Nuclear medicine molecular imaging techniques [Positron emission tomography (PET)/Single-photon emission computed tomography (SPECT)] play a pivotal role in precision oncology diagnostics and therapeutics by leveraging high sensitivity and functional imaging capabilities, with their efficacy critically dependent on the development of targeted molecular probes. Ephrin type-A receptor 2 (EphA2), a key member of the receptor tyrosine kinase family, is aberrantly overexpressed in multiple malignancies and has emerged as a promising therapeutic target. Through years of development, EphA2-targeting drugs primarily encompass three categories: antibody-based drugs, small-molecule drugs, and bicyclic peptide drugs. These agents exhibit distinct pharmacological profiles and demonstrate excellent diagnostic or therapeutic potential across diverse models. Encouraged by positive preclinical results, several agents have advanced to Phase Ⅰ clinical trials, showing preliminary efficacy while confronting challenges such as nonspecific accumulation and tumor heterogeneity. This review systematically summarized recent advances in EphA2-targeted nuclear molecular probes, analyzing optimization strategies and future research priorities for EphA2-targeted investigations.

Radionuclide imaging targeting fibroblast activation protein inhibitor (FAPI) has shown notable advantages in the early diagnosis and staging of various tumors. The continuous development of novel FAPI-based radiotracers has significantly enhanced imaging performance and propelled the advancement of FAPI-targeted radionuclide therapy (TRT). Preliminary clinical trials have confirmed the efficacy of various FAPI-TRT agents, demonstrating not only effective disease control but also synergistic benefits when combined with immunotherapy and chemotherapy. These advantages have contributed to the establishment of a closed-loop “diagnosis-treatment-reassessment” theranostic pathway. However, several challenges remain, including non-specific uptake in inflammatory and fibrotic lesions, rapid systemic clearance, and unintended radiation exposure to normal tissues. This review systematically summarized the latest applications and research progress of FAPI in pan-cancer theranostics, provided an in-depth discussion of existing challenges, and seeks to facilitate the full potential of FAPI, thereby expediting the establishment of a new paradigm for integrated radionuclide-based theranostics.

Trophoblast cell surface antigen 2 (TROP2) is a single-pass transmembrane glycoprotein encoded by the tumor-associated calcium signal transducer 2 (TACSTD2) gene. It is overexpressed in various malignant tumor cells and exhibits significant tissue specificity, with almost no expression in normal tissues. TROP2 is considered a potential ideal target for cancer diagnosis and treatment, particularly in antibody-drug conjugate (ADC) therapies. The efficacy of TROP2-targeted therapies is closely related to TROP2 expression levels. Detecting its expression levels can effectively screen potential patients who will benefit from treatment and predict therapeutic responses accurately, helping to improve efficacy, reduce toxicity, and lower the economic burden on patients. A key challenge that needs to be addressed is the systemic, real-time, dynamic, and visualized quantitative monitoring of the spatiotemporal heterogeneity of TROP2 expression. While immunohistochemistry staining is the gold standard for detecting its expression, it has limitations such as being localized, invasive, and difficult for repeated or multi-site biopsies. Non-invasive nuclear medicine molecular imaging based on radiopharmaceuticals provides the potential to overcome these limitations, offering valuable guidance for subsequent treatment strategies. Conventional radiopharmaceutical molecular probes targeting TROP2 include radiolabeled monoclonal antibodies, nanobodies, and nucleic acid aptamers, making the visualized and quantitative monitoring of TROP2 dynamic expression possible. This article systematically summarized the research progress of TROP2-targeted molecular imaging for tumor diagnosis and treatment, while also discussing current challenges and innovative approaches to overcome technical limitations and accelerate clinical implementation.

Objective: To develop a small-molecule-based theragnostic pair of vascular endothelial growth factor receptor (VEGFR)-targeted radionuclides, explore its diagnostic and therapeutic value in triple-negative breast cancer (TNBC), and provide a new therapeutic strategy for “immunogenically cold” TNBC. Methods: Based on the high-affinity VEGFR-targeting agent diZD, the diagnostic radionuclide ⁶⁴Cu and therapeutic radionuclide ¹⁷⁷Lu were conjugated via the DOTA chelator to prepare ⁶⁴Cu-DOTA-diZD [positron emission tomography (PET) imaging agent] and ¹⁷⁷Lu-DOTA-diZD (targeted radionuclide therapy agent). Cell-free binding assay and cell binding assay were used to verify the VEGFR binding affinity and specificity of the two agents. In the 4T1 TNBC tumor-bearing mouse model, small-animal PET/computed tomography (CT) was used to observe the tumor targeting and biodistribution of ⁶⁴Cu-DOTA-diZD. Tumor-bearing mice were divided into control group (saline), “cold” DOTA-diZD group, anti-programmed death-1 (PD-1) antibody group, and ¹⁷⁷Lu-DOTA-diZD group. The tumor volume change and median survival time of each group were compared, and the therapeutic mechanism was analyzed by immunohistochemistry and flow cytometry. Results: The Kd value of ⁶⁴Cu-DOTA-diZD and ¹⁷⁷Lu-DOTA-diZD for VEGFR2 was 0.45 nmol/L and (0.54±0.05) nmol/L, respectively. Specific blocking experiments confirmed that both could specifically bind to VEGFR. PET/CT showed that ⁶⁴Cu-DOTA-diZD gradually accumulated in 4T1 tumors over time, reaching a peak tumor uptake (2.25±0.09) at 48 h after injection, and could identify lung metastases. In the treatment experiment, the median survival time of mice in the ¹⁷⁷Lu-DOTA-diZD group (28 d) was significantly longer than that in the control group (18 d), “cold” DOTA-diZD group (20 d), and anti-PD-1 antibody group (16 d). Immunohistochemistry showed that the expression of the tumor proliferation marker Ki-67 in this group was significantly reduced, and flow cytometry confirmed that it could double the number of CD4⁺ and CD8⁺ T cells in the tumor. Conclusion: The ⁶⁴Cu/¹⁷⁷Lu-DOTA-diZD theranostic pair has good VEGFR targeting. It can not only achieve accurate diagnosis of TNBC and identification of metastases through ⁶⁴Cu-DOTA-diZD, but also effectively inhibit tumor growth and prolong survival through ¹⁷⁷Lu-DOTA-diZD, which is a potential strategy for the treatment of TNBC.

Objective: To construct a Claudin 18.2 (CLDN18.2)-targeted positron emission tomography (PET) probe, ⁶⁸Ga-THP-ACN376, using a site-specific labeling technique, and to evaluate its in vitro and in vivo biological performance. Methods: The chelator THP-Mal was site-specifically conjugated to the CLDN18.2-targeting nanobody ACN376 to synthesize the precursor compound THP-ACN376, which was subsequently radiolabeled with ⁶⁸Ga. The labeling efficiency, radiochemical purity, and in vitro stability of the probe were determined by radio thin layer chromatography (Radio-TLC). Micro-PET/computed tomography (CT) imaging was performed in BALB/c nude mice to assess its in vivo pharmacokinetic properties and targeting capability. Results: ⁶⁸Ga-THP-ACN376 exhibited high labeling efficiency (>99%), high radiochemical purity (>99%), and a high specific activity of 16 GBq/μmol. It remained stable in saline, 5% human serum albumin (HSA), and phosphate buffered saline (PBS). Micro-PET/CT results demonstrated that the metabolism of the probe in mice followed the typical profile of nanobodies, with tracer uptake in the stomach gradually increasing from 0.5 to 2 h. The maximum standardized uptake value SUV_max in the stomach at 2 h was 1.35±0.07. Conclusion: This study successfully constructed a CLDN18.2-targeted PET probe, ⁶⁸Ga-THP-ACN376, using a site-specific labeling method. The probe demonstrates high labeling efficiency, remarkable stability, and significant targeting ability, showing great potential for detecting CLDN18.2 protein expression levels.

Objective: ¹⁷⁷Lu-DOTA-ibandronate (IBA), a bone-targeting radiopharmaceutical, has demonstrated favorable therapeutic responses in bone metastases from various tumors. This study aimed to evaluate the efficacy and safety of ¹⁷⁷Lu-DOTA-IBA in the treatment of bone metastases from lung cancer. Methods: This prospective study included 45 patients with lung cancer and bone metastasis who were treated at the Affiliated Hospital of Southwest Medical University from November 2021 to May 2025 [24 males and 21 females, with an age of (55.0±10.5) years]. All patients underwent ⁶⁸Ga-DOTA-IBA PET/CT imaging before treatment, and those meeting the inclusion and exclusion criteria received ¹⁷⁷Lu-DOTA-IBA therapy. Functional status and pain intensity were assessed at baseline and at 3 days, 1 week, 2 weeks, 4 weeks, and 8 weeks post-treatment using the Eastern Cooperative Oncology Group (ECOG) performance status and the numerical rating scale (NRS) for pain. Hematologic and biochemical parameters (including complete blood count, liver function, and renal function) were evaluated at baseline and repeatedly within 8 weeks after treatment according to the Common Terminology Criteria for Adverse Events (CTCAE) version 5.0. At 8 weeks post-treatment, ⁶⁸Ga-DOTA-IBA PET/CT was repeated to assess therapeutic response. Data were analyzed using SPSS version 26.0, and differences before and after treatment were compared with the Wilcoxon signed-rank test. Results: Sixteen patients received one cycle of treatment, nine received two cycles, five received three cycles, five received four cycles, six received five cycles, one received six cycles, and three received seven cycles. The median administered activity of ¹⁷⁷Lu-DOTA-IBA per cycle was 1 110 MBq (range: 370-1 850 MBq). Compared with baseline, pain scores and ECOG performance status at 8 weeks after the final treatment showed an overall pain relief rate of 82.2% and an improvement rate in performance status of 64.4%. ⁶⁸Ga-DOTA-IBA PET/CT revealed a partial metabolic response in 37.8% of patients. The overall incidence of hematologic adverse events was 28.9%, while renal toxicity occurred in 6.7% of patients; no hepatic toxicity was observed. Conclusion: ¹⁷⁷Lu-DOTA-IBA, as a bone-targeting radiopharmaceutical reagent, demonstrated promising therapeutic efficacy and a favorable safety profile in patients with lung cancer bone metastases.

Objective: To investigate the imaging characteristics of ¹⁸F-FDG positron emission tomography (PET)/ computed tomography (CT) in prostate cancer patients at different treatment stages and their predictive value for prognosis. Methods: Prostate cancer patients who underwent ¹⁸F-FDG PET/CT at Peking Union Medical College Hospital from January 2016 to December 2022 were retrospectively analyzed. According to their treatment status, patients were divided into the pre-treatment group, the hormone-sensitive prostate cancer (HSPC) group, and the castration-resistant prostate cancer (CRPC) group. Clinical data and PET/CT semi-quantitative parameters, including maximum standard uptake value (SUV_max), metabolic tumor volume (MTV), total lesion glycolysis (TLG), and target-to-background ratio (TBR) were collected. Cut-off values for prognostic factors were determined using receiver operating characteristic (ROC) curve analysis. Intergroup comparisons were performed using the Mann-Whitney U test, and survival analysis was conducted using the Kaplan-Meier method with log-rank testing. All statistical analyses were performed using SPSS Statistics (version 22.0) and R, with a P value <0.05 considered statistically significant. Results: A total of 117 patients with prostate cancer were included, among which 78 cases were in the initial treatment group, 18 cases in the HSPC group, and 21 cases in the CRPC group. ¹⁸F-FDG PET/CT revealed significant metabolic heterogeneity among prostate cancer patients across different treatment stages and demonstrated substantial prognostic value. The vast majority (92.3%) of pre-treatment patients exhibited FDG-avid primary lesions. Following androgen deprivation therapy (ADT), the positivity rate significantly decreased to 5.6% in the HSPC group, although 23.8% of CRPC patients still showed primary lesion uptake. The SUV_max of primary lesions was significantly higher in patients with metastases than in those without (P<0.05). Within the treated cohorts, the FDG-positivity rate was significantly higher in the CRPC group than in the HSPC group (95.2% vs 11.1%, P<0.001). Patients receiving ADT combined with other therapies (radical surgery, chemotherapy, radiotherapy et al.) had a significantly higher FDG-positivity rate than those receiving ADT alone (85.7% vs 40%, P=0.006). At the lesion level, no significant differences in SUV_max or TBR were observed between the pre-treatment and CRPC groups, though both were significantly higher than those in the HSPC group (P<0.05). MTV and TLG were significantly higher in the pre-treatment group than in the CRPC group (P<0.01). Survival analysis indicated the poorest prognosis in the CRPC group, with a median overall survival (OS) of only 35 months, whereas the median OS was not reached in the pre-treatment and HSPC groups within 100 months of follow-up. The survival rate was significantly lower in the CRPC group, with 3-year survival rates of 86%, 100%, and 40%, and 5-year survival rates of 70%, 85.7%, and 0 for the pre-treatment, HSPC, and CRPC groups, respectively. Multivariate analysis identified prostate-specific antigen (PSA) ≤54.4 ng/mL, number of distant metastases >5, total number of lesions ≤6, MTV ≤84.8 mL, and TLG ≤134.9 g as significant independent prognostic factors in pre-treatment patients (all P<0.05). The presence of FDG-avid lesions after ADT was a significant predictor of poor prognosis, associated with markedly shortened median OS (P=0.001 9). Conclusion: ¹⁸F-FDG PET/CT parameters can be used to assess tumor burden and metabolic status in prostate cancer patients across different stages, serving as effective indicators for prognostic prediction.

Objective: To compare and analyze the diagnostic performance of ⁶⁸Ga-TBM-001 positron emission tomography (PET)/computed tomography (CT) and ^99mTc-MDP bone scintigraphy (BS) for bone metastases of prostate cancer. Methods: Patients with prostate cancer who underwent ⁶⁸Ga-TBM-001 PET/CT and ^99mTc-MDP BS from September 2021 to October 2023 at the Affiliated Hospital of Southwest Medical University were retrospectively included[Ethics No. KY2022114]. The sensitivity, specificity, accuracy, positive predictive value (PPV), and negative predictive value (NPV) of the two imaging were compared and analyzed. Results: A total of 43 patients with suspected bone metastases of prostate cancer were included, 27 patients were positive for bone metastasis, and 16 patients were negative for bone metastasis. A total of 389 suspicious lesions were included, 338 were bone metastases and 51 were non-bone metastases. ⁶⁸Ga-TBM-001 PET/CT accurately diagnosed 42 cases and 1 false positive,^99mTC-MDP BS accurately diagnosed 35 cases, 6 false positives, and 2 false negatives. ⁶⁸Ga-TBM-001 PET/CT accurately diagnosed 366 lesions, 5 false-positive lesions, and 18 false-negative lesions, ^99mTc-MDP BS accurately judged 324 lesions, 15 false-positive lesions, and 50 false-negative lesions. At the case level, the sensitivity, specificity, accuracy, PPV, and NPV of ⁶⁸Ga-TBM-001 PET/CT and ^99mTc-MDP bone scintigraphy were 100.00% vs 92.59%, 93.75% vs 62.50%, 97.67% vs 81.40%, 96.43% vs 80.65%, 100.00% vs 83.33%, and the AUC values of the two methods were 0.969 (95% CI 0.865-0.998) vs 0.775 (95% CI 0.623-0.888), except for sensitivity and NPV, the differences among all other indicators were statistically significant. At the lesion level, the sensitivity, specificity, accuracy, PPV, and NPV of ⁶⁸Ga-TBM-001 PET/CT versus ^99mTc-MDP bone scintigraphy were 94.67% vs 85.21%, 90.20% vs 70.59%, 94.09% vs 87.15%, 98.46% vs 95.05%, 71.88% vs 41.86%, and the AUC values were 0.924 (95% CI 0.893-0.949) vs 0.779 (95% CI 0.734-0.819), with differences that were statistically significant. Conclusion: ⁶⁸Ga-TBM-001 PET/CT demonstrates superior sensitivity, specificity, accuracy, PPV, NPV, and AUC values compared to ^99mTc-MDP BS in the diagnosis of bone metastases in prostate cancer, suggesting that ⁶⁸Ga-TBM-001 PET/CT may have higher clinical application value.

Objective: To assess the effectiveness of dual-layer detector spectral computed tomography (DLCT) in predicting the expression of programmed death ligand-1 (PD-L1) in patients with esophageal squamous cell carcinoma (ESCC) prior to surgical intervention. Methods: Patients with histopathologically confirmed esophageal squamous cell carcinoma (ESCC) at Liaoning Cancer Hospital between February 2022 and December 2023 were retrospectively collected. The combined positive score (CPS) was employed to evaluate PD-L1 expression in ESCC tumor cells, with scores ranging from 0-100. Patients with a CPS of 10 or greater were categorized into the high PD-L1 expression group, while those with a CPS of less than 10 were placed in the low PD-L1 expression group. Various quantitative spectral parameters, including Conventional CT values and CT values from 40 keV virtual monoenergetic images during both arterial and venous phases, iodine density (ID), normalized iodine density (NID), and effective atomic number (Z_eff) was record. Independent samples t-tests were utilized to compare the differences in parameters between the high and low expression groups. Additionally, receiver operating characteristic (ROC) curves were generated to assess the predictive capability of each parameter concerning CPS score expression, with comparisons of the area under the curve (AUC) conducted using the DeLong test. Results: Among 140 ESCC patients, 78 were classified as PD-L1 high-expression (CPS≥10) and 62 as low-expression (CPS<10). The PD-L1 high expression group showed significantly higher values in conventional CT, 40 keV CT, ID, NID, and Z_eff during both the arterial and venous phases compared to the PD-L1 low expression group, with statistically significant differences (P<0.05). The venous phase NID had the best predictive effect for ESCC PD-L1 expression, with an AUC of 0.954 (95% CI 0.916-0.992) and a cutoff value of 0.298, yielding a sensitivity of 94.8% and a specificity of 93.5%. The AUC of venous phase NID was statistically significantly different from the venous phase ID, conventional CT values, and 40 keV CT values (AUCs of 0.848, 0.779, and 0.830; Z=-3.504, 4.520, 3.828; all P<0.001), while the difference compared to the Z_eff in the venous phase (AUC=0.853, Z=3.401, P=0.001) was not statistically significant. The AUC of venous phase NID was also statistically significantly different from the arterial phase ID, Z_eff, conventional CT values, and 40 keV CT values (AUCs of 0.798, 0.784, 0.748, and 0.782; Z=-3.575, -3.826, -4.427, -3.843; all P<0.001), but the difference compared to arterial phase NID (AUC=0.833, Z=-2.938, P=0.003) was not statistically significant. Conclusion: The venous phase NID got form DLCT demonstrates a high predictive efficacy for the expression of PD-L1 in ESCC.

Objective: To evaluate the value of combining multi-region radiomics features from enhanced computed tomography (CT) with clinical risk factors for predicting preoperative lymphovascular invasion (LVI) in colon cancer. Methods: A retrospective analysis was conducted on pathologically confirmed colon cancer patients at Puning People’s Hospital from January 2018 to October 2024. Patients were randomly divided into a training set and a validation set. Radiomics features were extracted from venous-phase CT images by delineating the tumor core and peritumoral 3 mm region. Features were selected using t-tests, Spearman correlation analysis, and the least absolute shrinkage and selection operator (LASSO) regression to construct three radiomics models: tumor core, peritumoral, and combined dual-region models. Independent clinical risk factors were identified using logistic regression and integrated with the optimal radiomics model to develop a clinical-radiomics combined model. Model performance was evaluated using receiver operating characteristic (ROC) curve, with the area under curve (AUC), sensitivity, specificity, F1-score, and accuracy as metrics. Results: A total of 136 patients were included and randomly allocated to the training set (95 cases) and the validation set (41 cases) at a ratio of 7∶3. The dual-region radiomics model (Model 3) demonstrated the best performance, with a validation AUC of 0.863 (95% CI 0.720-1.000). The clinical-radiomics combined model further improved predictive performance, achieving a validation AUC of 0.873 (95% CI 0.740-1.000), with sensitivity and specificity of 0.808 and 0.800, respectively. Logistic regression identified gender (OR=2.446, 95% CI 1.056-5.665, P=0.037) and clinical stage (OR=3.838, 95% CI 1.045-14.096, P=0.043) as independent risk factors for LVI. Conclusion: The prediction model combining multi-region radiomics features from enhanced CT with clinical risk factors effectively assesses preoperative LVI status in colon cancer, providing valuable reference for individualized treatment decision-making.

Objective: To analyze the ultrasonic characteristics of jaw bone lesions and compare the diagnostic accuracy of high-frequency ultrasound with cone beam computed tomography (CBCT), and to explore the diagnostic value of ultrasound in jaw bone lesions. Methods: A retrospective analysis was conducted on patients diagnosed with jaw lesions at Jiangsu Provincial Hospital of Traditional Chinese Medicine from May 2019 to September 2022. According to the histopathological results of cysts and tumors/tumor-like lesions, the ultrasonic characteristics of the two groups of lesions were analyzed. The correlation between high-frequency ultrasound, CBCT, and histopathology was analyzed, and the diagnostic efficiency of high-frequency ultrasound and CBCT for jaw bone lesions was compared using receiver operating characteristic (ROC) curves. Results: Among the 30 patients, 12 cases were cysts, and 18 cases were tumors/tumor-like lesions. High-frequency ultrasound was correlated with the histopathological results of jaw bone lesions (P<0.05，r=0.747), and the correlation was positive. The comparison of the ROC curve areas of the two examination methods showed statistically significant differences (P<0.05), and the sensitivity and specificity of high-frequency ultrasound for the diagnosis of jaw bone lesions were 83.3% and 58.3%, respectively. Conclusion: High-frequency ultrasound has higher diagnostic efficiency than CBCT in evaluating soft tissue components within mandibular lesions, and can be used as a supplement to CBCT, providing effective reference for preoperative auxiliary diagnosis, monitoring treatment, or disease progression of clinical jaw bone lesions.

Objective: To explore the diagnostic value of ultra-microangiography (UMA) for the classification of breast tumors categorized as Breast Imaging Reporting and Data System (BI-RADS) category 4. Methods: Data of female patients with breast tumors who underwent conventional two-dimensional ultrasound and UMA examinations at the First Hospital of Shanxi Medical University from January 2023 to February 2024 were retrospectively reviewed and collected. The differences in UMA blood flow pixel ratios and blood flow characteristics between benign and malignant breast tumors were analyzed. Indicators with statistical significance and high positive predictive values were assigned scores. Based on these scores, receiver operating characteristic (ROC) curves were generated, and the optimal cutoff values were selected to adjust the BI-RADS grading for benign and malignant breast tumors. ROC curves before and after adjustment were plotted to investigate the differential diagnostic value of UMA for BI-RADS category 4 breast nodules. Results: A total of 101 breast patients were included. There were no statistically significant differences in the blood flow pixel ratios of cUMA, pUMA, and sUMA between benign and malignant tumors (Z-values were -0.194, -1.202, and -1.117, respectively; P-values were 0.846, 0.229, and 0.264, respectively). Among the UMA blood flow characteristics, the presence or absence of penetrating blood flow (χ²=8.394, P=0.004), whether the marginal blood flow course was distorted (χ²=7.317, P=0.007), the marginal blood flow distribution (χ²=55.733, P<0.001), whether the central blood flow course was distorted (χ²=4.178, P=0.041), and the presence or absence of central blood flow defects (χ²=6.051, P=0.014) showed statistically significant differences between benign and malignant breast tumors. One point was assigned when the tumor had penetrating blood flow, distorted marginal blood flow course, marginal blood flow distribution characterized by peripheral penetrating blood flow or radial blood flow, distorted central blood flow course, or central blood flow defects. The area under the curve (AUC) was 0.871, with a 95% CI of 0.799 to 0.943. After scoring, the AUC for BI-RADS category 4 breast tumors increased from 0.741 to 0.825. Conclusion: UMA can improve the diagnostic accuracy of benign and malignant BI-RADS 4 breast tumors and serve as an effective blood flow imaging modality for the diagnosis of breast cancer.

Objective: To investigate the predictive value of combining preoperative ultrasonographic imaging characteristics with BRAF V600E mutation testing for central lymph node metastasis (CLNM) risk in clinically node-negative (cN0) papillary thyroid microcarcinoma (PTMC) patients, aiming to guide individualized surgical decision-making. Methods: Clinical data from cN0 PTMC patients preoperatively diagnosed and undergoing thyroidectomy with central lymph node dissection at Shanxi Cancer Hospital from March 2022 to February 2024 were retrospectively analyzed. Preoperative color Doppler ultrasound was performed to evaluate tumor characteristics. Thyroid function indicators were detected, and BRAF V600E mutation status was detected via fine-needle aspiration biopsy specimens. Based on postoperative pathology, patients were categorized into non-metastatic and metastatic groups. Univariate and multivariate logistic regression analyses were performed to identify independent predictors of CLNM, followed by the construction of a combined prediction model. Receiver operating characteristic (ROC) curves were used to evaluate predictive performance, and DeLong’s test was applied to compare diagnostic efficacy between the combined model and individual predictors. Results: Univariate analysis revealed significant differences between the two groups in age, multifocality, tumor size >6 mm, capsular invasion, and BRAF V600E mutation status (P<0.05). Multivariate logistic regression identified age ≤45 years, multifocality, tumor size >6 mm, capsular invasion, and BRAF V600E mutation as independent predictors of CLNM. The combined prediction model achieved an area under the ROC curve (AUC) of 0.925 (95% CI 0.878-0.972), significantly higher than ultrasound features alone (AUC=0.681, 95% CI 0.588-0.774) or BRAF V600E mutation testing alone (AUC=0.618, 95% CI: 0.520-0.716) (P<0.05). Conclusion: The combination of preoperative ultrasonographic features and BRAF V600E mutation testing significantly improves the predictive accuracy for CLNM risk in cN0 PTMC patients. This integrated approach facilitates the identification of high-risk candidates for prophylactic central lymph node dissection, thereby optimizing surgical strategies and minimizing overtreatment risks.