Thyroid Cancers Detected by Imaging Are Not Necessarily Small or Early Stage

Abstract

Background:

The incidence of well-differentiated thyroid cancer (WDTC) in the United States is increasing rapidly. Much of this increase is due to the detection by imaging of small, nonpalpable tumors. The incidence of advanced WDTC is also increasing, suggesting a true increase in the incidence of WDTC. This study was performed to determine how WDTCs of all sizes and stages are initially detected.

Methods:

A retrospective chart review of 519 patients who underwent surgery for WDTC from January 1, 2007, through August 31, 2010, was performed. A total of 473 patients suitable for inclusion in this study were divided into three groups based upon the method by which the tumor was initially detected: tumors detected by an imaging study (184 patients—39%), those detected because a mass was felt in the neck (218 patients—46%), and those detected incidentally on pathological study of the surgical specimen (71 patients—15%). Method of detection was correlated with age and sex of the patient, and size, stage, and nodal status of the tumor.

Results:

Patients in the Palpation group were more likely to be female (79% vs. 67% vs. 74%), younger (46 vs. 51 vs. 52), and to have larger tumors than those in the Imaging or Incidental groups. In the Imaging group, the tumor was detected on thyroid sonogram in 98 (53%), computed tomography in 38 (21%), magnetic resonance imaging in 19 (10%), carotid duplex scan in 14 (8%), and positron-emission tomography or other imaging studies in 15 (8%). Thirty-three percent of tumors <1 cm, 51% 1–2 cm, 29% 2–4 cm, and 38% >4 cm were first detected on an imaging study. Forty-seven percent of Stage III and IV cancers in patients aged ≥45 years and 39% of patients with positive central nodes were in the Imaging group.

Conclusion:

This study demonstrates that while most tumors discovered by imaging were small and early stage, almost half of advanced (Stage III and IV) WDTCs were initially discovered by imaging studies. These findings are consistent with the hypothesis that the frequent use of imaging studies may explain not only the increasing incidence of early WDTC, but more advanced thyroid cancers as well.

Introduction

During the past 20 years, the incidence of well-differentiated thyroid cancer (WDTC) in the United States has increased more rapidly than any other cancer. The current rate of increase is 6.8% (1). Similar increases have been observed in other developed countries (2,3). In the United States, this is mostly due to an increase in the incidence of small papillary cancers (4) with an increasing proportion of patients older than 45 years (5). Autopsy series from around the world suggest that clinically occult thyroid cancers are present in 3–10% of patients (6 –11), although prevalence as high as 36% has been reported (12). This wide variability may be related to the technical details of how the thyroid is examined postmortem (13). This high incidence of occult thyroid cancer suggests that the observed increase in thyroid cancer may be due to the improved detection of subclinical disease rather than a real change in incidence. The role of increased detection of incidental microscopic thyroid cancers due to increased pathologic sampling of thyroids removed for benign disease may also contribute to the observed increased incidence (14).

In contrast, other studies have demonstrated an increase in the incidence of larger and more advanced thyroid cancers (15 –18), which suggests that not all of the observed increase in incidence is due to improved detection of subclinical disease. This conclusion assumes that larger more advanced tumors are not initially detected by imaging studies.

The percentage of thyroid cancers detected by imaging rather than by palpation has increased substantially over the past 20 years (19). Davies et al. state that screening and chance events are now responsible for the detection of almost one half of thyroid cancers (20). The role of imaging studies performed for other reasons in the detection of thyroid cancer has been discussed by many authors (21 –25).

The assumptions that most thyroid cancers detected incidentally by imaging studies are small and that the increasing incidence of larger cancers is proof of a real increase in incidence of WDTC may not be valid. This study was performed to identify the method of thyroid cancer detection in patients undergoing thyroid surgery, with particular attention to larger and more advanced cancers.

Materials and Methods

The medical records of all 519 patients who underwent thyroid surgery at NYU Langone Medical Center from January 1, 2007, through August 31, 2010, by the three members of NYU Endocrine Surgery Associates and who had a postoperative diagnosis of WDTC were reviewed. Forty-six patients were excluded because the method of initial detection of the cancer could not be determined from the patient's chart. The remaining 473 patients were divided into three groups based on the method of initial detection: Imaging group, Palpation group, and Incidental group. The Imaging group comprised 184 patients in whom the indication for surgery was a suspicious or malignant cytological finding on fine-needle aspiration biopsy (FNAB) of a tumor that was initially detected on an imaging study. The Palpation group comprised 218 patients in whom the indication for surgery was a suspicious or malignant cytological finding on FNAB and in whom the patient stated that further diagnostic studies were initiated because a physician had noted an abnormality on physical examination or because the patient or another nonprofessional had noted a mass in the neck. The Incidental group comprised the remaining 71 patients in whom incidental cancers were found on pathological study of the surgical specimen that were not related to the indication for thyroidectomy. In the Incidental group, indications for surgery included nodules with suspicious cytology that proved to be benign, symptomatic or enlarging nodules with benign cytology, multinodular goiters, and thyrotoxicosis.

The extent of thyroidectomy (lobectomy vs. total thyroidectomy) was determined by the operating surgeon based on preoperative evaluation, patient preference, and intraoperative findings. Central compartment lymph-node sampling or central compartment dissection was performed if suspicious nodes were identified at the time of surgery, or electively at the discretion of the surgeon.

Tumor size was measured by the pathologist from the unfixed, fresh surgical specimen. The largest tumor diameter was reported. In patients with multifocal cancer, the size of the largest tumor was reported. Pathology reports were reviewed for extrathyroidal extension and central compartment lymph-node metastases. This information was used to determine the stage of the tumor.

Statistical analysis was performed using GraphPad InStat v3.06 for Windows (GraphPad Software, San Diego, CA) and SPSS v20 for Windows (SPSS, Chicago, IL). Contingency tables were analyzed by Fisher's exact test, and comparison of means by a one-way analysis of variance (ANOVA). Comparison between groups was performed using binomial logistic regression analysis. p-Values ≤0.05 were considered significant.

This study was approved by the NYU Cancer Institute Protocol Review and Monitoring Committee and by the NYU Institutional Review Board.

Results

The characteristics of the patients and tumors in each of the three groups are summarized in Table 1. There were 184 patients (39%) in the Imaging group, 218 (46%) in the Palpation group, and 71 (15%) in the Incidental group. The patients ranged in age from 11 to 86 years (median 49±15). There were 345 (73%) women and 128 (27%) men. A total of 451 (95%) of the patients had papillary cancers, 12 (3%) had Hürthle cell cancers, and 10 (2%) had follicular cancers. Patients in the Imaging and Incidental groups were more likely to be older than those in the Palpation group. Patients in the Imaging group were more likely to be men than those in the Palpation group. The mean tumor size was largest in the Palpation group and smallest in the Incidental group.

Table 1.

Relationship of Method of Detection to Age and Sex of the Patient

	Imaging	Palpation	Incidental	p-Value
Mean age	51±14	46±15	52±13	^a
Female:male	124:60 (67% female)	172:46 (79% female)	49:22 (74% female)	^b
Mean size (mm)	16±12	21±14	5±6	^c

The comparisons of mean age between Imaging vs. Palpation (p<0.001) and between Palpation vs. Incidental (p<0.01) are significant. The comparison between Imaging vs. Incidental is not significant.

The comparison of sex distribution between Imaging vs. Palpation (p=0.0122) is significant. The comparisons between Imaging vs. Incidental and between Palpation vs. Incidental are not significant.

The comparisons of mean tumor size between Imaging vs. Palpation, Palpation vs. Incidental, and Imaging vs. Incidental are all significant (p<0.001).

The imaging studies that detected the tumors in the Imaging group are summarized in Table 2. The majority of these (53%) were detected by thyroid ultrasonography. It is important to note that the sonograms in the Imaging group were not performed to evaluate a palpable thyroid nodule or obvious thyroid-related symptoms, but were requested for numerous other reasons, some appropriate and some inappropriate, which are detailed in Table 3. In the Palpation group, 168 (77%) patients had a thyroid mass first noted by a healthcare provider, while in the remaining 48 (23%), the thyroid mass was initially noted by the patient or another medical nonprofessional. In two patients, the initial presenting sign was a palpable metastatic node in the neck.

Table 2.

Initial Imaging Study by Which the Cancer Was Detected in the Imaging Group

Imaging study	Number of patients
Thyroid sonogram	98 (53%)
CT	38 (21%)
MRI	19 (10%)
Carotid duplex	14 (8%)
PET	6 (3%)
Other	5 (3%)
Unknown	4 (2%)
Total	184

CT, computed tomography; MRI, magnetic resonance imaging; PET, positron-emission tomography.

Table 3.

Indication for Thyroid Ultrasonography

Indication	Number of patients (n=98)
Hashimoto's thyroiditis	17 (17%)
Family history of thyroid cancer	11 (11%)
Hyperthyroidism	11 (11%)
Nonspecific neck symptoms	10 (10%)
Hypothyroidism	9 (9%)
Nonspecific constitutional symptoms	6 (6%)
Gynecological exam	4 (4%)
History of radiation exposure	4 (4%)
Hyperparathyroidism	4 (4%)
Goiter	3 (3%)
Previous thyroid surgery (no malignancy)	2 (2%)
Hypercalcemia	2 (2%)
“Routine” sonogram	2 (2%)
Self-sonography	2 (2%)
Other lab abnormalities	2 (2%)
Surveillance of another thyroid nodule	1 (1%)
Unknown	8 (8%)

The relationship between the size of the tumor and the method of detection is summarized in Table 4. Smaller tumors were more likely to be detected by imaging or incidentally than larger tumors (p<0.001). Nevertheless, 38% of tumors larger than 4 cm were initially detected by imaging and not by palpation. The single tumor >4 cm in the Incidental group was a follicular variant of papillary carcinoma found in a massive multinodular goiter.

Table 4.

Relationship of Tumor Size to Method of Detection

	Imaging	Palpation	Incidental
Tumor <1 cm	52 (33%)	40 (26%)	64 (41%)
Tumor 1–2 cm	86 (51%)	78 (46%)	6 (4%)
Tumor 2–4 cm	32 (29%)	78 (71%)	0 (0%)
Tumor >4 cm	14 (38%)	22 (60%)	1 (3%)

The relationship between the method of detection and the age of the patient, cancer stage, and central nodal status is summarized in Table 5. Tumors in the Incidental group were usually small, low stage, and had negative nodes. None of the patients <45 years old presented with distant metastases, and all were therefore Stage I. Tumors in patients ≥45 years were more likely to be in the Imaging or Incidental groups than tumors in patients <45 years. Limiting the comparison to the Imaging and Palpation groups, in patients ≥45 years, there was no significant difference in the method of detection (Imaging vs. Palpation) when patients with less advanced (Stages I and II) were compared to those with more advanced cancers (Stages III and IV) or when tumors with negative central nodes were compared to those with positive central nodes. Forty-seven percent of Stage III and IV tumors and 39% of tumors with positive nodes were in the Imaging group.

Table 5.

Relationship of Age and Stage to Method of Detection

	Imaging	Palpation	Incidental	p-Value^a
Age <45 years	52 (30%)	102 (59%)	19 (11%)	0.0001
Age ≥45 years	132 (44%)	116 (39%)	52 (17%)
Stage I or II (age ≥45)	86 (43%)	66 (33%)	49 (24%)	0.6849
Stage III or IV (age ≥45)	46 (47%)	50 (51%)	3 (3%)
Negative nodes	129 (29%)	136 (41%)	67 (20%)	0.1137
Positive nodes	55 (39%)	82 (58%)	4 (3%)

Comparison of imaging and palpation. Incidental group was not included in the statistical analysis in this table.

At the time of initial surgical consultation, 84 of the 184 tumors in the Imaging group (46%) were found to be palpable. Larger tumors in this group were more likely to be palpable than smaller ones (Table 6).

Table 6.

Palpability of Cancers Initially Detected by Imaging

	Cancers initially detected by imaging	Palpable [n (%)]
Tumor <1 cm	52	11 (21%)
Tumor 1–2 cm	86	41 (48%)
Tumor 2–4 cm	32	20 (63%)
Tumor >4 cm	14	12 (86%)
Total	184	84 (46%)

Discussion

This study confirms the observations of others that a substantial percentage of WDTCs are initially detected by imaging studies. Including the 71 incidental cancers discovered at the time of thyroidectomy for other indications, only 46% of cancers in this study were discovered because the patient or physician felt a mass in the neck. As expected, small tumors were more likely to be detected incidentally or by imaging, and larger tumors were more likely to be detected by palpation. Almost all incidental tumors were small, low stage, and unlikely to have positive nodes. However, 38% of tumors larger than 4 cm, 47% of Stage III and Stage IV cancers, and 39% of tumors with positive nodes were initially detected by imaging studies. It is interesting to note that 46% of the tumors initially detected by imaging studies actually were palpable at the time of the initial surgical consultation. There is no way to determine if these tumors would have eventually been discovered if they had not been detected by imaging, nor is there any way to know if the nonpalpable tumors would have progressed to the point where they became either palpable or symptomatic.

Chen et al. examined the SEER database to calculate the annual percentage change (APC) in incidence of thyroid cancer in the United States (15). They showed that the APC increased from 1988 through 1998 and more rapidly from 1998 through 2005. In both men and women, APC increased for smaller and larger cancers and for patients with localized, regional, and distant disease. The greatest increase, however, was seen in patients with cancers of 1 cm or less. They suggested that while the increased detection of small cancers may explain a part of the increase in APC, the increase in more advanced cancers suggests other factors, including environmental, dietary, and genetic causes.

Enewold et al. reported similar conclusions based on their analysis of SEER data (16). They too noted an increasing incidence of papillary cancer in each stage, including regional and distant spread, but the greatest increase in incidence was in cancers ≤1 cm. They hypothesized that if all of the observed increase in incidence were due to improved disease detection, the incidence of more advanced tumors would decrease. They suggested that while a portion of the increased incidence is due to improved detection, some of the increase might be due to other factors. Possible additional factors they hypothesize include the pathologic reclassification of tumors once considered to be follicular neoplasms into follicular variants of papillary cancer, as well as radiation exposure due to the increased use of computed tomography scanning in adults and children.

Morris and Myssiorek also reported an increasing incidence of thyroid tumors >4 cm, as well as an increasing incidence of cancers with extrathyroidal extension and cervical lymph-node metastases (18). They stated that this suggests a true increase in the incidence of thyroid cancer, not simply a detection artifact. While this is certainly possible, other factors might explain these observations. Surgical approaches to thyroid cancer have changed, and elective central compartment neck dissections are more commonly performed. This results in the upstaging of cancers by the detection of occult micrometastases in cervical lymph nodes. Similarly, increased pathologic scrutiny may result in the increased detection of microscopic extrathyroidal spread that was not previously reported.

Li et al. studied the effect of socioeconomic status and tumor size on the increasing incidence of thyroid cancer (26). These authors, similar to Enewold et al. (16), posited that one would expect the incidence of larger cancers to decrease as a greater number of smaller cancers are discovered and treated. This hypothesis requires the assumption that all smaller tumors inevitably enlarge and that larger tumors are discovered because the patient or physician feels a mass in the neck or the patient becomes symptomatic. This assumption may not be true. Ito et al. have shown that papillary microcancers can be observed for years without treatment, with few enlarging significantly in size (27).

The patients included in this study were all treated in a single university teaching hospital in New York City and almost all of them live in the New York metropolitan area. It is possible that they represent a group with higher incomes, higher levels of education, and better access to medical care than patients treated in other parts of the United States or other countries in the world. The observations made in this study may not be equally valid elsewhere.

It is clear from our data that the assumption that a substantial number of larger, more advanced cancers are not initially detected by imaging is false. Larger, more advanced tumors are frequently detected by imaging. These findings support the hypothesis that increased diagnostic scrutiny may indeed by responsible for the increasing incidence of WDTCs of all sizes and stages. These data do not eliminate other possible explanations for our observations. For example, these findings would be consistent with a simultaneously co-occurring increase in the numbers of thyroid cancers, which in recent years were discovered on imaging because more imaging is occurring.

Conclusion

There may be many factors that contribute to the increasing incidence of WDTC. The increased detection of small tumors by various imaging studies is certainly one factor. This study demonstrates that this increased detection is not limited to small tumors and that a substantial number of tumors larger than 4 cm and more advanced stage cancers are initially detected by imaging studies as well. The assumption that the increasing incidence of larger and higher stage cancers is proof of a real increase in the incidence of WDTC rather than simply an artifact of increased diagnostic scrutiny needs to be reexamined.

Footnotes

Acknowledgments

The authors wish to thank Shiro Horiuchi, PhD, of the CUNY School of Public Health for his assistance in the statistical analysis of some of the data included in this paper.

Author Disclosure Statement

None of the authors has any competing financial interests to disclose.

References

National Cancer Institute 2012 SEER cancer statistics review, 1975–2009. Available at: http://seer.cancer.gov/csr/1975_2009_pops09 (accessed December 24, 2012 ).

Liu

, Semenciw

, Ugnat

, Mao

. 2001. Increasing thyroid cancer incidence in Canada, 1970–1996: time trends and age-period-cohort effects. Br J Cancer, 85:1335–1339.

Colonna

, Bossard

, Guizard

, Remontet

, Grosclaude

, Le Reseau

. 2010. Descriptive epidemiology of thyroid cancer in France: incidence, mortality and survival. Ann Endocrinol (Paris), 71:95–101.

Davies

, Welch

. 2006. Increasing incidence of thyroid cancer in the United States, 1973–2002. JAMA, 295:2164–2167.

Hughes

, Haymart

, Miller

, Gauger

, Doherty

. 2011. The most commonly occurring papillary thyroid cancer in the United States is now a microcarcinoma in a patient older than 45 years. Thyroid, 21:231–236.

Mortensen

, Bennett

, Woolner

. 1955. Incidence of carcinoma in thyroid glands removed at 1000 consecutive routine necropsies. Surg Forum, 5:659–663.

Tanriover

, Comunoglu

, Eren

, Comunoglu

, Turkmen

, Dogan

, Gündogmus

. 2011. Occult papillary thyroid carcinoma: prevalence at autopsy in Turkish people. Eur J Cancer Prev, 20:308–312.

Kovacs

, Gonda

, Vadasz

, Ludmany

, Uhrin

, Gorombey

, Kovács

, Hubina

, Bodó

, Góth

, Szabolcs

. 2005. Epidemiology of thyroid microcarcinoma found in autopsy series conducted in areas of different iodine intake. Thyroid, 15:152–157.

de Matos

, Ferreira

, Ward

. 2006. Prevalence of papillary microcarcinoma of the thyroid in Brazilian autopsy and surgical series. Endocr Pathol, 17:165–173.

10.

Neuhold

, Kaiser

, Kaserer

. 2001. Latent carcinoma of the thyroid in Austria: a systematic autopsy study. Endocr Pathol, 12:23–31.

11.

Chong

. 1994. Thyroid carcinomas in Singapore autopsies. Pathology, 26:20–22.

12.

Harach

, Franssila

, Wasenius

. 1985. Occult papillary carcinoma of the thyroid. A “normal” finding in Finland. A systematic autopsy study. Cancer, 56:531–538.

13.

Martinez-Tello

, Martinez-Cabruja

, Fernandez-Martin

, Lasso-Oria

, Ballestin-Carcavilla

. 1993. Occult carcinoma of the thyroid. A systematic autopsy study from Spain of two series performed with two different methods. Cancer, 71:4022–4029.

14.

Grodski

, Brown

, Sidhu

, Gill

, Robinson

, Learoyd

, Sywak

, Reeve

, Delbridge

. 2008. Increasing incidence of thyroid cancer is due to increased pathologic detection. Surgery, 144:1038–1043; discussion 43.

15.

Chen

, Jemal

, Ward

. 2009. Increasing incidence of differentiated thyroid cancer in the United States, 1988–2005. Cancer, 115:3801–3807.

16.

Enewold

, Zhu

, Ron

, Marrogi

, Stojadinovic

, Peoples

, Devesa

. 2009. Rising thyroid cancer incidence in the United States by demographic and tumor characteristics, 1980–2005. Cancer Epidemiol Biomarkers Prev, 18:784–791.

17.

Cramer

, Fu

, Harth

, Margevicius

, Wilhelm

. 2010. Analysis of the rising incidence of thyroid cancer using the Surveillance, Epidemiology and End Results national cancer data registry. Surgery, 148:1147–1152; discussion 52–53.

18.

Morris

, Myssiorek

. 2010. Improved detection does not fully explain the rising incidence of well-differentiated thyroid cancer: a population-based analysis. Am J Surg, 200:454–461.

19.

Heller

. 2007. Do all cancers need to be treated? The role of thyroglobulin in the management of thyroid cancer: the 2006 Hayes Martin lecture. Arch Otolaryngol Head Neck Surg, 133:639–643.

20.

Davies

, Ouellette

, Hunter

, Welch

. 2010. The increasing incidence of small thyroid cancers: where are the cases coming from?. Laryngoscope, 120:2446–2451.

21.

Mitchell

, Parangi

. 2005. The thyroid incidentaloma: an increasingly frequent consequence of radiologic imaging. Semin Ultrasound CT MR, 26:37–46.

22.

Shetty

, Maher

, Hahn

, Halpern

, Aquino

. 2006. Significance of incidental thyroid lesions detected on CT: correlation among CT, sonography, and pathology. Am J Roentgenol, 187:1349–1356.

23.

Jin

, Wilhelm

, McHenry

. 2009. Incidental thyroid nodule: patterns of diagnosis and rate of malignancy. Am J Surg, 197:320–324.

24.

Ahmed

, Horton

, Jeffrey

Jr , Sheth

, Fishman

. 2010. Incidental thyroid nodules on chest CT: review of the literature and management suggestions. Am J Roentgenol, 195:1066–1071.

25.

Shie

, Cardarelli

, Sprawls

, Fulda

, Taur

. 2009. Systematic review: prevalence of malignant incidental thyroid nodules identified on fluorine-18 fluorodeoxyglucose positron emission tomography. Nucl Med Commun, 30:742–748.

26.

, Du

, Reitzel

, Xu

, Sturgis

. 2013. Impact of enhanced detection on the increase in thyroid cancer incidence in the US: review of incidence trends by socioeconomic status within the Surveillance, Epidemiology, and End Results registry, 1980–2008. Thyroid, 23:103–110.

27.

Ito

, Miyauchi

, Inoue

, Fukushima

, Kihara

, Higashiyama

, Tomoda

, Takamura

, Kobayashi

, Miya

. 2010. An observational trial for papillary thyroid microcarcinoma in Japanese patients. World J Surg, 34:28–35.