Objectives: To evaluate the impact of surgical approach on cervical hematoma (CH) development and to determine the optimal timing for reoperation in patients developing CH. Study Design: We retrospectively analyzed consecutive patients who underwent thyroidectomy between January 2016 and March 2024. Twenty-seven cases developing CH were identified and matched with non-CH cases at a ratio of 1:4. Surgical-related factors were compared between the CH and the non-CH groups. Reoperation timing was categorized as early (within 120 minutes of symptom onset) and late, and clinical outcomes were assessed. Setting： CH is rare yet critical complication following thyroidectomy. Reducing the incidence and appropriate management remain clinical challenges. Methods: This study analyzed the data using methods including propensity score matching, t - test, chi - square test, and Mann - Whitney U test. Results: The incidence of CH was 0.33% (27/8102) in our center. No significant differences were observed in the extent of surgery or drainage placement between the CH and non-CH groups (p > 0.05). Postoperative cough was recorded in five patients developing CH, with four on the day of surgery and one on the second postoperative day. Among the 25 patients who underwent reoperation, early reoperation within 120 minutes of symptom onset was associated with a reduced need for intensive care unit (ICU) admission, without prolonging hospital stay. Conclusion: Perioperative airway management may play a more significant role in preventing CH than adjusting surgical plans. The optimal timing for reoperation falls within a critical 120-minute window after symptom onset.

Background:In recent years, the global incidence of thyroid cancer has been increasing. Objective:This study aimed to examine the association between Hashimoto thyroiditis (HT) and papillary thyroid cancer (PTC) progression under active surveillance (AS). Methods:Our retrospective study was conducted at Peking University Third Hospital and included 203 patients with PTC who underwent AS for ≥6 months before surgery. Patients were first categorized into 2 groups: the HT group (n=90) and the non-HT group (n=113). Cox proportional hazards models were then used to evaluate the association between HT and PTC progression during AS, adjusting for age, sex, baseline tumor size, BMI, pregnancy status, number of tumor foci, and thyroid-stimulating hormone level. Subgroup analyses stratified by the 6 covariates mentioned above were also applied to explore the potential effect modification. Results:No significant difference was observed between the HT and non-HT groups in PTC progression-free survival (hazard ratio [HR] 1.11, 95% CI 0.61-1.99; P=.74), tumor enlargement-free survival (HR 1.02, 95% CI 0.56-1.86; P=.95), or lymph node metastasis-free survival (HR 1.76, 95% CI 0.31-10.12; P=.52). Subgroup analyses revealed a potential interaction between HT and BMI. Among patients who were overweight or obese (BMI >24 kg/m²), HT was significantly associated with an increased risk of disease progression (HR 6.32, 95% CI 1.84-21.69; P=.003), while among patients with BMI ≤24 kg/m2, no association between HT and progression risk was observed (P=.01). Conclusions:We found no evidence of association between HT and PTC progression during AS. However, the relationship between HT and PTC progression may be modified by overweight or obesity status.

Segmentation of the pancreatic computed tomography (CT) image in the Internet of Medical Things (IoMT) environment faces dual challenges of feature robustness and accurate recognition of small organs. As a clinically critical but anatomically challenging organ, the pancreas exhibits small volume, high interpatient variability, and low contrast with surrounding tissues, making automatic pancreas segmentation a representative and difficult task in abdominal CT analysis. Existing automatic, machine-centric segmentation methods often perform unsatisfactorily across different medical institutions due to variations in imaging devices, changes in scanning protocols, as well as the irregular shape and blurred boundaries of the pancreas. To address this problem, this article proposes a SimCLR-based Contrastive Swin-UNet (ContSwinU) model, which integrates contrastive learning with the Swin Transformer architecture to achieve feature robustness learning and high-precision pancreas segmentation. Specifically, ContSwinU leverages SimCLR to learn robust feature representations, thereby enhancing the model's generalization capability across diverse scenarios. Additionally, by incorporating the hierarchical window attention mechanism of the Swin Transformer, the model effectively balances local texture and global structural information, improving segmentation accuracy for the pancreas as a small organ. Experimental results on a public pancreas CT dataset demonstrate that ContSwinU achieves an IoU of 0.8041, a Dice coefficient of 0.8872, and a recall of 0.9014, significantly outperforming mainstream baseline methods. These results indicate that the proposed framework is well suited for challenging small-organ segmentation tasks and has the potential to be extended to other organs and multicenter IoMT scenarios. This study provides an effective solution for pancreas segmentation in IoMT environments and has substantial clinical application value.

Background:Chronic inflammation is increasingly recognized as a fundamental driver of pancreatic ductal adenocarcinoma (PDAC) initiation and progression. Although numerous bioinformatics studies have characterized genetic alterations in PDAC, the key inflammatory regulators that bridge tumor cells and the immunosuppressive stroma remain unclear. Methods:We conducted an integrative multi-omics analysis of TCGA, GEO, and ArrayExpress datasets to define inflammation-associated molecular signatures in PDAC. Differentially expressed genes were analyzed through pathway enrichment, protein-protein interaction modeling, and immune infiltration profiling. Immunotherapeutic relevance was assessed using the IMvigor210 cohort and TIDE algorithm, while drug repurposing candidates were identified via molecular docking. Single-cell RNA sequencing and in vitro functional assays were employed to validate gene expression patterns and mechanistic functions within the PDAC microenvironment. Results:Our multi-cohort analysis revealed a robust inflammation-associated gene network in PDAC, with SERPINE1 emerging as a consistent central hub. Elevated SERPINE1 expression was tightly linked to a profoundly immunosuppressive tumor microenvironment and predicted diminished responsiveness to immunotherapy across datasets. Structure-based molecular docking further identified Lenvatinib and Dasatinib as previously unappreciated candidate inhibitors of SERPINE1, suggesting actionable therapeutic opportunities. Single-cell transcriptomic profiling resolved nine major cellular compartments and pinpointed fibroblasts as the principal stromal niche orchestrating SERPINE1-driven crosstalk between inflammation and immune evasion, a cellular origin that has not been systematically defined before. Translational analyses demonstrated consistently elevated SERPINE1 in tumor tissues, and functional validation using CRISPR-mediated knockout in PDAC cell lines significantly impaired proliferation and migration while inducing robust apoptosis, thereby establishing SERPINE1 as a previously underappreciated but essential driver of PDAC aggressiveness. Conclusions:This integrative multi-omics and single-cell analysis establishes SERPINE1 as a central orchestrator of inflammation-driven stromal remodeling and immune evasion in PDAC. Its strong prognostic power, combined with newly revealed druggability, positions SERPINE1 as a tractable therapeutic axis for precision immunotherapy and rational drug repurposing. These findings provide a mechanistically grounded and clinically actionable entry point into targeting the inflammatory tumor microenvironment of pancreatic cancer.

Pancreatic cancer is one of the most aggressive malignant solid tumors, and achieving early screening and diagnosis is the key to improving patient survival rates. Although deep learning has made significant progress in medical image analysis, the classification of pancreatic cancer computed tomography (CT) images remains highly challenging due to subtle interlesion differences and overlapping category distributions. Moreover, existing methods rely heavily on large amounts of high-quality annotations, which can lead to overfitting and hinder the effective exploitation of structural and semantic features within images. To address these challenges, we propose a contrastive Swin transformer with masked autoencoder (CSTMA) for pancreatic cancer CT image classification in the Internet of Medical Things (IoMT) environment. CSTMA leverages contrastive learning to enhance feature discriminability, while its multitask self-supervised architecture based on masked autoencoder (MAE) guides the model to learn both structural and semantic representations. We conduct comprehensive experiments on pancreatic cancer CT image classification tasks, and the results demonstrate that the proposed CSTMA model achieves superior performance across multiple evaluation metrics.

OBJECTIVE:The malignancy risk of follicular thyroid neoplasms (FTN) is variable, yet most studies rely on static, single-time-point assessments without monitoring dynamic changes. METHODS:We included patients who received two or more preoperative ultrasound examinations and were diagnosed with follicular thyroid carcinoma (FTC) or follicular thyroid adenoma (FTA) after surgery. We first identified 11 key predictors for the malignancy risk of FTN, comprising five static predictors (age, sex, body mass index, baseline mean tumor diameter, and surveillance duration) and six dynamic predictors [average thyroid-stimulating hormone (TSH) level, TSH instability, weighted rate and instability of mean diameter, and changes in tumor margin and calcifications over time]. Then, we applied logistic regression to evaluate the association between each dynamic predictor and the malignancy risk of FTN, adjusting for the static predictors as potential confounders. RESULTS:The results showed that a higher average TSH level was associated with a decreased risk of malignancy in FTN [odds ratio (OR): 0.67; 95% confidence interval (CI): 0.49, 0.92]. However, a shift in margin from circumscribed to irregular or maintaining an irregular margin was associated with a higher risk of malignancy [OR (95% CI): 4.18 (1.49, 11.71) and 5.33 (2.51, 11.35)]. We observed no association between the rate or instability of the mean tumor diameter and the malignancy risk of FTN. CONCLUSION:Changes in tumor margins (to irregular or lobulated) and lower average TSH levels were key predictors of a higher malignancy risk, while the rate or instability of tumor growth might play a minor role.

BACKGROUND In pancreatic neuroendocrine neoplasms (pNENs) with liver metastasis, marked upregulation of T-LAK cell-originated protein kinase (TOPK) is associated with poor prognosis. AIM To elucidate the role of TOPK in pNEN progression and metastasis and to explore the underlying mechanisms. The therapeutic potential of the TOPK inhibitor HI-TOPK-032 was further investigated to inform targeted treatment strategies. METHODS TOPK expression was assessed in tissue samples from 23 patients with pNENs, with and without liver metastasis. TOPK was knocked down in the BON-1 cell line to examine its effects on cell proliferation, epithelial-mesenchymal transition, and activation of the mitogen-activated protein kinase (MAPK) axis. RNA-seq analyzed gene expression changes following TOPK knockdown. The effects of HI-TOPK-032 on pNENs cell proliferation, migration, and invasiveness were also evaluated. RESULTS TOPK was significantly upregulated in pNENs with liver metastasis and correlated with poorer overall survival. TOPK knockdown in BON-1 cells reduced proliferative capacity and epithelial-mesenchymal transition-related protein expression and led to marked downregulation of MAPK pathway-associated genes. Treatment with HI-TOPK-032 demonstrated therapeutic potential by suppressing aggressive cellular phenotypes and inducing apoptosis. CONCLUSION TOPK plays a critical role in pNEN progression and liver metastasis through the MAPK axis. HI-TOPK-032 exhibits promising antitumor activity by targeting TOPK, suggesting its potential as a therapeutic option for pNENs with liver metastasis. Further in vivo and clinical validation is warranted.

OBJECTIVE:To develop and externally validate a prognostic nomogram for overall survival (OS) in resected duodenal adenocarcinoma (DA) using routinely available perioperative variables, thereby clarifying risk profiles and supporting clinical management. METHODS:Multicenter analysis of 2289 consecutive DA patients undergoing curative surgery (2012-2022) from China's National Cancer Center database. External validation used 335 patients from Zhejiang Provincial People's Hospital (2022-2024). LASSO-Cox regression selected variables from 89 perioperative factors to construct the nomogram, with web tool implementation. RESULTS:The LASSO-Cox model achieved 1-, 3-, and 5-year AUCs of 0.72 (95% CI, 0.68-0.77), 0.75 (95% CI, 0.72-0.77), and 0.76 (95% CI, 0.73-0.79), outperforming traditional Cox models (P < .01). External validation yielded AUCs of 0.76 (95% CI, 0.66-0.86) and 0.79 (95% CI, 0.74-0.86) for 1- and 3-year OS, and 0.81 (95% CI, 0.74-0.89) for estimated 5-year OS. The model stratified patients into low- and high-risk groups (cutoff 0.40), with low-risk patients showing superior survival. Eight predictors were selected, including modifiable surgical factors such as transfusion and operative time. CONCLUSIONS:We developed and externally validated a postoperative prognostic nomogram for DA using routinely available perioperative variables. In the present study, the model improved postoperative risk stratification and may support counseling, follow-up planning, and multidisciplinary discussion regarding adjuvant therapy; however, it should be viewed as complementary to standard staging and clinical judgment, and broader clinical implementation will require further validation.

INTRODUCTION AND OBJECTIVES:We initiated this study to explore the efficacy of camrelizumab combined with transcatheter arterial chemoembolization (TACE) plus sorafenib or lenvatinib versus TACE plus sorafenib or Lenvatinib for unresectable hepatocellular carcinoma (HCC). MATERIALS AND METHODS:From June 2019 to November 2022, 127 advanced HCC patients were retrospectively analyzed in this study. This consisted of 44 patients that received camrelizumab plus TACE plus sorafenib or lenvatinib (triple therapy group) and 83 patients that received TACE plus sorafenib or lenvatinib (double treatment group). The overall survival (OS), progression-free survival (PFS), objective response rate (ORR), and disease control rate (DCR) were compared between the two patient groups. RESULTS:Our findings demonstrated that patients received the triple therapy exhibited superior median OS (15.8 vs. 10.3 months, P=0.0011) and median PFS (7.2 vs. 5.2 months, P=0.019) compared to the double treatment group. In addition, the triple therapy group exhibited better 6-month (93.5% vs. 66.3%), 12-month (67.2% vs. 36.3%), and 24-month (17.2% vs. 7.6%) survival rates than the double treatment group. However, the ORR (43.2% vs. 28.9%, P = 0.106) and DCR (93.2% vs. 81.9%, P = 0.084) of the two groups were similar. Subgroup analysis showed that compared with the double treatment group, the triple therapy group had a better mOS for HCC with HBV (15.8 vs. 9.6 months, P = 0.0015) and tumor diameter ≥ 5cm (15.3 vs. 9.6 months, P = 0.00055). CONCLUSIONS:Camrelizumab plus TACE and sorafenib or lenvatinib may be a promising treatment approach for the clinical management of unresectable HCC patients.

The prognosis of gastroenteropancreatic neuroendocrine tumors (GEP-NETs) following metastasis is often poor. The efficacy of 177Lu-DOTATATE therapy and the subgroups that benefit from it remain unclear. Our objective is to characterize the composition of the tumor immune microenvironment in GEP-NETs and to identify predictive biomarkers associated with response and PFS following 177Lu-DOTATATE. Multiplex immunofluorescence (mIF) staining of tumor sections was used to characterize the cellular density and spatial organization of the microenvironment of 50 NET patients. The relationship between baseline immune microenvironment and 177Lu-DOTATATE efficacy or prognosis in 20 177Lu-DOTATATE-treated patients was explored. Patients with GEP-NET exhibited an immunosuppressive microenvironment. The overall response rate (ORR) to 177Lu-DOTATATE therapy was 60

Duodenal adenocarcinoma (DA) has a high recurrence rate, making the prediction of recurrence after surgery critically important. Our objective is to develop a machine learning-based model to predict the postoperative recurrence of DA. We conducted a multicenter, retrospective cohort study in China. 1830 patients with DA who underwent radical surgery between 2012 and 2023 were included. Wrapper methods were used to select optimal predictors by ten machine learning learners. Subsequently, these ten learners were utilized for model development. The model's performance was validated using three separate cohorts, and assessed by the concordance index (C-index), time-dependent calibration curve, time-dependent receiver operating characteristic curves, and decision curve analysis. After selecting predictors, ten feature subsets were identified. And ten feature subsets were combined with the ten machine learning learners in a permutation, resulting in the development of 100 predictive models, and the Penalized Regression + Accelerated Oblique Random Survival Forest model (PAM) exhibited the best predictive performance. The C-index for PAM was 0.882 (95

Objectives:We assess the incidence of exocrine pancreatic insufficiency (EPI) at different time points after pancreatic surgery and explore pancreatic enzyme replacement therapy's (PERT) efficacy.Background:EPI is characterized by inadequate pancreatic enzymes, resulting in maldigestion and abdominal symptoms. EPI is a common postoperative complication of pancreatic surgery, yet often overlooked by surgeons. There is no clear answer to when EPI occurs after pancreatic surgery nor the duration of PERT after partial pancreatectomy.Methods:Benign or borderline pancreatic tumor patients undergoing surgeries were recruited between December 2020 and November 2021 from 10 medical centers in China. The EPI Questionnaire (EPI-Q) was performed at discharge, and 3-, 6-, 9-, and 12-month follow-ups to evaluate the occurrence of EPI. Statistical analyses were performed to identify the occurrence of EPI and explore PERT's efficacy.Results:Of the 146 patients, 105 (71.9%) were female, and the median age was 49 (range 16-78 years). Symptoms of EPI patients were most pronounced within 3 months post-surgery (15.7%), while most patients recovered after 1 year (96.9%). There was a negative correlation between symptom score and time since surgery. Lack of PERT in the 3-month post-surgery was associated with higher symptom scores in pancreatoduodenectomy (PD) patients, while not in distal pancreatectomy (DP) patients.Conclusions:Generally, the high-occurrence period for postoperative EPI is within 3 months after resection in patients with benign or borderline pancreatic tumors, and most will gradually recuperate with time. PERT can improve symptoms in PD patients, while reduced PERT duration may be considered for DP patients.

Abstract Background Diagnosing and managing follicular thyroid neoplasms (FTNs) remains a significant challenge, as the malignancy risk cannot be determined until after diagnostic surgery. Objective We aimed to use interpretable machine learning to predict the malignancy risk of FTNs preoperatively in a real-world setting. Methods We conducted a retrospective cohort study at the Peking University Third Hospital in Beijing, China. Patients with postoperative pathological diagnoses of follicular thyroid adenoma (FTA) or follicular thyroid carcinoma (FTC) were included, excluding those without preoperative thyroid ultrasonography. We used 22 predictors involving demographic characteristics, thyroid sonography, and hormones to train 5 machine learning models: logistic regression, least absolute shrinkage and selection operator regression, random forest, extreme gradient boosting, and support vector machine. The optimal model was selected based on discrimination, calibration, interpretability, and parsimony. To address the highly imbalanced data (FTA:FTC ratio>5:1), model discrimination was assessed using both the area under the receiver operating characteristic curve and the area under the precision-recall curve (AUPRC). To interpret the model, we used Shapley Additive Explanations values and partial dependence and individual conditional expectation plots. Additionally, a systematic review was performed to synthesize existing evidence and validate the discrimination ability of the previously developed Thyroid Imaging Reporting and Data System for Follicular Neoplasm scoring criteria to differentiate between benign and malignant FTNs using our data. Results The cohort included 1539 patients (mean age 47.98, SD 14.15 years; female: n=1126, 73.16%) with 1672 FTN tumors (FTA: n=1414; FTC: n=258; FTA:FTC ratio=5.5). The random forest model emerged as optimal, identifying mean thyroid-stimulating hormone (TSH) score, mean tumor diameter, mean TSH, TSH instability, and TSH measurement levels as the top 5 predictors in discriminating FTA from FTC, with the area under the receiver operating characteristic curve of 0.79 (95% CI 0.77‐0.81) and AUPRC of 0.40 (95% CI 0.37-0.44). Malignancy risk increased nonlinearly with larger tumor diameters and higher TSH instability but decreased nonlinearly with higher mean TSH scores or mean TSH levels. FTCs with small sizes (mean diameter 2.88, SD 1.38 cm) were more likely to be misclassified as FTAs compared to larger ones (mean diameter 3.71, SD 1.36 cm). The systematic review of the 7 included studies revealed that (1) the FTA:FTC ratio varied from 0.6 to 4.0, lower than the natural distribution of 5.0; (2) no studies assessed prediction performance using AUPRC in unbalanced datasets; and (3) external validations of Thyroid Imaging Reporting and Data System for Follicular Neoplasm scoring criteria underperformed relative to the original study. Conclusions Tumor size and TSH measurements were important in screening FTN malignancy risk preoperatively, but accurately predicting the risk of small-sized FTNs remains challenging. Future research should address the limitations posed by the extreme imbalance in FTA and FTC distributions in real-world data.

The portal-mesenteric venous system surrounding the pancreas exhibits significant anatomical complexity and variability. Precise segmentation and assessment of these venous structures are crucial for preoperative planning in pancreatic surgery to reduce the risk of intraoperative hemorrhage and postoperative complications. To address this, we present the Venous Anatomy Analyzer for Pancreatic Surgery ($V A^{2} P S$), a novel framework that integrates 3D vein segmentation and Large Vision-Language Model (VLM), to provide surgeons with intuitive visualization and accurate analysis of the portalmesenteric venous system and pancreas. VA ${ }^{\mathbf{2}}$ PS first performs precise 3D segmentation of the venous system and pancreas from CT scans. Subsequently, it bridges the dimensional gap between segmentation result and VLM input by rendering these 3D structures into optimized 2D views. Finally, conditioned on the visual input and surgical instructions, $\text{VA}^{2}$ PS utilizes a VLM reasoning over a built-in repository of clinical knowledge to identify venous anatomical subtypes and generates surgical planning recommendations. Objective and subjective evaluations demonstrate that $\text{VA}^{2} \text{PS}$ provides accurate and efficient preoperative assessments for surgeons, holding the potential to improve surgical outcomes. Code and data are available at https://github.com/zengyue1376/VA2PS.

Background The clinicopathologic and prognostic significance of T-cell lymphoma invasion and metastasis 2 (TIAM2) in hepatocellular carcinoma (HCC) remains unclear.Methods TIAM2 expression was detected immunohistochemically in matched HCC and adjacent liver (AL) specimens from 168 patients with radical resection. The correlations between TIAM2 and clinicopathologic parameters, overall and disease-free survival were evaluated. The expression, prognostic value and genomic alterations of TIAM2 gene were explored in the online publicly available databases.Results TIAM2 was significantly overexpressed in HCC tissues, compared with AL tissues (P < 0.001). Its expression in multiple tumors was also statistically higher than that in solitary ones (P = 0.017). Moreover, TIAM2 overexpression was univariately associated with poor overall and disease-free survival (P = 0.0066 and 0.0060). In multivariate Cox regression analysis, TIAM2 overexpression was one of significant determinants of both overall and disease-free survival. In the Ualcan and Kaplan-Meier Plotter databases, overexpression and prognostic power of TIAM2 gene in HCC was confirmed, while its genetic alterations included mutation, amplification and deep deletion in the cBioPortal database. Some known tumor-related genes, such as FGD6, FGFR2 and FZD1, were strongly related to TIAM2 gene.Conclusions TIAM2 overexpression closely correlated with tumor multiplicity and poor prognosis in resected HCC, thus being a potential therapeutic target.

Pancreatic ductal adenocarcinoma (PDAC) includes local progression out of control and early systemic dissemination. Dysregulation of anoikis is closely related to PDAC metastasis, but its heterogeneity and different expression subtypes in the tumor microenvironment (TME) have not been fully clarified. Electrochemical sensors can achieve rapid and sensitive detection of biomolecules, especially in the detection of tumor markers, providing strong support for the early diagnosis and treatment monitoring of cancer. This study aims to analyze the expression heterogeneity of nest-lost apoptosis-related genes (ARGs) by integrating electrochemical sensor technology and single-cell transcriptome sequencing (scRNA-seq). Three RNA sequence datasets were obtained from databases such as The Cancer Genome Atlas (TCGA-PAAD) and the International Cancer Genome Consortium (ICGC-PAAD-CA & ICGC-PAAD-AU). And three other gene expression data were extracted from the Comprehensive Gene Expression Database (GEO), etc. for external validation. The single-cell transcriptome file has been downloaded. Through integrated analysis, genes related to nest-lost apoptosis (ARG) were identified, and a prognostic risk scoring model was constructed based on these genes. Combining the application advantages of electrochemical sensors in medical laboratory tests, this paper explores their potential application value in the diagnosis and treatment monitoring of PDAC. The study revealed three distinct subtypes of ARG, which exhibit significant differences in TME composition, metabolic reprogramming, immune landscape, and chemotherapy sensitivity. The prognostic risk scoring model constructed based on six core ARGs can effectively divide patients into high-risk and low-risk groups. There is a significant difference in overall survival rates between the two groups, and this scoring model has been consistently validated in independent internal and external cohorts. Research on the application of electrochemical sensors in medical testing shows that they have significant advantages in detection speed, sensitivity and portability, and can rapidly detect tumor markers in blood, providing a new technical means for the early diagnosis and treatment monitoring of PDAC.

Background Surgeons often face challenges in distinguishing between benign and malignant follicular thyroid neoplasms (FTNs), particularly small tumors, until diagnostic surgery is performed. Objective This study aimed to identify the size-specific predictors for the malignancy risk of FTNs preoperatively. Methods A retrospective cohort study was conducted at Peking University Third Hospital in Beijing, China, from 2012 to 2023. Patients with a postoperative pathological diagnosis of follicular thyroid adenoma (FTA) or follicular thyroid carcinoma (FTC) were included. FTNs were classified into small- and large-sized categories based on the cutoff value of the tumor diameter derived from spline regression, which indicated the turning point of malignancy risk. We identified the 5 most important predictors from 22 variables including demography, sonography, and hormones, using machine learning methods. We also calculated the odds ratios (OR) with 95% CI for these predictors in both small- and large-sized FTNs. Results Altogether, we included 1494 FTNs, comprising 1266 FTAs and 228 FTCs. FTNs with a maximum diameter less than 3.0 cm were grouped as small-sized tumors (n=715), while those with larger diameters were categorized as large-sized tumors (n=779). In the small-sized group, tumors with macrocalcification (OR 2.90, 95% CI 1.50-5.60), those with peripheral calcification (OR 4.50, 95% CI 1.50-13.00), and those in younger patients (OR 1.33, 95% CI 1.05-1.69) showed a higher malignancy risk. In the large-sized group, tumors presenting with a nodule-in-nodule appearance (OR 3.30, 95% CI 1.30-7.90) exhibited a higher malignancy risk. In both groups, lower thyroid-stimulating hormone levels (OR 1.49, 95% CI 1.20-1.85 for small-sized FTNs; OR 1.61, 95% CI 1.37-1.96 for large-sized FTNs) and a larger mean diameter (OR 1.40, 95% CI 1.10-1.70 for small-sized FTNs; OR 1.50 95% CI 1.20-1.70 for large-sized FTNs) were associated with the malignancy risk of FTNs. Conclusion This study identified size-specific predictors for malignancy risk in FTNs, highlighting the importance of stratified prediction based on tumor size.

Background:. Pancreatic neuroendocrine tumors (pNETs) with synchronous liver metastasis (NELM) are associated with poor prognosis. Radical surgery is the only curative treatment option for these patients. However, the prognostic indicators for NELM patients who underwent radical surgery remain unclear. This study aimed to evaluate the prognostic value of liver metastasis burden (LMB) and other general clinicopathological indicators in patients who underwent radical resection. Methods:. A retrospective analysis was conducted on a cohort of 36 patients who underwent surgery between January 2016 and December 2023. The LMB indicator was defined as the ratio of metastatic volume to liver volume (RML). The optimal cutoff value of RML was identified via X-tile software, dividing the population into high and low groups. Results:. Kaplan-Meier (KM) curves for progression-free survival (PFS) revealed a significant difference between high and low groups of RML (P = .004), but there were no statistical differences in overall survival (OS) between the 2 groups (P = .309). Cox regression univariable analysis indicated that only high RML was associated with poor PFS (hazard ratio = 4.42, 95% confidence interval = 1.48–13.22; P = .008). However, it was not significantly associated with OS (P = .331). High RML emerged as an independent risk factor for PFS in these patients (hazard ratio = 4.73, 95% confidence interval = 1.59–14.03; P = .005). In the high RML group, patients receiving preoperative chemotherapy or chemotherapy plus somatostatin analogs (SSA), had lower recurrence rates compared with those without chemotherapy or without chemotherapy plus SSA, respectively. No such benefit was observed in the low RML group. Conclusion:. RML may serve as an effective prognostic indicator of PFS for patients with NELM who undergo curative surgery. For these patients with high liver metastatic burden, preoperative treatment regimens containing chemotherapy should be recommended.