Background: Currently, follicular thyroid carcinoma (FTC) has a relatively low incidence with a lack of effective preoperative diagnostic means. To reduce the need for invasive diagnostic procedures and to address information deficiencies inherent in a small dataset, we utilized interpretable foreground optimization network deep learning to develop a reliable preoperative FTC detection system. Methods: In this study, a deep learning model (FThyNet) was established using preoperative ultrasound images. Data on patients in the training and internal validation cohort (n=432) were obtained from Ruijin Hospital, China. Data on patients in the external validation cohort (n=71) were obtained from four other clinical centers. We evaluated the predictive performance of FThyNet and its ability to generalize across multiple external centers and compared the results yielded with assessments from physicians directly predicting FTC outcomes. In addition, the influence of texture information around the nodule edge on the prediction results was evaluated. Results: FThyNet had a consistently high accuracy in predicting FTC with an area under the receiver operating characteristic curve (AUC) of 89.0% [95% CI 87.0–90.9]. Particularly, the AUC for grossly invasive FTC reached 90.3%, which was significantly higher than that of the radiologists (56.1% [95% CI 51.8–60.3]). The parametric visualization study found that those nodules with blurred edges and relatively distorted surrounding textures were more likely to have FTC. Furthermore, edge texture information played an important role in FTC prediction with an AUC of 68.3% [95% CI 61.5–75.5], and highly invasive malignancies had the highest texture complexity. Conclusion: FThyNet could effectively predict FTC, provide explanations consistent with pathological knowledge, and improve clinical understanding of the disease.

Background: The prognosis for colon adenocarcinoma (COAD) today remains poor. Changes in mitochondria-related genes and metabolic reprogramming are related to tumor growth, metastasis, and immune evasion and are key factors in tumor genesis and development.Methods: TCGA database was used to analyze the differentially expressed mitochondrial energy metabolism pathway-related genes (MMRGs) in COAD patients, and the mutation of MMRG in tumor cells, the biological processes involved, and the correlation with tumor immunity were also analyzed. Then, MMRG and MMRG-related genes were used to divide COAD patients into different subtypes, and immunocorrelation analysis and survival analysis were performed. Finally, univariate regression analysis and LASSO regression analysis were used to construct a prognostic risk model for COAD patients, which was verified by the GEO database and evaluated by Kaplan–Meier (K-M) and receiver operating characteristic (ROC) curves, and the correlation between the risk model and immunity and clinical subtypes based on MMRG was analyzed.Results: In this study, the MMRG patterns and tumor immune microenvironment characteristics in COAD patients were systematically evaluated by clustering the expression of 188 MMRGs. We identified two subtypes of COAD with different clinical and immunological characteristics. Eight of the 28 differentially expressed MMRG genes were used to construct risk scores. ROC and K-M curves suggested that the risk model could well predict the prognosis of COAD patients, and the risk model was related to immune cell infiltration and immune function.Conclusions: The two COAD subtypes identified by MMRG are helpful for the clinical differentiation of patients with different prognoses and tumor progressions, and the risk score can assist the clinical evaluation of patient prognosis. Our results suggest that CPT2 contributes to the recruitment and regulation of neutrophils in COAD. CPT2 may act as a valuable biomarker for COAD immunotherapy.

Purpose s: To quantitatively predict the risk of neck lymph node metastasis for unilateral and bilateral papillary thyroid microcarcinomas (PTMC) that may guide individual treatment strategies for the neck region. Methods: A total of 717 PTMC patients from three medical centers were enrolled for analysis. Results: Bilateral PTMCs were demonstrated to be more aggressive with a much higher cervical lymph node metastasis rate including for both central (CLNM) and lateral lymph node metastasis (LLNM) when being compared to unilateral PTMCs. In unilateral PTMC, five (age < 55 years old, male, maximum tumor diameter (MTD) ≥ 0.5 cm, and the presence of thyroid capsular invasion (TCI) and multifocality) and three (maximum diameter of positive CLN (MDCLN) > 0.5 cm, the presence of multifocality and nodular goiter, iNG) factors were identified as independent risk factors for CLNM and LLNM, respectively. In bilateral PTMC, three (age < 55 and presence of TCI and multifocality in at least one side of thyroid lobe) and two (MDCLN > 0.5 cm and presence of nodular goiter (iNG)) factors were identified as independent factors for CLNM and LLNM, respectively. Predictive models of CLNM and LLNM for patients with unilateral disease and of CLNM for patients with the bilateral disease were established based on the described risk factors. Bilateral patients with positive CLNM were also stratified into different subgroups according to the presence and absence of independent risk factors. Conclusion: An evaluation system based on independent factors of CLNM and LLNM for PTMC patients with bilateral and unilateral disease was established. Our newly established evaluation system can efficaciously quantify risks of CLNM and LLNM for PTMC patients with bilateral and unilateral disease and may guide individual treatment strategy including both surgical and postoperative adjuvant treatment of the neck region for these patients.

Objective: To quantitatively predict the probability of lateral lymph node metastasis (LLNM) for papillary thyroid carcinomas (PTC) patients with central lymph node metastasis (CLNM) in order to guide postoperative adjuvant treatment. Methods: Five hundred and three PTC patients with CLNM from three medical centers were retrospectively analyzed. Results: The LLNM rate for all patients was 23.9% (120 in 503), with 15.5% (45 in 291) and 35.4% (75 in 212) for patients with papillary thyroid microcarcinoma (PTMC) and large papillary thyroid carcinoma (LPTC), respectively. Patients with no fewer than five positive central lymph nodes (CLN) exhibited a higher risk of LLNM. For patients with fewer than five positive CLN, a maximum diameter of positive CLN > 0.5 cm and the presence of ipsilateral nodular goiter were identified as independent risk factors of LLNM for papillary thyroid microcarcinoma (PTMC) patients. The independent risk factors of LLNM for large papillary thyroid carcinoma (LPTC) patients included a tumor located in the upper portion of thyroid, maximum tumor diameter ≥ 2.0 cm, maximum diameter of positive CLN > 0.5 cm, and the presence of thyroid capsular invasion. Predictive nomograms were established based on these risk factors for PTMC and LPTC patients, respectively. The accuracy and validity of our newly built models were verified by C-index and calibration curves. PTMC and LPTC patients with fewer than five positive CLN were each stratified into three subgroups based on their nomogram risk scores, and a detailed risk stratification flow chart was established for a more accurate evaluation of LLNM risk in PTC patients. Conclusions: A detailed stratification flow chart for PTC patients with CLNM to quantitatively assess LLNM risk was established, which may aid in clinical decision-making for those patients.

Copper participates in biological processes such as oxygen metabolism and iron uptake, and is a key factor in immune regulation. Based on the transcription data, mutation data and clinical data of colon adenocarcinoma (COAD) patients from The Cancer Genome Atlas (TCGA) database and Gene Expression Profiling Interactive Analysis (GEPIA2) database, the expression and mutation of copper metabolization-related genes in COAD patients and their correlation with tumor immune microenvironment were analyzed. Copper metabolization-related genes (CMRGs) were used to construct COAD subtypes and prognostic risk models for COAD patients. Furthermore, Kaplan-Meier (K-M) curve and receiver operating characteristic (ROC) curve were used to analyze the clinical value of COAD subtypes and genotyping models in distinguishing clinical characteristics of patients, and the immune infiltration of patients with different genotypes was analyzed. Finally, the clinical tissue samples from COAD patients were used to analyze the mRNA expression of genes in risk model between tumor and normal tissues by the method of Polymerase Chain Reaction (PCR). Of the 479 CMRGs, 68 genes were differentially expressed in normal and tumor tissues of COAD patients in TCGA and GEPIA2. Two subtypes with different clinical and immunological characteristics were identified by using 482 genes related to copper metabolism. Finally, a prognostic risk model consisting of five CMRGs was constructed, which could not only predict the prognosis of patients, but also correlated with COAD subtypes. In addition, some genes (glutathione S-transferase mu 1, cyclin D1and cytochrome P450 family 2 subfamily S member 1) in risk model was show significant difference between normal and tumor tissues. The COAD subtypes identified by CMRGs can help clinically distinguish patients with different prognosis and tumor progression, and the risk score can assist in clinical evaluation of patient prognosis, serving as a valuable biomarker for COAD immunotherapy.

RNA therapy represents an innovative approach for cancer treatment, with several RNA-based therapeutics having received approval from the US Food and Drug Administration. Circular RNA (circRNA), a closed-loop RNA molecule characterized by its high stability, plays a significant role in regulating biological processes by modulating gene expression and facilitating protein translation. Given its unique structure and diverse functionalities, the delivery of exogenous circRNA has emerged as a novel strategy for cancer therapy. This review examines the mechanism underlying circRNA-mediated tumor therapy, emphasizing its various biological roles, including that of an RNA sponge, aptamer, gene editing tool, and facilitator of protein translation, and explores the therapeutic potential in oncology. The review provides a comprehensive discussion on the synthesis strategies of exogenous circRNA, based on T4 DNA ligase and the permuted introns-exons (PIE) method, as well as elucidating the purification techniques. This article also reviews prominent carriers currently employed for circRNA delivery, such as lipid nanoparticle (LNP), exosomes, and virus-like particle (VLP), with a particular emphasis on their application in cancer-targeted therapies. Finally, the review summarizes key challenges currently in the field along with viable solutions. It highlights the prospective role of artificial intelligence in enhancing circRNA delivery to facilitate precise cancer treatment based on exogenous circRNA.