Combined Detection of Preoperative Serum Calcitonin and Carcinoembryonic Antigen in Medullary Thyroid Carcinoma: A Retrospective Multicenter Cohort Study Authors: Yu Yang, Yufei He, Bo Lin, Yuanyuan Zheng, Xinyi Deng, Wenxin Chen, Zhenbang Ke, Yue Qiu, Yuting Hong, Jiayuan Zhang, Guojie Dong, Siqi Qiao, Ziyi Xin, Xuyang Li, Tianyi Xu, Wenke Chen, Xinwen Chen, Yingtong Hou, Fenghua Lai, Jianyan Long, Zhanqiang Zhang, Xiangdong Xu, Yihao Liu Published: 2026 ABSTRACT ABSTRACTBackgroundThe aim of this study was to investigate the clinical value of combined detection of preoperative serum calcitonin (Ctn) and carcinoembryonic antigen (CEA) for diagnosing medullary thyroid carcinoma (MTC), as well as predicting lymph node metastases and prognosis, compared with the value of detection of either marker alone.MethodsThis retrospective multicenter cohort study included patients with thyroid nodules who underwent preoperative measurements of serum Ctn and CEA at six medical centers between 2012 and 2024. Receiver operating characteristic (ROC) curves were analyzed to compare the diagnostic performances of Ctn, CEA, and their combination.ResultsA total of 13,272 patients were included in the study, comprising 115 with MTC, 10,845 with papillary thyroid carcinoma, and 2312 with benign nodules. For diagnosing MTC, the area under the ROC curve (AUC) values for Ctn and CEA alone were 0.992 and 0.983, respectively, while the combination of the two markers achieved the highest AUC of 0.999. Ctn had a higher positive predictive value (PPV) for MTC diagnosis than CEA (51% vs. 31%), and the combination did not improve the PPV (46%) compared with Ctn alone. For predicting cervical and lateral neck lymph node metastases in MTC, the combination had a higher AUC than Ctn or CEA alone (0.735 vs. 0.708 vs. 0.655 and 0.769 vs. 0.707 vs. 0.675). The combination also had a higher AUC for predicting recurrence than Ctn or CEA alone (0.734 vs. 0.642 vs. 0.611).ConclusionsCompared with the detection of either marker alone, the combined detection of Ctn and CEA improved the diagnostic accuracy for MTC but did not enhance the PPV. The combined detection also improved the predictive accuracy for cervical and lateral lymph node metastases and disease recurrence. Abbreviations American Thyroid Association area under the ROC curve carcinoembryonic antigen confidence interval computed tomography calcitonin fine‐needle aspiration cytology hazard ratio interquartile range medullary thyroid carcinoma negative predictive value overall survival positive predictive value papillary thyroid carcinoma recurrence‐free survival receiver operating characteristic standard deviation. tumor‐node‐metastasis Background Arising from the calcitonin (Ctn)‐secreting parafollicular C cells of the thyroid gland, medullary thyroid carcinoma (MTC) is more aggressive and has a poorer prognosis than differentiated thyroid carcinoma [1, 2]. Thus, the current guidelines recommend more aggressive surgical approaches and more intensive postoperative surveillance in patients with MTC [3, 4, 5, 6, 7]. Consequently, it is crucial to establish a preoperative diagnosis of MTC and conduct a postoperative prognostic assessment to enable the formation of surgical plans and postoperative follow‐up strategies in MTC patients. However, preoperative diagnosis of MTC remains a clinical challenge. Current routine diagnostic methods, including cervical ultrasonography and fine‐needle aspiration cytology (FNAC), have suboptimal efficacy for the diagnosis of MTC [8, 9]. As a characteristic tumor marker for MTC, Ctn screening exhibits high sensitivity for MTC detection, outperforming both ultrasound and FNAC [10, 11, 12]. Nevertheless, the application of Ctn for MTC screening is associated with false‐positive results [13], leading to unnecessary or overly extensive surgeries. Although several studies have also indicated that preoperative serum Ctn is correlated with lymph node metastases and prognosis, its predictive performance remains suboptimal [14, 15, 16, 17, 18]. Carcinoembryonic antigen (CEA) is a classical tumor marker that is elevated in various malignancies, including MTC [19, 20]. While the diagnostic and prognostic utility of CEA has been extensively investigated in gastrointestinal cancers [21, 22, 23, 24, 25], its role in MTC remains underexplored. Small‐scale studies have shown that serum CEA is associated with tumor differentiation, tumor size, tumor‐node‐metastasis (TNM) stage, extent of lymph node metastasis, distant metastasis, biochemical cure rates, and mortality in MTC [26, 27]. However, no studies have systematically evaluated the performance and applicability of combining preoperative serum Ctn and CEA measurements for diagnosing MTC, predicting lymph node metastasis, and assessing prognosis. This retrospective multicenter study aimed to determine the diagnostic and prognostic performance of combined detection of preoperative serum Ctn and CEA for MTC compared with the performance of detection of either marker alone. Methods Patients and Study Design This retrospective multicenter cohort study consecutively enrolled patients with thyroid nodules at six medical centers between September 2012 and December 2024. The participating institutions were the First Affiliated Hospital of Sun Yat‐sen University, the Sun Yat‐sen University Cancer Center, the Seventh Affiliated Hospital of Sun Yat‐sen University, the First Affiliated Hospital of Guangxi Medical University, Shunde Hospital of Southern Medical University, and the Guangzhou Women and Children's Medical Center. The inclusion criteria were as follows: (1) simultaneous detection of baseline serum Ctn and CEA levels before surgery; (2) performance of surgical treatment; and (3) definitive pathological diagnosis after surgery. The exclusion criteria were as follows: (1) recurrent disease; and (2) history of other malignant tumors or neuroendocrine tumors. A flowchart for the patient inclusion and exclusion is shown in Supporting Information S1: Figure S1. The study was approved by the Ethics Committee of the First Affiliated Hospital of Sun Yat‐sen University. As a retrospective study, the requirement for informed consent was waived. Other hospitals have recognized the ethical approval granted by the First Affiliated Hospital of Sun Yat‐sen University. Data Collection Clinical data, including age, sex, preoperative serum Ctn and CEA levels, postoperative histopathology, primary tumor size (diameter of the largest tumor for multifocal tumors), lymph node metastasis status and sites, distant metastasis, and TNM stage, were extracted from the electronic medical record systems of the institutions. Tumor staging was conducted using the TNM staging criteria in the 8th edition of the American Joint Committee on Cancer's cancer staging manual. Patients were followed up until the date of death or the end of the study (March 31, 2025). Detection of Serum Ctn and CEA Levels Preoperative serum Ctn and CEA levels were measured by electrochemiluminescence immunoassay (Roche Diagnostics, Mannheim, Germany). The upper limits of the normal ranges were defined as 18 pg/mL for Ctn, with analytical sensitivity of 2 pg/mL, and 5 ng/mL for CEA, with analytical sensitivity of 0.5 pg/mL. Levels below the lower limit of detection were recorded as the lowest detection limit. Outcomes The study outcomes were recurrence‐free survival (RFS) and overall survival (OS). RFS was calculated from the date of surgery to the first recurrence or death from any cause. OS was calculated from the date of surgery to death from any cause. Statistical Analysis All statistical analyses were performed using R version 4.3.1 (R Foundation, Vienna, Austria). Continuous variables were presented as mean ± standard deviation for those with a normal distribution or as median (interquartile range [IQR]) for those with a skewed distribution. Categorical variables were presented as frequency and percentage. Comparisons among groups of independent samples were conducted by analysis of variance (normal distribution) and the Kruskal–Wallis test (skewed distribution) for continuous variables and by the χ 2 test for categorical variables. Correlations were evaluated by Spearman's rank correlation coefficient analysis. Receiver operating characteristic (ROC) curves were constructed to evaluate the diagnostic performances of Ctn, CEA, and their combination for diagnosing MTC, as well as their ability to predict lymph node metastases and prognosis. Multivariable Cox regression models were used to analyze the relationships between preoperative serum Ctn and CEA levels and OS or RFS in MTC patients, presented as the hazard ratio (HR) with 95% confidence interval (CI). All analyses were two‐sided tests. Statistical significance was set at p < 0.05. Results Demographic Data and Comparisons of Serum Ctn and CEA Levels Among Groups A total of 13,272 patients with thyroid nodules who underwent simultaneous measurements of preoperative serum Ctn and CEA levels were included in the study, comprising 115 with MTC, 10,845 with papillary thyroid carcinoma (PTC), and 2312 with benign nodules. The clinical characteristics of the patients are summarized in Table 1. The median preoperative serum Ctn and CEA levels in the MTC group were 2001.00 (IQR: 374.00–5001.00) ng/L and 39.26 (IQR: 9.64–126.80) μg/L, respectively, being significantly higher than the corresponding levels in the PTC and benign nodule groups (all p < 0.0001). Spearman rank correlation coefficient analysis revealed a significant positive correlation between the serum Ctn and CEA levels in the MTC group, but not in the PTC and benign nodule groups (Supporting Information S1: Table S1). Table 1Clinical characteristics of the patients.VariablesPatients with MTCPatients with PTCPatients with benign nodules n 11510,8452312SexMale63 (54.8)2874 (26.5)518 (22.4)Female52 (45.2)7971 (73.5)1794 (77.6)Age43.61 ± 13.9141.49 ± 11.9146.93 ± 13.43Tumor diameter (mm)22.0 (12.0, 33.0)9.0 (6.0, 14.0)29.0 (11.0, 42.0)Lymph node metastasisYes72 (62.6)4046 (37.3)0 (0.0)No43 (37.4)6799 (62.7)2312 (100.0)Location of lymph node metastasisCentral24 (33.3)2691 (66.4)0 (0.0)Lateral11 (15.3)339 (8.4)0 (0.0)Central + Lateral37 (51.4)1016 (25.1)0 (0.0)Distant metastasisYes7 (6.1)21 (0.2)0 (0.0)No108 (93.9)10824 (99.8)2312 (100.0)CtnMedian (IQR)2001.00 (374.00, 5001.00)0.98 (0.50, 1.99)0.87 (0.49, 1.99)Patients with elevated Ctn111 (96.5)203 (1.9)30 (1.3)Patients with normal Ctn4 (3.5)10642 (98.1)2282 (98.7)CEAMedian (IQR)39.26 (9.64, 126.80)1.34 (0.95, 1.93)1.44 (0.98, 2.07)Patients with elevated CEA99 (86.1)134 (1.2)48 (2.1)Patients with normal CEA16 (13.9)10711 (98.8)2264 (97.9)Surgical approachTotal thyroidectomy100 (87.0)4288 (39.5)351 (15.2)Lobectomy15 (13.0)6557 (60.5)1961 (84.8)Extent of lymph node dissectionNo dissection7 (6.1)3851 (35.5)2110 (91.3)Only central47 (40.9)5387 (49.7)182 (7.9)Only lateral2 (1.7)239 (2.2)11 (0.5)Central + Lateral59 (51.3)1368 (12.6)9 (0.4) Note: Values are presented as n (%) or mean ± SD or median (IQR).Abbreviations: CEA, carcinoembryonic antigen; Ctn, calcitonin; IQR, interquartile range; MTC, medullary thyroid carcinoma; PTC, papillary thyroid carcinoma; SD, standard deviation.John Wiley & Sons, Ltd. Clinical characteristics of the patients. Note: Values are presented as n (%) or mean ± SD or median (IQR). Abbreviations: CEA, carcinoembryonic antigen; Ctn, calcitonin; IQR, interquartile range; MTC, medullary thyroid carcinoma; PTC, papillary thyroid carcinoma; SD, standard deviation. Diagnostic Performances of Preoperative Serum Ctn and CEA Levels for MTC For diagnosing MTC, the area under the ROC curve (AUC) values for Ctn and CEA alone were 0.992 and 0.983, respectively, while the combination of the two markers achieved the highest AUC of 0.999. The sensitivity, specificity, and negative predictive value (NPV) were 95%, 99%, and 100% for Ctn alone, 90%, 98%, and 100% for CEA alone, and 97%, 99%, and 100% for the combination, respectively. Although Ctn demonstrated a higher positive predictive value (PPV) for MTC diagnosis than CEA (51% vs. 31%), the combination did not improve PPV (46%) compared with Ctn alone (Table 2 and Supporting Information S1: Figure S2). Table 2Predictive value of serum Ctn or CEA levels alone and their combination for the diagnosis of MTC.BiomarkerSensitivitySpecificityPPVNPVROC‐AUCCtn0.95 (0.89, 0.98)0.99 (0.99, 0.99)0.51 (0.44, 0.57)1 (1.00, 1.00)0.992CEA0.90 (0.84, 0.95)0.98 (0.97, 0.98)0.31 (0.26, 0.36)1 (1.00, 1.00)0.983Ctn + CEA0.97 (0.93, 0.99)0.99 (0.99, 0.99)0.46 (0.40, 0.52)1 (1.00, 1.00)0.999 Note: Values are presented as point estimate (95% CI) or point estimates.Abbreviations: AUC, area under the ROC curve; CEA, carcinoembryonic antigen; Ctn, calcitonin; CI, confidence interval; MTC, medullary thyroid carcinoma; NPV, negative predictive value; PPV, positive predictive value; ROC, receiver operating characteristic.John Wiley & Sons, Ltd. Predictive value of serum Ctn or CEA levels alone and their combination for the diagnosis of MTC. Note: Values are presented as point estimate (95% CI) or point estimates. Abbreviations: AUC, area under the ROC curve; CEA, carcinoembryonic antigen; Ctn, calcitonin; CI, confidence interval; MTC, medullary thyroid carcinoma; NPV, negative predictive value; PPV, positive predictive value; ROC, receiver operating characteristic. Prediction of Lymph Node Metastases in MTC MTC patients with cervical or lateral compartment lymph node metastases had significantly higher preoperative serum Ctn and CEA levels than patients without metastases (all p < 0.01; Supporting Information S1: Table S2). For predicting cervical lymph node metastases, the AUC values for CEA and Ctn alone were 0.655 and 0.708, respectively. The combination of the two markers improved the predictive AUC to 0.735, with sensitivity of 56%, specificity of 86%, PPV of 87%, and NPV of 54% (Table 3 and Supporting Information S1: Table S3 and Figure S3). For predicting lateral lymph node metastases, the AUC values for CEA and Ctn alone were 0.675 and 0.707, respectively. The combination of the two markers increased the AUC to 0.769, with sensitivity of 42%, specificity of 100%, PPV of 100%, and NPV of 71% (Table 3, Supporting Information S1: Table S3, and Supporting Figure S4). Table 3Predictive value of serum Ctn or CEA levels alone and their combination for cervical and lateral lymph node metastases in MTC patients.BiomarkerSensitivitySpecificityPPVNPVROC‐AUCPrediction of cervical lymph node metastasesCtn0.50 (0.39, 0.61)0.86 (0.73, 0.93)0.86 (0.72, 0.93)0.51 (0.39, 0.62)0.708CEA0.81 (0.70, 0.88)0.49 (0.35, 0.63)0.72 (0.62, 0.81)0.60 (0.44, 0.74)0.655Ctn + CEA0.56 (0.44, 0.66)0.86 (0.73, 0.93)0.87 (0.74, 0.94)0.54 (0.42, 0.65)0.735Prediction of lateral lymph node metastasesCtn0.58 (0.44, 0.71)0.79 (0.68, 0.87)0.67 (0.52, 0.79)0.73 (0.61, 0.82)0.707CEA0.79 (0.66, 0.88)0.52 (0.40, 0.64)0.54 (0.43, 0.65)0.78 (0.64, 0.87)0.675Ctn + CEA0.42 (0.29, 0.56)1.00 (0.95, 1.00)1.00 (0.84, 1.00)0.71 (0.61, 0.79)0.769 Note: Values are presented as point estimate (95% CI) or point estimates.Abbreviations: AUC, area under the ROC curve; CEA, carcinoembryonic antigen; Ctn, calcitonin; CI, confidence interval; MTC, medullary thyroid carcinoma; NPV, negative predictive value; PPV, positive predictive value; ROC, receiver operating characteristic.John Wiley & Sons, Ltd. Predictive value of serum Ctn or CEA levels alone and their combination for cervical and lateral lymph node metastases in MTC patients. Note: Values are presented as point estimate (95% CI) or point estimates. Abbreviations: AUC, area under the ROC curve; CEA, carcinoembryonic antigen; Ctn, calcitonin; CI, confidence interval; MTC, medullary thyroid carcinoma; NPV, negative predictive value; PPV, positive predictive value; ROC, receiver operating characteristic. Prognostic Performances of Preoperative Serum Ctn and CEA Levels for MTC Among the MTC patients, 98 (85.2%) exhibited elevation of both Ctn and CEA, three (2.6%) exhibited no elevation of either biomarker, one (0.9%) had elevation of CEA only, and 13 (11.3%) had elevation of Ctn only. The baseline clinical characteristics of the patients in these four subgroups are presented in Supporting Information S1: Table S4. The median follow‐up duration in the MTC group was 61.88 (IQR: 35.15, 86.90) months. During follow‐up, no recurrence was observed in patients with elevation of neither biomarker or elevation of CEA only. 7.7% (1 out of 13) of patient with elevation of Ctn only experienced recurrence, while 15.3% (15 out of 98) of patients with elevation of both Ctn and CEA had recurrence. No deaths were recorded in any of the four groups by the end of the follow‐up period. Compared with the patients in the other three groups, patients with elevation of both Ctn and CEA had a higher risk of recurrence (HR: 2.143; 95% CI: 0.282–16.289) (Table 4). Table 4Clinical outcomes of MTC patients with or without elevated Ctn and/or CEA levels.Biomarker n Follow‐up (Month)DeadRecurrenceHR a (95% CI)Ctn−/CEA−32.51 (0.86, 92.61)0/3 (0.0)0/3 (0.0)1.000Ctn−/CEA+122.41 (22.41, 22.41)0/1 (0.0)0/1 (0.0)Ctn+/CEA−1344.06 (40.66, 53.60)0/13 (0.0)1/13 (7.7)Ctn+/CEA+9866.73 (40.17, 89.41)0/98 (0.0)15/98 (15.3)2.143 (0.282–16.289) Note: Values are presented as median (IQR) or n (%) or point estimate (95% CI).Abbreviations: CEA, carcinoembryonic antigen; CI, confidence interval; Ctn, calcitonin; HR, hazard ratio; MTC, medullary thyroid carcinoma. a Because of the small sample sizes in the groups where neither Ctn nor CEA was elevated or only one of the markers was elevated, these three groups were combined into a “non‐Ctn+/CEA+” group for Cox proportional hazards regression analysis and compared with the “Ctn+/CEA+” group. Furthermore, because there were no deaths in any of the groups, the HR was calculated based on the risk of recurrence.John Wiley & Sons, Ltd. Clinical outcomes of MTC patients with or without elevated Ctn and/or CEA levels. Note: Values are presented as median (IQR) or n (%) or point estimate (95% CI). Abbreviations: CEA, carcinoembryonic antigen; CI, confidence interval; Ctn, calcitonin; HR, hazard ratio; MTC, medullary thyroid carcinoma. Because of the small sample sizes in the groups where neither Ctn nor CEA was elevated or only one of the markers was elevated, these three groups were combined into a “non‐Ctn+/CEA+” group for Cox proportional hazards regression analysis and compared with the “Ctn+/CEA+” group. Furthermore, because there were no deaths in any of the groups, the HR was calculated based on the risk of recurrence. The AUC values for predicting recurrence were 0.642 for Ctn alone and 0.611 for CEA alone. The combination of the two markers improved the AUC to 0.734, with sensitivity of 75%, specificity of 79%, PPV of 36%, and NPV of 95% (Table 5 and Supporting Information S1: Figure S5). Table 5Predictive value of serum Ctn or CEA levels alone and their combination for the prognosis in MTC patients.BiomarkerSensitivitySpecificityPPVNPVROC‐AUCCtn0.50 (0.28, 0.72)0.78 (0.69, 0.85)0.27 (0.14, 0.44)0.91 (0.83, 0.95)0.642CEA0.31 (0.14, 0.56)0.98 (0.93, 0.99)0.71 (0.36, 0.92)0.90 (0.83, 0.94)0.611Ctn + CEA0.75 (0.51, 0.90)0.79 (0.70, 0.86)0.36 (0.22, 0.53)0.95 (0.88, 0.98)0.734 Note: Values are presented as point estimate (95% CI) or point estimates.Abbreviations: AUC, area under the ROC curve; CEA, carcinoembryonic antigen; CI, confidence interval; Ctn, calcitonin; MTC, medullary thyroid carcinoma; NPV, negative predictive value; PPV, positive predictive value; ROC, receiver operating characteristic.John Wiley & Sons, Ltd. Predictive value of serum Ctn or CEA levels alone and their combination for the prognosis in MTC patients. Note: Values are presented as point estimate (95% CI) or point estimates. Abbreviations: AUC, area under the ROC curve; CEA, carcinoembryonic antigen; CI, confidence interval; Ctn, calcitonin; MTC, medullary thyroid carcinoma; NPV, negative predictive value; PPV, positive predictive value; ROC, receiver operating characteristic. Discussion This retrospective multicenter study evaluated the utility of preoperative serum Ctn and CEA levels for diagnosing MTC and predicting lymph node metastases and prognosis. The combined detection of both markers improved the diagnostic accuracy for MTC but did not enhance the PPV. Compared with detection of either marker alone, combined detection improved predictive accuracy for cervical and lateral lymph node metastases and disease recurrence. Although Ctn is a characteristic biomarker for MTC, its routine preoperative screening remains controversial. European and Chinese guidelines recommend routine Ctn screening in patients with thyroid nodules [3, 28], while the American Thyroid Association (ATA) guidelines leave its testing to the clinician discretion [5]. The National Comprehensive Cancer Network guidelines do not recommend routine preoperative Ctn screening in patients with thyroid nodules, instead suggesting that Ctn testing should only be conducted when histopathological findings raise suspicion of MTC [7]. CEA, initially identified in colon cancer [29], is elevated in multiple cancers, including MTC. Studies have shown that serum CEA is correlated with tumor differentiation, tumor size, TNM staging, extent of lymph node metastasis, distant metastasis, biochemical cure, and mortality in MTC [26, 27]. However, because CEA is not a specific tumor marker for MTC, the ATA guidelines recommend concurrent measurement of Ctn and CEA once MTC has been diagnosed to assess the disease status and establish the baseline levels for postoperative monitoring [5]. Early and accurate diagnosis, followed by appropriate surgical treatment, is crucial for improving the prognosis of MTC patients [30, 31]. Ctn is elevated in nearly all MTC patients, and its sensitivity for preoperative screening and diagnosis of MTC ranges from 83% to 100% [13]. However, elevated Ctn can also be observed in C‐cell hyperplasia, autoimmune thyroiditis, hyperparathyroidism, chronic renal failure, other neuroendocrine tumors, and some differentiated thyroid carcinomas [32, 33, 34, 35, 36, 37, 38]. Consequently, Ctn screening has the potential for false‐positive results, leading to unnecessary or excessive surgical procedures [39]. In previous studies, the highest PPV for diagnosis of MTC using Ctn alone was slightly above 60% [40], and the PPV observed in the present study was only 51%. For patients with thyroid nodules who exhibit elevated serum Ctn, levels between 10 and 100 pg/mL are generally considered to fall within a diagnostic “gray zone” for MTC. In our cohort, 145 patients had serum Ctn levels within this range (10–100 pg/mL), including 10 (6.90%) with MTC, 119 (82.07%) with PTC, and 16 (11.03%) with benign nodules. Therefore, further investigations are required to improve the performance of Ctn for MTC diagnosis, especially in patients whose Ctn levels fall within the gray zone. Although CEA is not a specific marker for MTC, it has a high rate of positive expression in MTC patients, and elevated serum CEA is positively correlated with serum Ctn [26, 41]. The current guidelines do not recommend using serum CEA alone as a screening indicator for MTC. Consistent with the findings reported by Bei Cai et al. [42], we found that the combination of Ctn and CEA did not improve the PPV compared with Ctn alone. From a health economics perspective, the value of detecting CEA in all patients with thyroid nodules is questionable. The issue of false‐positive findings in Ctn‐based MTC diagnosis requires resolution through alternative methods. The presence of cervical lymph node metastases, particularly in the lateral compartment, is an important factor influencing the surgical approach and prognosis of MTC patients. However, the sensitivities of current first‐line modalities for assessing lymph node metastases—ultrasound and computed tomography (CT) examination—are unsatisfactory [43, 44]. Preoperative serum Ctn levels have been demonstrated to reflect the extent of regional lymph node metastases in MTC patients [14, 16, 18]. For patients with MTC and no ultrasonographic evidence of neck metastases, the ATA and European Society for Medical Oncology guidelines now recommend that prophylactic dissection of lymph nodes in the lateral compartments may be considered based on the serum Ctn level [5, 6]. Likewise, several studies have shown that preoperative serum CEA is associated with the extent of lymph node metastases in MTC [26, 27]. In the present study, we found that the AUC for predicting lateral cervical lymph node metastases was higher for the combined detection of Ctn and CEA than for the detection of either marker alone. Our findings support the notion that the combined detection of Ctn and CEA enhances the prediction of lateral lymph node metastases, potentially guiding the extent of surgery. Prognostic assessment is essential for tailoring follow‐up strategies. In our previous study, we found that preoperative Ctn screening was associated with more comprehensive surgeries and improved survival [45]. Several studies have further shown that preoperative serum Ctn can predict disease recurrence in MTC [15, 17, 18]. The prognostic value of serum CEA in various malignancies has been widely reported, with studies demonstrating that elevated CEA is associated with tumor recurrence and poor clinical outcomes in colorectal, gastric, pancreatic, breast, and non‐small cell lung cancers [21, 22, 24, 25, 46, 47, 48, 49]. For MTC, only one study involving 33 MTC patients has indicated a relationship between preoperative serum CEA and prognosis, with the finding that serum CEA > 500 ng/mL was significantly correlated with increased mortality (67%) [26]. In the present study, we found that elevation of both Ctn and CEA was associated with a higher risk of recurrence. Moreover, the combined detection of Ctn and CEA improved the predictive accuracy for recurrence compared with the detection of either marker alone. Thus, the combined detection of Ctn and CEA may provide more robust guidance for individualized treatment planning and follow‐up strategies in MTC patients. Because we found that the combined detection of Ctn and CEA did not improve the PPV for MTC diagnosis, but did enhance the accuracy for MTC diagnosis and prediction of cervical and lateral lymph node metastases, as well as disease recurrence, and because of the low incidence of MTC, we do not recommend routine combined screening of serum Ctn and CEA for patients with thyroid nodules from a cost‐effectiveness perspective. However, for patients with elevated preoperative Ctn and suspected MTC, we recommend additional CEA testing. This study has several limitations. The retrospective study design introduces potential biases. Although the study is the largest multicenter study to date to investigate the combined detection of CEA and Ctn in MTC, the sample size remains limited because of the rarity of MTC. The analyses of prognostic factors were underpowered because of the small numbers in certain categories. In addition, the data were sourced from hospitals in southern China, which may limit the generalizability of the findings to other populations. Further large‐scale, multicenter, high‐quality clinical studies are warranted to provide more robust evidence and to enhance the generalizability of the research findings. Conclusions Our study demonstrates that the combined detection of Ctn and CEA does not improve the PPV for diagnosing MTC compared with the detection of either marker alone. However, the combined detection improves the accuracy for predicting cervical and lateral lymph node metastases, as well as disease recurrence. These findings suggest that the combined detection of preoperative serum Ctn and CEA has a certain clinical value for the diagnosis, treatment, and prognostic evaluation of MTC patients. Author Contributions Yu Yang: conceptualization, writing – original draft, data curation, formal analysis. Yufei He: conceptualization, data curation. Bo Lin: conceptualization, data curation, formal analysis, project administration. Yuanyuan Zheng: data curation. Xinyi Deng: data curation. Wenxin Chen: data curation. Zhenbang Ke: data curation. Yue Qiu: data curation. Yuting Hong: data curation. Jiayuan Zhang: data curation. Guojie Dong: data curation. Siqi Qiao: data curation. Ziyi Xin: data curation. Xuyang Li: data curation. Tianyi Xu: data curation. Wenke Chen: data curation. Xinwen Chen: data curation. Yingtong Hou: data curation. Fenghua Lai: data curation. Jianyan Long: formal analysis. Zhanqiang Zhang: supervision, writing – review and editing, project administration. Xiangdong Xu: supervision, writing – review and editing, project administration. Yihao Liu: supervision, writing – review and editing, project administration. Ethics Statement This study was approved by the Research Ethics Committee of the First Affiliated Hospital of Sun Yat‐sen University ([2024]235), and it was compliant with the Helsinki Declaration of 1975, as revised in 2024. Other hospitals have recognized the ethical approval granted by the First Affiliated Hospital of Sun Yat‐sen University. Consent Based on the retrospective nature of this study, the Research Ethics Committee of the First Affiliated Hospital of Sun Yat‐sen University waived the requirement for patient‐informed consent. Conflicts of Interest The authors declare no conflicts of interest. Supporting information Supporting File 1