Abstract
Background/Aim: Many studies indicate that multifocality is associated with high-risk features of differentiated thyroid cancer (DTC). Herein, we evaluated the impact of the unilateral multifocality on post-lobectomy recurrence in patients with DTC. Patients and Methods: We retrospectively analyzed 1,684 patients with DTC who underwent thyroid lobectomy from 2008 to 2015 using logistic regression models to calculate the relative risk on post-lobectomy recurrence. Results: Tumor diameter increased from 4.9 mm to 8.1 mm and the proportion of extrathyroidal extension (ETE) and unilateral multifocality progressively increased from 2008 to 2015 (2.1% to 24.3% and 4.2% to 22.8%, respectively). During the 88.6-month follow-up period, 67 (3.98%) recurrences and 2 (0.12%) deaths were observed. There were 269 (16.0%) multifocal DTC cases. There was no significant difference between the multifocal and unifocal groups in terms of the proportion of recurrences (5.2% vs. 3.7%) and distant metastasis (0.4% vs. 0.1%). Logistic regression analysis revealed age <42.5 years (OR=1.83), tumor diameter greater than 7.5 mm (OR=1.89), and N1a (OR=2.04) were potent risk factors for post-lobectomy recurrence. Conversely, male sex (OR=0.77; p=0.407), ETE (OR=1.16; p=0.698) and multifocality (OR=1.23; p=0.526) were not risk factors for recurrence after thyroid lobectomy. A positive node ratio (PNR) ≥42.0%, N1a stage, a tumor diameter ≥7.5mm and age <42.5 years were significant risk factors for recurrence (Log-rank p=0.001, p=0.001, p=0.004 and p=0.009, respectively). Contrariwise, multifocality and ETE were proven to not be risk factors for DTC recurrence after thyroid lobectomy (Log-rank p=0.099 and p=0.126, respectively). Conclusion: Unilateral multifocality was not a risk factor for DTC recurrence after thyroid lobectomy and could not be considered an indication for immediate completion or total thyroidectomy.
There have been major changes in the management of low-risk differentiated thyroid cancer (DTC) over the period of 2000 to 2018, consistent with the American Thyroid Association (ATA) guidelines, which suggest less extensive surgical management of low-risk tumors (1, 2). The proportion of thyroid lobectomy rather than total thyroidectomy slightly increased during the 2014-2018 period (1). However, the magnitude of this increment after 2015 was unexpectedly more subtle (1). Many researchers have sought to identify the risk factors of unfavorable outcomes. Multifocality is defined as the presence of two or more cancer foci within the thyroid gland (3). Furthermore, collision cancers, which are mixed with other types of carcinomas, are also often found in multifocal thyroid cancer (4, 5).
Many studies indicate that multifocality is associated with high-risk features of papillary thyroid cancer (PTC) (6, 7). Multifocality has been considered as a predictive factor for PTC progression (3). Moreover, it has been shown to be an independent risk factor for recurrence (3, 8) and decreased overall and cancer-specific survival (9, 10). Nevertheless, there are opposite opinions on the role of multifocality as an independent risk factor for recurrence. Feng et al. suggested that multifocality has less clinical importance in patients with papillary thyroid microcarcinoma (PTMC), unlike those with macro-PTC with diameters greater than 1.0 cm (11). They proposed that multifocal PTMC should be treated using strategies different from those of multifocal macro-PTC (11). Harries et al. (12) and Woo et al. (13) further suggested that multifocal PTC should not be considered an indication for completion thyroidectomy. Geron et al. revealed that multifocality is not an independent prognostic factor, as proven by propensity score-matching analysis (14).
We presumed that the risk of recurrence of unilateral multifocal PTMC is probably different from that of larger PTC and sought to investigate the impact of multifocal DTC and collision cancers on recurrence, especially in patients who have undergone lobectomy, through a retrospective study.
Patients and Methods
Patient selection. From 2008 to 2015, 9,174 thyroid surgeries were performed by five surgeons in Gwangju and Hwasun Chonnam National University Hospitals. Of these patients, 2,845 (31.0%) underwent thyroid lobectomy. We excluded 1,161 patients who had previous thyroid surgery, benign histology, poorly differentiated thyroid cancer, diffuse large B-cell lymphoma, thyroid gland metastasis originating from lung cancer and some patients with insufficient clinical data. Finally, we enrolled 1,684 patients with DTC.
Variables. Patient factors including age, sex, and whether to take levothyroxine (LT4) or not; pathological factors including histologic types of DTC, tumor-node-metastasis staging, and the positive node ratio (PNR); and laboratory factors including thyroglobulin (TG) and thyroid stimulating hormone (TSH) levels were analyzed. The cut-off values of the continuous variables were identified via the area under the receiver operating characteristic curve as 42.5 years for age; 7.5 mm for tumor diameter; and 42.0% for PNR on recurrence.
Surgical approaches. The three surgical approaches for thyroid lobectomy on patients with low-risk DTC performed at our institutions were the classic, minimal invasive, and endoscopic or robotic approaches. We have tried to implement prophylactic ipsilateral central neck dissection (CND) if possible and intraoperative frozen biopsy. The prophylactic ipsilateral CND and frozen biopsy were performed if the disease was upstaged into intraoperative cN1a even in preoperatively cN0 DTC patients. We also performed frozen section examination, as it seems to be a safe and effective strategy to decrease the need for immediate completion thyroidectomy or recurrence. We chose to immediately convert to total thyroidectomy, if the frozen section was positive for macro-metastasis. Conversely, wider compartmental CND or immediately converted to total thyroidectomy was favored if the frozen section had proven positive nodes with micro-metastasis.
Ethical approval. This study was approved by the International Review Board of Chonnam National University Hospital (CNUH-2022-387), which waived the need for informed consent.
Follow-up. We included 1,684 patients with DTC and 269 (16.0%) unilateral multifocal DTC. Out of the 269 cases with multifocal DTC, 265 (98.5%) were PTCs and 4 (1.5%) were collision tumors of PTC and thyroid follicular cancer (TFC). Patients were followed up every three to six months during the first postoperative years and every six months or annually after that. Thyroid hormone, TG, and anti-TG antibody levels were checked at every visit. Ultrasonography was performed at six-month or one-year intervals. LT4 tapering or cessation trial was attempted during every visit if the risk of recurrence had decreased after dynamic risk stratification. Recurrence was confirmed by aspiration cytology, washout TG, and postoperative histology. Even if the recurrence was confirmed by immediate completion within 6 months, the cases of residual cancer were presumed by inappropriate thyroid lobectomy and classified as recurrence regardless of the duration of reoperation.
Statistical analysis. Continuous variables are presented as mean±standard deviation (SD) or mean (range) and categorical as frequencies and percentages. We used logistic regression models to investigate the association between possible predictive factors and post-lobectomy recurrence. The recurrence rate was calculated using the Kaplan-Meier analysis and differences were compared using the log-rank test. Statistical significance was set at a p-value <0.05. Analyses were performed using IBM Statistical Product and Service Solutions Statistics version 27.0 (IBM Corp., Armonk, NY, USA).
Results
Baseline characteristics of patients who had undergone thyroid lobectomy. Table I shows the characteristics of the 1,684 patients who underwent thyroid lobectomy between 2008 and 2015. The surgical approaches used were the classic method (699; 41.5%), minimal incision with less than 3 cm incision (661; 39.3%), and endoscopic or robotic approaches (324; 19.2%). The distribution of T stage was as follows: 1,443 (85.1%) T1a, 66 (3.9%) T1b, 19 (1.1%) T2, and 166 (9.9%) T3 stage with ETE. Central lymph node dissection (CND) was performed in 1,120 (66.5%) patients and the N stage was as follows: 859 (51.0%) N0, 564 (33.5%) Nx, and 261 (15.5%) N1a. The mean number of harvested central nodes was 2.2 (median=1.0, range=0-19) and positive nodes were 0.32 (median=0, range=0-12).
Baseline characteristics of 1,684 patients who underwent thyroid lobectomy.
The largest and most aggressive subtype of thyroid cancer was considered the main tumor. The composition of the main tumor was as follows: 1,579 (93.8%) classic, 83 (4.9%) less aggressive [follicular variant (FV), cribriform morular variant (CMV), oncocytic variant (OV)], and 10 (0.6%) aggressive type [diffuse sclerosing variant (DSV), Hobnail, tall cell variant (TCV)] PTC. Follicular cancer was composed of 10 (0.6%) minimal invasive TFC (TFCmi), and 2 (0.1%) wide invasive TFC (TFCwi). There were 269 multifocal DTCs composed of 265 multifocal PTC and 4 collision cancers. During the mean 88.6-month follow-up, 67 (3.98%) recurrences were occurred: 15 out of 33 immediate completion thyroidectomies and 52 out of 53 delayed reoperations with completion thyroidectomy. The total 90 recurrence sites in the 67 patients involved 44 (2.6%) contralateral thyroids, 35 (2.1%) central compartments, and 11 (0.7%) lateral compartments.
Histology of 269 unilateral multifocal cancers. The most common combination was classic-classic PTC (88.5%) in the following order: FV-classic 17 (6.3%), FV-FV 6 (2.2%), TFCmi-classic 3 (1.1%), DSV-classic 2 (0.7%), TCV-FV 1 (0.4%), and TFCmi-FV 1 (0.4%) (Table II). There were 4 (1.5%) collision cancers composed of different histological subtypes with TFCmi and PTC.
Histology of unilateral multifocal PTC (n=265) and collision DTC (n=41).
Characteristics of the multifocal DTC. Table III shows the characteristics of the multifocal (n=269, 16.0%) and unifocal (n=1,415; 84.0%) DTC groups. The proportions of patients aged ≥42.5 years (61.7% vs. 49.1%, p=0.001) were significantly different between the groups; however, the proportions of females (77.3% vs. 76.9%, p=0.877) were not different. The main tumor diameter in the multifocal group was larger (0.6 cm vs. 0.5 cm, p=0.001), and the proportion of DTC tumor diameter greater than 7.5 mm was significantly higher than that in the unifocal group (28.6% vs. 19.9%, p=0.001). The proportion of N1a patients was significantly higher in the multifocal group (27.1% vs. 13.3%, p=0.001). However, the proportion of PNRs ≥42.0% did not differ between the two groups (20.0% vs. 17.9%; p=0.500). LT4 tapering or cessation trial was attempted in every visit, if the risk of recurrence was decreased after dynamic risk stratification. Over the study period, 661 (39.4%) patients were able to discontinue LT4, and 1,017 (60.6%) patients continued LT4 supplementation. The proportion of LT4 supplementation was not significantly different between the groups (56.9% vs. 61.3%, p=0.180).
Characteristics of the patients with unilateral multifocal differentiated thyroid cancer.
The distribution of total recurrences (5.2% vs. 3.7%, p=0.262), thyroid (2.6% vs. 2.6%, p=0.513), central (3.3% vs. 1.8%, p=0.112), lateral recurrence (1.1% vs. 0.6%, p=0.305), and distant metastasis (0.4% vs. 0.1%, p=0.189) were not significantly different between the two groups. The median values of the last postoperative TG (6.7 vs. 6.8 ng/ml, p=0.964) and TSH (2.1 vs. 2.0 mIU/l, p=0.357), and preoperative TG (11.8 vs. 11.5 ng/ml, p=0.893), and TSH (1.65 vs. 1.75 mIU/l, p=0.151) also did not differ between the groups.
There were two deaths in patients with distant metastasis. A 70-year-old female 3.3 cm TFCwi patient with multiple bone and lung metastasis died 116 months later, despite immediate completion thyroidectomy with radioactive iodine treatment. Another 80-year-old female patient with 1.5 cm TFCwi with lung, bone, and adrenal metastasis died 53 months later; her case was misdiagnosed as FVPTC during the initial thyroid lobectomy.
Risk factors for recurrence after thyroid lobectomy. Logistic regression analysis was performed to identify the predictive factors for recurrence after thyroid lobectomy (Table IV). On univariable analysis, age <42.5 years [odds ratio (OR)=1.92], tumor diameter ≥7.5 mm (OR=2.30), and N1a (OR=2.62) were significant risk factors. Conversely, male sex (OR=0.80; p=0.471), ETE (OR=1.64; p=0.160), and unilateral multifocality (OR=1.41; p=0.264) were not proved as risk factors for recurrence. On multivariable analysis, age <42.5 years (OR=1.83; p=0.023), tumor diameter ≥7.5 mm (OR=1.94; p=0.055), N1a (OR=2.04; p=0.014) were independent risk factors for recurrence after thyroid lobectomy.
Logistic regression analysis on risk factors of recurrence after thyroid lobectomy.
Overall survival according to risk factors. We further analyzed the above factors using the Kaplan-Meier method and arranged them in the order of high OR values (Figure 1). A PNR ≥42.0% and central lymph node metastasis were potent risk factors for recurrence after thyroid lobectomy (Log-rank p=0.001). Tumor diameter ≥7.5 mm (Log-rank p=0.004) and age <42.5 years (Log-rank p=0.009) were equally significant risk factors. Conversely, unilateral multifocality (Log-rank p=0.099) and ETE (Log-rank p=0.126) were proven to not be risk factors for recurrence after thyroid lobectomy.
Kaplan-Meier analysis of the recurrence rate after thyroid lobectomy. (A) More than 42.0% of positive node ratio (Log-rank p=0.001), (B) N1a stage rather than N0 stage (Log-rank p=0.001), (C) DTC larger than 7.5mm (Log-rank p=0.004), and (D) young age less than 42.5 years (Log-rank p=0.009) were good predictor on recurrence after thyroid lobectomy. Contrary, (E) unilateral multifocality (Log-rank p=0.099) and (F) extrathyroidal extension (Log-rank p=0.126) were poor classifiers on post-lobectomy recurrence.
The trends of tumor diameter, ETE, and unilateral multifocality in DTC. Between 2008 and 2015, thyroid lobectomy was expanded from 15% to 45% and the main tumor diameter increased from 4.9 mm to 8.1 mm. Thyroid lobectomies increased in selected patients with T3 or multifocal DTC. The annual proportion of ETE decreased from 10.2% to 2.1% in 2012 and then progressively increased to 24.3% in 2015. The proportion of unilateral multifocality also increased from 5.1% to 10.7% until 2012, and then it exhibited a significant increment from 2014 to 2015. Contrariwise, there was little change in the satellite tumor diameter, which ranged between 3.0 and 3.6mm (Figure 2).
The annual trends of tumor diameter, extrathyroidal extension (ETE) and unilateral multifocality. The annual trends of main and satellite tumor diameter of 265 multifocal PTC and four collision cancers among the 1,684 patients are depicted. Thyroid lobectomy was expanded from 15% to 45% in 2015. The main cancer size increased from 4.9 mm to 8.1 mm. The annual proportion of ETE (10.2% to 24.3%) and unilateral multifocality (5.1% to 22.8%) progressively increased until 2013, and then exhibited a significant increment during the 2014-2015 period. On the other hand, there was little change in the satellite tumor diameter, which ranged between 3.0 and 3.6 mm.
Discussion
Many studies have indicated that bilateral multifocality is associated with high-risk features (6, 7), disease progression (3), independent risk factor for recurrence (3, 8), and decreased overall survival of DTC (9, 10). In contrary, Geron et al. have pointed out that unilateral or bilateral multifocality is not an independent prognostic factor for papillary thyroid cancer, as proven by propensity score-matching analysis (14). In the present retrospective cohort study, we evaluated the risk factors of post-lobectomy recurrence with logistic regression. Young age (<42.5 years old), tumor size larger than 7.5 mm, and N1a stage were identified as significant risk factors for post-lobectomy recurrence.
In our study, logistic regression analysis revealed that unilateral multifocality was not a risk factor for post-lobectomy recurrence, and moreover, the proportion of recurrences and distant metastases did not significantly differ between the multifocal and unifocal groups. Feng et al. suggested that the number of cancers has less clinical importance regarding recurrence in patients with multifocal PTMC, unlike those with macro-PTC larger than 1.0 cm (11). In our retrospective cohort study, we also proved that multifocality was not associated with post-lobectomy recurrence and could not be considered an indication for immediate completion or total thyroidectomy. Harries et al. suggested that multifocal PTC should not be an indication for completion thyroidectomy and showed that selected multifocal patients managed with lobectomy alone have comparable recurrence and survival rates with patients with unifocal PTC (12). Our study also suggested that multifocal thyroid micro-carcinomas (<1.0 cm) should be treated using different strategies from those of multifocal macro-carcinomas.
Thyroid cancer is one of the most common malignancies worldwide, and its incidence has steeply risen over the last decades (1, 15). James et al. suggested that the incidence of thyroid lobectomy has not increased despite evidence of equivalency with total thyroidectomy and recommendations from the ATA encouraging thyroid lobectomy for patients with small-size PTC (16). This can be explained by the fact that clinical behavior commonly lags behind new professional recommendations. Ongoing population-based studies are also needed to monitor future changes in practice (16). Our study also clearly showed the above transitional lags. A steep rise and about three-fold increment over a period of 7 years of thyroid lobectomy was observed (15% in 2008 vs. 43% in 2015). Then, the rate of increment slowed down; however, it steadily increased to 57.0% in 2020. During this transition period, the main tumor diameter increased from 4.9 mm to 8.1 mm. Furthermore, the proportion of ETE and unilateral multifocality progressively increased (2.1% to 24.3% and 4.2% to 22.8%, respectively).
Furthermore, we were able to present additional findings on the trend of main and satellite tumor diameter by year. There was a slight change in the satellite tumor diameter (3.0-3.6 mm) during the study period, even though the main cancer size had increased from 4.9 mm to 8.1 mm. Most satellite tumors were small tumors which were incidentally found, and further investigation is needed regarding the natural evolution and specific clinical significance of the size of satellite tumors.
Limitations. This is a retrospective study with inherent limitations. First, the sample population may not be representative of the larger population, leading to limitations in the generalizability of the results. Second, this study may have more confounding variables than other studies, especially on nodal metastasis. Harris et al. excluded all T3 or N1a stage and included only T1T2N0M0 stage patients (12). Conversely, we performed thyroid lobectomy in selected T3 or N1a stage patients who were upstaged mostly into N1mic even in the preoperative cN0 stage. Thus, T3 or N1a patients accounted for 9.9% and 15.5%, respectively. We presumed that our selection of patients is more realistic because it reflects many recent changes in guidelines and treatment methods.
Conclusion
There was no significant difference between the multifocal and unifocal groups in post-lobectomy recurrences. Unilateral multifocality was not a risk factor for post-lobectomy recurrence in DTC and could not be considered an indication for immediate completion in patients with selected DTC patients who do not have cancer foci in the opposite lobe with precise preoperative screening. Further studies with a long-term follow-up and large patient samples are needed to assess the long-term outcomes of multifocal DTC.
Footnotes
Authors’ Contributions
JS Cho conceptualized and designed the hypothesis, collected, and analyzed the data, and wrote the manuscript. HK Kim revised the hypothesis, analyzed data, validated statistical errors, supervised, and reviewed the manuscript.
Conflicts of Interest
The Authors have no financial or personal relationships that could potentially influence this work. We report no potential conflicts of interest relevant to this article.
- Received January 13, 2023.
- Revision received April 24, 2023.
- Accepted May 2, 2023.
- Copyright © 2023 The Author(s). Published by the International Institute of Anticancer Research.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.