Surgical extent, tumor size, and calcitonin levels in medullary thyroid carcinoma: impact on oncologic outcomes in a retrospective cohort study Authors: Ekin Yiğit Köroğlu, Berna Evranos Öğmen, Muzaffer Serdar Deniz, Barış Saylam, Hilal Ahsen, Cevdet Aydin, Oya Topaloğlu, Reyhan Ersoy, Bekir Çakir Published: 2026 ABSTRACT AbstractBackgroundMedullary thyroid carcinoma (MTC) treatment guidelines recommend surgical extent based primarily on calcitonin levels, yet the oncologic impact of deviating from these protocols remains incompletely characterized. This study evaluates whether individualized treatment approaches based on comprehensive tumor characterization compromise disease outcomes compared to guideline-adherent management, and examines the relative importance of tumor size, calcitonin levels, and surgical extent in predicting persistent disease.MethodsThis retrospective cohort study included 91 patients with histologically confirmed MTC followed at a tertiary center from January 2019 to March 2025. Surgical procedures were compared against ATA-2015, ESMO-2019, and NCCN-2025 guideline recommendations and classified as concordant, less extensive, or more extensive. Primary outcomes included persistent disease, defined as biochemical or structural incomplete response occurring within 6 months of initial treatment; recurrent disease, defined as the reappearance of biochemical or structural disease more than 6 months after treatment in patients who had previously achieved an excellent response; reoperation rates; and postoperative complications. Associations between preoperative calcitonin levels and lymph node (LN) metastasis patterns were analyzed. Multivariable logistic regression identified independent predictors of persistent disease.ResultsSurgical procedures deviated from guidelines in 32.3% (ATA), 70.8% (ESMO), and 40.0% (NCCN) of cases. Deviation from guideline-recommended extent did not significantly impact persistent disease rates, reoperation requirements, or complication rates. Overall, 23.1% of patients had persistent disease at final follow-up. On multivariable analysis, independent predictors of persistent disease were higher preoperative calcitonin levels (OR 1.02 per 100 pg/mL increase, 95% CI 1.00-1.04, p = 0.042) and presence of LN metastasis (OR 4.52, 95% CI 1.01–20.24, p = 0.049), but not surgical extent classification. Primary tumor size correlated with metastatic involvement across all cervical compartments, while preoperative calcitonin levels predicted only ipsilateral lateral LN metastasis. A calcitonin threshold of 189.5 pg/mL predicted ipsilateral lateral metastasis with 92.3% sensitivity and 61.8% specificity.ConclusionsIndividualized surgical approaches based on comprehensive preoperative assessment do not compromise oncologic outcomes when compared to guideline-adherent protocols. Disease persistence is determined by tumor biology rather than adherence to standardized surgical algorithms. Integration of tumor size with calcitonin measurements may improve risk stratification and enable treatment personalization while maintaining disease control.Keywords: Medullary thyroid cancer, Calcitonin, Lymph node dissection, Tumor size Background Medullary thyroid carcinoma (MTC) is a neuroendocrine tumour originating in the thyroid gland’s C cells, also known as the parafollicular cells. It accounts for around one to two per cent of thyroid cancers [1]. Patients with MTC can only be treated with complete resection of the thyroid tumor and any local and regional metastases [1]. Guidelines guide the extent of surgery and the need for further treatment. For this purpose, the American Thyroid Association (ATA) (2015) [1], European Society for Medical Oncology (ESMO) (2019) [2] and National Comprehensive Cancer Network (NCCN) (2025) [3] guidelines are the most commonly used and up-to-date guidelines. Although these guidelines provide direction for treatment, many studies support less extensive interventions, particularly regarding the extent of surgery. For example, in a study by Yamashita et al., 18 out of 110 MTC patients without lateral metastatic lymph node (LN) findings underwent prophylactic LN dissection (LND), while 92 did not. No significant difference was found between the two groups in terms of recurrence and median survival during the 5-year follow-up period [4]. Similarly, Spanheimer et al. also found that prophylactic LND based on preoperative calcitonin (CT) levels had no additional effect on recurrence or survival in their study [5]. In recent years, an approach that narrows the surgical width according to patient-based findings has gained prominence, replacing the previous strategy based on CT levels without radiological evidence. A 2023 study evaluating this approach examined 235 patients undergoing surgery for MTC. The study investigated the safety of preserving the node-negative side. It demonstrated that loco-regional recurrence in MTC does not affect overall survival or metastatic disease, even if a second surgery is required [6]. In a study published in 2024, Tao et al. suggested that routine prophylactic central LND may not be necessary in T1 stage tumors [7]. In our clinical practice, when scheduling surgery for patients with MTC, we often adopt a patient-based approach, meaning that the surgical decision may be more or less extensive than the guideline recommendations. This study aims to demonstrate how performing surgery that deviates from guideline recommendations impacts clinical endpoints and to evaluate how CT levels may influence preoperative planning. Methods Participants Patients followed for MTC at our clinic from January 2019 to March 2025 were examined retrospectively. Patients who underwent surgery during or before the study period, had a histopathologically confirmed MTC diagnosis, and had at least 6 months of postoperative follow-up were included. The study excluded patients who could not obtain surgical pathology results, those under 18 years old, and those with less than 6 months of postoperative follow-up. Covariates The data collected included age at diagnosis, gender, preoperative CT levels, and preoperative ultrasound findings, such as the size and volume of the MTC nodule confirmed on histopathology after surgery, along with the presence and location of metastatic LNs. Surgical variables included the procedures performed, those recommended by current guidelines (ATA 2015, ESMO 2019, and NCCN 2025) [1–3], and adherence to guideline-based recommendations. Adherence categories were defined by comparing the surgery performed against the guideline-recommended extent of resection: compatible denotes surgery concordant with current recommendations; more extensive denotes the inclusion of additional lymph node dissections beyond those recommended; and less extensive denotes the omission of one or more recommended dissections. These comparisons regarding surgical extent were made by classifying and evaluating surgical procedures as follows: lobectomy; total thyroidectomy (TTx); TTx + ipsilateral central LND (CLND); TTx + bilateral CLND; TTx + bilateral CLND + ipsilateral lateral LND (LLND); and TTx + bilateral CLND + bilateral LLND. The extent of surgery was determined through an individualized, multidisciplinary approach for each patient, integrating clinical, biochemical, and imaging findings. Clinical factors included tumor size on palpation, LN palpability, hereditary versus sporadic status, and RET mutation status (performed in all patients). Biochemical markers, particularly preoperative serum CT levels, were considered in the surgical planning. Imaging findings included ultrasonographic tumor characteristics such as size and multifocality, ultrasonographic LN involvement in central and lateral compartments, and cross-sectional imaging when performed. Postoperative histopathological data included the number and total size of MTC foci within the thyroid gland, bilateral tumor involvement, confirmed metastatic LN status with their anatomical locations and numbers, and AJCC (8th edition) [8] TNM classification. The proposed response to therapy for MTC was based on previous suggestions [9, 10]. Patients were divided into three groups based on their response to initial treatment: excellent response, biochemical incomplete response, and structural incomplete response [11]. An excellent response was defined as undetectable CT levels and no evidence of structural disease on imaging. The presence of detectable CT characterized a biochemical incomplete response without evidence of structural disease. A structural incomplete response was defined as the presence of structural disease following initial therapy. Patients who were identified as having biochemical or structural incomplete responses within 6 months during the follow-up were considered to have ‘persistent disease’ [11]. Recurrence was defined as the development of biochemical or structural incomplete disease during follow-up in a patient whose initial response to treatment was excellent for more than 6 months. Follow-up duration, persistent disease status, requirement for reoperation, and RET gene mutation analysis were also recorded. Radiological, clinical, and pathological parameters associated with preoperative CT levels were statistically analyzed. The impact of surgical decisions based on guidelines (whether consistent, less extensive, or more extensive) on the need for re-surgery and persistent disease was examined. Additionally, the clinicopathological features of patients with and without RET mutations were compared. The ellipsoid volume formula (width × height × length × π/6), based on preoperative ultrasonographic measurements, was used to calculate the volumes of the thyroid nodules. A Siemens IMMULITE 2000 automated chemiluminescence immunoassay analyzer, with a sensitivity of 2 pg/mL, was employed for the CT analysis, with reference values up to 5 pg/mL for women and 8.4 pg/mL for men. The analysis was conducted using a standard assay kit designed for in vitro diagnostics, supplied by Siemens Healthcare Diagnostics Products Limited. Statistical analysis The SPSS 25.0 software package (IBM Corp., Armonk, NY, USA) was used for statistical analysis. Descriptive statistics are presented as median (minimum-maximum) for non-normally distributed variables and mean ± standard deviation for normally distributed variables, while qualitative variables are expressed as absolute and relative (%) frequencies. Comparisons between categorical variables were performed using the chi-squared test. The Student’s t-test was applied for parametric variables, and the Mann-Whitney U test was used for non-parametric variables. The Kruskal-Wallis test was employed to compare differences among more than two groups for non-parametric variables. Multivariable logistic regression was performed to identify independent predictors of persistent disease and postoperative complications. Variables with p < 0.2 on univariable analysis or with clinical relevance were included in the multivariable model. P < 0.05 was considered statistically significant. Spearman’s correlation test was used for data that were not normally distributed. Receiver Operating Characteristic (ROC) curve analysis was conducted to assess the diagnostic performance and determine the optimal cut-off value of relevant parameters. Compliance with ethical standards Approval from the Ankara Bilkent City Hospital ethics committee was obtained in accordance with the ethical standards outlined in the Declaration of Helsinki. Informed consent was obtained from all participants. Results A total of 91 patients were included in the study. The mean age of the patients was 53.19 ± 13.12 years, and 56 of them (61.5%) were female. Patients were followed up for a median of 30.0 (range 6.0–221.0) months. According to the AJCC 8th edition TNM staging, 49 patients (53.8%) were stage I, 8 (8.8%) were stage II, 10 (11.0%) were stage III, 16 (17.6%) were stage IVa, 1 (1.1%) was stage IVb, and 7 (7.7%) were stage IVc. Baseline characteristics of the patients are shown in Table 1. Table 1Baseline characteristics of the patients with MTCCharacteristicValue (n = 91)Age at diagnosis (years)53.19 ± 13.12Gender (female)56 (61.5%)Follow-up duration (months)30.0 (6.0-221.0)Primary tumor diameter (cm)1.35 (0.11-11.00)Multifocality18 (19.8%)Bilateral tumor foci11 (12.1%)Microcarcinoma31 (34.1%)Cervical LN metastasis pN061 (67.0%) pN1a12 (13.2%) pN1b18 (19.8%)TNM stage (AJCC 8th edition) I49 (53.8%) II8 (8.8%) III10 (11.0%) IVa16 (17.6%) IVb1 (1.1%) IVc7 (7.7%)Response to initial therapy Excellent response50 (54.5%) Biochemical incomplete response27 (29.7%) Structural incomplete response14 (15.4%) Preoperative serum calcitonin (pg/ml)168.00 (0.50-16800.00) Postoperative serum calcitonin (pg/ml) (n = 41)70.30 (2.60-21168.00)Parametric variable (age) was compared using Student’s t-test, and non-parametric variable (tumour diameter) was compared using the Mann-Whitney U test; categorical variables were analysed using the chi-square testMTC Medullary thyroid carcinoma, LN Lymph node, AJCC American Joint Cancer Committee Baseline characteristics of the patients with MTC Parametric variable (age) was compared using Student’s t-test, and non-parametric variable (tumour diameter) was compared using the Mann-Whitney U test; categorical variables were analysed using the chi-square test MTC Medullary thyroid carcinoma, LN Lymph node, AJCC American Joint Cancer Committee During preoperative evaluations of patients, the median CT level was 168.00 (0.50-16800.00) pg/ml. In preoperative neck ultrasounds, the median longest size of thyroid nodules with MTC foci was 15.40 mm (range 2.60–73.50), and the median nodule volume was 0.99 cm³ (range 0.01-108.08). The correlation analysis performed to identify factors associated with preoperative CT levels is shown in Table 2. Preoperative CT levels were within the normal range for four patients. Of these, one had undergone lobectomy and three had undergone TTx. Microcarcinoma was found in all these patients, with tumor sizes of 0.1, 0.1, 0.2, and 0.3 cm. Postoperative CT levels were undetectable in all four patients. Table 2Correlation analysis of clinical and pathological parameters with preoperative calcitonin levelsVariableSpearman’s rp-valueSex0.1720.149Age at diagnosis0.0360.766Number of MTC nodules0.1470.237Sonographic diameter of the MTC nodule 0.657 < 0.001 Sonographic volume of the MTC nodule 0.656 < 0.001 Postop metastatic LN presence 0.279 0.020 Ipsilateral central LN presence0.2110.084Ipsilateral lateral LN presence 0.364 0.002 Contralateral central LN presence0.1620.186Contralateral lateral LN presence0.1780.146Number of metastatic LNs 0.344 0.004 Number of tumor foci0.1060.387Total tumor diameter 0.822 < 0.001 Bilateral MTC-0.0880.473Microcarcinoma -0.615 < 0.001 Statistical analysis was performed using Spearman’s rank correlation coefficientLN Lymph node, MTC Medullary thyroid carcinoma Correlation analysis of clinical and pathological parameters with preoperative calcitonin levels Statistical analysis was performed using Spearman’s rank correlation coefficient LN Lymph node, MTC Medullary thyroid carcinoma Of the 91 patients, 83 were referred for surgery with a preoperative known diagnosis of MTC. The preferred surgical procedures for all patients are outlined in Table 3. When patients with an unknown MTC diagnosis before surgery were evaluated separately, two patients underwent lobectomy, four patients underwent TTx, one patient underwent TTx and bilateral CLND, and one patient underwent TTx and ipsilateral CLND. When surgical procedures were evaluated according to the ATA 2015, ESMO 2019, and NCCN 2025 guidelines, 67.7% of surgeries were concordant with the ATA 2015 recommendations, while 27.7% were less extensive and 4.6% were more extensive than recommended. According to the ESMO 2019 guidelines, only 29.2% of the procedures were in concordance with the recommendations, whereas 67.7% were less extensive and 3.1% were more extensive. Evaluation based on the NCCN 2025 guidelines showed that 60.0% of surgeries were concordant, 6.2% were less extensive, and 33.8% were more extensive than the guideline recommendations. Table 3Types of surgery performed on patientsSurgical Procedure n %Lobectomy22.2TTx2426.3TTx + ipsilateral central LND22.2TTx + bilateral central LND2931.9TTx + bilateral central LND + ipsilateral lateral LND2224.2TTx + bilateral central LND + bilateral lateral LND1213.2Total 91 100.0 TTx Total thyroidectomy, LND Lymph node dissection Types of surgery performed on patients TTx Total thyroidectomy, LND Lymph node dissection A total of 30 patients (33.0%) showed metastatic LN involvement. Metastasis was located in the ipsilateral central LN (CLN) in 27 patients (90.0% of those with metastasis; 29.7% of all patients), in the contralateral CLN in 10 patients (33.3% of those with metastasis; 11.0% of all patients), in the ipsilateral lateral LN (LLN) in 18 patients (60.0% of those with metastasis; 19.8% of all patients), and in the contralateral LLN in 7 patients (23.3% of those with metastasis; 7.7% of all patients). The distribution of LN metastases is shown in Table 4. All patients with contralateral LLN metastasis had ipsilateral LLN metastasis. Among those with contralateral CLN metastasis, all but two had ipsilateral LLN metastasis. Among patients with ipsilateral LLN metastasis, 10 (55.5%) had no contralateral CLN metastasis. Table 4Distribution of lymph node metastasis locations in patients with cervical metastasisLymph node involvementNumber of patientsn (%)Ipsilateral central only10 (33.3%)Ipsilateral central + Ipsilateral lateral9 (30.0%)Ipsilateral central + Contralateral central + Ipsilateral lateral + Contralateral lateral6 (20.0%)Ipsilateral lateral only1 (3.3%)Ipsilateral central + Contralateral central + Ipsilateral lateral1 (3.3%)Ipsilateral central + Contralateral central1 (3.3%)Contralateral central + Ipsilateral lateral + Contralateral lateral1 (3.3%)Contralateral central only1 (3.3%)Total 30 Distribution of lymph node metastasis locations in patients with cervical metastasis In the preoperative ultrasound evaluation of 30 patients with histopathologically confirmed LN metastasis, metastatic LN was identified in only 19 (63.3%) patients. Examining the factors associated with metastatic LN presence revealed a connection with preoperative CT level (rho: 0.279, p = 0.02), ultrasonographic nodule volume (rho: 0.267, p = 0.039), number of tumor foci (rho: 0.210, p = 0.048), and the longest tumor size in pathology (rho: 0.525, p < 0.001). Similarly, the number of metastatic LNs was linked to preoperative CT level (rho = 0.344, p = 0.04), ultrasonographic nodule volume (rho = 0.310, p = 0.016), number of tumor foci (rho = 0.251, p = 0.018), and longest tumor size in pathology (rho = 0.571, p < 0.001). Analyzing the relationship between metastatic LN compartments and preoperative CT level and longest tumor size in pathology showed that the longest tumor size in pathology was connected to the presence of metastatic LNs in all four compartments. However, the preoperative CT level was only associated with ipsilateral LLN metastasis (r = 0.364, p = 0.002). The results of this analysis are summarized in Table 5. ROC analysis of the effectiveness of preoperative CT level in predicting ipsilateral LLN metastasis is also shown in Fig. 1. This analysis found that ipsilateral LLN could be expected with 92.3% sensitivity and 61.8% specificity using a CT level cutoff value of 189.5 pg/mL. Table 5Correlation analyses between metastatic lymph node compartments and preoperative calcitonin level and primary tumor diameterMetastatic LN compartmentVariableSpearman’s rp-valueIpsilateralcentralPreop. calcitonin0.2110.084Tumor diameter 0.455 < 0.001 Contralateral centralPreop. calcitonin0.1620.186Tumor diameter 0.365 0.001 IpsilaterallateralPreop. calcitonin 0.364 0.002 Tumor diameter 0.505 < 0.001 Contralateral lateralPreop. calcitonin0.1780.146Tumor diameter 0.367 0.001 Statistical analysis was performed using Spearman’s rank correlation coefficientLN Lymph node Correlation analyses between metastatic lymph node compartments and preoperative calcitonin level and primary tumor diameter Ipsilateral central Ipsilateral lateral Statistical analysis was performed using Spearman’s rank correlation coefficient LN Lymph node ROC analysis of the usability of calcitonin level in predicting ipsilateral lateral lymph node metastasis Within six months of initial treatment, 50 patients (54.5%) had achieved an excellent response, while the remaining 41 patients (45.1%) were classified as having persistent disease, comprising 27 (29.7%) with a biochemical incomplete response and 14 (15.4%) with a structural incomplete response. Of the 27 patients with biochemically incomplete disease, structural disease was detected in 15 of them (55.6%) during follow-up. Of the 41 patients who did not achieve an excellent response, 26 underwent re-surgery during the follow-up period. One patient who initially achieved an excellent response developed a recurrence; the others continued to be monitored as excellent responders. In this patient, who initially showed an excellent response but later experienced recurrence, structural recurrence was observed 26 months after surgery. The initial surgery of this patient was performed at a width consistent with the recommendations of all three guidelines. During the follow-up period, this structural disease was treated with thermal ablation, resulting in an excellent response. At the final follow-up, 21 patients (23.1%) remained in a biochemical or structural incomplete state, all of whom had persistent disease following the initial treatment. Figure 2 summarises the follow-up process for patients. Seven patients had distant metastases: two had liver metastases, three had bone metastases, one had both liver and bone metastases, and one had liver, bone, and lung metastases. Summary of patient follow-up and clinical status Spearman’s correlation analysis revealed that preoperative CT levels (r = 0.392, p = 0.001), tumor volume (r = 0.290, p = 0.021), number of tumor foci (r = 0.257, p = 0.014), tumor diameter (r = 0.520, p < 0.001), and bilateral MTC (r = 0.274, p = 0.010) were significantly correlated with the persistent disease, whereas sex (p = 0.309) and age at diagnosis (p = 0.219) showed no significant correlation. On multivariable logistic regression analysis adjusting for tumor size, preoperative CT level, LN metastasis, age, gender, and surgical extent was not independently associated with persistent disease (OR 2.18, 95%CI 0.52–9.13, p = 0.285 and OR 0.85, 95%CI 0.18–4.01, p = 0.837, respectively). Independent predictors of persistent disease included higher preoperative CT levels (OR 1.02 per 100 pg/mL increase, 95%CI 1.00–1.04, p = 0.042) and presence of LN metastasis (OR 4.52, 95%CI 1.01–20.24, p = 0.049) (Table 6). Table 6Multivariable logistic regression analysis for predictors of persistent diseaseVariableAdjusted OR95% CIp-valueSurgical extent (reference: guideline-concordant)Less extensive2.180.52–9.130.285More extensive0.850.18–4.010.837Tumor size (per 1-cm increase)1.280.95–1.720.104Preoperative calcitonin (per 100 pg/mL increase)1.021.00–1.04 0.042 Lymph node metastasis4.521.01–20.24 0.049 Age (per year)0.980.93–1.030.382Male gender1.450.42–5.040.552Surgical extent categories were based on American Thyroid Association (ATA) 2015 guideline recommendationsThe model was adjusted for all variables listed in the tableOR Odds ratio, CI confidence intervalBold values indicate statistical significance (p < 0.05) Multivariable logistic regression analysis for predictors of persistent disease Surgical extent categories were based on American Thyroid Association (ATA) 2015 guideline recommendations The model was adjusted for all variables listed in the table OR Odds ratio, CI confidence interval Bold values indicate statistical significance (p < 0.05) Postoperative complications occurred in 28 patients (30.8%). Vocal cord paralysis developed in nine patients (9.9%), with transient hoarseness in four patients (4.4%) and persistent hoarseness requiring intervention in five patients (5.5%). All cases of hoarseness were associated with unilateral vocal cord paralysis confirmed by laryngoscopic examination. Hypoparathyroidism occurred in 26 patients (28.6%), of which seven (7.7%) experienced transient hypoparathyroidism that resolved within six months, while 19 patients (20.9%) developed persistent hypoparathyroidism. A comparative analysis was performed to evaluate how surgeries that were either in line with, less extensive than, or more extensive than the recommendations of the ATA, ESMO, and NCCN guidelines affected clinical outcomes in patients with MTC. No statistically significant differences were observed in complication rates, including overall postoperative complications, vocal cord paralysis, or hypoparathyroidism, regardless of surgical extent classification according to ATA, ESMO, or NCCN guidelines (all p > 0.05). Similarly, reoperation rates did not differ significantly based on guideline concordance (ATA: p = 0.584; ESMO: p = 0.398; NCCN: p = 0.533). Regarding persistent disease, a borderline association was observed with surgical concordance as per the ATA guidelines (p = 0.074); however, this finding was not statistically significant. No significant differences in persistent disease rates were identified according to ESMO (p = 0.206) or NCCN (p = 0.942) guidelines. No significant differences were found between biochemical incomplete disease response to initial treatment and surgical approach for any of the three guidelines (p > 0.05). Germline mutations in the RET proto-oncogene were identified in 13 patients (14.3%). Patients with hereditary MTC were diagnosed at a significantly younger age compared to those with sporadic MTC (43.13 ± 16.86 vs. 55.01 ± 11.69 years, p = 0.002). The median tumor diameter was also larger in the hereditary group (2.30 (1.00–6.00) cm) than in the sporadic group (1.20 (0.10–11.00) cm), (p = 0.005). Furthermore, bilateral MTC was markedly more frequent in hereditary cases (38.5%) compared to sporadic cases (7.7%), (p = 0.001). During follow-up, three patients died. Two deaths were unrelated to MTC, whereas one patient died from locally invasive MTC at month 11 after initial surgery; this patient had undergone guideline-concordant surgery according to all three guideline recommendations evaluated in this study. Discussion This study showed that using either less or more extensive surgical techniques, based on patients’ clinical and radiological features rather than strictly following guideline recommendations, did not negatively impact disease prognosis. Additionally, it was found that ultrasonographic and histopathological tumor sizes influenced CT levels at least as much as LN metastasis. The recommendations in the guidelines, particularly the ESMO guidelines [2], state that an increase in CT levels is directly related to the extent of the disease, and that the initial surgical treatment plan should be based on this assessment. However, although the CT level is related to LN metastasis, it is not the only clinical feature associated with it. Similar to the study presented, a study conducted by Park et al. found that serum CT levels were related to both the primary tumor size and the number of metastatic LNs [12]. Cohen et al. also stated that preoperative CT levels can predict tumor size and postoperative CT normalization [13]. However, current guidelines contain a significant gap: they provide no specific recommendations regarding tumor size assessment. This creates a clinical paradox where patients presenting with identical serum CT levels but different primary tumor sizes are considered to have similar disease extent. While serum CT levels do increase with tumor size, elevated CT levels alone are less reliable than tumor size itself for predicting the burden of lymph node metastasis. This limitation suggests that treatment planning based solely on CT levels may overlook crucial prognostic information that tumor size could provide. Beyond descriptive comparisons, multivariable analysis demonstrated that surgical extent—whether less or more extensive than guideline recommendations— did not independently predict persistent disease. Rather, preoperative CT levels and LN metastasis emerged as independent predictors, suggesting that tumor biology rather than surgical protocol adherence drives outcomes. This supports individualized surgical planning based on comprehensive preoperative assessment. In our study, preoperative CT level was only linked to ipsilateral LLN metastasis, whereas histopathologically, tumor size was related to metastatic LN in four compartments. One possible reason for this could be the relatively small number of patients with contralateral CLN and LLN metastasis. It can be concluded that tumor size should be considered along with CT levels before undergoing prophylactic LND. The cut-off value we determined to predict ipsilateral LLN metastasis was well above that recommended in the guidelines [2] and previous studies for prophylactic LLND [14]. Recent studies support higher calcitonin thresholds for prophylactic lateral neck dissection than current guideline recommendations, with values ranging from 240 to 693.9 pg/mL for lateral compartment involvement [15–17]. These discrepancies may partly reflect differences in assay generations and calibration across studies, as well as variations in statistical methodology, since guideline thresholds were largely derived to minimize missed nodal disease, whereas our ROC-based approach aimed to optimize overall discriminative performance [17]. In addition, as a tertiary referral center, our cohort may include a higher proportion of patients with advanced tumor burden, which could shift the statistically optimal threshold upward. In light of all these data and our own, it may be considered that guidelines for prophylactic LLND should move away from CT-based assessment alone or recommend higher CT thresholds. The AJCC staging distribution of the patients was similar to the literature, with stage I and stage IV dominance [9, 18]. When we consider the responses to initial treatment, the excellent response rate of 54.5% obtained in the present study varies in the literature. For example, Kwon et al. reported a rate of 70% in their own population [9], whereas Lindsey et al. reported a rate of 27% [19]. The structural incomplete response rate was 15.4% in this study, whereas the other two articles reported rates of 7% and 38%, respectively. The LN metastasis status in the TNM classification was also similar to the literature [9]. In our patient group, LN metastasis was most frequently seen in the ipsilateral central compartment. Ipsilateral LLN metastasis appeared more common than contralateral CLN metastasis. In addition, more than half of the patients with ipsilateral LLN metastasis did not have contralateral CLN metastasis. Machens et al. also described a similar pattern of LN metastasis in their research in 2010 [14]. Another study, published by Machens et al. in 2024, evaluated 640 unilateral MTC patients and revealed the same lymphatic spread pattern [20]. Another study, conducted in 2019, found that ipsilateral LLN metastasis was present in all patients with contralateral CLN metastasis. However, contralateral CLN metastasis was absent in patients without ipsilateral LLN metastasis [21]. In current clinical practice, contralateral CLND is prioritised over ipsilateral LLND in prophylactic surgeries [1–3]. However, it can be assumed with the available data that ipsilateral lateral spread may occur earlier than spread to the contralateral central compartment. Based on this, it can be assumed that performing ipsilateral CLND rather than bilateral CLND in patients without suspected metastatic LN in the lateral compartments may not negatively affect the disease prognosis and may also protect patients from complications such as iatrogenic hypoparathyroidism and recurrent laryngeal nerve injury. However, these assumptions need to be confirmed with more extensive data. Current treatment approaches recommend considering patient-based clinical and ultrasonographic findings when determining the surgical width, rather than relying solely on CT levels. Based on studies conducted with this perspective, a less extensive surgical approach is advised [4, 5]. In our study, preoperative ultrasonography failed to detect metastatic spread in about one-third of patients with metastatic LN. This suggests that relying mainly on radiological findings for treatment decisions may increase the likelihood of requiring repeat surgery. However, a second surgery does not seem to result in a worse prognosis compared to an unnecessarily extensive first surgery. In a study of 235 patients, Papachristos et al. demonstrated that local recurrence does not increase the risk of survival or distant recurrence, and that a second surgery is safe in terms of complications [6]. In a 2021 study, 44 out of 89 MTC patients with preoperative CT levels above 200 pg/ml underwent prophylactic LND, while 45 did not, and again, no significant difference was observed between the two groups in terms of recurrence or survival [5]. Additionally, there are factors that affect the risk of recurrence more than the initial surgical method. In an article published in 2024, Abou Azar et al. showed that recurrence is not related to the initial surgical method but to clinical, radiological, genetic and histopathological findings [22]. Of the surgical treatments provided to our patients, about one in four involved a less extensive approach than recommended by the ATA guidelines. In clinical practice, the choice that would most differ from ATA recommendations would be to perform prophylactic contralateral LND based on CT levels in cases with ipsilateral LLN metastasis. Contralateral LLN metastasis is the rarest form of cervical metastasis in both our study and the literature [14, 20] Even if metastatic lymph nodes in that compartment, which were undetectable by radiology, were not removed by surgery, it has been shown that long-term follow-up survival rates for patients with a biochemical incomplete response are very high [5, 11, 19]. Compared to the ESMO guideline recommendations, two-thirds of patients in this study underwent less extensive surgery. The ESMO guideline recommends ipsilateral LLND when the CT level is greater than 50 pg/mL, and bilateral lateral LND when the CT level exceeds 200 pg/mL [2]. Current studies have shown that the prevalence of LN metastasis is influenced not only by the CT level, but also by tumor size, age at diagnosis, spread to other LN regions, and the number of metastatic LNs [20, 21, 23]. Therefore, all these factors should be evaluated together, rather than relying solely on the CT level. Although more extensive surgery was performed in one-third of patients in this study compared to the NCCN guideline recommendations, no improvement in clinical outcomes was achieved. An additional aspect is that a more extensive surgical approach may increase morbidity. Therefore, expanding the dissection field should be individualized and supported by concordant high-risk features, rather than a single biomarker threshold alone. In parallel with this individualized approach, Gambardella et al. highlighted that readily available systemic inflammatory indices can complement risk stratification in thyroid malignancy work-up [24], while their report of calcitonin- and CEA-negative MTC underscores the limitation of relying on biomarkers in isolation [25]. Similarly, principles drawn from more aggressive thyroid cancer management support reserving treatment intensification for appropriately selected patients within a balanced risk–benefit framework [26]. Limitations include the retrospective design of the study. While we reported surgical complications including vocal cord paralysis and hypoparathyroidism, the retrospective nature of data collection may have led to underreporting of minor or transient complications. Additionally, we did not stratify complications by surgeon experience or surgical volume, factors that may influence complication rates independently of surgical extent. The relatively limited median follow-up duration may have reduced our ability to detect differences in long-term prognostic outcomes. In conclusion, when making a surgical decision for MTC, keeping the surgical extent less extensive does not negatively impact the clinical course. When including LN compartments for prophylactic surgery, it is important to note that ipsilateral LLN metastasis may occur before contralateral CLN metastasis. Furthermore, CT cut-off values higher than those recommended in guidelines can be used to predict the presence of ipsilateral LLN metastasis. Additionally, considering tumor size may also help when using CT levels in the treatment of MTC. These findings require validation through multicentre, prospective studies. Acknowledgements None. Abbreviations American Joint Committee on Cancer American Thyroid Association Confidence Interval Central Lymph Node Central Lymph Node Dissection Calcitonin European Society for Medical Oncology Lateral Lymph Node Lateral Lymph Node Dissection Lymph Node Lymph Node Dissection Medullary Thyroid Carcinoma National Comprehensive Cancer Network Odds Ratio REarranged during Transfection (proto-oncogene) Receiver Operating Characteristic Tumor, Node, Metastasis (staging system) Total Thyroidectomy Authors’ contributions EYK, BEO, and BC conceptualized the study. EYK, BEO, MSD, BS, HA, and CA performed data curation. EYK and BEO conducted the formal analysis. EYK, BEO, MSD, and CA performed the investigation. EYK, BEO, and BC developed the methodology. EYK, BEO, BS, and HA administered the project. OT, RE, and BC provided resources. EYK performed software operations. OT, RE, and BC supervised the study. BEO and BC validated the results. EYK and BEO performed visualization. EYK wrote the original draft. BEO and BC reviewed and edited the manuscript. All authors read and approved the final manuscript. Funding This research received no specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Data availability Data are available from the corresponding author upon request. Declarations Ethics approval and consent to participate Ankara Bilkent City Hospital ethical committee approval was obtained on 11.06.2025 (TABED-2-25-1300) in accordance with the ethical standards of the Declaration of Helsinki. Informed consent was obtained from all participants. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement Data are available from the corresponding author upon request.