The role of surgery in small differentiated thyroid cancer

The incidence of small, differentiated thyroid cancer (DTC) cases has been increasing in the United States and the world mainly due to incidental detection because of widespread use of diagnostic modalities. While the option of active surveillance instead of surgical resection is getting more popular, there is still an open discussion about the best approach in these cases. The National Cancer Database was queried for patients diagnosed with non-metastatic small T1/N0 DTC between 2004 and 2016, who have known surgical status and Charlson comorbidity index of two or less. We evaluated the overall survival (OS) based on the surgery status using Kaplan–Meier estimates and multivariable cox regression analyses. A total of 98,501 patients with non-metastatic small DTC were included, within which 96,612 (98.1%) were treated with surgery, and 1889 (1.9%) were not treated with surgery or other ablative modalities. We found that patients who were treated with surgery had better OS compared to patients who were not treated with surgery (mean OS 171 months vs 134.1 months, P < 0.001, median OS was not reached). This difference was still statistically significant even after we used propensity score matching for age, gender, race, Charlson-Deyo score, tumor size, and histology. On multivariate analysis, surgery was associated with better OS (HR 0.218; 95% CI: 0.196–0.244; P < 0.001). Same trend was found in subgroup analysis when we split the cohort according to tumor size (<1 and ≥1 cm), histology (follicular, papillary and Hurthle cell carcinoma), and age (<55 years vs ≥55 years). Patients with non-metastatic small DTC who were treated with surgery had significant improvement in OS compared to patients who were not treated with surgery. Notwithstanding the limitations of the current analysis, these results call for caution prior to recommending routine surveillance for all patients with small DTC. Thyroid cancer is the most common endocrine malignancy. Thyroid cancer incidence rate is increasing, especially small thyroid tumors. The oncologic safety of active surveillance versus surgical resection in small, differentiated thyroid cancer is still controversial. Surgical resection improves the OS in small DTC. The survival benefit of surgical treatment is seen in all histologic types of DTC. Thyroid cancer is the most common endocrine malignancy. Thyroid cancer incidence rate is increasing, especially small thyroid tumors. The oncologic safety of active surveillance versus surgical resection in small, differentiated thyroid cancer is still controversial. Surgical resection improves the OS in small DTC. The survival benefit of surgical treatment is seen in all histologic types of DTC.


Introduction
Thyroid cancer is the most common endocrine malignancy with an estimated 44,280 new cases and 2200 deaths in the United States (US) alone in 2021 and 586,202 new cases and about 43,000 deaths worldwide in 2020. The majority of thyroid cancer cases diagnosed in the US are differentiated thyroid cancer (DTC) which include papillary, follicular and Hurthle cell carcinoma [1][2][3][4]. The widespread use of diagnostic modalities including ultrasonography and fine-needle aspiration (FNA) has led to a rapid increase of thyroid cancer incidence rate in the US and worldwide in the last few decades. This increase is largely due to the rise in the diagnosis of small thyroid tumors, (mainly small papillary thyroid cancers [5][6][7]). Despite the controversial findings, some studies suggested some environmental factors as partial cause for the thyroid cancer incidence rate increase. These factors include the increased exposure to irradiation from diagnostic X-rays and computed tomography (CT) scans, iodine intake imbalance and exposure to heavy metals [8][9][10][11][12][13].
In recent years, active surveillance has been adopted more frequently as an acceptable option in managing small papillary thyroid cancer (PTC) as more studies showed comparable outcomes between active surveillance and surgical resection approaches [14,15]. Despite the rapid adoption of active surveillance approach, there is still an open debate about the best approach in small thyroid cancer cases. This debate is getting ground for two main reasons. First, majority of the studies that compared between AS and surgical resection focused on assessing the difference in tumor size increase and developing regional or distant metastases, but very few studies compared the survival outcomes between these two approaches. Second, there is a lack of large randomized controlled trials that compared the outcomes between the two approaches [16][17][18][19][20].
In this study, we aimed to evaluate the difference in overall survival (OS) in patients with small DTC who had surgical resection vs the patients who did not have surgical resection.

Data source
The National Cancer Database (NCDB) is a nationwide database supported by the Commission on Cancer (CoC) of the American College of Surgeons and the American Cancer Society. About 70% of all cancer cases diagnosed in the United States from more than 1500 cancer centers are included in this database. Details of patients' demographics, malignancy staging, and histological characteristics in addition to treatment and outcome information are provided in this database [21]. The CoC's NCDB and the hospitals participating in the CoC's NCDB are the sources 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 of the current study.
After we obtained the approval on our proposed protocol and the letter of support from the cancer committee chair at Cleveland Clinic, the NCDB was queried for patients diagnosed with non-metastatic small DTC between 2004 and 2016. Survival follow-up for the studied cohort is available till December 2016.

Patient selection
For our study, we selected the patients who were diagnosed with histologically proven non-metastatic differentiated thyroid cancer at age of 18 or older. We included patients with tumor size of 2 cm or less with clinical stage T1, N0, M0 only. We excluded patients with unknown surgical status to the primary site in addition to patients with Charlson-Deyo score of 3 and above. Also, we excluded patients with lost follow up which is defined by patients with missing details about their last contact status as alive or dead, and those with missing information about their time from diagnosis to the last contact. We

Variables
Using the NCDB, we included the following patient demographic variables: Age, gender, race, median income, education level, insurance status, facility type (which includes community cancer program, comprehensive community cancer program, academic/research program, and integrated network cancer program) in addition to the geographic area which was classified as metropolitan, urban, or rural location. Also, we collected the following disease-related variables: tumor size, histological type, surgery status and Charlson-Deyo comorbidity score.

Outcome
The primary outcome of this study was overall survival (OS) in months defined as the time from diagnosis to the time of death for any reason.

Statistical analysis
We compared the baseline demographics and characteristics between patients who had surgical resection to the primary tumor versus patients who did not have surgery using the Pearson Chi-square test.
For OS comparisons, we split the cohort into two separate groups according to the surgical status, the first group who had surgical resection to the primary tumor and the second group who did not have surgery or other ablative modalities. Then, we compared the OS between the two groups.
We also did subgroup analyses. In the first subgroup analysis, we split the cohort according to the tumor size into <1 and ≥1 cm groups and compared the OS between patients who had surgical resection and patients who did not have surgery in each group. In the second subgroup analysis, we split the cohort according to the histological type into follicular, papillary, Hurthle and combined papillary & follicular groups and compared the OS between patients who had surgical resection and patients who did not have surgery in each group. In the third subgroup analysis, we split the cohort according to age cutoff of <55 years vs ≥55 years. The reason for selecting this age cutoff is the use of this cutoff in the current American Joint Committee on Cancer (AJCC) Cancer Staging Manual for DTC. Moreover, giving that most of the previous studies evaluated the active surveillance in papillary microcarcinoma (PMC), and that the NCCN thyroid cancer guideline specifically recommended active surveillance as an optional approach for low-risk PMC, we did a subgroup analysis that included only patients with papillary thyroid cancer, diagnosed at age less than 75 with tumor size <1 cm and Charlson-Deyo score of 0-1.
OS analyses were evaluated using the Kaplan-Meier survival method. A log-rank test was used to evaluate survival differences between groups. We used Cox regression analysis to conduct multivariable analyses to evaluate the factors associated with improved OS; hazard ratios (HRs) with associated 95% confidence intervals (CI) were accordingly generated, statistical significance was defined as a P value less than 0.05 for all analyses. The following factors were included in the multivariable model: facility type, age, gender, race, median household income, education level, insurance status, area, Charlson score, histology type, tumor size, and surgery status.
All the above statistical analyses were performed using the SPSS Statistics (Version 27.0, SPSS Inc.) To further reduce any potential biases related to the variation in baseline characteristics between the patient group that had surgical resection and the one that did not have surgery, we used nearest neighbor, 1:1, propensity score matching using RStudio Version 1.3.1093 (MA, USA). We then excluded patients with Charlson-Deyo score of 3 (N = 665), patients with unknown surgery status (N = 76) and patients who lost the follow up for OS (N = 1273). We ended up with 98,501 patients who had non-metastatic small DTC with known surgery status; 96,612 (98.1%) had surgery and 1889 (1.9%) did not have surgery. The baseline characteristics comparison between the two groups is summarized in Table 1.

Baseline characteristics
We also checked the percentage of patients who had surgery and patients who did not have surgery between 2004 and 2016 in yearly basis. We found the percentage of patients who did not have surgery increased over the years which could be a reflection on the increase adoption of active surveillance approach (Supplementary Fig. 1).

Survival outcome
The entire cohort We compared OS between patients who were treated with surgery versus patients who did not have surgery and found that patients who were treated with surgery had better OS compared to patients who did not have surgery (mean OS 171 months vs 134.1 months, P < 0.001, median OS was not reached) as illustrated in the Kaplan-Meier survival curves (Fig. 2).
Propensity score matching yielded 3778 patients for analysis: 1889 patients in the surgery group and 1889 patients in the no surgery group. Comparison of baseline characteristics between the groups after propensity score matching is shown in Table 2.
We compared the OS between patients who had surgery versus patients who did not have surgery using Kaplan-Meier analysis on the propensity score-matched groups and found that patients who had surgery had better OS than those who did not have surgery (mean OS 165.8 months vs 134.1 months, P < 0.001, median OS was not reached) as illustrated in the Kaplan-Meier survival curves (Fig. 3).

Subgroup analysis
We did four subgroup analyses: Based on tumor size We split the cohort according to the tumor size into <1 and ≥1 cm groups to compare OS between patients who had surgery and patients who did not have surgery in each tumor size group. We found patients who had surgery had a statistically significant better OS than those who did not have surgery in both groups; <1 and ≥1 cm tumor size (mean OS 169.3 months vs 138.3 months, P < 0.001, and mean OS 171.6 months vs 131 months, P < 0.001, respectively. median OS was not reached) as illustrated in the Kaplan-Meier survival curves ( Supplementary Fig. 2a, b).
In the patient group with tumor size <1 cm, propensity score matching yielded 1062 patients for analysis: 531 in the surgery group and 531 patients in the no surgery group. We compared the OS between patients who had surgery and patients who did not have surgery using Kaplan-Meier analysis on the propensity score-matched groups and found that patients who had surgery had statistically significant better OS compared to patients who did not have surgery (mean OS 155.5 months vs 138.3 months, P < 0.001, median OS was not reached) as illustrated in the Kaplan-Meier survival curves ( Supplementary Fig. 3a).
In the patient group with tumor size ≥1 cm, propensity score matching yielded 2716 patients for analysis: 1358 in   Fig. 6). Also, on multivariate analysis we found that surgery was associated with better OS (HR 0.204, 95% CI 0.153-0.273, P < 0.001).

Multivariable analysis
Multivariate analysis was done to assess factors affecting OS. Factors associated with worse OS were older age (HR  for surgery vs no surgery 0.218, 95% CI 0.196-0.244, P < 0.001) ( Table 3).

Discussion
While active surveillance is being considered an option in managing low-risk small PTC, there is still a debate about the best treatment approach in these cases. And despite this debate, there is a lack of large and controlled trials that compared the outcomes between surgical management and active surveillance, especially in the US. To our knowledge, this is the largest real-world study that compared the survival outcome in small DTC between patients who had surgical resection versus patients who did not have surgical resection. And while most of the studies that looked at the active surveillance in DTC focused on papillary thyroid cancer, in our study we evaluated the survival benefit of surgical resection in all histologic types of DTC. We found that patients who were treated with surgical resection had better OS compared to the ones who did not have surgery. The survival benefit of surgery was seen in all histologic types of DTC and different tumor sizes (<1 and 1-2 cm).
The active surveillance (AS) approach was first introduced by Miyauchi and his team after his observation of increase thyroid cancer incidence rate without an increase in mortality rate around the world, in addition to having different studies showing frequent detection of PMC, defined as well-differentiated tumors 1 cm or less in size with no lymph node or distant metastasis and no extrathyroidal involvement, on autopsy studies of patients who died of other reasons than thyroid cancer [20,23,24]. This proposal was a trigger to start experimental observation with no surgical resection for low risk PMC in two hospitals in Japan and multiple studies reported very low percentage of tumor size increase and lymph node metastases development [17][18][19]25].
Ito and Miyauchi, along with other researchers, conducted several studies evaluating the outcome of AS in PMC in Japan. Ito et al. [19] between 1993 and 2004 followed a total of 1395 patients with PMC; 340 patients underwent observation and 1055 patients had immediate surgical treatment for an average of 74 months. Both groups had the same rate of lymph node metastases. It was noted that 32% of the observation group patients required surgery during the study period for various reasons mainly due to tumor enlargement or tumor location, but none of these patients had disease recurrence after surgery. More recently, Ito et al. [26] conducted a study of 1235 low-risk PMC patients who were followed with AS between 1993 and 2011. It was reported that 8.0% of the patients had tumor enlargement and 3.8% developed lymph node metastases at 10 years (more significant in less than 60 years old patients), but there was no death or distant metastases in any of the study patients. Sugitani et al. [18] followed 230 patients with asymptomatic PMC for a mean time of 5 years. They found only 7% of tumors increased in sizes with no extrathyroidal invasion or distant metastasis. Oda et al. [27] also compared the outcomes between AS and surgical resection in 2153 patients with low-risk PMC (defined as <1 cm tumor size with no nodal or distant metastasis, macroscopic extrathyroidal extension, high-grade malignancy on cytology, or evidence of progression) in Japan. There was no significant difference in Fig. 3 Mean overall survival in the whole cohort of nonmetastatic small DTC patients according to surgery status after propensity score matching-the median overall survival was not reached developing lymph node metastases between the two groups (0.5% in AS group and 0.2% in surgical treatment group) and none of the 2153 patients developed distant metastases or died of thyroid cancer. In addition to that, few studies found AS to be cost-effective compared to early surgical resection in PMC [28,29]. The reported findings of the mentioned studies played a significant role in adopting AS approach in Japan in 2010 by the Japanese Association of Endocrine Surgeons and the Japanese Society of Thyroid Surgeons [30,31].
In the US, very few studies evaluated the AS approach. In 2017, Tuttle et al. [32] in a study of 291 patients with low-risk PTC (defined as intrathyroidal tumors < 1.5 cm) were followed for a median time of 25 months. The findings were very similar to the Japanese studies, 3.8 % of the patients had tumor growth of 3 mm or more and no regional or distant metastases reported during the active surveillance. Wang et al. [33] on the other had reported a retrospective study of 29,512 patients with PTMC using SEER database. They split the cohort into three groups according to the treatment approach they had: 1.4% did not have surgery, 25.2% had partial thyroidectomy and 73.4% had total thyroidectomy. There was no significant difference between the two surgical approaches, but patients who did not have surgery had lower diseasespecific survival. Also 5-year OS was 25% in the observation group compared to 97.6 % for patients who underwent either partial or total thyroidectomy (p < 0.001). In the last few years, the AS approach started getting more popular in the US and the 2015 ATA guideline mentioned active surveillance as an option in selected patients with PMC who have low-risk tumors (no metastases or local extension and no cytological indication of aggressive disease) [34][35][36][37].
Our study has some limitations that are worth mentioning. First, lack of details if patients who did not have surgical resection were followed with active surveillance or not. Second, other than overall survival, NCDB doesn't have details about other oncologic endpoints like disease-free survival and cancer-specific survival. Therefore, one might argue that higher mortality in the no surgery group is derived by non-thyroid cancer deaths rather than thyroid cancer deaths (particularly given the imbalance in baseline characteristics between both groups with older age and more comorbidity in the no surgery group). It is possible that factors like age and comorbidity were part of the reason why some of those patients did not have surgery in the first place. In order to mitigate the impact of this limitation, we conducted multivariable analyses (to adjust for age, comorbidity and other non-thyroid cancer factors), propensity score matching, as well as multiple subgroup analyses. Third, we did not have the details about all high-risk features, like aggressive histological subtypes or history of prior neck irradiation. Fourth, the retrospective nature of the study and data collection within the NCDB carry different types of bias that might affect the accuracy of the analyses and result interpretation. Despite these limitations, this study represents the largest observational study to compare the survival outcomes in small DTC between patients who were managed with surgical resection and patients who did not have surgical resection. Further prospective randomized studies and/or multi-center studies are needed to further compare the outcomes between surgical approach and active surveillance in small DTC.

Conclusion
Patients with non-metastatic small DTC who were treated with surgical resection had statistically significant improvement in OS compared to patients who did not have surgical resection. The benefit of the surgical approach is seen in all histologic types, size and age categories. We cannot confirm though whether this excess mortality associated with an observation strategy for those patients is totally related to thyroid cancer or is caused by other nonthyroid cancer factors. While the limitations of this retrospective analysis are acknowledged, these results raise the importance of conducting further studies to compare the outcomes between active surveillance and surgical resection in patients with small DTC.
Further large-scale studies that look at disease-specific survival are needed to evaluate the benefit of active surveillance compared to surgical resection in small DTC.

Data availability
The data we used in this study is available from the National Cancer Database.
Author contributions FB: data analysis, data interpretation, manuscript writing. OA-R: study concept, manuscript editing, critical revision.