed after lenvatinib or sorafenib (up to two prior VEGFR-targeting agents allowed) and for which radioactive iodine has been used or is not suitable, how does cabozantinib compare with dabrafenib plus trametinib in progression-free survival (the registry's primary outcome)? Registry record: randomised, open-label phase 3, 264 estimated, sponsor ECOG-ACRIN Cancer Research Group; the record labels dabrafenib plus trametinib as the active comparator and cabozantinib as the experimental arm. Registry status: recruiting; start 2024-12-27; registry primary completion date 2030-09-30 (estimated); no results posted; record last updated 2026-09-24.
INTERMEDIATE NCT04290663
In intermediate-risk differentiated thyroid cancer after total thyroidectomy (risk defined in the record by TNM 2017 criteria), how does a strategy that decides on radioiodine after a post-operative work-up (serum thyroglobulin and diagnostic radioiodine scintigraphy) compare with systematic radioiodine administration in the rate of patients with an excellent tumoral response 36 months after randomisation (the registry's primary outcome)? Registry record: randomised, open-label phase 3, 368 estimated, sponsor Centre Francois Baclesse. Registry status: recruiting; start 2020-03-02; registry primary completion date 2031-02 (estimated); no results posted; record last updated 2026-06-25.
ESTIMABL3 NCT03570021
In clinically node-negative papillary thyroid cancer with a nodule of 11 to 40 mm, how does total thyroidectomy alone compare with total thyroidectomy plus prophylactic bilateral central compartment (level VI) neck dissection in the rate of patients in complete remission 1 year after randomisation (the registry's primary outcome)? Registry record: randomised, open-label non-inferiority phase 3 with thyroidectomy alone as the experimental group, 352 enrolled (actual), sponsor Gustave Roussy. Registry status: active, not recruiting; start 2018-08-08; registry primary completion date 2026-12 (estimated); no results posted; record last updated 2026-03-06.
AL2846 versus placebo after VEGFR-targeted therapy NCT06860971
In locally advanced or metastatic radioiodine-refractory differentiated thyroid cancer that has progressed after no more than two lines of VEGFR-targeted tyrosine kinase inhibitor therapy, how does AL2846 compare with placebo in progression-free survival assessed by an independent review committee (the registry's primary outcome)? Registry record: randomised, double-blind, placebo-controlled phase 3, 144 estimated, participants aged 18 to under 75 years, sponsor Chia Tai Tianqing Pharmaceutical Group; the record describes AL2846 as a multi-target tyrosine kinase inhibitor acting on c-MET, c-KIT, VEGFR1 and RET and gives no generic name. Registry status: recruiting; start 2025-04-18; registry primary completion date 2027-01 (estimated); no results posted; record last updated 2025-12-31.
Adjuvant radiotherapy in high-risk locally advanced differentiated thyroid cancer NCT06558981
After total or near-total thyroidectomy for differentiated thyroid cancer with R1 or R2 resection (residual tumour volume under 2 cm3 in R2) and no distant metastasis, does adding external-beam radiotherapy to radioiodine differ from radioiodine alone in local-regional recurrence-free survival at 5 years after surgery (the registry's primary outcome)? Registry record: randomised 1:1, open-label phase 3, 124 estimated, participants aged 14 years to under 80, sponsor Fudan University. Registry status: recruiting; start 2024-07-11; registry primary completion date 2031-06 (estimated); no results posted; record last updated 2026-02-27.
CAST-MTC NCT07383246
In newly diagnosed medullary thyroid carcinoma without distant metastasis on conventional imaging, how does surgery guided by 68Ga-CTR-FAPI PET-CT compare with surgery chosen by the investigator in a biochemical outcome rate measured 1 month after surgery (the registry's primary outcome)? Registry record: randomised 2:1, open-label non-inferiority phase 3, 150 estimated, sponsor Cancer Institute and Hospital, Chinese Academy of Medical Sciences. Registry status: recruiting; start 2025-08-25; registry primary completion date 2028-09-01 (estimated); no results posted; record last updated 2026-02-03.

Pioneers

Theodor Kocher
1841-1917 · Surgeon, Nobel laureate in Physiology or Medicine 1909
Samuel M. Seidlin
1895-1955 · Physician and investigator of radioiodine therapy for metastatic thyroid carcinoma
Saul Hertz
Physician at the Massachusetts General Hospital; early radioiodine research

References

Show all 105 references

Loading references… Open the list as a page

Every figure was checked against its primary source to the extent possible. Items that could not be fully confirmed are marked “needs review”. Educational history, not medical advice.

Report an error

Links open the primary source. A flag marked ⚑ needs review means a value or claim has not yet been independently confirmed.

2023ReversalTargeted

notableLocally advanced

NRG/RTOG 0912: pazopanib plus chemoradiotherapy in anaplastic cancer

The trial report describes anaplastic thyroid cancer as having no standard radiotherapy-based local treatment. A US randomised trial added pazopanib or placebo to paclitaxel plus radiotherapy. Survival was not significantly better with pazopanib (median 5.7 versus 7.3 months). The authors report the approach was feasible and that the data may generate hypotheses.

NRG/RTOG 0912 (double-blind phase 2; 71 eligible of 89 enrolled): paclitaxel plus IMRT 66 Gy with pazopanib (n=36) vs placebo (n=35). Median OS 5.7 months (95% CI 4.0 to 12.8) vs 7.3 (4.3 to 10.6); HR 0.86 (95% CI 0.52 to 1.43; one-sided log-rank p=0.28). 1-year OS 37.1% vs 29.0%. Grade 3-5 adverse events 88.9% vs 85.3% (p=0.73).

Standard before

The trial report describes anaplastic thyroid cancer as rare and aggressive, with no standard radiotherapy-based local treatment.

What changed

Adding pazopanib to concurrent paclitaxel and radiotherapy did not significantly improve overall survival; the report concludes the combination was feasible and safe in a multicentre cooperative-group setting and generated hypotheses that might warrant further investigation.

Why it mattered

NRG/RTOG 0912 was a randomised, double-blind, placebo-controlled test of adding pazopanib to chemoradiotherapy in anaplastic thyroid cancer. It reports no significant overall survival improvement in a rare disease.

Caveats

Phase 2 trial with small groups; the hazard ratio confidence interval (0.52 to 1.43) is wide and includes both benefit and harm. The p-value is a one-sided log-rank test.
The report states that 89 patients were enrolled to the randomised phase 2 trial, of whom 71 were eligible; the primary endpoint was overall survival in eligible intention-to-treat randomised patients. The full text says 18 of 89 (20%) were ineligible, almost all because laboratory or imaging studies were missing or outside protocol limits, and that the design used a one-sided significance level of 0.15 (0.1379 after a blinded protocol amendment); the observed one-sided p-value was 0.28.
Alanine aminotransferase increases of grade 3 or 4 occurred in 12 of 36 patients (33%) on pazopanib and none on placebo; one treatment-related death occurred in each group.