Medullary thyroid carcinoma (MTC) is a life-threatening disease that accounts for approximately 5%–10% of all thyroid malignancies worldwide and 13.4% of deaths from thyroid cancer are attributed to MTC (1, 2). The prominent feature of MTC is relatively slow tumor growth, but early cervical lymph node metastasis (LNM) occurs, with LLNM occurring in approximately 40.0%-66.7% of patients diagnosed with MTC for the first time (3, 4). In particular, lateral neck lymph node metastasis (LLNM) was very common in MTC. The American Thyroid Association (ATA) recommended that the curative therapy for MTC was primary tumor resection and cervical lymph node dissection (5). However, the extent of initial surgery for cervical lymph node dissection is still controversial (6), especially whether lateral lymph node dissection (LND) is necessary in MTC (7). Some surgeons prefer more proactive management, such as performing lateral LND or prophylactic dissection, to improve postoperative biochemical recovery rates. However, more aggressive prophylactic LND may increase the risk of severe neurological damage and hypoparathyroidism without significant survival benefits (8).

Preoperative imaging plays an important guiding role in the diagnosis and staging of MTC. Ultrasound (US) is the main method for evaluating lymphatic status of MTC before surgery. However, the diagnostic sensitivity for cervical lymph node metastasis (LNM) of US is only about 20%-40% (9). Many MTC patients experienced occult LLNM, which were with clinically negative (cN0) lateral cervical lymph nodes. Considering the existence of occult LLNM that is difficult to detect preoperatively, some patients undergoing thyroidectomy may already have some metastatic lymph nodes in the lateral compartment of the neck, which leading poor prognosis and high mortality. Therefore, accurately assessing the status of lateral cervical lymph node in MTC patients is all-important for choosing appropriate surgical strategy and making clinical decisions.

Serum calcitonin (Ctn) was reported to be a predictive factor for the need of prophylactic lateral lymph node dissection in MTC patients. A high preoperative Ctn level was related to the extent of LNM and poor prognosis in MTC (10). However, serum Ctn has a high false-positive rate, which had a non-specific increase in various physiological and pathological conditions (11). The 2023 European Thyroid Association (ETA) guidelines for thyroid nodule management recommends measuring Ctn in specific scenarios, such as in patients scheduled for surgery or thermal-ablation, or with nodules with indeterminate cytology or suspicious sonographic pattern, or with family history of MTC or multiple endocrine neoplasia (MEN) (12). The 2015 ATA guidelines confirmed the role of preoperative serum Ctn in determining whether lateral neck dissection should be performed in patients with cN0 and cN1a (5), but did not provide a clear threshold.

In this study, we aimed to establish a multivariate logistic regression model based on neck ultrasound examination, serum Ctn levels and other prognostic factors to predict the risk of LLNM. This model was further presented as a nomogram and provided individual recommendations for whether lateral neck lymph node dissection should be performed in MTC patients.

In this research, we constructed and validated the MTC nomogram based on clinical and ultrasound characteristics to predict the probability of LLNM in patients with MTC. The MTC nomogram effectively categorized patients based on their risk of LLNM, and achieved excellent performance in both internal and external testing datasets. Therefore, the preoperative probability of LLNM can be estimated individually and noninvasively. Our study has several advantages: (I) To the best of our knowledge, this is the largest scale retrospective consecutive multicenter study to construct and evaluate a nomogram to predict status of LLNM in MTC. This study included MTC patients within 10 years of Tianjin medical cancer institute and hospital, and contained the results of more than one year of follow-up for all the patients (II) Different from published studies based on radiomics, contrast enhanced ultrasound (CEUS), or TNM staging of the postoperative pathology, this novel nomogram only incorporated clinical and gray-scale US factors preoperatively, which increasing the general applicability of the model. It was particularly important for MTC patients in underdeveloped countries and regions. (III) Comparing with other machine learning (ML) models which were not visual in previous studies, the MTC nomogram we proposed had a better interpretability and maneuverability in clinical practice.

Routine measurement of basal serum Ctn is reported as an important part of the diagnosing and staging of MTC. There is a discrepancy between the different guidelines. The American Association of Clinical Endocrinologists/Associazione Medici Endocrinologi/European Thyroid Association (AACE/AME/ETA) guidelines have suggested examining CT only in subjects with a familial history of MTC and patients with cytology suggestive of MTC or undergoing surgery for goiter (13). On the other hand, according to ATA recommendations, serum Ctn measurement should not be performed routinely, mainly for cost-effectiveness reasons. Usually, preoperative diagnosis is based on cytological examination. However, Jassal et al. showed that preoperative serum Ctn measurement can improve the evaluation of MTC with indeterminate cytology and be helpful in planning surgery (14). It has been shown that, in the diagnosis of MTCs, the measurement of serum Ctn concentration has a higher sensitivity than cytological evaluation of FNA biopsy samples, so it is possible that in the future, serum Ctn concentration measurement will be used in routine diagnosis not only in Europe but also elsewhere.

Calcitonin concentration is directly proportional to the degree of progression. With a serum Ctn values of 60–100 pg/mL, there is a strong probability of the presence of a C-cell proliferative process. Zhu et al. (15) thought MTC patients with a pre-operative serum Ctn level >302.50 pg/mL were more likely to have CLNM. Some researchers indicated that the thresholds of preoperative serum calcitonin according to disease extent were 20 pg/mL for ipsilateral lateral LNM, 200 pg/mL for contralateral lateral LNM, and 500 pg/mL for distant metastases (16). Many guidelines indicated that the preoperative serum calcitonin level and ultrasound findings can be helpful for determining the need for lateral node dissection (5, 17). However, there is currently no globally uniform and standard cutoff value of serum Ctn to indicate LLNM. In our study, there is a positive correlation between preoperative serum calcitonin levels and the risk of lateral cervical lymph node metastasis, and the cutoff value of serum Ctn is 317 pg/mL. The main limitation of serum Ctn is the risk of false-positive results. For example, false-positive results of serum Ctn measurement may be caused by ectopic production of Ctn by neuroendocrine tumors. Increased serum calcitonin levels are also observed in hypercalcemia, during the use of proton pump inhibitors, in renal failure, chronic obstructive pulmonary disease and hypothyroidism (18, 19). In our study, hypothyroidism caused by Hashimoto’s thyroiditis may be a major cause of false-positive results.

Consistent with previous studies, male sex was identified as independent risk factor predicting LLNM in MTC patients in our study. But the independent predictors of LLNM in MTC are not in agreement in different researches. Jin et al. reported that only multifocal lesions and suspected LLNM in preoperative US were independent risk factors of LLNM in MTC patients (20). Zhou et al. believed that sex, tumor size, multifocality, extrathyroidal extension, and distant metastasis were identified to be associated with LLNM in MTC (21). Wu et al. showed that patients with MTC (≤1cm) have a lower risk of lateral lymph node metastasis than those with tumor size greater than 1cm (22). However, multifocality and tumor size were not risk factors of LLNM in our study, which may need to be further verified by prospective researches of patients.

Previous studies have found that extrathyroidal extension often reflects the tumor aggressiveness of MTC. Once the malignant tissue infiltrates the capsule, it is easy to enter the lymphatic circulation system, which makes it extremely prone to lymph nodes. Studies showed that extrathyroidal extension was also an independent risk factor for LLNM and recurrence (21, 23). However, our result showed that not only extrathyroidal extension but the position of thyroid nodule close to capsule (≤2mm) was also an independent predictive indicator to LLNM. When the nodule was near the capsule, the thyroid capsule might have received invasion.

Previous studies have demonstrated that nomograms have been useful in predicting LLNM risk in PTC patients (24, 25), but there were relatively few reports for predicting LLNM risk in MTC. Previous nomograms for estimating LLNM of MTC were constructed basing on different risk parameters, such as suspected LLNM in preoperative ultrasound and distant metastasis (20). However, the existence of occult LLNM is not easily detected in ultrasound and the incidence of distant metastases is relatively low, making this index not applicable for the vast majority of MTC. Some scholars predicted LLNM in MTC by the SEER database, but it lacked an important data of calcitonin, which benefited a lot in predicting LLNM in MTC (21). Besides, previous studies of MTC are all single-center studies with a small number of samples (3), so the applicability is limited to a certain extent.

There are several limitations in the present study. First, it was designed retrospectively, and the assessment of static US images has an inherent limitation to the precision of US interpretation. Second, this study did not distinguish between the subtypes of MTC. Whether the model is applicable to hereditary MTC requires further verification. Third, given that LND was only performed in patients with high suspicion of LLNM based on preoperative imaging and FNAC, microscopic metastases might be overlooked. Last, the external validation dataset contains a small number of samples, and further confirmation with a large sample and prospective clinical trial is needed in the future.

In conclusion, we comprehensively identified individual predictive factors of LLNM for patients with MTC, including gender (male), the relationship with the thyroid capsule and serum Ctn. The nomograms were proved to be accurate, efficient, and clinically beneficial in preoperative guiding the clinical diagnosis and treatment process of MTC patients.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving humans were approved by The Ethics Committee of Tianjin Medical University Cancer Institute and Hospital. The studies were conducted in accordance with the local legislation and institutional requirements. The ethics committee/institutional review board waived the requirement of written informed consent for participation from the participants or the participants’ legal guardians/next of kin because Patient consent was waived due to the retrospective nature of the study. Written informed consent was obtained from the individual(s) for the publication of any potentially identifiable images or data included in this article.

JLZ: Writing – review & editing, Writing – original draft, Funding acquisition, Conceptualization. TG: Writing – original draft, Methodology. SG: Writing – original draft, Methodology, Funding acquisition. LC: Writing – original draft, Validation, Methodology, Data curation. JZ: Writing – original draft, Data curation. XQW: Writing – original draft, Data curation. XW: Writing – review & editing, Writing – original draft, Project administration, Conceptualization.