C o n t ro v e r s y O v e r Radioiodine Ablation In T h y ro i d C a n c e r : Who Benefits? Don C. Yoo, MDa,*, Richard B. Noto, MDa, Peter J. Mazzaglia, MDb KEYWORDS Thyroid cancer Radioiodine ablation RIA Risk stratification Overdiagnosis KEY POINTS Since the early 1970s, the incidence of thyroid cancer, mainly papillary, in the United States has almost tripled with unchanged mortality. The extent of surgery performed for small thyroid cancers is controversial but only patients who have undergone total thyroidectomy are candidates for radioiodine ablation (RIA). RIA in low-risk patients has not been shown to decrease mortality and may not be indicated. Intermediate-risk patients may benefit from selective administration of RIA. Retrospective data support the use of iodine-131 (I-131) in high-risk patients, resulting in decreased recurrence and improved mortality. Lower doses of I-131 have been proven effective for RIA of remnant thyroid tissue after thyroidectomy.
INTRODUCTION
In the United States, the first use of radioactive iodine therapy as an alternative to surgery for toxic goiter was in 1943. Its subsequent use in the treatment of thyroid cancer was documented in 1948 in a patient with hyperthyroidism due to pulmonary metastases.1 In the 1950s and 1960s, the ability of neoplastic thyroid tissue to concentrate radioactive iodine and cause subsequent destruction was elucidated. Soon there were reports of improved survival in patients suffering with metastatic thyroid cancer who were treated with radioiodine ablation (RIA).2 Over the last 70 years, the use of RIA in well-differentiated thyroid cancer has evolved and become the standard of care in many cases. The concept of thyroid remnant ablation was introduced in the 1970s3
a Department of Diagnostic Imaging, Rhode Island Hospital, The Warren Alpert Medical School of Brown University, 593 Eddy Street, Providence, RI 02903, USA; b Department of General Surgery, Rhode Island Hospital, The Warren Alpert Medical School of Brown University, 593 Eddy Street, Providence, RI 02903, USA * Corresponding author. E-mail address: [email protected]
Surg Clin N Am 94 (2014) 573–586 http://dx.doi.org/10.1016/j.suc.2014.03.004 surgical.theclinics.com 0039-6109/14/$ – see front matter Ó 2014 Elsevier Inc. All rights reserved.
574
Yoo et al
and, in an effort to eradicate thyroglobulin production, is now cited as one of the main goals of therapy.4 Currently, about 50% of patients diagnosed with thyroid cancer will receive adjuvant RIA.5 However, there is still considerable controversy about the appropriate role of RIA in thyroid cancer. In the United States, recommendations for appropriate use of RIA come from the organizations American Thyroid Association (ATA) and National Comprehensive Cancer Network (NCCN), with consensus guidelines predominantly based on retrospective cohort studies.4,6 Until recently, large prospective randomized studies evaluating the use of RIA had not been published. The controversy concerning RIA is complicated by the ever-increasing numbers of small, low-risk thyroid cancers being diagnosed. To better understand some of the controversies, the evolving epidemiology of the disease must be examined. EPIDEMIOLOGY
The incidence of thyroid cancer in the United States has almost tripled since the early 1970s and nearly doubled during the last decade, especially in women. Thyroid cancer is now the fifth most common cancer in women in the United States.7 Despite this continued increase in incidence, the mortality rate from thyroid cancer has not significantly changed.7 The 2013 US statistics for thyroid cancer estimate 60,220 new cases (45,310 women and 14,910 men). However, the number of deaths estimated are only 1850 (1040 women and 810 men).8 The reasons for such a steep increase in thyroid cancer incidence were not well understood until recently. The study by Chen and colleagues9 investigated the increasing incidence of differentiated thyroid cancer in the United States from 1988 to 2005 and found the highest rate of increase was for small, localized thyroid cancers. It suggested that the increase was partly due to better cancer detection by ultrasonography and other imaging techniques. They also found an increased incidence across all tumor sizes and suggested that increased diagnostic scrutiny is not the sole explanation. Other possible causes, including environmental influences, should be considered. Another study by Nikiforov10 hypothesized that increased exposure to ionizing radiation, a known risk factor, especially for papillary thyroid cancer (PTC), may be playing a role. Historically, increased incidence of papillary thyroid carcinomas have occurred in areas of high radiation exposure. Specific examples include children in the Marshall Islands after atomic bomb testing and in the Ukraine after the Chernobyl nuclear accident.11,12 However, ionizing radiation is an unlikely explanation for the marked increase in incidence of thyroid cancer in the United States,13 where the highest amount of radiation is from background exposure to radon, followed by medical radiation, including CT scans.13,14 Although there has been increasing attention paid to rising levels of medical radiation, the attributable cancer risk from medical imaging is relatively low and certainly would not be expected to account for a significant component of the rising thyroid cancer incidence. Another explanation for the explosion of thyroid cancer diagnoses is increased detection during pathologic evaluation of surgical specimens. In a 40-year retrospective study of 2260 thyroidectomies for retrosternal goiter without known history of thyroid cancer, Grodski and colleagues15 showed that the percentage of thyroid cancer increased from 3.6% to 7.5% over time (P1 cm), equal rates of multifocality, 35% in each group, were reported. They also found cervical lymph node metastases in 19.7% of cancers greater than 1 cm and 9.7% of PTMCs.26 Total thyroidectomy also allows for the option of RIA and easier surveillance with neck ultrasounds and thyroglobulin. Other investigators argue that, because most studies do not show a survival benefit for total thyroidectomy in low-risk patients, lobectomy should be sufficient. The largest retrospective review of total thyroidectomy versus lobectomy was performed by Bilimoria and colleagues.25 They evaluated 52,173 patients (43,227 had total thyroidectomy and 8946 had lobectomy) from the National Cancer Data Base who underwent surgery for PTC in the United States from 1985 to 1998. For patients with tumors less than 1.0 cm, there was no difference in recurrence or survival between total
The Growing Incidence of Papillary Thyroid Cancer
thyroidectomy and lobectomy. However, for patients with tumors greater than 1.0 cm, lobectomy was associated with a 15% higher risk of recurrence and 31% higher risk of death. They concluded that total thyroidectomy improved survival in thyroid cancers greater than 1 cm but the survival benefit was unchanged in patients with tumors less than 1 cm. Hay and colleagues,27 in a retrospective review of 900 cases of PTMC observed over 60 years, reported that the extent of thyroid surgery (total or near-total thyroidectomy vs unilateral lobectomy) did not affect recurrence rates. The recurrence rates at 10, 15, and 20 years in patients who had lobectomy was 5.7%, 5.7%, and 9.8%, respectively, compared with recurrence rates of 4.5%, 4.7%, and 5.5%, respectively, in patients who had total or near-total thyroidectomy.24 There are no current published studies in the United States comparing long-term follow-up of patients with PTMC who are observed instead of immediately having surgery. However, in Japan, Ito and colleagues28 followed 340 patients with PTMC who chose to be observed instead of having immediate surgery at time of thyroid cancer diagnosis. The proportions of patients whose PTMC showed enlargement by 3 mm or more were 6.4% at 5-year follow-up and 15.9% on 10-year follow-up. Of these, 109 patients ended up having surgery after observation primarily due to tumor enlargement. None of these patients showed tumor recurrence after surgery on follow-up (average follow-up was 76 months). The investigators concluded observation can be a viable option instead of surgery for PTMC without unfavorable features. In light of these ongoing questions into whether or not a total thyroidectomy is necessary for small PTC, it is logical to question the indications for RIA in most patients at low-risk for thyroid cancer who have excellent long-term disease-free survival. ADVANTAGES OF RADIOIODINE TREATMENT
In 2014, most patients with thyroid cancer will undergo total thyroidectomy and at least 50% will undergo adjuvant RIA.5 The traditional arguments for postoperative administration of radioiodine have been (1) to decrease the risks for cancer recurrence and mortality; (2) to potentially diagnose and treat possible additional sites of microscopic tumor in the remnant thyroid tissue, cervical lymph nodes, and distant sites; and (3) to facilitate surveillance with thyroglobulin monitoring by ablating any viable remnant thyroid tissue in the thyroid bed.29 Thyroglobulin is more sensitive than a diagnostic iodine scan to detect residual or recurrent disease.30 However, when considering these arguments, one must not forget the excellent prognosis for most patients with PTC. The disease-specific mortality in well-differentiated, low-risk PTC ranges from 0.4% to 1.7%.31,32 In the series published by Memorial Sloan Kettering in 1994, looking at 1038 patients, there was 99% 20-year survival and 4% recurrence for low-risk patients treated with lobectomy and isthmusectomy, and long-term hormonal suppression.33 In light of these data, the justifications for RIA deserve reexamination. RECENT GUIDELINES
Currently, there are numerous published sets of guidelines for risk stratification and determination of which patients should or should not receive RIA. Lang and colleagues34 showed that as many as 16 different staging systems for thyroid cancer have been proposed. The most widely used guidelines are from the NCCN (2012) and ATA (2009). They differ slightly and the ATA is scheduled to publish new guidelines in June, 2014. As of this writing, the ATA guidelines state that a large number of retrospective studies show significant reductions in disease recurrence rates and
577
578
Yoo et al
cause-specific mortality, in support of RIA.4 However, subgroup analyses have shown these advantages to be limited to higher risk patients. Previously, the traditional staging systems AMES (age, metastases, extent, size), AGES (age, grade, extent, size), and MACIS (metastases, age, completeness of resection, invasion, size),35–37 classified low-risk patients as younger patients with histologically classic or follicular-variant PTC, without extrathyroidal extension or distant metastases. Notably, none of these staging systems included the presence of lymph node metastases as a prognostic factor. These low-risk patients have typically comprised 70% to 85% of diagnosed thyroid cancer.4 However, this percentage has grown given the disproportionate increase in the detection of smaller thyroid malignancies.5 Most pathologists and clinicians are now using the more universal TNM (tumor, nodes, metastases) staging system published by the American Joint Committee on Cancer (AJCC). The TNM staging for thyroid cancer is unique because patients who are younger than 45 years old are limited to stage I unless they have distant metastases, whereas those 45 years or older are classified in the more traditional TNM method of staging (AJCC 2002). Based on the available data in 2009, the ATA recommended RIA for a broad group of patients, including all stage III and IV, virtually all stage II, and selected stage I patients. These stage I patients are described as “those with multifocal disease, nodal metastases, extrathyroidal or vascular invasion, and/or more aggressive histologies.”4 The NCCN guidelines updated in 2013 also recommend that a large proportion of patients with PTC receive RIA. Essentially, all patients are recommended for RIA unless they meet all of the following criteria: primary tumor less than 1 cm, classic PTC, no extrathyroidal extension, no vascular invasion, postoperative stimulated thyroglobulin less than 1 ng/mL, and no clinical lymph node or distant metastases.6 As a consequence of the broadly described categories of patients who may benefit from and are, therefore, candidates for RIA, in both publications there has been little motivation to curb the use of RIA. This is despite scant data supporting its use in lowrisk patients who have had a complete tumor resection and who make up most current new thyroid cancer diagnoses. WHO BENEFITS? High-Risk Patients
Currently, there is little argument that RIA benefits patients at high risk for thyroid cancer recurrence. Known predictors of recurrence and metastasis include tumor size greater than 4 cm, extrathyroidal extension, residual tumor, gross lymph node metastases, as well as thyroglobulinemia out of proportion to what is seen on the posttreatment scan.4 Numerous studies have shown a benefit of RIA in these patients. Age of the patient is not an absolute indication for whether a patient should receive I-131; however, patients who are younger and older usually receive higher risk stratification. Tuttle and colleagues38 stratify patients younger than 20 years or older than 60 years as high-risk. Other investigators stratify patients older than 45 years as high-risk and may not define younger patients as being high-risk.39 In the meta-analysis by Sacks and colleagues,40 examining 79 studies published from 1966 to 2008 that compared radioiodine treatment versus no treatment for thyroid cancer, high-risk patients had significantly reduced recurrence rates and improved survival after treatment with RIA. One of the earlier and most quoted studies is by Mazzaferri and Jhiang.41 They demonstrated a 30% relative risk for thyroid cancer recurrence rates in 1355 subjects
The Growing Incidence of Papillary Thyroid Cancer
treated with RIA and thyrotropin suppression with mean follow-up of 15.7 years. Similar retrospective studies published in the early 1990s by DeGroot and colleagues,42 and Samaan and colleagues,43 strongly supported the use of RIA. Accordingly, both the NCCN and ATA guidelines recommend RIA for all such patients.4,6 NCCN also recommends radioiodine treatment of all patients with gross residual disease after surgery.6 However, in all of these studies, the benefits were restricted to subjects with tumors greater than 1.5 cm or with residual disease. Lack of Benefit for Low-risk Patients
One of today’s primary goals for the administration of radioactive iodine is ablation of the very small amount of thyroid tissue that surgeons may leave behind in cases when it is necessary for recurrent laryngeal nerve protection. However, several investigators have questioned the legitimacy of the claim that ablation of very small normal remnant tissue will decrease recurrences. They have also challenged the necessity of mandating an undetectable postoperative thyroglobulin versus accepting a low level that can be followed.27 The following studies have shown no benefit for RIA in low-risk subjects. Hay and colleagues,24 reviewing long-term outcomes in 2444 subjects treated over 60 years, showed that RIA did not change local recurrence or disease specific mortality in subjects with MACIS scores lower than 6, regardless of lymph node status. In a 2004 Lahey Clinic study, RIA added no survival benefit in 727 subjects stratified by AMES as 585 low-risk and 142 high-risk. The 20-year survival in low-risk, younger high-risk, and older high-risk patients receiving RIA versus no RIA was 100% versus 97.6% (P 5 .24), 64.2% versus 73.2% (P 5 .53), and 44.7% versus 44.4% (P 5 .53), respectively.44 In a case control study using the Swedish Cancer Registry, Lundgren and colleagues45 compared 595 subjects dying of thyroid cancer with 595 matched controls and found that the administration of I-131 was associated with poorer survival, even in early-stage disease. The most important prognostic factor was complete tumor removal. The meta-analysis by Sawka and colleagues46 examined studies for a benefit of RIA in low-risk subjects, for either recurrence or disease-specific survival. Incremental benefit was unclear and improved survival could not be established. A metaanalysis by Sacks and colleagues,40 looking at 79 studies, found the preponderance of evidence suggests that RIA is not associated with improved survival in subjects with low-risk thyroid cancer. None of the studies used a randomized controlled trial design and overwhelmingly used a retrospective cohort design. Neither the NCCN nor the ATA recommend RIA for patients with either unifocal or multifocal papillary microcarcinomas (
INTRODUCTION
In the United States, the first use of radioactive iodine therapy as an alternative to surgery for toxic goiter was in 1943. Its subsequent use in the treatment of thyroid cancer was documented in 1948 in a patient with hyperthyroidism due to pulmonary metastases.1 In the 1950s and 1960s, the ability of neoplastic thyroid tissue to concentrate radioactive iodine and cause subsequent destruction was elucidated. Soon there were reports of improved survival in patients suffering with metastatic thyroid cancer who were treated with radioiodine ablation (RIA).2 Over the last 70 years, the use of RIA in well-differentiated thyroid cancer has evolved and become the standard of care in many cases. The concept of thyroid remnant ablation was introduced in the 1970s3
a Department of Diagnostic Imaging, Rhode Island Hospital, The Warren Alpert Medical School of Brown University, 593 Eddy Street, Providence, RI 02903, USA; b Department of General Surgery, Rhode Island Hospital, The Warren Alpert Medical School of Brown University, 593 Eddy Street, Providence, RI 02903, USA * Corresponding author. E-mail address: [email protected]
Surg Clin N Am 94 (2014) 573–586 http://dx.doi.org/10.1016/j.suc.2014.03.004 surgical.theclinics.com 0039-6109/14/$ – see front matter Ó 2014 Elsevier Inc. All rights reserved.
574
Yoo et al
and, in an effort to eradicate thyroglobulin production, is now cited as one of the main goals of therapy.4 Currently, about 50% of patients diagnosed with thyroid cancer will receive adjuvant RIA.5 However, there is still considerable controversy about the appropriate role of RIA in thyroid cancer. In the United States, recommendations for appropriate use of RIA come from the organizations American Thyroid Association (ATA) and National Comprehensive Cancer Network (NCCN), with consensus guidelines predominantly based on retrospective cohort studies.4,6 Until recently, large prospective randomized studies evaluating the use of RIA had not been published. The controversy concerning RIA is complicated by the ever-increasing numbers of small, low-risk thyroid cancers being diagnosed. To better understand some of the controversies, the evolving epidemiology of the disease must be examined. EPIDEMIOLOGY
The incidence of thyroid cancer in the United States has almost tripled since the early 1970s and nearly doubled during the last decade, especially in women. Thyroid cancer is now the fifth most common cancer in women in the United States.7 Despite this continued increase in incidence, the mortality rate from thyroid cancer has not significantly changed.7 The 2013 US statistics for thyroid cancer estimate 60,220 new cases (45,310 women and 14,910 men). However, the number of deaths estimated are only 1850 (1040 women and 810 men).8 The reasons for such a steep increase in thyroid cancer incidence were not well understood until recently. The study by Chen and colleagues9 investigated the increasing incidence of differentiated thyroid cancer in the United States from 1988 to 2005 and found the highest rate of increase was for small, localized thyroid cancers. It suggested that the increase was partly due to better cancer detection by ultrasonography and other imaging techniques. They also found an increased incidence across all tumor sizes and suggested that increased diagnostic scrutiny is not the sole explanation. Other possible causes, including environmental influences, should be considered. Another study by Nikiforov10 hypothesized that increased exposure to ionizing radiation, a known risk factor, especially for papillary thyroid cancer (PTC), may be playing a role. Historically, increased incidence of papillary thyroid carcinomas have occurred in areas of high radiation exposure. Specific examples include children in the Marshall Islands after atomic bomb testing and in the Ukraine after the Chernobyl nuclear accident.11,12 However, ionizing radiation is an unlikely explanation for the marked increase in incidence of thyroid cancer in the United States,13 where the highest amount of radiation is from background exposure to radon, followed by medical radiation, including CT scans.13,14 Although there has been increasing attention paid to rising levels of medical radiation, the attributable cancer risk from medical imaging is relatively low and certainly would not be expected to account for a significant component of the rising thyroid cancer incidence. Another explanation for the explosion of thyroid cancer diagnoses is increased detection during pathologic evaluation of surgical specimens. In a 40-year retrospective study of 2260 thyroidectomies for retrosternal goiter without known history of thyroid cancer, Grodski and colleagues15 showed that the percentage of thyroid cancer increased from 3.6% to 7.5% over time (P1 cm), equal rates of multifocality, 35% in each group, were reported. They also found cervical lymph node metastases in 19.7% of cancers greater than 1 cm and 9.7% of PTMCs.26 Total thyroidectomy also allows for the option of RIA and easier surveillance with neck ultrasounds and thyroglobulin. Other investigators argue that, because most studies do not show a survival benefit for total thyroidectomy in low-risk patients, lobectomy should be sufficient. The largest retrospective review of total thyroidectomy versus lobectomy was performed by Bilimoria and colleagues.25 They evaluated 52,173 patients (43,227 had total thyroidectomy and 8946 had lobectomy) from the National Cancer Data Base who underwent surgery for PTC in the United States from 1985 to 1998. For patients with tumors less than 1.0 cm, there was no difference in recurrence or survival between total
The Growing Incidence of Papillary Thyroid Cancer
thyroidectomy and lobectomy. However, for patients with tumors greater than 1.0 cm, lobectomy was associated with a 15% higher risk of recurrence and 31% higher risk of death. They concluded that total thyroidectomy improved survival in thyroid cancers greater than 1 cm but the survival benefit was unchanged in patients with tumors less than 1 cm. Hay and colleagues,27 in a retrospective review of 900 cases of PTMC observed over 60 years, reported that the extent of thyroid surgery (total or near-total thyroidectomy vs unilateral lobectomy) did not affect recurrence rates. The recurrence rates at 10, 15, and 20 years in patients who had lobectomy was 5.7%, 5.7%, and 9.8%, respectively, compared with recurrence rates of 4.5%, 4.7%, and 5.5%, respectively, in patients who had total or near-total thyroidectomy.24 There are no current published studies in the United States comparing long-term follow-up of patients with PTMC who are observed instead of immediately having surgery. However, in Japan, Ito and colleagues28 followed 340 patients with PTMC who chose to be observed instead of having immediate surgery at time of thyroid cancer diagnosis. The proportions of patients whose PTMC showed enlargement by 3 mm or more were 6.4% at 5-year follow-up and 15.9% on 10-year follow-up. Of these, 109 patients ended up having surgery after observation primarily due to tumor enlargement. None of these patients showed tumor recurrence after surgery on follow-up (average follow-up was 76 months). The investigators concluded observation can be a viable option instead of surgery for PTMC without unfavorable features. In light of these ongoing questions into whether or not a total thyroidectomy is necessary for small PTC, it is logical to question the indications for RIA in most patients at low-risk for thyroid cancer who have excellent long-term disease-free survival. ADVANTAGES OF RADIOIODINE TREATMENT
In 2014, most patients with thyroid cancer will undergo total thyroidectomy and at least 50% will undergo adjuvant RIA.5 The traditional arguments for postoperative administration of radioiodine have been (1) to decrease the risks for cancer recurrence and mortality; (2) to potentially diagnose and treat possible additional sites of microscopic tumor in the remnant thyroid tissue, cervical lymph nodes, and distant sites; and (3) to facilitate surveillance with thyroglobulin monitoring by ablating any viable remnant thyroid tissue in the thyroid bed.29 Thyroglobulin is more sensitive than a diagnostic iodine scan to detect residual or recurrent disease.30 However, when considering these arguments, one must not forget the excellent prognosis for most patients with PTC. The disease-specific mortality in well-differentiated, low-risk PTC ranges from 0.4% to 1.7%.31,32 In the series published by Memorial Sloan Kettering in 1994, looking at 1038 patients, there was 99% 20-year survival and 4% recurrence for low-risk patients treated with lobectomy and isthmusectomy, and long-term hormonal suppression.33 In light of these data, the justifications for RIA deserve reexamination. RECENT GUIDELINES
Currently, there are numerous published sets of guidelines for risk stratification and determination of which patients should or should not receive RIA. Lang and colleagues34 showed that as many as 16 different staging systems for thyroid cancer have been proposed. The most widely used guidelines are from the NCCN (2012) and ATA (2009). They differ slightly and the ATA is scheduled to publish new guidelines in June, 2014. As of this writing, the ATA guidelines state that a large number of retrospective studies show significant reductions in disease recurrence rates and
577
578
Yoo et al
cause-specific mortality, in support of RIA.4 However, subgroup analyses have shown these advantages to be limited to higher risk patients. Previously, the traditional staging systems AMES (age, metastases, extent, size), AGES (age, grade, extent, size), and MACIS (metastases, age, completeness of resection, invasion, size),35–37 classified low-risk patients as younger patients with histologically classic or follicular-variant PTC, without extrathyroidal extension or distant metastases. Notably, none of these staging systems included the presence of lymph node metastases as a prognostic factor. These low-risk patients have typically comprised 70% to 85% of diagnosed thyroid cancer.4 However, this percentage has grown given the disproportionate increase in the detection of smaller thyroid malignancies.5 Most pathologists and clinicians are now using the more universal TNM (tumor, nodes, metastases) staging system published by the American Joint Committee on Cancer (AJCC). The TNM staging for thyroid cancer is unique because patients who are younger than 45 years old are limited to stage I unless they have distant metastases, whereas those 45 years or older are classified in the more traditional TNM method of staging (AJCC 2002). Based on the available data in 2009, the ATA recommended RIA for a broad group of patients, including all stage III and IV, virtually all stage II, and selected stage I patients. These stage I patients are described as “those with multifocal disease, nodal metastases, extrathyroidal or vascular invasion, and/or more aggressive histologies.”4 The NCCN guidelines updated in 2013 also recommend that a large proportion of patients with PTC receive RIA. Essentially, all patients are recommended for RIA unless they meet all of the following criteria: primary tumor less than 1 cm, classic PTC, no extrathyroidal extension, no vascular invasion, postoperative stimulated thyroglobulin less than 1 ng/mL, and no clinical lymph node or distant metastases.6 As a consequence of the broadly described categories of patients who may benefit from and are, therefore, candidates for RIA, in both publications there has been little motivation to curb the use of RIA. This is despite scant data supporting its use in lowrisk patients who have had a complete tumor resection and who make up most current new thyroid cancer diagnoses. WHO BENEFITS? High-Risk Patients
Currently, there is little argument that RIA benefits patients at high risk for thyroid cancer recurrence. Known predictors of recurrence and metastasis include tumor size greater than 4 cm, extrathyroidal extension, residual tumor, gross lymph node metastases, as well as thyroglobulinemia out of proportion to what is seen on the posttreatment scan.4 Numerous studies have shown a benefit of RIA in these patients. Age of the patient is not an absolute indication for whether a patient should receive I-131; however, patients who are younger and older usually receive higher risk stratification. Tuttle and colleagues38 stratify patients younger than 20 years or older than 60 years as high-risk. Other investigators stratify patients older than 45 years as high-risk and may not define younger patients as being high-risk.39 In the meta-analysis by Sacks and colleagues,40 examining 79 studies published from 1966 to 2008 that compared radioiodine treatment versus no treatment for thyroid cancer, high-risk patients had significantly reduced recurrence rates and improved survival after treatment with RIA. One of the earlier and most quoted studies is by Mazzaferri and Jhiang.41 They demonstrated a 30% relative risk for thyroid cancer recurrence rates in 1355 subjects
The Growing Incidence of Papillary Thyroid Cancer
treated with RIA and thyrotropin suppression with mean follow-up of 15.7 years. Similar retrospective studies published in the early 1990s by DeGroot and colleagues,42 and Samaan and colleagues,43 strongly supported the use of RIA. Accordingly, both the NCCN and ATA guidelines recommend RIA for all such patients.4,6 NCCN also recommends radioiodine treatment of all patients with gross residual disease after surgery.6 However, in all of these studies, the benefits were restricted to subjects with tumors greater than 1.5 cm or with residual disease. Lack of Benefit for Low-risk Patients
One of today’s primary goals for the administration of radioactive iodine is ablation of the very small amount of thyroid tissue that surgeons may leave behind in cases when it is necessary for recurrent laryngeal nerve protection. However, several investigators have questioned the legitimacy of the claim that ablation of very small normal remnant tissue will decrease recurrences. They have also challenged the necessity of mandating an undetectable postoperative thyroglobulin versus accepting a low level that can be followed.27 The following studies have shown no benefit for RIA in low-risk subjects. Hay and colleagues,24 reviewing long-term outcomes in 2444 subjects treated over 60 years, showed that RIA did not change local recurrence or disease specific mortality in subjects with MACIS scores lower than 6, regardless of lymph node status. In a 2004 Lahey Clinic study, RIA added no survival benefit in 727 subjects stratified by AMES as 585 low-risk and 142 high-risk. The 20-year survival in low-risk, younger high-risk, and older high-risk patients receiving RIA versus no RIA was 100% versus 97.6% (P 5 .24), 64.2% versus 73.2% (P 5 .53), and 44.7% versus 44.4% (P 5 .53), respectively.44 In a case control study using the Swedish Cancer Registry, Lundgren and colleagues45 compared 595 subjects dying of thyroid cancer with 595 matched controls and found that the administration of I-131 was associated with poorer survival, even in early-stage disease. The most important prognostic factor was complete tumor removal. The meta-analysis by Sawka and colleagues46 examined studies for a benefit of RIA in low-risk subjects, for either recurrence or disease-specific survival. Incremental benefit was unclear and improved survival could not be established. A metaanalysis by Sacks and colleagues,40 looking at 79 studies, found the preponderance of evidence suggests that RIA is not associated with improved survival in subjects with low-risk thyroid cancer. None of the studies used a randomized controlled trial design and overwhelmingly used a retrospective cohort design. Neither the NCCN nor the ATA recommend RIA for patients with either unifocal or multifocal papillary microcarcinomas (
