Recent data have shown an apparent increase in the incidence of thyroid cancer in the United States.¹ The annual percent change (for both men and women) from 1980 – 1997 was a 2.4% increase per year, but from 1997 to 2009, this increased to 6.6% per year.¹ There is disagreement on the exact cause of this increase – particularly whether it is an accurate reflection on thyroid cancer cases actually increasing, or rather due to increased screening.² Literature reviews suggests controversy, with evidence that could be interpreted to support either idea. Studies have shown that increased detection of the preclinical manifestations of thyroid cancer are a key contribution factor of the increased incidence rate.² However, others have reported evidence of a true increase in incidence, that cannot be fully explained simply by an increase in diagnosis – and thus suggesting that the increase could be due to lifestyle and environmental factors, as well.³

The carcinomas of the thyroid are demarcated and named for the primary cell type within the thyroid that is affected.⁴ Typically, they are all referred to as thyroid cancer. Thyroid cancer is a relatively rare cancer, but it is the most common cancer of the endocrine system.⁵ The four subtypes of this cancer are papillary carcinoma, follicular carcinoma, medullary carcinoma, and anaplastic carcinoma.⁶ Papillary carcinoma, which comprises approximately 80 – 85% of all thyroid cancers, has the highest prognosis, with ten-year survival being more than 95%.⁶ Although frequently diagnosed at early stages, papillary carcinoma can spread to the cervical lymph nodes, and is highly associated with exposure to ionizing radiation.⁶ Follicular carcinoma comprises ten percent of thyroid cancer cases, and is one of the few types among all cancer types to spread through hematogenous routes instead of lymph pathways. Medullary carcinoma (comprising five percent of thyroid cancer cases) is associated with hereditary mutations of the RET oncogene. Lastly, anaplastic (or undifferentiated) carcinomas – which have the poorest prognosis – comprise less than five percent of all thyroid cancer cases. Generally, anaplastic carcinomas afflict the elderly far more than those of younger ages. These carcinomas are typically fatal within a few years.⁶

The American Cancer Society suggests routine screening as part of a periodic physical exam and health counseling to include thyroid cancer screening for men and women 20 years and older.¹⁰ The USPSTF is currently reviewing and updating their thyroid cancer screening recommendations.¹¹ However, in 1996, their published recommendations for asymptomatic thyroid cancer screening using palpation or ultrasound was a grade D, meaning they recommend against the use of these screenings. Asymptomatic screening for those with childhood head and neck irradiation was a grade C, meaning it is suggested that these screenings are selectively offered based on patient preference and provider judgment.¹¹

The purpose of this study was to analyze thyroid cancer incidence and mortality across age and racial groups in Oklahoma against Surveillance, Epidemiology, and End Results (SEER) program data.

To determine incidence rates of thyroid cancer in Oklahoma, we used existing data from the Oklahoma Central Cancer Registry (OCCR), which includes data on all cancer cases diagnosed in Oklahoma residents from 1997–2012. The OCCR is a participant in the National Program of Cancer Registries and follows all guidelines of the North American Association of Central Cancer Registries. To obtain data on incidence of thyroid cancer in the US, we used the Surveillance, Epidemiology, and End Results (SEER) program data from 1997–2012. We restricted this analysis to those diagnosed with thyroid cancer using the SEER site recodes for thyroid cancer, which included International Classification of Diseases of Oncology, Third Edition/World Health Organization (ICDO-3/WHO) 2008 site codes of C739, excluding histology codes of 9050-9055, 9140, 9590-9992 (lymphomas, Kaposi sarcoma, and mesothelioma).¹² We used SAS v. 9.4 (Cary, NC) to calculate age-adjusted incidence rates (AAIR) per 100,000 by the direct method using the 2000 standard population.¹³ To calculate AAIR for the US, we used SEER*Stat software (v. 8.2.1) using the SEER 13 region data, which included more detailed information on race and Hispanic ethnicity.¹⁴ We compared AAIR and 95% confidence intervals over time from 1997–2012, by demographic groups of sex, and race/ethnicity.¹⁴ We classified race/ethnicity as white non-Hispanic (NH), African American NH, American Indian/Alaska Native (AI/AN) NH, Asian/Pacific Islander NH, and Hispanic. To reduce misclassification of the AI/AN population, we linked the cancer records with the Indian Health Service (IHS) records. For this study, AI/ANs who either reported their primary race as AI/AN or who linked with an IHS record regardless of the primary race selected, were reclassified as AI/AN. Age-specific incidence rates were calculated in five-year age groups, combining those under 20 years of age and 75 years and older for stability and confidentiality reasons. We also evaluated stage at diagnosis using the SEER Summary Stage 2000 for those diagnosed between 1998 and 2003 and the derived SEER Summary Stage 2000 for those diagnosed between 2004 and 2012 for Oklahoma and the Summary Stage 2000 for the SEER data from 1998–2012. We excluded cancers reported on a death certificate only or at autopsy for analysis of staging.

To evaluate survival from thyroid cancer, we used data from the OCCR and SEER using SEER*Stat software. We calculated five-year relative survival by age, sex, race, stage, and year of diagnosis for Oklahoma and SEER data using the expected survival life tables provided by SEER.¹⁶ By comparing the observed and expected survival in the population, relative survival allows for the estimation of excess deaths due to cancer in the population without the need for underlying cause of death.¹⁵ Expected survival was only available for the racial groups of white, African American, and other (AI/NA and Asians/Pacific Islanders). Hispanic ethnicity was also not available. We grouped year of diagnosis into 1997–2000, 2001–2003, and 2004–2008 for stability purposes. We used the same SEER Summary Stage classification as in the incidence estimates. The OCCR conducts routine follow-up on all cancer cases and links with the Oklahoma Mortality Data, the Social Security Death Index, and the National Death Index to determine mortality. Follow-up was conducted through November 1, 2014, thus we limited our analysis of survival from 1997–2008 to allow for more complete five-year survival for all cases of thyroid cancer.

This study was approved by the Institutional Review Boards of the University of Oklahoma Health Sciences Center and the Oklahoma State Department of Health.

Thyroid cancer is not a particularly common cause of cancer, even with the recent apparent increases in incidence. Historically thyroid cancer survivability has been very high – among the highest of all types of cancer, with the five-year relative rate having been consistently above ninety percent.¹ Currently, that rate is almost ninety-eight percent in the US.¹ Though the incidence continues to rise, there does not appear to be a concurrent effect on survival, as evidence by our data. The relationship between this admittedly minute uptick and the recent increases in incidence have not yet been conclusively demonstrated. The high overall relative survival suggests that the cases being detected are either not clinically important or are being effectively managed with the treatment options currently at hand.

Women in general, across all races, have a higher incidence of thyroid cancer than men.¹ In the United States, whites and Asians/Pacific Islanders have a higher incidence rate of thyroid cancer, with African Americans having a particularly low incidence rate.¹ However, this may not be reflective of actual incidence rates, and may be more reflective of increased detection (medical surveillance bias), in turn stemming from increased access to healthcare – which itself is correlated with increased wealth.¹⁷

The clinical stage at which thyroid cancer is first detected can have quite a pronounced effect on survivability. As is the case with most all cancer scenarios, an earlier diagnosis yields higher rates of survival. Metastasis to regional lymph nodes, which can make the cancer more likely to spread to other parts of the body,⁶ does not decrease survivability considerably. However, metastasis to more distant sites does dramatically reduce survivability (from nearly one hundred percent to almost half at fifty-four percent).¹

All the characteristic-specific rates in Oklahoma are lower than the corresponding rates in the United States. This can be because of a truly lower incidence across all groups, or from a lower rate of diagnosis across all groups – or even a combination of both. As mentioned previously, there is conflicting evidence and marked controversy about the recent increase in thyroid cancer incidence.^2,3 That same uncertainty can be present in Oklahoma as well. This disparity presents an intriguing avenue for future study. Furthermore, future studies should consider evaluating specific subtypes of thyroid cancer by race/ethnicity to understand whether disparities in incidence and survival exist. Oklahoma’s survivability figures were less than those of the United States in virtually every characteristic. This may be due to Oklahoma’s relative poverty compared to the United States as a whole.¹⁸ As established above, socioeconomic status (SES) is correlated with relative increases in diagnosis and incidence of thyroid cancer, but SES is also correlated with increased use of medical care.^17,19 Poverty may be associated with a relative decreases in diagnosis and incidence of thyroid cancer.²⁰ Following diagnosis, the various treatment options for thyroid cancer all require significant medical care by specialists.^21–23 Since specialist medical care is disproportionately available to and used by the affluent,¹⁷ an economically disadvantaged population will have less access to such care. This will have a negative effect on survivability.^21–23 Thus, a state like Oklahoma, with generally lower SES, can be expected to have both lower incidence rates and lower survivability for thyroid cancer.

Although thyroid cancer is not an extensive cause of cancer mortality, the increasing incidence rate makes it of great interest. The exact cause of increase has not yet been explained and is probably mutlifactorial. Prevention guidelines have not been put forth as evidence-based strategies have not yet been published. Further research is needed to identify the factors contributing to the increasing incidence of thyroid cancer, both in the United States and Oklahoma. Further, the deviation from national data for thyroid cancer incidence across racial groups is also a possible avenue for further study. The reasons for Oklahoma’s lower incidence rates and survivability are also as yet inconclusive, and certainly merit further study. This information will assist in establishing solid prevention recommendations which may decrease the incidence of thyroid cancers in the future. Additionally, it can guide the recommendations for diagnosis treatment so as not to over treat subclinical cancers.