The incidence of thyroid cancer (TC) has continued to rise in the United States over the last four decades (1), and TC was reported to be the fifth most common malignant tumor in women in 2018 (2). Based on the recent data, the American Cancer Society has predicted that the number of new TC cases and deaths due to TC will become the sixth most common cause of malignant tumors for women in 2019 (3). Among several variants of TC, papillary thyroid cancer (PTC) takes up 85-90% of all cases (4).

PTC and follicular thyroid cancer (FTC) are classified as differentiated TCs that both arise from thyroid follicular cells and generally have excellent prognoses, while the poorly differentiated thyroid cancers have a poorer prognosis which probably arises from either PTCs or FTCs. Medullary thyroid cancer (MTC) has been derived from the parafollicular C cells, accounting for about 1-2% of TCs (5). Anaplastic thyroid cancer (ATC) originates from follicular cells and is an aggressive, undifferentiated, and rapidly fatal tumor (6).

The American Joint Committee on Cancer (AJCC) Staging Manual 6th Edition (7) sets out the lymph node (N) stages of cancer as follows: no metastatic nodes (N0); nodes not assessed at surgery (NX); metastases to level VI [pretracheal, paratracheal, and prelaryngeal/Delphian lymph nodes] (N1a); metastasis to unilateral, bilateral, or contralateral cervical or superior mediastinal mode metastases (N1b). In the 8th edition, lymph node stage of N0 were divided into 2 groups with more detailed information, which referred to N0a (One or more cytologic or histologically confirmed benign lymph node) and N0b (no radiologic or clinical evidence of local regional lymph node metastasis). However, the N1a stage in the 8th edition, which can be unilateral or bilateral disease, covers not only the lymph node VI but also the lymph node VII (the upper mediastinal lymph nodes). Additionally, in the 8th edition, the N1b stage referred to metastasis to levels I, II, III, IV, or V (unilateral, bilateral, or contralateral lateral neck lymph nodes) or retropharyngeal lymph nodes, which include all disease in the lateral neck (8). Compared with 6th edition, the other N stages of 8th edition remain unchanged. In the United States, it has been reported that 5% of TC patients have distant metastasis (9), while the most commonly observed metastatic sites have been reported as lung and bone, followed by brain and liver (10).

Although previous studies have shown that TC patients have favorable survival outcomes, with a mortality rate of ~10% and a propensity to have lymph node metastases (11, 12), certain lymph node involvement was reported to have an adverse impact on the prognosis for TC (13). Moreover, the prognosis for NX patients and the association between N stage and distant metastasis remain unknown. In addition, most of these studies were based on data derived from a small sample size population and were limited due to the types of statistical analyses undertaken.

The objective of this study is to investigate the prognosis of TC patients at the NX stage and to determine the association between N stages and distant metastasis in TC based on the surveillance, epidemiology, and end results (SEER) database.

In this study, we have investigated the association between the N stage and disease-specific mortality in TC patients with initial distant metastasis. Our data indicated that TC patients with NX had the highest mortality risk, while the N0 group had the lowest mortality risk. In addition to this, our results also suggest that the N1 group had a higher probability of lung metastasis.

The TNM stage system was designed empirically to predict the survival of most TC patients and to guide therapeutic methods and clinical follow-up setup. An interesting retrospective study by Adam et al. found that the cervical lymph node metastases were not in correlation with a favorable but compromised prognosis in PTC patients younger than 45 years (14). Compared with the 7th AJCC Staging Manual, the 8th AJCC Staging Manual has a remarkable change in the definition of N1a and N1b, which may result in better accuracy of central and lateral neck disease (8). However, there is no special attention paid to the NX stage. Previous studies have investigated the risk factors and clinical outcomes of the NX stage in TC and showed its association with poor prognosis (15–17). A study by Baek et al. retrospectively analyzed data concerning 189 PTC patients who had undergone surgery and confirmed that lymph node metastasis was a factor for increasing the risk of recurrence (18, 19). Although it has been reported that lymph node involvement has prognostic significance for TC patients (14, 17, 20), the significance of different N stages still remains controversial. In a study conducted in 1961, TC patients whose disease extended beyond the capsule of the thyroid gland to the regional lymph nodes, surrounding organs or tissues, etc., had a much higher survival rate than that of other groups at 20 years after surgery (21). On the contrary, recent study results suggest that the lymph node involvement have a bad effect on the prognosis of TC (22, 23). Whereas in such a situation, there is still not much attention paid to the investigation on the survival outcomes of patients with different lymph node stages, especially for the NX stage. Liu et al. reported that differentiated TC patients with the NX stage had an unexpectedly poor prognosis via evaluating the survival outcomes (16). They have found that either cancer-specific mortality or all-cause mortality for NX stage patients was higher than N1 and N0 stage patients. Meanwhile, they ascertained that the prognosis for NX stage patients was worse compared to N0 patients. After propensity score matching for relevant confounding factors, patients with N1 stage sustained similar results, which may result from insufficient use of thyroidectomy. In our study, we found that the NX patients had the highest mortality followed by N1b, N1NOS, and N1a groups in PTC, which was consistent with the results of Liu et al. However, the NX did not show any prognostic value when analyzing FTC population alone. To our knowledge, previous studies have noted the risk factors of prognosis for TC patients, such as BRAF mutation (24) and multifocality (25), and categorized the disease stages using the N stage of N0-N1 to assess the clinical outcomes of TC (8). However, the studies focused on the prognosis of NX stage TC patients with distant organ metastasis at diagnosis are scarce. Our study filled the research gap by investigating the prognosis of NX in TC patients, which may provide new clinical-relevant information for improving the treatment strategies. The incidence of TC is increasing rapidly with its diagnosis becoming more and more convenient with the development of images, computed tomography, magnetic resonance imaging, and other medical auxiliary inspection technologies. Even a small percentage change in the mortality rate is likely to result in a big difference in disease-specific mortality. More attention should be given to the NX patients in terms of clinical diagnosis and treatment because of its poor prognosis compared with other N stages. Similar results could be obtained in PTC patients, which may be the consequence of the fact that PTC accounts for ~85-90% of all TC cases (4). Conversely, in the FTC group, our study did not show results similar to the whole TC group and PTC group, which indicates that the N1a was a risk factor for FTC-specific death instead of the NX and N1b. A study by Witte et al. reported that the lymph node metastasis is an important factor affecting FTC patients' survival, while no detail information such as the lymph node location was provided despite the N stage (26). Thus, it still needs further investigation on the correlation between the N stage and the prognosis of patients with FTC.

A high potential survival risk has been reported for patients with TC and distant organ metastasis (27). The four most common distant metastasis organs identified in the 2010 AJCC Staging Manual were bone, brain, liver, and lung. The previous study has shown that lungs and bones were more likely to be metastatic sites than brains and livers for TC patients (28). Additionally, organ metastases were reported to have a correlation with worse disease-specific survival in TC (29). We have investigated the association between the N stage and the two organ metastases, lung, and bone. The metastatic rate for lung is 51.2% (720/1,407), which is consistent with the results of a study by Hirsch et al., the study reported that the lung was the most common distant metastatic organ with an incidence rate of 85.5% (118/138) (28). According to the previous reports the most common cause of death for TC patients with lung metastasis to be respiratory failure due to the progression of lung metastasis (30, 31). Therefore, chest computed tomography should be used to diagnose TC patients, especially PTC patients with N1 stage. Bone is regarded as a common metastatic site. Though Slook et al. reported that bone metastasis is linked to high mortality risk as a risk factor having an impact on the survival of differentiated TC patients (32), there was no previous study focused on the association between N stage and bone metastasis. Our study indicated that N1 patients were more likely to have lung metastasis than bone metastasis. TC patients with regional lymph node metastasis should carefully undergo ultrasonography or biopsy of lymph nodes must be performed to potentially identify clinical lung metastasis rather than bone metastasis, even though bone metastasis has been reported to lead to high morbidity rates for skeletal-related events such as pathologic fractures, spinal cord compression, the requirement for bone irradiation, or bone surgery (7). In terms of treatments, even though radioactive iodine therapy has great significance in TC patients with lung metastasis, more than half of the cases showed progressive metastasis (33). For TC patients with radioactive iodine ablation-refractory metastasis, Moneke et al. (34) reported the effectiveness of surgical resection. More aggressive treatment could be arranged for individuals to balance the clinical benefits and harms and reach the optimal results.

This study had some limitations. Though the SEER database is a well-validated dataset, this was a retrospective study. Moreover, the SEER database does not provide information which may be required for further prospective studies, such as gene mutations and types of therapy. Besides, our study only obtained the data between 2004 and 2015 from the SEER database to investigate the association between N stage and distant organ metastases of lungs and bones. Future studies on the other two common distant metastases sites, brains and livers, and multi-organ metastases are being conducted.

TC Patients with NX stage had the poorest survival, and the highest disease-specific mortality risk, followed by patients with N1b, N1a, and N0 lymph node stages. The lung was found to be the most common metastatic site. Among different lymph node stages, N1 patients were more prone to lung metastasis. Therefore, we recommend chest computed tomography screening during follow-ups for TC patients, particularly PTC patients with N1 stage. Our study provided new information regarding the treatment of patients with NX stage differentiated TC.

Data on TC patients were retrieved from the SEER database (https://seer.cancer.gov/), which is maintained by the National Cancer Institute. The datasets generated for this study are available on request to the corresponding author.

JZ, XC, LS, and SQ: conception and design. All authors contributed to acquisition, statistical analysis or interpretation of the data, drafting of the manuscript. All authors reviewed and approved the final version of the manuscript.