Abstract
Background
Anaplastic thyroid cancer is an aggressive malignancy, for which prompt treatment is necessary. Surgical resection is the recommended initial treatment for resectable cases and most patients require postoperative radiation therapy (PORT). NCCN Guidelines state that adjuvant radiation should ideally begin 2–3 weeks postoperatively, however, limited evidence exists to support this guideline. Herein, we examine the National Cancer Database (NCDB) to assess practice patterns with regard to PORT timing, and to assess whether the time interval between surgery and the start of RT impacts overall survival (OS).
Methods
An NCDB data file was obtained and condensed to only include stage IVa and IVb anaplastic thyroid cancer cases during the years 2004 to 2018. Patients were included if they were treated with surgery followed by radiation (6000 cGy-7020 cGy), and the radiation commenced no later than 90 days after surgery. Patients were grouped according to PORT: Early PORT (≤ 21 days after surgery) and Late PORT (> 21 days after surgery). Several variables were tested using univariate and multivariate binary logistic regression to determine their effect on whether patients received Early PORT vs. Late PORT. These variables were also used to conduct univariate and multivariate Cox regression to determine effects on survival for the cohort. The covariate variables included: presence or absence of a chemotherapy regimen in the treatment plan, total thyroidectomy or alternate surgery, type of treatment facility (Academic vs. Other), resection margin status (R0/R1 vs. R2), Charlson-Deyo comorbidity score, tumor stage, age, race/ethnicity and sex.
Results
The analysis included 277 patients after exclusions. The Early PORT group included 57 patients, and the Late PORT group included 220 patients. Multivariate binary logistic regression analysis showed that patients with R2 resection margins were less likely to receive Late PORT compared to patients with R0/R1 resection margins (OR: 0.306 (0.123–0.764)), and patients with stage IVb disease were less likely to receive Late PORT compared to patients with stage IVa disease (OR: 0.496 (0.257–0.957)). Multivariate cox regression analysis displayed survival advantages for patients who received Late PORT (HR: 0.563 (0.401–0.790)) and patients with R0/R1 resection margins (HR: 1.901 (1.161–3.114)), while treatment at an academic center trended toward significance (p = 0.053).
Discussion
The majority of patients in this cohort received Late PORT, later than the proposed guidance by the NCCN. Resection margin status and disease stage appears to influence how physicians select the timing of PORT. Within the limitations of our retrospective study, delayed PORT beyond NCCN recommendations is not associated with a detriment to OS. Further prospective analyses may be necessary to identify the ideal timing of adjuvant radiation therapy.
Keywords: Anaplastic, Thyroid, Cancer, Radiation, Postoperative
Background
Thyroid cancer is the most common endocrine cancer, and incidence rates have been rising for both well-differentiated and anaplastic thyroid cancer (ATC) [1–3]. Although well-differentiated thyroid cancer is the most common and most curable, ATC is rare (1–5% of thyroid cancers) and the most aggressive. Recent data indicates that the median overall survival (OS) for ATC is approximately 5 months [4, 5]. Approximately 50% of patients with ATC are found to have distant metastasis at diagnosis [4]. Also, given the location of the thyroid in proximity to the trachea, esophagus, and critical blood vessels, the aggressive nature of ATC can lead to very rapid clinical deterioration when critical structures are invaded [6, 7].
Due to the aggressive nature of ATCs, clinicians have promoted the rapid initiation of treatment for these patients. Surgery is the mainstay of treatment for operable patients. National Comprehensive Cancer Network (NCCN) guidelines recommend post-operative radiation therapy (PORT) with the option for concurrent radiosensitizing chemotherapy for suitable patients [8]. More specifically, these guidelines suggest that the ideal timeline for initiation of PORT in patients with R0 or R1 resection margins following thyroid resection should be 2 to 3 weeks, and the guidelines also suggest that patients with gross residual disease (R2 resection) following thyroid resection should receive radiation therapy “as quickly as possible” [8]. Since no prospective studies have explored the optimal timing of PORT for resected ATC, these recommendations are based on expert guidance in the context of a tumor that displays rapid growth characteristics [9].
In this study, we utilized the National Cancer Database (NCDB) to assess practice patterns and survival results regarding the use of Early PORT (≤ 21 days after surgery) versus Late PORT (> 21 days after surgery). This work has been presented in part at a recent meeting of the American Association for Cancer Research (AACR) [10].
Methods
An NCDB datafile containing de-identified patient information for patients diagnosed with thyroid cancer between 2004 and 2018 at participating institutions in the United States was obtained. Research involving publicly available datasets, such as this NCDB datafile, does not require institutional review board (IRB) approval at our institution. Histological (ICD-0-3) codes in the dataset were analyzed to exclude patients who were not diagnosed with ATC. Patients with the following ICD-0-3 codes were selected for further analysis: 8020, 8021, and 8022 [11]. There were 3,951 patients with ATC or grade 4 thyroid cancer. Staging in this analysis was conducted using the sixth and seventh editions of the AJCC Staging Manual, which did not differ in relation to criteria for stage IVa and IVb ATC [12, 13]. Patients were included if they had stage IVa or IVb disease, received surgery followed by PORT with doses of 6000–7020 cGy, and if their PORT was less than 90 days after surgery (Fig. 1). The 90-day cutoff was used to avoid an immortality bias with the aim of including patients who had PORT as part of the initial treatment plan. These exclusions resulted in a cohort of 277 patients. The distribution of patient characteristics between the Early and Late PORT groups was assessed by chi square analysis for categorical variables and t-test for continuous variables.
Fig. 1.
Consort Flow diagram detailing inclusion and exclusion criteria for patients
Univariate binary logistic regression was conducted to assess factors that may have predisposed patients to receive Early PORT or Late PORT. The independent variables used in this regression analysis included the following factors: resection margin status (R0/R1 vs. R2), staging (IVa vs. IVb), type of surgery (total thyroidectomy vs. alternative procedure), chemotherapy regimen (regimen initiated vs. not initiated), type of treatment facility (academic vs. non-academic), race/ethnicity (white vs. others), sex, age (< 65 vs. >65), and Charlson-Deyo Score (0 vs. >0). Multivariate binary logistic regression was conducted for variables that displayed statistically significant associations on univariate analysis (if p < 0.05). Next, these same variables, in conjunction with PORT timing groups (early vs. late) were used to conduct a univariate cox regression analysis to determine variables affecting survival. Variables with statistically significant effects on survival in univariate analysis were used in a multivariate cox regression analysis of survival.
Subsequently, univariate Kaplan-Meier survival curves were constructed for survival and included the variables of time from surgery to radiation, post-surgical margin status, treatment facility type (academic/research vs. other), and the use of chemotherapy. The Academic/Research programs are designated by the Commission on Cancer (CoC) Accreditation Program and include Academic Comprehensive Cancer Programs and NCI-Designated Comprehensive Cancer Center Programs [14]. The analyses involving the use of chemotherapy divided patients into groups based on whether chemotherapy was administered at any point in their treatment course. Stage of disease was analyzed as IVa vs. IVb.
Results
There were 277 patients with ATC who were included for analysis in this study and 57 patients received Early PORT while 220 received Late PORT. The balance of patient characteristics with respect to Early vs. Late PORT was assessed and is shown in Table 1. Mean time from date of surgery to initiation of PORT was 16.8 days for the Early PORT cohort and 38.9 days for the Late PORT cohort, and these cohorts received chemotherapy in 84.2% and 80% of cases respectively (P = 0.471). There was also similarity in the proportion of patients who underwent a total thyroidectomy: 61.4% in the Early PORT group and 68.2% in the Late PORT group. Statistically significant differences in resection margin status (P = 0.003), and stage (P = 0.015) were noted between patient cohorts.
Table 1.
Characteristics of patients in the Early PORT and Late PORT Cohorts
| Characteristics of the ≤ 21 Day Post-Operative Radiation Timing Cohort (Early PORT): | Characteristics of the > 21 Day Post-Operative Radiation Timing Cohort (Late PORT): | P-Value | ||
|---|---|---|---|---|
| Total Patients | 57 Patients | 220 Patients | ||
| Mean Days from Surgery to Radiation | 16.82 +/- 4.36 Days | 38.96 +/- 13.12 Days | P = < 0.001 | |
| Mean Age | 65.63 +/- 11.24 Years | 65.06 +/- 11.37 Years | P = 0.6391 | |
| Sex | Male | 31 Patients (54.4%) | 105 Patients (47.7%) | P = 0.370 |
| Female | 26 Patients (45.6%) | 115 Patients (52.3%) | ||
| Race | White | 51 Patients (89.5%) | 194 Patients (88.2%) | P = 0.786 |
| Non-White | 6 Patients (10.5%) | 26 Patients (11.8%) | ||
| Chemotherapy Regimen Receipt | 48 Patients (84.2%) | 176 Patients (80.0%) | P = 0.471 | |
| Resection Margins | R0/R1 | 43 Patients (75.4%) | 193 Patients (87.7%) | P = 0.003 |
| R2 | 10 Patients (17.5%) | 12 Patients (5.5%) | ||
| Unknown | 4 Patients (7.0%) | 15 Patients (6.8%) | ||
| Type of Surgery: Total Thyroidectomy | 35 Patients (61.4%) | 150 Patients (68.2%) | P = 0.333 | |
| Stage | 4 A | 15 Patients (26.3%)* | 97 Patients (44.1%)* | P = 0.015 |
| 4B | 42 Patients (73.7%)* | 123 Patients (55.9%)* | ||
| Facility Type | Academic Program | 36 Patients (63.2%) | 121 Patients (55.0%)** | P = 0.268 |
| Other Program | 21 Patients (36.8%) | 99 Patients (45.0%) | ||
| Alive or Dead at Last Contact | Alive | 9 Patients(15.8%) | 65 (29.5%) | P = 0.036 |
| Dead | 48 Patients (84.2%) | 155 (70.5%) | ||
| Mean Survival Time (Months from Diagnosis) | 15.73 +/- 19.26 | 30.66 +/- 34.92 | P = 0.0021 | |
Univariate binary logistic regression showed that patients with R2 resection margins following surgery and patients with stage IVb disease were less likely to receive Late PORT compared to patients with R0/R1 resections or patients with IVa disease (P = 0.004 and P = 0.016 respectively) (Table 2). Additionally, multivariate binary logistic regression was conducted for both variables with statistically significant results on univariate analysis, and it was found that patients with R2 resection margins following surgery and patients with stage IVb disease were less likely to receive Late PORT on multivariate analysis as well (P = 0.011 and P = 0.036 respectively, Table 2).
Table 2.
Binary logistic regression assessing likelihood of receiving late PORT
| Univariate Binary Logistic Regression | ||
| Variables used in Univariate Analysis | Odds Ratio (95% CI) | P-Value |
|
Resection Margins: R0/R1 vs. R2 (Reference Category: R0/R1) |
0.267 (0.108–0.659) | 0.004 * |
|
Staging: IVa vs. IVb (Reference Category: IVa) |
0.453 (0.237–0.865) | 0.016* |
|
Type of Surgery: Total Thyroidectomy vs. Other Procedure (Reference Category: Other Procedure) |
1.347 (0.736–2.464) | 0.334 |
|
Chemotherapy: Initiated vs. Not Initiated (Reference Category: Not Initiated) |
0.750 (0.342–1.644) | 0.473 |
|
Facility of Treatment: Academic vs. Non-Academic (Reference Category: Non-Academic) |
0.713 (0.391–1.299) | 0.269 |
|
Race: White vs. Other (Reference Category: Other) |
0.786 (0.445–2.916) | 0.786 |
|
Sex: Male vs. Female (Reference Category: Male) |
1.306 (0.728–2.343) | 0.371 |
|
Age: <65 vs. >65 (Reference Category: <65) |
1.152 (0.643–2.065) | 0.634 |
|
Charlson-Deyo (CD) Score: 0 (No comorbidities) vs. 1–3 (Reference Category: CD Score: 0) |
1.250 (0.617–2.533) | 0.536 |
| Multivariate Binary Logistic Regression | ||
| Variables used in Multivariable Analysis | Odds Ratio (95% CI) | P-Value |
|
Resection Margins: R0/R1 vs. R2 (Reference Category: R0/R1) |
0.306 (0.123–0.764) | 0.011* |
|
Staging: IVa vs. IVb (Reference Category: IVa) |
0.496 (0.257–0.957) | 0.036* |
Several univariate Kaplan-Meier analyses were performed on variables of interest. Improved cumulative survival was observed in the patients who belonged to the Late PORT cohort compared to the use of Early PORT on univariate Kaplan-Meier analysis (P = < 0.001, Fig. 2a). This trend in improved cumulative survival was observed in the Late PORT cohort for patients with stage IVa disease (P = < 0.001, Fig. 2b) and patients with stage IVb disease (P = 0.045) (Fig. 2c). When comparing overall survival of patients who received chemotherapy with patients who did not receive chemotherapy, there was a slight advantage for the use of chemotherapy, but it was not significant (P = 0.439, Fig. 3). Patients with R0/R1 resections had improved overall survival compared to R2 resections (Fig. 4a), and Late PORT was associated with improved survival in patients with R0/R1 resections (P = 0.002), but not in the group with R2 resections (P = 0.331, Fig. 4b and c). Since others have shown improved results for the treatment of thyroid cancers at large tertiary/academic centers [15, 16], additional survival analyses were performed to determine variations in survival outcomes for patients who were treated at Academic/Research programs vs. other facilities (Fig. 5). Amongst the entire patient population, there was a significant survival advantage for patients who were treated at Academic/Research programs (Fig. 5a, P = 0.048). Regardless of whether patients were treated at an Academic/Research programs or at other facilities, there was a survival advantage for Late PORT compared to Early PORT (Fig. 5b and c).
Fig. 2.
For the overall group, Late PORT showed improved survival compared to Early PORT (A) and this was true for the stage IVa group (B) and the stage IVb group (C)
Fig. 3.
The use of chemotherapy did not show a survival advantage
Fig. 4.
Patients with R0/R1 resections showed improved survival compared to R2 resections (A). Late PORT showed improved survival compared to Early PORT for the patients who had R0/R1 resections (B), but not those who had R2 resections (C)
Fig. 5.
Patients treated at academic/research centers had improved survival compared to other centers (A) and Late PORT was associated with improved survival regardless of the type of center (B, C)
Several treatment-related and demographic factors were analyzed relative to survival using cox regression analysis. First, univariate cox regression analysis (Table 3) demonstrated an association with improved survival for the following factors as compared to their counterparts: Late PORT (HR: 0.513, P = < 0.001), R0 or R1 resection margins (HR: 2.554, P = < 0.001), stage IVa disease (HR: 1.403, P = 0.020), and patients who were treated at an academic facility (HR: 0.758, P = 0.049). Multivariate cox regression analysis was conducted on variables with significant results (p < 0.05) on univariate analysis as shown in Table 3. On multivariate analysis, Late PORT (HR: 0.563, P < 0.001) and R0/R1 resection margins (HR: 1.901, P = 0.011) were associated with improved survival. Notably, on multivariate analysis, patients who were treated at academic centers were found to have a possible survival advantage with results approaching significance (HR: 0.756, P = 0.053).
Table 3.
Cox regression analysis evaluating factors associated with overall survival
| Univariate Cox Regression Analysis | ||
| Variables used in Univariate Analysis | Hazard Ratios (95% CI) | P-Value |
|
Early PORT (≤ 21 Days) vs. Late PORT (> 21 Days) (Reference Category: Early PORT) |
0.513 (0.370–0.710) | < 0.001* |
|
Resection Margins: R0/R1 vs. R2 (Reference Category: R0/R1) |
2.554 (1.598–4.081) | < 0.001* |
|
Staging: IVa vs. IVb (Reference Category: IVa) |
1.403 (1.056–1.864) | 0.020* |
|
Type of Surgery: Total Thyroidectomy vs. Other Procedure (Reference Category: Other Procedure) |
0.789 (0.592–1.053) | 0.107 |
|
Chemotherapy: Initiated vs. Not Initiated (Reference Category: Not Initiated) |
0.873 (0.620–1.231) | 0.440 |
|
Facility of Treatment: Academic vs. Non-Academic (Reference Category: Non-Academic) |
0.758 (0.575–0.998) | 0.049* |
|
Race: White vs. Other (Reference Category: Other) |
1.032 (0.674–1.582) | 0.884 |
|
Sex: Male vs. Female (Reference Category: Male) |
0.854 (0.648–1.125) | 0.262 |
|
Age: <65 vs. >65 (Reference Category: <65) |
1.240 (0.941–1.634) | 0.127 |
|
Charlson-Deyo (CD) Score: 0 (No comorbidities) vs. 1–3 (Reference Category: CD Score: 0) |
1.202 (0.878–1.645) | 0.251 |
| Multivariate Cox Regression Analysis | ||
| Variables use in Multivariate Analysis | Hazard Ratios (95% CI) | P-Value |
|
Early PORT (≤ 21 Days) vs. Late PORT (> 21 Days) (Reference Category: Early PORT) |
0.563 (0.401–0.790) | < 0.001* |
|
Resection Margins: R0/R1 vs. R2 (Reference Category: R0/R1) |
1.901 (1.161–3.114) | 0.011* |
|
Staging: IVa vs. IVb (Reference Category: IVa) |
1.231 (0.913–1.659) | 0.173 |
|
Facility of Treatment: Academic vs. Non-Academic (Reference Category: Non-Academic) |
0.756 (0.569–1.004) | 0.053 |
Discussion
This analysis focused on assessing variations in the timing of the initiation of PORT for patients with surgically resected ATC. The results of this analysis showed that in certain patient populations, patients who initiated radiation treatments at a time greater than 3 weeks after their initial surgery (Late PORT) demonstrated improved survival as compared to patients who initiated radiation treatments within a 3-week timeframe following surgery (Early PORT). It is noteworthy that the NCCN guidelines recommend Early PORT based on the aggressive nature of tumors. In this study, when analyzing practice patterns related to timing of PORT, and it was found that patients with IVb disease and patients with R2 resection margins were less likely to receive Late PORT than patients with IVa disease or R0/R1 resection margins, respectively. Utilizing univariate results, a subsequent multivariate Cox regression analysis demonstrated a significant survival advantage for patients who received Late PORT and patients with R0/R1 resection margins while patients treated at academic facilities were found to have a survival advantage that was trending toward significance (P = 0.053). These findings build on the work of others and provide further information with respect to a better understanding of the optimal treatment timeline for patients diagnosed with ATC [17, 18].
The timing of different segments of the diagnosis/treatment algorithm for ATC has been analyzed in a variety of formats. Following diagnosis, appropriate staging and other assessments such as genetic testing (important for treatment selection) can delay treatments [17–19]. These delays may be harmful in rapidly progressing disease, but they need to be done as quickly as possible to help decide subsequent steps [20]. To this end, one effort that has focused on this portion of the treatment timeline has been the FAST (Facilitating Anaplastic Thyroid Cancer Specialized Treatment) program, which was initiated at MD Anderson and is focused on reducing the amount of time between referral and initial consultation [19]. This program resulted in the mean-time from referral to initial consultation to be reduced from 8.5 days to 0.5 days at their institution [19]. Efforts such as this program can be effective for streamlining patient care prior to the initiation of multimodal therapy, but once a treatment course is initiated, it can be difficult to optimize timelines between different modalities for a myriad of reasons [21, 22]. Also, one study found that approximately 40% of ATC cases were not diagnosed pathologically prior to an initial surgery, which causes variability in the start of timelines [21].
When one reviews our results, the limitations of retrospective analyses must be considered. Postoperative surgical complications may not be captured in the NCDB, and these complications could have implications regarding the efficacy of PORT relative to the timing of PORT [11]. Therefore, it is possible that our results reflect poor outcomes for the Early PORT group relative to the Late PORT group as Late Port may allow for better healing and improved oxygenation of the tissues. Alternatively, there may have been patient characteristics such as preoperative rapid and aggressive growth of the tumor that is not recorded in the NCDB, which may have contributed to poor results in the Early PORT group.
However, the findings related to timing of PORT in this analysis are likely multi-faceted, especially in relation to stage of disease. The multivariate Cox regression showed only a trend towards worse survival for stage IVb disease compared to stage IVa disease. Both stages showed an improved survival with late PORT, but staging in relation to PORT deserves additional inspection. Patients with stage IVb disease were less likely to receive Late PORT than patients with stage IVa disease (Table 2). These findings could lead to several future explorations. For instance, it is important to remember that the AJCC 6th and 7th staging systems were used to differentiate stage IVa and stage IVb disease in this study. When using these staging systems, the only difference between stage IVa and IVb disease is extra-thyroidal extension that is present in IVb disease. Unlike the 8th edition of the staging system, these staging systems classify all anaplastic thyroid cancer cases as T4 disease and do not account for aggressive characteristics such as lymph node metastasis to help differentiate patients with IVa or IVb disease. Also, the magnitude of extrathyroidal extension cannot be inferred from the database and may impact these results. Therefore, it is plausible that unreported but aggressive tumor findings at surgery prompted the implementation of Early PORT. This scenario may indicate selection bias for patients receiving Early PORT, which could harm interpretation of survival results for Early PORT vs. Late PORT patients. This potential selection bias could lead to interpretations that overestimate the possible detrimental effects of Early PORT or beneficial effects of Late PORT. This possibility requires further study with more specific parameters for tumor characteristics prior to treatment initiation.
Treatment facility type has been a topic of discussion in previous studies related to ATC treatment [15, 16]. Programs such as FAST at MD Anderson are examples of how the treatment facility can play a large role in the quality of care that patients receive [18]. Also, patients treated at high-volume academic tertiary centers may benefit from the efficiency and expertise associated with this greater volume of patients [19, 20, 23]. In our analysis, patients treated at Academic/Research programs were found to have a survival advantage over patients treated at other facilities on univariate cox regression analysis (HR: 0.758; P = 0.049, Table 3), and on multivariate cox regression analysis, there was a survival advantage that was trending toward significance (HR: 0.756; P = 0.053). This is consistent with previous findings in the literature regarding treatment outcomes at tertiary centers [20]. However, it is interesting that there was not a significant difference between the proportion of patients who received Late PORT based on the type of facility (Table 1). Also, Late PORT was found to be advantageous regardless of facility type (Fig. 5).
The NCCN guidelines group patients with R0 and R1 resection margins into the same category when recommending PORT within 2 to 3 weeks and R2 should receive PORT as quickly as possible [8]. In our study population, the 236 patients who had R0 or R1 resection margins showed improved survival over the 22 patients who had R2 resections (Table 3), but it is notable that patients with R2 resections were more likely to receive Early PORT (Table 2). Similar to the findings for patients with stage IVb (vs. IVa) disease noted above, it is possible that the R2 patients may have had advanced subclinical disease (or other poor prognostic findings), and therefore, the timing of PORT may be less relevant to survival. Notably, amongst the 236 patients with R0/R1 resection margins who were included in this analysis, only 43 patients, or 18.2%, of the R0/R1 cohort received Early PORT (Table 1). Therefore, the large majority of patients in this group received Late PORT, which could be related to various facility-specific logistical constraints or physician preferences.
The type of surgery performed is another important potential confounding factor related to the interpretation of PORT timing and survival. It is known that total thyroidectomy is the preferred treatment modality for patients with disease burden that is amenable to this type of procedure. In this dataset and analysis, patients were grouped into subpopulations that either received total thyroidectomy or “other procedure” (lobectomy, isthmusectomy, debulking). This analysis format allowed for binary analyses, but it may have overly simplified the relationship between surgery type and survival data. The decision to format surgery type groups in this way may have biased results toward a survival advantage for Late PORT patients if patients with more aggressive disease were more likely to receive subtotal resections and Early PORT. Therefore, further analyses that account for more specific data related to disease burden and factors affecting type of surgical approach would help determine whether surgical approach plays a more important role in survival data when compared to the results of this study.
Several studies have suggested that the use of chemotherapy with PORT in ATC patients is beneficial with respect to local control and survival [8, 24, 25]. The report herein showed that there was only a slight trend for improved survival with the use of chemotherapy on univariate cox regression analysis (P = 0.440, Table 3; Fig. 3), but the comparison was limited by the fact that the vast majority of patients (> 80%) received chemotherapy. Also, our analysis was not able to account for the type of chemotherapy regimen or timing of chemotherapy. In a recent SEER analysis, a significant cancer-specific survival benefit was displayed amongst patients who were placed on a chemotherapy plus a radiation regimen compared to radiation alone, and this survival advantage was consistent in multivariable analysis accounting for surgical resection, distant metastasis, and tumor extension [24]. The types of systemic therapy utilized for ATC include doxorubicin, combinations of Paclitaxel with Pazopanib or Carboplatin as well as promising new targeted therapies/immunotherapies such as combination therapy of dabrafenib and trametinib with or without pembrolizumab [9, 25–30].
Another important limitation of the NCDB is the suboptimal information on radiation fractionation. Many patients had total dose reported, but the type of fractionation and the frequency of breaks were not reported. Further analyses will likely be necessary to provide helpful information related to the optimization of radiation fractionation [10, 31–33]. For patients with ATC who have undergone surgery with R0 or R1 resection margins, NCCN guidelines currently propose dose and fractionation schedules of “60–66 Gy in 1.2 Gy twice-daily fractions or 1.8-2 Gy daily fractions”[ 8 ].
Conclusions
This analysis provides insight into practice patterns as well as outcomes related to the multimodal treatment of anaplastic thyroid cancer, with a specific focus on the timeframe between surgery and the initiation of radiation. Surgical resection with consideration of multimodality treatments are the standard options for ATC [34, 35]. This analysis showed that current practice recommendations largely favor the initiation of PORT at a time greater than three weeks after surgery, but there are factors such as advanced tumor staging and R2 resection margin status that increase the likelihood that physicians will implement PORT within a three-week timeframe. These survival results from this NCDB analysis are limited by lack of information on the type of chemotherapy, radiation, fractionation, preoperative tumor growth rate, and lack of detailed comorbidity information. These survival results will be useful as a starting point for further analyses to determine whether factors such as delayed surgical healing time or molecular characteristics of disease could explain the significant improvement in survival for patients who received Late PORT compared to Early PORT. We did not find a benefit for Early PORT in any subgroups, but the timing of PORT needs to be a clinical decision based on the individual clinical scenario.
Acknowledgements
None to report.
Author contributions
AB performed data collection, data analysis, and manuscript preparation. AK supervised AB throughout the data collection and analysis process and provided key contributions during manuscript preparation. HC, SS, and CW aided in interpreting data and provided feedback on manuscript preparation.
Funding
No funding sources to declare.
Data availability
Data for this study obtained by applying for a participant user file from the American College of Surgeons website. This file contained de-identified patient information for patients diagnosed with thyroid cancer between 2004 and 2018 at participating institutions in the United States. The website to view the application for these datafiles is included here: [https://www.facs.org/quality-programs/cancer-programs/national-cancer-database/puf/](https://www.facs.org/quality-programs/cancer-programs/national-cancer-database/puf/)The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request. Email address of corresponding author: [abjbonner@gmail.com](mailto: abjbonner@gmail.com).
Declarations
Ethics approval and consent to participate
At our institution, research involving publicly available datasets do not require institutional review board approval.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Presented at the American Association for Cancer Research – American Head and Neck Society (AACR-AHNS) Head and Neck Cancer Conference in Montreal, Canada on August 3, 2023.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Bonner A, et al. The association of socioeconomic factors and well-differentiated thyroid cancer. J Surg Res. 2023;283:973–81. 10.1016/j.jss.2022.11.033. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Lin B, et al. The incidence and survival analysis for anaplastic thyroid cancer: a SEER database analysis. Am J Transl Res. 2019;11:5888–96. [PMC free article] [PubMed] [Google Scholar]
- 3.Janz TA, Neskey DM, Nguyen SA, Lentsch EJ. Is the incidence of anaplastic thyroid cancer increasing: A population-based epidemiology study. World J Otorhinolaryngol Head Neck Surg. 2019;5:34–40. 10.1016/j.wjorl.2018.05.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Schmied M, et al. Radio(chemo)therapy in anaplastic thyroid cancer—high locoregional but low distant control rates—a monocentric analysis of a tertiary referral center. Strahlenther Onkol. 2022;198:994–1001. 10.1007/s00066-022-01943-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Jacob J, Vordermark D, Lorenz K, Medenwald D. Prognostic factors in radiotherapy of anaplastic thyroid carcinoma: a single center study over 31 years. Radiat Oncol. 2023;18:71. 10.1186/s13014-023-02249-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.B H. Rapid recurrence of anaplastic carcinoma: advocacy towards non-surgery. Endocrinol Metab Int J. 2017;5. 10.15406/emij.2017.05.00134
- 7.Shaha AR. Airway management in anaplastic thyroid carcinoma. Laryngoscope. 2008;118:1195–8. 10.1097/MLG.0b013e3181726d36. [DOI] [PubMed] [Google Scholar]
- 8.National Comprehensive Cancer Network (NCCN). Clinical Practice Guidelines in Oncology: Thyroid carcinoma [Internet]. 2021 [cited 2025 May 30]. Available from: https://www.nccn.org/professionals/physician_gls/pdf/thyroid.pdf
- 9.Rao SN, Smallridge RC. Anaplastic thyroid cancer: an update. Best Pract Res Clin Endocrinol Metab. 2023;37:101678. 10.1016/j.beem.2022.101678. [DOI] [PubMed] [Google Scholar]
- 10.Bonner AB, Chen H, Spencer SA, Willey CD, Kole AJ. Abstract PO-056: the effect of timing of post-operative radiation on outcomes in anaplastic thyroid cancer. Clin Cancer Res. 2023;29(18Suppl). 10.1158/1557-3265.aacrahns23-po-056. [DOI] [PMC free article] [PubMed]
- 11.World Health Organization. International classification of diseases for oncology (ICD-O). 3rd ed., 1st revision. Geneva: World Health Organization; 2013.
- 12.Greene FL, Page DL, Fleming ID, Fritz AG, Balch CM, Haller DG, Morrow M. AJCC cancer staging manual. 6th ed. New York: Springer; 2002. [Google Scholar]
- 13.Edge SB, Byrd DR, Compton CC, Fritz AG, Greene FL, Trotti A. AJCC cancer staging manual. 7th ed. New York: Springer; 2010. [Google Scholar]
- 14.American College of Surgeons. About cancer program categories. Internet. 2023;cited 2025 May 30. Available from: https://www.facs.org/quality-programs/cancer-programs/commission-on-cancer/coc-accreditation/categories/.
- 15.Melfa G, et al. Surgeon volume and hospital volume in endocrine neck surgery: how many procedures are needed for reaching a safety level and acceptable costs? A systematic narrative review. G Chir. 2018;39:5–11. 10.11138/gchir/2018.39.1.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Al-Qurayshi Z, Randolph GW, Srivastav S, Kandil E. Outcomes in endocrine cancer surgery are affected by racial, economic, and healthcare system demographics. Laryngoscope. 2016;126:775–81. 10.1002/lary.25606. [DOI] [PubMed] [Google Scholar]
- 17.De Leo S, Trevisan M, Fugazzola L. Recent advances in the management of anaplastic thyroid cancer. Thyroid Res. 2020;13:17. 10.1186/s13044-020-00091-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Cabanillas ME, Zafereo M, Gunn GB, Ferrarotto R. Anaplastic thyroid carcinoma: treatment in the age of molecular targeted therapy. J Oncol Pract. 2016;12:511–8. 10.1200/jop.2016.012013. [DOI] [PubMed] [Google Scholar]
- 19.Cabanillas ME, et al. Facilitating anaplastic thyroid cancer specialized treatment: A model for improving access to multidisciplinary care. Head Neck. 2017;39:1291–5. 10.1002/hed.24784. [DOI] [PubMed] [Google Scholar]
- 20.Jannin A, et al. Anaplastic thyroid carcinoma: an update. Cancers (Basel). 2022;14. 10.3390/cancers14041061. [DOI] [PMC free article] [PubMed]
- 21.Tiedje V, et al. Anaplastic thyroid carcinoma: review of treatment protocols. Endocr Relat Cancer. 2018;25:R153–61. 10.1530/ERC-17-0435. [DOI] [PubMed] [Google Scholar]
- 22.Goffredo P, Thomas SM, Adam MA, Sosa JA, Roman SA. Impact of timeliness of resection and thyroidectomy margin status on survival for patients with anaplastic thyroid cancer: an analysis of 335 cases. Ann Surg Oncol. 2015;22:4166–74. 10.1245/s10434-015-4742-6. [DOI] [PubMed] [Google Scholar]
- 23.Maniakas A, et al. Evaluation of overall survival in patients with anaplastic thyroid carcinoma, 2000–2019. JAMA Oncol. 2020;6:1397–404. 10.1001/jamaoncol.2020.3362. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Zhou W, Yue Y, Zhang X. Radiotherapy plus chemotherapy leads to prolonged survival in patients with anaplastic thyroid cancer: a retrospective population study. Front Endocrinol (Lausanne). 2021;12. 10.3389/fendo.2021.748023. [DOI] [PMC free article] [PubMed]
- 25.Sherman EJ, et al. 1914MO randomized phase II study of radiation therapy and Paclitaxel with pazopanib or placebo: NRG-RTOG 0912. Ann Oncol. 2020;31:S1085. 10.1016/j.annonc.2020.08.1402. [Google Scholar]
- 26.Alobuia W, Gillis A, Kebebew E. Contemporary management of anaplastic thyroid cancer. Curr Treat Options Oncol. 2020;21:78. 10.1007/s11864-020-00776-2. [DOI] [PubMed] [Google Scholar]
- 27.Bible KC, et al. 2021 American thyroid association guidelines for management of patients with anaplastic thyroid cancer. Thyroid. 2021;31:337–86. 10.1089/thy.2020.0944. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Ryder M, et al. Genetic and Pharmacological targeting of CSF-1/CSF-1R inhibits tumor-associated macrophages and impairs BRAF-induced thyroid cancer progression. PLoS ONE. 2013;8:e54302. 10.1371/journal.pone.0054302. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Keam B, et al. Updated efficacy and safety data of Dabrafenib and Trametinib in patients with BRAF V600E–mutated anaplastic thyroid cancer. Ann Oncol. 2018;29:viii645–6. 10.1093/annonc/mdy302.002. [Google Scholar]
- 30.Evans LK, et al. Improved survival of advanced-stage anaplastic thyroid cancer with systemic therapy. Laryngoscope. 2024. 10.1002/lary.31712. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Stavas MJ, et al. Short course high dose radiotherapy in the treatment of anaplastic thyroid carcinoma. J Thyroid Res. 2014;2014:764281. 10.1155/2014/764281. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.De Crevoisier R, et al. Combined treatment of anaplastic thyroid carcinoma with surgery, chemotherapy, and hyperfractionated accelerated external radiotherapy. Int J Radiat Oncol Biol Phys. 2004;60:1137–43. 10.1016/j.ijrobp.2004.05.032. [DOI] [PubMed] [Google Scholar]
- 33.Sun XS, et al. Indications of external beam radiation therapy in non-anaplastic thyroid cancer and impact of innovative radiation techniques. Crit Rev Oncol Hematol. 2013;86:52–68. 10.1016/j.critrevonc.2012.09.007. [DOI] [PubMed] [Google Scholar]
- 34.Ullah A, et al. Anaplastic thyroid carcinoma: interplay of predictive factors, treatment challenges, and survival insights. Expert Rev Endocrinol Metab. 2025;20(2):129–38. 10.1080/17446651.2025.2467660. [DOI] [PubMed] [Google Scholar]
- 35.Zhang K, et al. Stratified impact of therapies on anaplastic thyroid cancer outcomes in the pre-gene-targeted therapy era. Ann Surg Oncol. 2025;32(4):2732–42. 10.1245/s10434-024-16852-y. [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.
Data Availability Statement
Data for this study obtained by applying for a participant user file from the American College of Surgeons website. This file contained de-identified patient information for patients diagnosed with thyroid cancer between 2004 and 2018 at participating institutions in the United States. The website to view the application for these datafiles is included here: [https://www.facs.org/quality-programs/cancer-programs/national-cancer-database/puf/](https://www.facs.org/quality-programs/cancer-programs/national-cancer-database/puf/)The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request. Email address of corresponding author: [abjbonner@gmail.com](mailto: abjbonner@gmail.com).