Thyroid cancer: An epidemic of disease or an epidemic of diagnosis?

La Vecchia et al.1 published in this Journal a comprehensive review of trends in thyroid cancer (TC) mortality rates for all available countries worldwide and incidence rates in selected countries. They showed that incidence rates were increasing in nearly all countries driven by the rise in papillary carcinomas (i.e., the by far most frequent TC type nowadays). Conversely, TC mortality rates have declined or, at least, stabilized at lowest risk. They concluded that the declines in mortality reflect reductions in iodine deficiency (see, for instance, the declines of TC mortality in Switzerland since the 1920s after the introduction of salt iodization2) and improvements in TC diagnosis and treatment. However, the same diagnostic changes led to large rises in observed TC incidence. We would like to further address the question of whether we are witnessing an epidemic of disease or an epidemic of diagnosis. Figure 1 shows trends in age-standardized (world population) incidence rates of TC in women in a selection of European and extra-European very high-income countries (http://ci5.iarc.fr) already included in La Vecchia et al.1 Contrary to these authors, however, we plotted incidence rates on the same scale to better visualize the large between-country differences in the magnitude of the TC epidemic. TC incidence rates were relatively low and increased little in Denmark (1968–2007, nationwide) and the United Kingdom (1988–2007, seven registries from England and all Scotland) whereas steep upward trends were seen in France (1988–2007, eight registries) and Italy (1988–2007, nine registries) (Fig. 1a). By far the largest increase (5-fold) was found in the Republic of Korea (1993–2007, Seoul) (Fig. 1b). Upward trends were also present in the United States (1973–2007, nine SEER registries), and Australia (1983–2007, seven registries) but not in Japan (1978–2007, Miyagi and Osaka). The 5-year prevalence of TC per 100,000 women in 2012 was estimated to vary from <30 in Denmark and the United Kingdom to over 100 in Italy and the United States. In Korea, 6 out of 1,000 women were estimated to be 5-year TC-survivors (http://globocan.iarc.fr). Age-standardized incidence rates (ASR) of thyroid cancer in women in selected countries. TC incidence trends in men in individual countries were consistent with those in women although TC incidence is approximately 3-fold higher in women (http://ci5.iarc.fr). TC diagnosis at an earlier age in women has been attributed to the greater use of health services before middle age by women than men due to events related to reproduction and peri-menopausal symptoms.3 Until the 1970s, most TCs were found in patients who presented with nodules causing compression symptoms or visible neck masses.4 Only relatively big nodules (>20 mm) were assessed by palpation and biopsy. The advent of neck ultrasonography in the 1980s and then ultrasound-guided biopsy enabled detection and biopsy of nodules as small as 2 mm. Additional increases in TC diagnoses have resulted from the spread of portable ultrasound machines and, after 2000, new imaging technologies used for a variety of medical conditions. In the United States, nearly 16% of computed tomography and magnetic resonance images show incidental thyroid nodules, of which around three quarters are <15 mm.4 Today, more patients receive a TC diagnosis after an evaluation of an incidentally found thyroid nodule than after evaluation of a symptomatic or palpable nodule.4 A contribution of known TC risk factors (environmental or medical radiation,5 deficit or excess of iodine intake,2 body mass, including being overweight or tall6) or suspected risk factors (dietary aspects,7 nitrates and nitrites, certain microelements in the diet or drinking water,8 etc.) to TC incidence increases has been neither proven nor ruled out. However, it seems difficult to conceive any modern lifestyle or environmental risk factor that may have abruptly affected TC incidence in Australia, France, Italy, Korea, and the United States so heavily, and, in the same period, have relatively spared Nordic countries, the United Kingdom, and Japan. It seems to us that the organization of the health systems and the penetration of new diagnostic and screening practices are key elements that nowadays differ substantially even across the most developed countries. The international pattern of TC incidence strongly suggests that these can have a large impact on the detection of the large reservoir of indolent thyroid tumours. Small papillary (low-risk) TCs are very unlikely to cause morbidity or premature mortality especially in young women. In women age 15–44 years, the incidence/mortality ratio for TC (all carcinomas included) was 203 in the most developed countries and 19 in less developed countries (vs., for instance, 7 and 4, respectively, for breast cancer).9 In an observational study conducted in Japan since 1993, patients of both sexes with low-risk papillary TC were given the choice of surgery or active surveillance with ultrasonography.10 None of the 1,235 patients who chose active surveillance, including 191 patients who eventually underwent surgery for clinical progression, showed distant metastases or died of TC. Total thyroidectomy (performed in the vast majority of TC patients) is associated with postoperative complications, such as permanent hypoparathyroidism and vocal cord paralysis, and implies lifelong thyroid replacement therapy and monitoring.11 The traditional and now controversial use of supraphysiological doses of levothyroxine to suppress thyroid-stimulating hormone causes mild thyrotoxicosis that affects life quality and is associated with cardiovascular side-effects and bone fractures.11 Despite recommendations in favour of a selective use, approximately half of the TC patients in the United States received neck lymph node dissection and radiation treatment between 1975 and 2009.12 Different ways to address the problem of overdiagnosis and overtreatment of low-risk TC can be envisaged.11, 12 They include: (i) avoidance of TC screening activities; (ii) reclassification of low-risk TC using terms other than cancer. This reclassification approach was useful in cervical cancer screening9 (i.e., cervical intraepithelial neoplasia grade 1–3) and has been proposed for in situ ductal carcinoma of the breast and (iii) establishment of randomized clinical trials or observational cohorts of watchful-waiting approaches. Yours sincerely, Silvia Franceschi Salvatore Vaccarella