Seaberg RM, Eski S, Freeman JL. Influence of Previous Radiation Exposure on Pathologic Features and Clinical Outcome in Patients With Thyroid Cancer. Arch Otolaryngol Head Neck Surg. 2009;135(4):355-359. doi:10.1001/archoto.2009.13
To determine whether previous radiation exposure to the head and neck is related to less favorable pathologic and clinical outcome in patients after surgical management of thyroid cancer.
Retrospective chart review.
Academic teaching hospital (referral center).
All patients with diagnosed thyroid cancer who had been exposed to radiation before surgical treatment were retrospectively identified from the thyroid cancer database at our institution (1963-2007). One hundred twenty-five patients (95 women and 30 men) were included. Inclusion criteria included surgical treatment for thyroid cancer and a history of exposure to radiation at least 3 years before diagnosis of the disease.
Main Outcome Measures
Pathologic features and data related to disease recurrence, distant metastasis, and survival.
Mean (range) age at first exposure to radiation was 19.4 (1-65) years, and mean lag time to diagnosis of disease was 28.7 (3-60) years. Patients were treated surgically with either total or near-total thyroidectomy (83%) or partial or subtotal thyroidectomy (17%). Pathologic diagnoses included 111 papillary carcinomas (89%). Sixty-three percent of patients had multifocal disease, 12% had lymphovascular tumor invasion, and 26% had direct extrathyroid extension of disease. Twenty-five percent of patients had metastases to cervical lymph nodes, and 9% had distant metastases. Sixteen percent of patients experienced local recurrence of disease. At last follow-up, 86% of patients were alive and free of disease, 8% were alive with disease, 4% had died of thyroid cancer, and 2% had died of an unrelated cause. Compared with other patients with thyroid cancer, this radiation-exposed cohort was more likely to undergo total thyroidectomy, multiple operative procedures, and external radiotherapy. A higher percentage had multifocal disease, extrathyroid extension, stage IV disease, and distant metastases. At follow-up, fewer patients were free of disease, and more patients had died of thyroid disease.
Patients who have been exposed to radiation have more aggressive disease and worse clinical outcome than other patients with thyroid cancer.
Thyroid cancer is one of the well-known malignant neoplasms associated with radiation exposure. It often induces characteristic histologic changes in thyroid tissue, and it is a well-established risk factor for both benign and malignant thyroid tumors. This is supported by epidemiologic studies in atomic bomb survivors1 and in children living in contaminated areas around Chernobyl, Ukraine, after the 1986 nuclear reactor accident.2 Studies in patients who received therapy because of external radiation also support the causative role of irradiation in the development of thyroid cancers3; in Canada, benign childhood diseases of the head and neck were treated with external radiotherapy from the 1930s to the1960s.4 The objective of this study was to determine whether patients with previous radiation exposure have thyroid cancer with more aggressive pathologic features and clinical behavior, and ultimately a worse clinical outcome, compared with patients with thyroid cancer without a history of radiation exposure.
A retrospective medical record review of the thyroid carcinoma database at Mount Sinai Hospital, Toronto, Ontario, Canada (1963-2007) was undertaken to identify all surgical patients who had a history of exposure to radiation. Patients considered positive for a history of radiation exposure included those who at least 3 years before diagnosis of thyroid cancer had received direct external radiation to the head and neck for therapeutic purposes (eg, treatment of acne), direct external radiation to another body site (eg, treatment of breast cancer), occupational exposure (eg, unprotected radiography technician or dental assistant), diagnostic exposure (eg, repeated medical imaging of the head and neck), environmental exposure (eg, Chernobyl disaster), or therapeutic exposure to radioactive iodine. Patients were classified into 5 groups according to these exposure categories. Paper and electronic medical records were reviewed for each patient. Of 134 patients with a history of radiation exposure, 9 were excluded because the exposure had occurred less than 3 years before diagnosis. Of these, 8 patients had been exposed 1 year or less before diagnosis. It has been documented that radiation-induced thyroid tumors are associated with a latent period of 4 to 45 years after exposure.5 The remaining 125 patients were included in the study.
Data collected from patient medical records included epidemiologic features (sex and age at diagnosis), details of clinical findings (clinically evident lymph nodes or vocal cord paralysis), radiation exposure(s) (reason for exposure, nature of exposure, and patient age at exposure), treatment details (timing and type of surgical procedure(s), adjuvant radioactive iodine ablation, and adjuvant external radiotherapy), pathologic features (tumor size, histologic diagnosis and variant, multifocality, involvement of cervical lymph nodes, extracapsular invasion, extrathyroidal spread, and lymphovascular invasion), and clinical outcome (date of last follow-up, disease recurrence, distant metastasis, and death from disease). Cancer stage (American Joint Commission for Cancer) and MACIS (metastases, age, completeness of resection, invasion, and size) scores (in patients with papillary thyroid carcinoma) were calculated for each patient. One-way analysis of variance was carried out with post hoc Tukey tests to compare the ages of the 5 radiation exposure groups using commercially available software (Statistica 6 for Windows; StatSoft, Inc, Tulsa, Oklahoma).
One hundred twenty-five patients with thyroid cancer who had a history of radiation exposure were identified from the thyroid cancer database at the Department of Otolaryngology–Head and Neck Surgery, Mount Sinai Hospital. Mean (range) patient age was 48.2 (18-79) years. The patient group consisted of 95 women (76%) and 30 men (24%). Most patients (89%) had a clinically evident (palpable) thyroid nodule, whereas fewer patients had thyroid nodules that were noted at ultrasonography to be enlarging (7%) or had palpable cervical lymph nodes.
Mean age at first exposure to radiation was 19.4 (1-65) years, and mean (SD) lag time to diagnosis of thyroid disease was 28.7 (3-60) years. Patients with exposure occurring less than 3 years before diagnosis of the disease were excluded from the study. Most patients (56%) had a history of direct external radiation to the head and neck, often as therapeutic radiotherapy to treat benign conditions. Other exposures included direct external radiation to other body sites (6%), occupational or diagnostic exposures (eg, unprotected radiography technicians, dental assistants, or patients exposed to repeated radiographic imaging of the head and neck) (23%), environmental exposure (eg, the Chernobyl disaster) (11%), and radioactive iodine treatment (4%). Patient and radiation exposure data are given in Table 1.
All patients underwent surgical treatment of thyroid cancer including total or near-total thyroidectomy in 83% and subtotal or partial thyroidectomy in 17%. Thirty-eight percent of patients underwent formal neck dissection, and 23% of patients required at least 1 additional operative procedure. Of these procedures, 50% were neck dissections, 25% were completion thyroidectomies, 14% were both completion thyroidectomies and neck dissections, 7% were excisions of single lymph nodes, and 4% were repeated resections of disease in the thyroid bed combined with neck dissection. Seventy-four percent of patients underwent adjuvant radioactive iodine ablation, and 6% underwent adjuvant external radiotherapy. Treatment data are given in Table 2.
Final surgical pathologic results revealed that 89% of patients had papillary thyroid carcinoma. Fewer patients had follicular (6%), Hürthle cell (2%), or medullary (2%) carcinomas; 1 patient had a hyalinizing trabecular tumor. In terms of histologic variants, 72% of tumors exhibited a classic pattern. There were also follicular (22%), diffuse sclerosing (5%), and insular (1%) variants represented in this patient population. No tall cell or columnar variants were reported.
Mean (range) tumor size was 1.8 (0.1-7) cm. Sixty-three percent of patients had multifocal disease, with a mean (range) of 1.94 (1-7) foci. Twelve percent of tumors demonstrated lymphovascular invasion, and 26% demonstrated extrathyroid spread. Lymph nodes were positive for disease in 25% of cases; of these, 36% exhibited extranodal invasion. Pathologic data are given in Table 1.
Patients were followed up for a mean (range) of 10.6 (0.08-35.10) years. Clinical stage (American Joint Commission for Cancer) and clinical outcome data are given in Table 2. MACIS scores were calculated for all patients with papillary thyroid carcinoma. Most patients had stage I disease (64%) and MACIS scores lower than 6 (81%). Sixteen percent of patients had local recurrence of disease, and 9% had distant metastases. At last follow-up, most patients (86%) were alive and free of disease; fewer patients were alive with local recurrence (4%) or distant metastases (4%). Four percent of patients had died of thyroid cancer, and 2% had died of an unrelated cause.
This patient cohort was compared with a group of general thyroid carcinoma patients, that is, patients who were not identified on the basis of previous radiation exposure. This group included 574 patients who were similar for age and sex distribution and were identified from the same thyroid cancer database located in our department.6 Treatment, pathologic findings, and clinical outcome data for the 2 groups are given in Table 3.
Although statistical calculations are impossible, there are some notable results evident with this type of comparison. The radiation-exposed group was more likely to undergo total or near-total thyroidectomy (83% vs 38%) and more likely to require additional operative procedures (23% vs 2%) compared with the general thyroid cancer group. In addition, these patients were more likely to require adjuvant external radiotherapy for treatment of malignant lesions of the thyroid gland (6% vs <1%). The radiation-exposed patients were more likely to have multifocal tumors (63% vs 36%) and tumors exhibiting extrathyroid spread (26% vs 8%). More radiation-exposed patients had stage IV disease compared with the general thyroid cancer group (16% vs 5%); they were also more likely to have distant metastases (9% vs 2%), to have disease at follow-up (8% vs 3%), or to have died of thyroid malignant neoplasm (4% vs 1.5%).
In an effort to determine whether the type of previous radiation exposure had an effect on tumor pathologic findings or clinical outcome, the radiation-exposed patients were divided into 5 subgroups according to type of radiation exposure. These subgroups were compared for age at exposure, cancer stage, MACIS score, local recurrence, distant metastases, and likelihood of death from disease. The subgroup exposed to direct external radiotherapy to the head and neck accounted for 56% of all patients with stage IV disease and 60% of all patients with MACIS scores higher than 8. In addition, this subgroup accounted for 63% of patients who experienced local recurrence, 55% of those with distant metastases, and 80% of those who died of thyroid cancer. This subgroup also represents most patients (56%) in the current study. Further, these patients were exposed to radiation at a statistically significant younger age (mean age, 14.9 years) than each of the other subgroups (mean age, 36.3, 28.7, and 34.2 years) (P < .005) except the group exposed to occupational or diagnostic sources of radiation (mean age, 19.9 years) (Table 4).
Most patients with well-differentiated thyroid cancer can expect good disease-specific outcomes. Age and sex are important prognostic factors,7 as is family history.8 Other factors such as distant metastasis, completeness of resection, local invasion, and tumor size are used in prognostic scoring systems such as the MACIS score.9 The objective of the present study was to investigate whether, in addition to being a risk factor for thyroid malignant neoplasms, radiation exposure is also related to more aggressive tumor pathology and, ultimately, worse prognosis.
This relationship has been investigated in other studies, and it has been suggested that tumors in radiation-exposed patients are more likely to involve both thyroid lobes and demonstrate extrathyroid spread.10 Some have suggested an increase in multifocality11 and incidental microcarcinomas.12 In contrast, others have reported no difference in multifocality or lymph node disease and have suggested that radiation-exposed patients have smaller tumors.12 Even those studies reporting differences in pathologic features have not reported a difference in clinical outcome for local recurrence, distant metastases, or survival.10- 13
The results of the present study suggest that there are different pathologic features in radiation-exposed patients. Although there were not an unexpectedly high number of aggressive variants of thyroid cancer, more multifocal disease and extrathyroid spread was noted. More patients underwent total or near-total thyroidectomy compared with a general population of patients with thyroid cancer.6 Further, more patients required additional operative procedures, most commonly neck dissection. This is in contrast to a previous study that suggested that the extent of thyroid surgery was similar in these 2 populations.12 Although the distribution of MACIS scores in the present study was similar to that noted in the original MACIS patient cohort,9 radiation-exposed patients in the present study were more likely to have stage IV disease and, thus, distant metastases compared with the general thyroid cancer group. Fewer patients were free of disease at follow-up, and more patients had died of thyroid cancer. This is in contrast to previous studies that reported no difference in clinical outcome between patient groups exposed or not exposed to radiation.10- 13 In addition to the well-documented association in children between radiation exposure at Chernobyl and worse pathologic and clinical outcome,14 to our knowledge, the present study is the first to suggest worse prognosis in radiation-exposed patients. It is possible that this difference can be explained, at least in part, by the relatively larger number of radiation-exposed patients included in the present study (n = 125) compared with other recent studies.11,12
The present study further sought to determine whether the type of previous radiation exposure was related to more aggressive pathologic features and worse clinical outcome. A wide variety of exposure types were included in this study, such as exposure to radioactive iodine for therapeutic purposes. While some have proposed that this exposure does not increase the risk of thyroid carcinomas, the number of patients exposed during vulnerable times such as childhood has been too small for firm conclusions to be drawn.15 Similarly, in the present study, this was the smallest subgroup of patients, and most were not exposed during childhood.
The subgroup of patients exposed to direct external radiotherapy to the head and neck accounted for most patients in the current study with stage IV disease, MACIS scores higher than 8, local recurrences, distant metastases, and death from thyroid disease. However, this observation needs to be interpreted with caution because this was also the largest subgroup in the study and the youngest at the time of radiation exposure. It is known that radiation-induced thyroid tumors are associated with a strong inverse correlation with age at the time of exposure.3,5
There are several other limitations to this study that are important. The heterogeneity of patient exposures to radiation limits the ability to draw conclusions about the risk of various types of exposures. Previous studies have reported that the lowest dose with an observed risk is 10 cGy.3 However, with retrospective patient data, it is difficult to determine the radiation exposure dose. Other limitations include treatment by multiple surgeons and nonstandardized reporting by multiple pathologists. Most important, although the present study design enabled inclusion of a large number of radiation-exposed patients, it does not permit statistical comparisons. Despite this, the differences observed between groups (Table 4) must be taken seriously, and they suggest the need not only for further careful study but also for consideration of more aggressive treatment of malignant thyroid nodules in patients who were exposed to radiation compared with those who were not.
It is interesting to speculate on the mechanism by which radiation exposure may cause more aggressive tumor behavior and, ultimately, worse clinical outcome. The ret proto-oncogene (ret/PTC) (OMIM 188550) is unique to papillary carcinomas of the thyroid gland16 and has been observed in significantly higher frequency in radiation-associated thyroid cancers.17,18 These rearrangements, in turn, have been associated with more aggressive tumor behavior,19 which suggests a possible explanation for the more aggressive pathologic features and worse clinical outcome noted in radiation-exposed patients in the present study.
Compared with other patients with thyroid cancer, patients who have a history of exposure to radiation are more likely to be treated surgically with total thyroidectomy, multiple operative procedures, and external radiotherapy compared with the general thyroid cancer patient population. A higher percentage of radiation-exposed patients have multifocal and extrathyroid extension of disease, stage IV disease, and distant metastases. At follow-up, fewer of these patients are free of disease and more have died of thyroid disease. Therefore, this study suggests that patients who have been exposed to radiation have more aggressive disease and worse clinical outcome than other patients with thyroid cancer and, therefore, may require more aggressive treatment.
Correspondence: Jeremy L. Freeman, MD, FRCSC, Department of Otolaryngology–Head and Neck Surgery, Mount Sinai Hospital, 600 University Ave, Ste 401, Toronto, ON M5G 1X5, Canada (firstname.lastname@example.org).
Submitted for Publication: July 1, 2008; final revision received October 5, 2008; accepted October 7, 2008.
Author Contributions: Dr Seaberg had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Seaberg and Freeman. Acquisition of data: Seaberg and Eski. Analysis and interpretation of data: Seaberg and Freeman. Drafting of the manuscript: Seaberg and Freeman. Critical revision of the manuscript for important intellectual content: Seaberg, Eski, and Freeman. Statistical analysis: Seaberg. Obtained funding: Freeman. Administrative, technical, and material support: Seaberg, Eski, and Freeman. Study supervision: Freeman.
Financial Disclosure: None reported.
Funding/Support: Dr Freeman holds the Temmy Latner/Dynacare Chair in Head and Neck Oncology.
Previous Presentation: This study was presented at the Seventh International Conference on Head and Neck Cancer of the American Head and Neck Society; July 22, 2008; San Francisco, California.