Abstract
Objective:
Although lobectomy is acceptable for patients with small, low-risk differentiated thyroid cancer (DTC), gross extrathyroidal extension (ETE) remains an indication for total thyroidectomy (TT). Here we investigate evolving trends in extent of surgery for +ETE DTC.
Methods:
Patients with +ETE DTC who underwent resection from 2010 to 2020 were identified using the National Cancer Database. The primary outcome was performance of TT versus lobectomy.
Results:
Among 5,851 patients, most were female (79.7%), white (80.0%), and had minimal ETE (91.8%). Ninety-two percent of patients received TT. Year of treatment was influential (p<0.001), with increasing lobectomy rates in later years. On multivariable analyses, a decreased likelihood of TT was seen in years 2015 through 2020.
Conclusions:
Most patients with +ETE DTC underwent guideline-concordant TT, but lobectomy rates doubled over the study period. These findings may reflect increased preference for lobectomy in low-risk DTC, but could undertreat patients with high-risk features.
Keywords: extrathyroidal extension, differentiated thyroid cancer, lobectomy, thyroidectomy, extent of surgery
Introduction
Although differentiated thyroid cancer is recognized as a generally indolent form of thyroid cancer, extrathyroidal extension is a known risk factor for disease recurrence and decreased disease-specific survival within this otherwise favorable histologic subtype. Due to this increased risk, expert guidelines from the American Thyroid Association (ATA) for DTC have recommended total thyroidectomy (TT) for gross (i.e. macroscopic) ETE.1,2 By contrast, for low-risk DTC 1 to 4 cm in size, the 2015 ATA guidelines stated that lobectomy alone or total thyroidectomy are both appropriate options, resulting in an increase in rates of lobectomy for patients in this low-risk cohort.3 It remains unclear how these shifts toward minimal surgery for low-risk DTC patients may have affected guideline-concordant care with total thyroidectomy for higher-risk DTC patients with ETE. Thyroid lobectomy is associated with fewer short- and long-term risks than TT.4,5
In this study, we aimed to evaluate contemporary trends in extent of surgery for +ETE DTC in a 10-year cohort of patients from the National Cancer Database to evaluate factors associated with differences in extent of surgery and the impact of 2015 ATA guidelines on guideline-concordant care for ETE as it was classified at time of diagnosis. We hypothesized that the 2015 ATA guideline recommendation for either lobectomy or total thyroidectomy for low-risk DTC may have led to an unintentional decrease in guideline-concordant total thyroidectomy for patients with +ETE DTC.
Methods
Study Overview
Retrospective data from the National Cancer Database (NCDB), a collaborative database from the American College of Surgeons (ACS) Commission on Cancer (CoC) and American Cancer Society, was used to identify a study cohort of patients in the United States treated at facilities accredited by the CoC. Based on prior estimates, the NCDB Participant User Files (PUF) captures approximately 80% of all newly diagnosed thyroid cancers in the United States.6 Study protocols were submitted to the University of Alabama at Birmingham Institutional Review Board, who determined this study to be exempt. No direct funding was received for conduct of this study. Study methods and findings are reported according to the STROBE guidelines for observational studies.7
Cohort Selection
The cohort included adult patients with differentiated thyroid cancer between 1 to 4 cm in size with extrathyroidal extension who underwent lobectomy or total thyroidectomy from 2010 to 2020. A diagnosis of DTC was determined based on ICD-O-3 histology codes consistent with well-differentiated papillary or follicular thyroid cancers per the North American Association of Central Cancer Registries. Cases with other high-risk primary tumor features such as multifocal tumors and grades of moderately differentiated, poorly differentiated, or undifferentiated were excluded. Patients with nodal or distant metastases were excluded, as well as those who underwent surgical procedures for palliative intent.
Extrathyroidal Extension
Patients with American Joint Commission on Cancer (AJCC) 7th and 8th edition pathologic T stage less than pT3 were excluded (i.e. pT0, pT1, pT2, pTX). Extrathyroidal extension was determined using AJCC pathologic T stage based on year of diagnosis due to the change in how ETE was classified in each edition (7th edition for study years 2010–2017, 8th edition for 2018–2020). Extent of ETE was subdivided into two categories of minimal and extensive; minimal was ETE including the perithyroidal soft tissues or strap muscles alone, and extensive ETE included other surrounding structures including parathyroid glands, nerves, airway and vascular structure, and musculoskeletal structures other than strap muscles. For patients diagnosed in 2010–2017, minimal ETE was considered to be pT3 (as tumors greater than 4 cm were already excluded), and extensive ETE for patients with pT4 disease. For patients diagnosed in 2018–2020, pT3 and pT3b disease considered to be minimal ETE, and pT4, pT4a, or pT4b considered to be extensive.8
Surgical Outcomes
The primary outcome for this study was extent of surgical resection (i.e. thyroid lobectomy versus total thyroidectomy). Secondary outcomes included annual rates of lobectomy relative to total thyroidectomy, and patient, disease, and facility factors associated with performance of total thyroidectomy. Extent of surgery was determined based on NCDB PUF codes for Surgical Procedure of Primary Site, based on thyroid site-specific codes for lobectomy (with or without isthmectomy) and total thyroidectomy (including near-total and sub-total thyroidectomy). The NCDB requires reporting of the most extensive surgical procedure for the primary site, therefore patients who initially received a lobectomy and then underwent completion thyroidectomy would be coded as total thyroidectomy. Due to this reporting methodology of a single surgery code for the primary site, it was not possible to determine which patients received upfront total thyroidectomy versus those who underwent lobectomy followed by completion thyroidectomy. For the purposes of this study, only patients who underwent surgical resection at the diagnosing facility were included so that associations between facility type and extent of surgery could be studied.
Statistical Analysis
Descriptive statistics were performed with categorical variables reported as frequencies with percentages and continuous variables reported as median values with interquartile ranges. Bivariate analysis using Chi-squared, Student’s t, and ANOVA tests were performed to evaluate differences in patient, disease, and facility variables between lobectomy and total thyroidectomy groups. Next, multivariable logistic regression was used to assess the likelihood of total thyroidectomy based on included covariates. Covariates were selected a priori and included year of diagnosis, patient age, sex, race and ethnicity, extent of ETE, tumor size category, facility type, facility volume tertile, and receipt of radioactive iodine (RAI) therapy. Pathologic stage was excluded as a covariate due to multicollinearity with patient age, likely due to the age-stratified AJCC staging system for thyroid cancer. Facility volume tertiles were created by totaling all cases in the analytic dataset treated at each facility, then stratifying into tertiles based on cases per facility and matching facility tertiles back to individual cases. A subgroup analysis was performed to compare treatment for patients with minimal versus extensive ETE. A sensitivity analysis was conducted to study the influence of patients who were documented to have received both lobectomy alone and RAI after identification of this group on bivariate analysis. Statistical analyses were performed using SAS 9.4 (SAS Institute Inc, Cary, NC) and Stata 18 (StataCorp LLC, College Station, TX) statistical software.
Results
A cohort of 5,851 patients with +ETE DTC 1 to 4 cm in size was identified. Descriptive statistics are detailed in Table 1. This group had a median age of 56 years (IQR 45–66 years). This cohort was predominantly female (79.7%), white (80.0%), and not Hispanic or Latino (83.4%). The majority of the cohort had minimal ETE (90.8%) versus those with extensive ETE (9.2%). Most patients were treated at either an Academic/Research Program (35.2%) or at a Comprehensive Community Cancer Program (CCCP) (27.7%), and 61.2% received RAI. Overall, 7.4% of this +ETE cohort received lobectomy alone. When evaluating the annual rates of lobectomy in this cohort across the study period, the rate of lobectomy doubled over the course of the study period, from 5.1% in 2010 to 10.7% in 2020 (Figure 1A).
Table 1.
Cohort demographic data for patient, disease, and facility characteristics. Variables represented as n (percentage) for categorical measures, and median (interquartile range) for continuous measures.
| Variable | Study Cohort (n=5,851) |
|---|---|
|
| |
| Age (years) | 56 (45–66) |
|
| |
| Sex | |
| Female | 4,662 (79.7) |
| Male | 1,189 (20.3) |
|
| |
| Race | |
| White | 4,681 (80.0) |
| Asian or Pacific Islander | 538 (9.2) |
| Black | 390 (6.7) |
| Other or unknown | 218 (3.7) |
| Native American | 24 (0.4) |
|
| |
| Ethnicity | |
| Non-Hispanic | 4,879 (83.4) |
| Hispanic | 801 (13.7) |
| Unknown | 171 (2.9) |
|
| |
| Extent of surgery | |
| Total Thyroidectomy | 5,420 (92.6) |
| Lobectomy | 431 (7.4) |
|
| |
| Extrathyroidal extension | |
| Minimal (limited to strap muscles or perithyroid soft tissue) | 5,310 (90.8) |
| Extensive (other surrounding structures) | 541 (9.2) |
|
| |
| Tumor size | |
| 1.0 to 2.0 cm | 3,478 (59.4) |
| 2.1 to 3.0 cm | 1,507 (25.8) |
| 3.1 to 4.0 cm | 866 (14.8) |
|
| |
| Pathologic stage† | |
| Stage I | 1,398 (23.9) |
| Stage II | 441 (7.5) |
| Stage III | 2,676 (45.7) |
| Stage IV | 257 (4.4) |
| Unknown | 1,079 (18.4) |
|
| |
| Insurance payor | |
| Private Insurance | 3,449 (58.9) |
| Medicare | 1,656 (28.3) |
| Medicaid | 435 (7.4) |
| Not Insured | 166 (2.8) |
| Unknown | 94 (1.6) |
| Other Government | 51 (0.9) |
|
| |
| Facility type | |
| Academic | 2,061 (35.2) |
| Comprehensive Community Cancer Program | 1,622 (27.7) |
| Unknown* | 994 (17.0) |
| Integrated Network Cancer Program | 990 (16.9) |
| Community Cancer Program | 184 (3.1) |
|
| |
| Received RAI | |
| Yes | 3,581 (61.2) |
| No | 2,270 (38.8) |
Pathologic stage determined by American Joint Committee on Cancer 7th (2010–2017) or 8th edition (2018–2020) based on year of diagnosis.
Facility type is suppressed by the National Cancer Database for patients under age 40.
Figure 1.
Temporal trend of percentage of lobectomy among study cohort (A) and likelihood of total thyroidectomy (B) over the study period. Likelihood of total thyroidectomy (B) represented as odds ratio with 95% confidence interval error bars; regression model covariates include patient age, sex, race, ethnicity, extent of ETE, tumor size, pathologic stage, receipt of RAI, facility type and volume tertile.
Patient, disease, and facility factors differed between the lobectomy and total thyroidectomy groups (Table 2). Statistically significant differences between these groups were seen with patient sex, with a higher proportion of female patients receiving total thyroidectomy (93.1%) versus male patients (90.9%) (p=0.011). More patients treated at CCCPs and Integrated Network Cancer Programs underwent total thyroidectomy (p=0.002). Within the facility volume tertiles, there was a higher proportion of lobectomy patients treated at low volume facilities (12.8%) compared to the total thyroidectomy group (8.9%), and higher proportions of mid volume or high volume facilities among the total thyroidectomy group (overall p=0.018). More patients with stage III disease underwent total thyroidectomy compared to patients with stage I disease (p<0.001). Finally, more patients treated in later study years underwent lobectomy, a trend that was first observed in 2015 (p<0.001). There were no statistically significant differences detected based on patient age, race, ethnicity, minimal versus extensive ETE, tumor size, or insurance status.
Table 2.
Bivariate analysis of lobectomy versus total thyroidectomy. Data presented as n (group %) or median (interquartile range).
| Variable | Lobectomy (n=431) | Total Thyroidectomy (n=5,420) | P-value |
|---|---|---|---|
|
| |||
| Age (years) | 55 (41–68) | 56 (45–66) | 0.67 |
|
| |||
| Sex | |||
| Female | 323 (74.9) | 4,339 (80.1) | 0.011 * |
| Male | 108 (25.1) | 1,081 (19.9) | |
|
| |||
| Race | |||
| White | 328 (76.1) | 4,353 (80.3) | 0.15 |
| Asian or Pacific Islander | 48 (11.1) | 490 (9.0) | |
| Black | 33 (7.7) | 357 (6.6) | |
| Other or unknown | 22 (5.1) | 196 (3.7) | |
| Native American | 0 (0.0) | 24 (0.4) | |
|
| |||
| Ethnicity | |||
| Non-Hispanic | 371 (86.1) | 4,508 (83.2) | 0.11 |
| Hispanic | 45 (10.4) | 756 (13.9) | |
| Unknown | 15 (3.5) | 156 (2.9) | |
|
| |||
| Extrathyroidal extension | |||
| Minimal (strap muscles or perithyroid soft tissue) | 401 (93.0) | 4,909 (90.6) | 0.089 |
| Extensive (other surrounding structures) | 30 (7.0) | 511 (9.4) | |
|
| |||
| Tumor size category | |||
| 1.0 to 2.0 cm | 248 (57.5) | 3,230 (59.6) | 0.56 |
| 2.1 to 3.0 cm | 112 (26.0) | 1,395 (25.7) | |
| 3.1 to 4.0 cm | 71 (16.5) | 795 (14.7) | |
|
| |||
| AJCC Pathologic stage† | |||
| Stage I | 125 (29.0) | 1,273 (23.5) | <0.001 * |
| Stage II | 36 (8.4) | 405 (7.5) | |
| Stage III | 156 (36.2) | 2,520 (46.5) | |
| Stage IV | 11 (2.6) | 246 (4.5) | |
| Unknown | 103 (23.9) | 976 (18.0) | |
|
| |||
| Received RAI | |||
| Yes | 131 (30.4) | 3,450 (63.7) | <0.001 * |
| No | 300 (69.6) | 1,970 (36.3) | |
|
| |||
| Insurance payor | |||
| Private Insurance | 241 (55.9) | 3,208 (59.2) | |
| Medicare | 128 (29.7) | 1,528 (28.2) | |
| Medicaid | 38 (8.8) | 397 (7.3) | 0.55 |
| Not Insured | 11 (2.6) | 155 (2.9) | |
| Unknown | 10 (2.3) | 84 (1.5) | |
| Other Government | 3 (0.7) | 48 (0.9) | |
|
| |||
| Facility type | |||
| Academic | 154 (35.7) | 1,907 (35.2) | |
| Comprehensive Community Cancer Program | 94 (21.8) | 1,528 (28.2) | 0.002 * |
| Unknown* | 100 (23.2) | 894 (16.5) | |
| Integrated Network Cancer Program | 67 (15.5) | 923 (17.0) | |
| Community Cancer Program | 16 (3.7) | 168 (3.1) | |
|
| |||
| Facility volume (analytic cohort) | |||
| Low | 55 (12.8) | 482 (8.9) | 0.018 * |
| Mid | 64 (14.8) | 936 (17.3) | |
| High | 312 (72.4) | 4,002 (73.8) | |
Pathologic stage determined by American Joint Committee on Cancer (AJCC) 7th (2010–2017) or 8th edition (2018–2020) based on year of diagnosis.
Facility type is suppressed by the National Cancer Database for patients under age 40.
As the significant difference in extent of surgery by pathologic stage may have been more attributable to patient age given the age-stratified AJCC staging (with all patients under age 45 in this non-metastatic, +ETE cohort expected to be Stage 1), further analysis was performed to evaluate differences between lobectomy and total thyroidectomy groups among patients younger than age 45. Results from this analysis are detailed in Supplemental Table 1, with the only significant difference between surgery groups being receipt of RAI, with a higher proportion of total thyroidectomy patients receiving RAI than lobectomy patients (820/1,323, 62.0% versus 36/132, 27.3%; p<0.001).
Multivariable regression was performed to evaluate factors associated with increased likelihood of total thyroidectomy (Table 3). When controlling for co-variates listed in Table 3, years of diagnosis of 2015 through 2020 were associated with a decrease in likelihood of total thyroidectomy (Figure 1B). Receipt of RAI was associated with increased odds of total thyroidectomy (OR 3.81, 95% CI 3.04–4.79). No other patient or disease factors, including age, sex, race, extent of ETE, tumor size, or pathologic T stage, were associated with a statistically significant difference in likelihood of total thyroidectomy. Among facility characteristics, patients treated at hospitals with an unknown facility type (OR 0.55, 95% CI 0.36–0.83) relative to the reference group (Academic/Research Program) were less likely to undergo total thyroidectomy, and patients treated at mid volume (OR 1.65, 95% CI 1.08–2.54) or high volume facilities (OR 1.51, 95% CI 1.04–2.19) were more likely to receive total thyroidectomy relative to those treated at low volume facilities.
Table 3.
Multivariable logistic regression modeling likelihood of total thyroidectomy based on patient, disease, and facility characteristics. Diagnosis year also included as covariate in multivariable model (odds ratios presented in Figure 1B). Data presented as odds ratio (OR) with 95% confidence intervals (CI).
| Variable | Likelihood of Total Thyroidectomy (OR, 95% CI) | p-value |
|---|---|---|
|
| ||
| Age | OR 0.99, 0.98–1.00 | 0.134 |
|
| ||
| Sex | ||
| Male | reference | |
| Female | OR 1.27, 0.99–1.63 | 0.064 |
|
| ||
| Race | ||
| White | reference | |
| Asian or Pacific Islander | OR 0.90, 0.63–1.28 | 0.548 |
| Black | OR 0.84, 0.56–1.25 | 0.384 |
| Other or unknown | OR 1.03, 0.49–2.17 | 0.931 |
| Native American | excluded due to low observations | N/A |
|
| ||
| Ethnicity | ||
| Non-Hispanic | reference | |
| Hispanic | OR 1.37, 0.96–1.95 | 0.081 |
|
| ||
| Extent of extrathyroidal extension | ||
| Minimal | reference | |
| Extensive | OR 1.47, 0.97–2.22 | 0.069 |
|
| ||
| Tumor size | ||
| 1.0 to 2.0 cm | reference | |
| 2.1 to 3.0 cm | OR 0.95, 0.74–1.23 | 0.713 |
| 3.1 to 4.0 cm | OR 0.85, 0.63–1.16 | 0.310 |
|
| ||
| Facility type | ||
| Academic | reference | |
| Community Cancer Program | OR 1.09, 0.57–2.406 | 0.800 |
| Comprehensive Community Cancer Program | OR 1.25, 0.91–1.72 | 0.166 |
| Integrated Network Cancer Program | OR 1.08, 0.76–1.53 | 0.669 |
| Unknown* | OR 0.55, 0.36–0.83 ** | 0.005 |
|
| ||
| Facility volume (analytic cohort) | ||
| Low | reference | |
| Mid | OR 1.65, 1.08–2.54 ** | 0.02 |
| High | OR 1.51, 1.04–2.19 ** | 0.03 |
|
| ||
| RAI | ||
| None | reference | |
| Received | OR 3.81,3.04–4.79 ** | <0.001 |
Facility type is suppressed by the National Cancer Database for patients under age 40;
indicates statistical significance.
In the subgroup analysis of patients with minimal versus extensive ETE (Table 4), patients with extensive ETE were more likely to be older (p<0.001), Black (p=0.028), Hispanic (p=0.044), have larger tumors (p<0.001), insured by Medicare (p<0.001), treated at Academic or Comprehensive Community Cancer Programs (p=0.002), and treated at low- or mid-volume facilities (p=0.002). Interestingly, there was no statistically significant difference in extent of surgery between patients who had minimal versus extensive ETE (92.4% TT vs 94.5% TT, p=0.089), or in receipt of RAI (60.9% vs 64.5%, p=0.098). There was also no difference based on patient sex.
Table 4.
Subgroup analysis of minimal versus extensive ETE groups. Data presented as n (group %) or median (interquartile range).
| Variable | Minimal ETE (n=5,310) | Extensive ETE (n=541) | P-value |
|---|---|---|---|
|
| |||
| Age (years) | 56 (44–66) | 62 (49–73) | <0.001 * |
|
| |||
| Sex | |||
| Female | 4,234 (79.7) | 428 (79.1) | 0.73 |
| Male | 1,076 (20.3) | 113 (20.9) | |
|
| |||
| Race | |||
| White | 4,251 (80.1) | 430 (79.5) | |
| Asian or Pacific Islander | 492 (9.3) | 46 (8.5) | 0.028* |
| Black | 346 (6.5) | 44 (8.1) | |
| Other or unknown | 203 (3.8) | 15 (2.7) | |
| Native American | 18 (0.3) | 6 (1.1) | |
|
| |||
| Ethnicity | |||
| Non-Hispanic | 4,435 (83.5) | 444 (82.1) | 0.044* |
| Hispanic | 713 (13.4) | 88 (16.3) | |
| Unknown | 162 (3.1) | 9 (1.7) | |
|
| |||
| Surgery | |||
| Total Thyroidectomy | 4,909 (92.4) | 511 (94.5) | 0.089 |
| Lobectomy | 401 (7.6) | 30 (5.5) | |
|
| |||
| Tumor size category | |||
| 1.0 to 2.0 cm | 3,210 (60.5) | 268 (49.5) | <0.001* |
| 2.1 to 3.0 cm | 1,338 (25.2) | 169 (31.2) | |
| 3.1 to 4.0 cm | 762 (14.4) | 104 (19.2) | |
|
| |||
| AJCC Pathologic stage† | |||
| Stage I | 1,285 (24.2) | 113 (20.9) | |
| Stage II | 439 (8.3) | 2 (0.4) | <0.001* |
| Stage III | 2,570 (48.4) | 106 (19.6) | |
| Stage IV | 2 (0.0) | 255 (47.1) | |
| Unknown | 1,014 (19.1) | 65 (12.0) | |
|
| |||
| Received RAI | |||
| Yes | 3,232 (60.9) | 349 (64.5) | 0.098 |
| No | 2,078 (39.1) | 192 (35.5) | |
|
| |||
| Insurance payor | |||
| Private Insurance | 3,209 (60.4) | 240 (44.4) | |
| Medicare | 1,432 (27.0) | 224 (41.4) | |
| Medicaid | 385 (7.3) | 50 (9.2) | <0.001* |
| Not Insured | 150 (2.8) | 16 (3.0) | |
| Unknown | 89 (1.7) | 5 (0.9) | |
| Other Government | 45 (0.8) | 6 (1.1) | |
|
| |||
| Facility type | |||
| Academic | 1,863 (35.1) | 198 (36.6) | |
| Comprehensive Community Cancer Program | 1,446 (27.2) | 176 (32.5) | 0.002* |
| Unknown* | 927 (17.5) | 67 (12.4) | |
| Integrated Network Cancer Program | 912 (17.2) | 78 (14.4) | |
| Community Cancer Program | 162 (3.1) | 22 (4.1) | |
|
| |||
| Facility volume (analytic cohort) | |||
| Low | 467 (8.8) | 70 (12.9) | 0.002* |
| Mid | 899 (16.9) | 101 (18.7) | |
| High | 3,944 (74.3) | 370 (68.4) | |
Pathologic stage determined by American Joint Committee on Cancer (AJCC) 7th (2010–2017) or 8th edition (2018–2020) based on year of diagnosis.
Facility type is suppressed by the National Cancer Database for patients under age 40.
When analyzing receipt of RAI by extent of surgery group, 131 (30.4%) of the lobectomy patients were noted to have received RAI, compared to 3,450 (63.7%) of the total thyroidectomy group (p<0.001) (Table 2). As this was an unexpected finding given that RAI typically requires complete thyroid ablation prior to administration, these 131 patients were further compared to the non-Lobe+RAI group. The Lobe+RAI patients were not statistically significantly different from the rest of the cohort with the exception of larger tumor sizes, with increased proportions of tumors 2.1–3.0cm (30.5% vs 25.6%) and 3.1–4.0cm (21.4% vs 14.7%) compared to the higher proportion of smaller 1.0–2.0cm tumors among the non-Lobe+RAI group (59.7% vs 48.1%) (p=0.018). The only other statistically significant difference between groups was a higher proportion of care at low volume facilities in the Lobe+RAI group versus non-Lobe+RAI group (15.3% vs 9.0%) compared to mid volume (16.8% vs 17.1%) or high volume facilities (67.9% vs 73.9%) (p=0.049).
Next, this Lobe+RAI cohort was excluded in a sensitivity analysis given the small but nonzero possibility of miscoded T-stage or misallocation of surgical treatment (Supplemental Table 2). Exclusion of these patients removed approximately half of the patients in the original cohort who received lobectomy and had extensive ETE. The multivariable logistic regression again showed a decreased likelihood of TT in later years (2015 through 2020). Additional factors were associated with total thyroidectomy including extensive ETE (OR 1.90, 95% CI 1.11–3.26), female sex (OR 1.43, 95% CI 1.07–1.92), and treatment at a Comprehensive Community Cancer Program (OR 1.53, 95% CI 1.03–2.28) or Unknown facility type (OR 0.47, 95% CI 0.29–0.76) versus Academic reference group. There were no other significant differences based on patient race, tumor size or facility volume.
Discussion
There is large variation in access and treatment for thyroid cancer.9,10 Though the 2015 ATA guidelines supported lobectomy alone for low-risk DTC, extrathyroidal extension remains a poor prognostic factor for which total thyroidectomy is indicated.11,12 In this retrospective analysis of contemporary trends in extent of surgery in a national cohort of DTC patients with ETE, 7.4% received lobectomy alone, with a notable two-fold increase in the rate of lobectomy for +ETE DTC across the study period from 5.1% in 2010 to 10.7% in 2020. Multivariable logistic regression demonstrated a decreased likelihood of TT for diagnosis years 2015 through 2020 relative to 2010, which may reflect an unintentional decrease in guideline-concordant TT for +ETE DTC.2
Recognizing that selecting an operative approach involves a variety of factors that cannot be fully captured in a retrospective observational study, our results suggest that the majority of this +ETE cohort received guideline-concordant surgical resection with total thyroidectomy, but with a notable decrease in TT starting in 2015, indicating possible alignment with the 2015 ATA guidelines. Interestingly, patients with extensive ETE or larger tumor size had no statistically significant increase in likelihood of TT on multivariable analysis.
Taken together, these findings suggest that the trend toward minimal surgery for low-risk DTC following release of the 2015 ATA guidelines may have led to an unintentional decrease in guideline-concordant TT for +ETE DTC. While implementation of clinical guidelines is a notoriously difficult and often lengthy undertaking,14,15 it is important to recognize potential scenarios for misapplication of guidelines in high-risk groups that could have clinical ramifications. Our findings support that while the vast majority of +ETE DTC patients in this contemporary cohort received guideline-concordant TT, concern for undertreatment of this high-risk group remains, with a statistically significant decreased likelihood of TT noted in the last half of the study period when controlling for other patient and disease factors. Other reasons for this increase in lobectomy rates apart from the guideline change may have been increasing surgeon, endocrinologist, or patient preference for lobectomy given the reduced risk of need for thyroid hormone supplementation or nerve injury, though long-term quality of life outcomes are likely equivalent between surgical approaches.13 While the retrospective observational design of this study limited our ability to investigate alternative explanations for these trends, future qualitative work may be warranted to further understand underlying reasons for departures from guideline-concordant care of this population.
Classification of ETE has undergone important changes in the past decade as the prognostic differences of microscopic versus gross ETE have been elucidated, with multiple studies showing no differences in important clinical outcomes based on microscopic ETE alone.16–18 In in the 7th edition of AJCC staging system, “minimal” ETE to the strap muscles or perithyroidal soft tissues was classified as pT3, without specification for microscopic versus gross extension to these tissues.19 After differences in microscopic and macroscopic ETE were further recognized, the 8th edition AJCC T staging system (released in 2018) eliminated microscopic ETE from T staging and instead classified gross ETE to strap muscles as T3b, with more severe extension to other structures as T4a or T4b.8,20 While a portion of the lobectomy group in the 2010–2017 cohort may have had microscopic ETE alone as it was not possible to parse out this group based on available variables, lobectomy rates in the 2018–2020 group ranged from 8.4–10.7% (versus 2010–2015 rates of 4.8–6.1%), suggesting that this trend in extent of surgery persisted even after changes in the classification of ETE in AJCC T staging to restrict to gross ETE alone.
An important question moving forward that was unable to be addressed in this study will be understanding whether total thyroidectomy affords a clinically meaningful benefit for patients with microscopic or minimal gross ETE. Multiple studies have highlighted the lack of increased risk of disease-specific mortality or tumor recurrence for patients with microscopic ETE in DTC,16,17 which likely influenced changes in the 8th edition AJCC T staging that eliminated microscopic ETE as an element of T3 stage and updated T3b to represent gross strap muscle invasion.8,21 Debate remains, however, over the prognostic implications of gross ETE with extension into the perithyroidal soft tissue or strap muscles alone, termed by some as “minimal” gross ETE.22,23 In one large institutional retrospective study, ETE to strap muscles alone had a 3.3 times increased likelihood of disease-specific mortality in patients with PTC and age 55 or greater compared to those without macroscopic ETE.12 Contrastingly, another study of over 2,000 DTC patients found no difference in overall or disease-specific survival between patients with no ETE, ETE into perithyroidal soft tissue, or ETE into the strap muscles; however, there was an increase in locoregional recurrence for both of the ETE groups compared to non-ETE patients.24 Similarly, a meta-analysis from Diker-Cohen et al demonstrated no difference in disease-specific mortality for patients with minimal gross ETE, although a small increase in the absolute risk of recurrence was noted (3.5% with ETE, 2.2% without ETE).22
While not the primary focus of this study, rates of radioactive iodine (RAI) administration were lower than expected for this population of +ETE patients, with only 61.2% of the overall cohort receiving RAI, including 63.7% of the TT group. As RAI administration is currently recommended by ATA guidelines for patients with gross ETE, it was somewhat surprising that nearly four-tenths of patients in this contemporary cohort did not receive this therapy. One explanation for these findings may be the inclusion of non-gross ETE (i.e. microscopic ETE) based on pathologic T staging in the 2010–2017 subgroup. Based on 2015 ATA guidelines, RAI may be considered for microscopic ETE and is “[g]enerally favored based on risk of recurrent disease.”2 The 2009 ATA guidelines for DTC, which would have been in effect for the first half of this study period, recommended RAI for any gross ETE, and recommended “selective use” for those with “minimal” ETE (pT3).1 Therefore, it is feasible that RAI may still have been selected for use in patients with microscopic ETE alone. Further explanation about reasoning for or against RAI use for individual patients is not available in the NCDB and limits a more detailed interpretation of the rate of RAI use in this +ETE cohort.
Furthermore, a small subgroup of patients were noted to have undergone lobectomy alone and RAI. This phenomenon has been previously discussed by Kiernan et al, who studied an NCDB cohort (1998–2011) who underwent lobectomy for DTC, with 24% of this cohort receiving RAI.25 Although this practice is not recommended by ATA guidelines, the authors of that retrospective study noted a statistically significant overall survival advantage for postoperative RAI after lobectomy compared to lobectomy alone. Without further details about thyroid or surgical history as well as patient and surgeon preference that are not available in the NCDB, it is difficult to make definitive statements about the appropriateness of this practice based on this type of study.26 Our findings suggest that the group of patients who underwent lobectomy and RAI in this cohort were more likely to have larger tumor sizes and treated at lower volume facilities.
This study has several limitations. While the NCDB captures the majority of newly diagnosed thyroid cancers in the United States, important differences in extent of surgery for +ETE could exist between patients treated at CoC-accredited hospitals versus those treated at non-CoC accredited facilities. The noted differences in AJCC T staging as related to ETE classification changed in the middle of this study period. While this likely created an uneven representation of ETE before and after the change (i.e. both microscopic and gross ETE cases in the 2010–2017 group but only gross ETE in 2018–2020), the intention of this analysis was to evaluate overarching trends in extent of surgery for ETE as it was classified at the time of diagnosis and treatment, including the periods before and after the 8th edition AJCC staging system in 2018 as well as impact from the 2015 ATA Guidelines.
Additional notable limitations include the lack of available data in the NCDB detailing other important factors that may have influenced surgeons’ decision making about an operative approach in this cohort, such as patient preference, history of prior neck operations or recurrent laryngeal nerve injury, need for lifelong thyroid hormone supplementation, socioeconomic or transportation barriers to returning for completion thyroidectomy, or risk of primary or nodal recurrence.27 Identification of completion thyroidectomy cases is not available in the NCDB and limits our ability to make inferences about extent of surgery in a more nuanced manner. As NCDB does not provide cancer recurrence data, we were unable to evaluate whether the observed deviations from guideline-concordant care – though they may have been clinically appropriate depending on other factors mentioned above – lead to meaningful differences in locoregional recurrence or other relevant oncologic outcomes for DTC. While the intent of this study was to examine trends in extent of surgery for +ETE DTC in a contemporary U.S. cohort, further analysis of the relationship of surgical approach for +ETE DTC and surgical outcomes such as cancer recurrence and disease-specific survival is needed to evaluate the long-term implications of differences in extent of surgery for this population.
Conclusion
While most patients with +ETE DTC received guideline-concordant total thyroidectomy in this contemporary sample, rates of lobectomy doubled during the study period and likelihood of total thyroidectomy was significantly decreased starting in 2015, suggesting that an unintentional decrease in guideline-concordant TT for high-risk +ETE DTC may have been influenced by the 2015 ATA guideline recommendations for low-risk DTC. Updates in the 8th edition AJCC T staging related to ETE may assist with appropriate selection of surgical approach for evidence-based care of this high-risk DTC population. Further study is needed to understand influential factors involved in the implementation of national guidelines across different disease risk levels.
Supplementary Material
Highlights:
Gross extrathyroidal extension in DTC is associated with worse outcomes
7.4% of this cohort with ETE received lobectomy alone
Rates of lobectomy nearly doubled over the course of the study period
With trends toward lobectomy, high-risk DTC patients may be undertreated
Acknowledgements:
The National Cancer Database (NCDB) is a joint project of the Commission on Cancer (CoC) of the American College of Surgeons and the American Cancer Society. The CoC’s NCDB and the hospitals participating in the CoC’s NCDB are the source of the de-identified data used herein; they have not verified and are not responsible for the statistical validity of the data analysis or the conclusions derived by the authors.
Financial Support:
No direct funding for this work was received from funding agencies in the public, commercial, or not-for-profit sectors. K.B. Montgomery receives support from the Agency for Healthcare Research and Quality on grant T32 HS013852, and K.K. Broman from the American College of Surgeons, the American Society of Clinical Oncology, and the National Institutes of Health (National Center for Advancing Translational Sciences) on grant KL2 TR003097.
Footnotes
Conflicts of Interest: The authors have no relevant financial conflicts of interest to declare.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.American Thyroid Association (ATA) Guidelines Taskforce on Thyroid Nodules and Differentiated Thyroid Cancer, Cooper DS, Doherty GM, et al. Revised American Thyroid Association management guidelines for patients with thyroid nodules and differentiated thyroid cancer. Thyroid. 2009;19(11):1167–1214. doi: 10.1089/thy.2009.0110 [DOI] [PubMed] [Google Scholar]
- 2.Haugen BR, Alexander EK, Bible KC, et al. 2015 American Thyroid Association Management Guidelines for Adult Patients with Thyroid Nodules and Differentiated Thyroid Cancer: The American Thyroid Association Guidelines Task Force on Thyroid Nodules and Differentiated Thyroid Cancer. Thyroid. 2016;26(1):1–133. doi: 10.1089/thy.2015.0020 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Ullmann TM, Gray KD, Stefanova D, et al. The 2015 American Thyroid Association guidelines are associated with an increasing rate of hemithyroidectomy for thyroid cancer. Surgery. 2019;166(3):349–355. doi: 10.1016/j.surg.2019.03.002 [DOI] [PubMed] [Google Scholar]
- 4.Pradhan MC, Young CD, Thomas SM, Stang MT, Kazaure HS, Scheri RP. Outcomes of concurrent parathyroidectomy and thyroidectomy among CESQIP surgeons. The American Journal of Surgery. 2022;224(5):1190–1196. doi: 10.1016/j.amjsurg.2022.03.041 [DOI] [PubMed] [Google Scholar]
- 5.Gillis A, Chen H. There’s no such thing as a free lunch: Outcomes of concurrent parathyroidectomy and thyroidectomy among CESQIP surgeons. Am J Surg. 2022;224(5):1188–1189. doi: 10.1016/j.amjsurg.2022.05.025 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Boffa DJ, Rosen JE, Mallin K, et al. Using the National Cancer Database for Outcomes Research: A Review. JAMA Oncol. 2017;3(12):1722–1728. doi: 10.1001/jamaoncol.2016.6905 [DOI] [PubMed] [Google Scholar]
- 7.von Elm E, Altman DG, Egger M, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Epidemiology. 2007;18(6):800–804. doi: 10.1097/EDE.0b013e3181577654 [DOI] [PubMed] [Google Scholar]
- 8.Amin MB, Edge SB, Greene FL, et al. , eds. AJCC Cancer Staging Manual Eigth Edition. 8th ed. Springer; New York, NY; 2016. Accessed January 19, 2023. https://link.springer.com/book/9783319406176 [Google Scholar]
- 9.Luff MK, Kim J, Tseng CH, Livhits MJ, Yeh MW, Wu JX. Racial/ethnic disparities in thyroid cancer in California, 1999–2017. Am J Surg. Published online September 24, 2022:S0002–9610(22)00590–6. doi: 10.1016/j.amjsurg.2022.09.041 [DOI] [PubMed] [Google Scholar]
- 10.Goodsell K, Ermer J, Amjad W, Swisher-McClure S, Wachtel H. External beam radiotherapy for thyroid cancer: Patients, complications, and survival. Am J Surg. Published online January 11, 2023:S0002–9610(23)00010–7. doi: 10.1016/j.amjsurg.2023.01.009 [DOI] [PubMed] [Google Scholar]
- 11.Youngwirth LM, Adam MA, Scheri RP, Roman SA, Sosa JA. Extrathyroidal Extension Is Associated with Compromised Survival in Patients with Thyroid Cancer. Thyroid. 2017;27(5):626–631. doi: 10.1089/thy.2016.0132 [DOI] [PubMed] [Google Scholar]
- 12.Harries V, McGill M, Yuan A, et al. Does macroscopic extrathyroidal extension to the strap muscles alone affect survival in papillary thyroid carcinoma? Surgery. Published online September 30, 2021. doi: 10.1016/j.surg.2021.08.041 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Chen W, Li J, Peng S, et al. Association of Total Thyroidectomy or Thyroid Lobectomy With the Quality of Life in Patients With Differentiated Thyroid Cancer With Low to Intermediate Risk of Recurrence. JAMA Surg. 2022;157(3):200–209. doi: 10.1001/jamasurg.2021.6442 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Fischer F, Lange K, Klose K, Greiner W, Kraemer A. Barriers and Strategies in Guideline Implementation—A Scoping Review. Healthcare (Basel). 2016;4(3):36. doi: 10.3390/healthcare4030036 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Grol R, Grimshaw J. From best evidence to best practice: effective implementation of change in patients’ care. Lancet. 2003;362(9391):1225–1230. doi: 10.1016/S0140-6736(03)14546-1 [DOI] [PubMed] [Google Scholar]
- 16.Hay ID, Johnson TR, Thompson GB, Sebo TJ, Reinalda MS. Minimal extrathyroid extension in papillary thyroid carcinoma does not result in increased rates of either cause-specific mortality or postoperative tumor recurrence. Surgery. 2016;159(1):11–19. doi: 10.1016/j.surg.2015.05.046 [DOI] [PubMed] [Google Scholar]
- 17.Arora N, Turbendian HK, Scognamiglio T, et al. Extrathyroidal extension is not all equal: Implications of macroscopic versus microscopic extent in papillary thyroid carcinoma. Surgery. 2008;144(6):942–948. doi: 10.1016/j.surg.2008.07.023 [DOI] [PubMed] [Google Scholar]
- 18.Watanabe A, Nabata K, Wiseman SM. Synoptic operative reporting in cancer surgery: A scoping review. Am J Surg. Published online January 7, 2023:S0002–9610(23)00004–1. doi: 10.1016/j.amjsurg.2023.01.003 [DOI] [PubMed] [Google Scholar]
- 19.Edge SB, Byrd DR, Compton CC, Fritz AG, Greene FL, Trotti A, eds. AJCC Cancer Staging Manual Seventh Edition. 7th ed. Springer; New York, NY; 2010. Accessed January 12, 2023. https://link.springer.com/book/9780387884400 [Google Scholar]
- 20.Nam SH, Bae MR, Roh JL, et al. A comparison of the 7th and 8th editions of the AJCC staging system in terms of predicting recurrence and survival in patients with papillary thyroid carcinoma. Oral Oncology. 2018;87:158–164. doi: 10.1016/j.oraloncology.2018.11.003 [DOI] [PubMed] [Google Scholar]
- 21.Tuttle RM, Haugen B, Perrier ND. Updated American Joint Committee on Cancer/Tumor-Node-Metastasis Staging System for Differentiated and Anaplastic Thyroid Cancer (Eighth Edition): What Changed and Why? Thyroid. 2017;27(6):751–756. doi: 10.1089/thy.2017.0102 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Diker-Cohen T, Hirsch D, Shimon I, et al. Impact of Minimal Extra-Thyroid Extension in Differentiated Thyroid Cancer: Systematic Review and Meta-analysis. J Clin Endocrinol Metab. Published online March 1, 2018. doi: 10.1210/jc.2018-00081 [DOI] [PubMed] [Google Scholar]
- 23.Ji YB, Song CM, Kim D, et al. Efficacy of hemithyroidectomy in papillary thyroid carcinoma with minimal extrathyroidal extension. Eur Arch Otorhinolaryngol. 2019;276(12):3435–3442. doi: 10.1007/s00405-019-05598-z [DOI] [PubMed] [Google Scholar]
- 24.Amit M, Boonsripitayanon M, Goepfert RP, et al. Extrathyroidal Extension: Does Strap Muscle Invasion Alone Influence Recurrence and Survival in Patients with Differentiated Thyroid Cancer? Ann Surg Oncol. 2018;25(11):3380–3388. doi: 10.1245/s10434-018-6563-x [DOI] [PubMed] [Google Scholar]
- 25.Kiernan CM, Parikh AA, Parks LL, Solórzano CC. Use of radioiodine after thyroid lobectomy in patients with differentiated thyroid cancer: does it change outcomes? J Am Coll Surg. 2015;220(4):617–625. doi: 10.1016/j.jamcollsurg.2014.12.014 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Watanabe A, Wiseman SM. Quality indicators for thyroid cancer care: What should surgeons know? Am J Surg. Published online February 10, 2023:S0002–9610(23)00057–0. doi: 10.1016/j.amjsurg.2023.02.007 [DOI] [PubMed] [Google Scholar]
- 27.Schneider DF, Mazeh H, Chen H, Sippel RS. Lymph node ratio predicts recurrence in papillary thyroid cancer. Oncologist. 2013;18(2):157–162. doi: 10.1634/theoncologist.2012-0240 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.