Multistage Carcinogenesis and the Incidence of Thyroid Cancer in the US by Sex, Race, Stage and Histology

Rafael Meza; Joanne T. Chang


BMC Public Health. 2015;15(789) 

In This Article


Thyroid cancer incidence worldwide has increased dramatically during the past three decades. Specifically, global age-standardized thyroid cancer incidence rates have increased 3-fold for both women and men since the 1970s, although with geographical variations.[1] Multiple descriptive studies have reported the upward trend of thyroid cancer.[2–7] However, the underlying causes are still under debate, with some attributing the increase to the widespread use of ionizing radiation therapy for head and neck benign conditions among children and adolescents back in the 1920–1950s,[8,9] while others attributing it to improvements in diagnostic tests and increase surveillance.[10,11]

The etiology of thyroid cancer is not fully understood, nonetheless it has been shown that gender and exposures to high-levels of ionizing radiation are major risk factors.[12–14] In terms of gender, thyroid cancer has become the fifth most common cancer among women in the United States,[15] with a female to male ratio of 3:1. The gender imbalance has been attributed among other reasons to female hormonal and reproductive factors, which correlate with the age-specific rise in female thyroid cancer incidence around the age of menarche and the decrease or slow-down after menopause.[1,16] With regards to ionizing radiation, previous studies have shown that atomic bomb survivors and children living in contaminated areas around Chernobyl in 1986 experienced particularly high rates of thyroid cancer. In addition, other studies have suggested that levels of iodine could be a risk factor for thyroid cancer. For example, patients with goiter, which suffer from iodine deficiency, have high rates of thyroid cancer.[17] There is also evidence that low levels of iodine in the diet associate with follicular thyroid cancers, whereas high levels are associated with papillary thyroid cancers.[12]

Thyroid cancer has the fastest growing incidence in the US. Various authors have suggested that the rise in thyroid cancer incidence in the US is predominantly due to increases in surveillance and diagnostic improvements for detecting smaller tumors.[2–4] Other interpretations suggest that diagnostic scrutiny is not fully responsible for the increasing trends, but rather that there could be due to other factors like increases in obesity and changes in diet and physical activity[5–7] or exposures to environmental agents like radiation, Bisphenol A, and polybrominated biphenyl ethers.[18–20] Independently of the underlying reasons, thyroid cancer has become the fifth most common cancer in US women and has become significant public health issue.

Significant differences in thyroid cancer risk by race in the US have been reported,[4,21,22] with whites having about twice the incidence than blacks. Healthcare access and socioeconomic status (SES) have been shown to be positively associated with thyroid cancer incidence, likely due to better detection and surveillance.[4,23] Thus several authors have suggested that differences in SES by race may explain the lower thyroid cancer incidence in blacks.[4,23,24] However, it has been shown that differences in health care access between blacks and whites only explain partly the racial gap in thyroid cancer,[24] and that increasing trends are similar between whites and blacks.[16,25] Therefore other mechanisms, like racial variations in susceptibility and differences in exposures to environmental agents could be also responsible for the observed disparities.[24,26]

Several studies have previously examined thyroid cancer trends in SEER using descriptive analyses, joinpoint regression and APC models.[2,3,16,25] Here we extend those analyses and complement them using multistage modeling that provides additional insights in terms of the biologicals mechanism of initiation, promotion and malignant conversion.[27–33]

In this paper we investigated thyroid cancer incidence trends by sex, race, and stage in the US using multistage carcinogenesis models and age-period-cohort (APC) analysis. Multistage thyroid carcinogenesis models were used to investigate potential differences in initiation, promotion and malignant conversion rates by sex, race, stage and histology. Models were adjusted for period and birth-cohort trends to investigate the contributions of each of these factors, and to examine whether birth- or diagnosis-year better correlate with observed incidence patterns.