Personalized radiomics signature to screen for KIT-11 mutation genotypes among patients with gastrointestinal stromal tumors: a retrospective multicenter study.

OBJECTIVES: Gastrointestinal stromal tumors (GISTs) carrying different KIT exon 11 (KIT-11) mutations exhibit varying prognoses and responses to Imatinib. Herein, we aimed to determine whether computed tomography (CT) radiomics can accurately stratify KIT-11 mutation genotypes to benefit Imatinib therapy and GISTs monitoring. METHODS: Overall, 1143 GISTs from 3 independent centers were separated into a training cohort (TC) or validation cohort (VC). In addition, the KIT-11 mutation genotype was classified into 4 categories: no KIT-11 mutation (K11-NM), point mutations or duplications (K11-PM/D), KIT-11 557/558 deletions (K11-557/558D), and KIT-11 deletion without codons 557/558 involvement (K11-D). Subsequently, radiomic signatures (RS) were generated based on the arterial phase of contrast CT, which were then developed as KIT-11 mutation predictors using 1408 quantitative image features and LASSO regression analysis, with further evaluation of its predictive capability. RESULTS: The TC AUCs for K11-NM, K11-PM/D, K11-557/558D, and K11-D ranged from 0.848 (95% CI 0.812-0.884), 0.759 (95% CI 0.722-0.797), 0.956 (95% CI 0.938-0.974), and 0.876 (95% CI 0.844-0.908), whereas the VC AUCs ranged from 0.723 (95% CI 0.660-0.786), 0.688 (95% CI 0.643-0.732), 0.870 (95% CI 0.824-0.918), and 0.830 (95% CI 0.780-0.878). Macro-weighted AUCs for the KIT-11 mutant genotype ranged from 0.838 (95% CI 0.820-0.855) in the TC to 0.758 (95% CI 0.758-0.784) in VC. TC had an overall accuracy of 0.694 (95%CI 0.660-0.729) for RS-based predictions of the KIT-11 mutant genotype, whereas VC had an accuracy of 0.637 (95%CI 0.595-0.679). CONCLUSIONS: CT radiomics signature exhibited good predictive performance in estimating the KIT-11 mutation genotype, especially in prediction of K11-557/558D genotype. RS-based classification of K11-NM, K11-557/558D, and K11-D patients may be an indication for choice of Imatinib therapy.

Current Status and Analysis of Machine Learning in Hepatocellular Carcinoma.

Hepatocellular carcinoma (HCC) is a common tumor. Although the diagnosis and treatment of HCC have made great progress, the overall prognosis remains poor. As the core component of artificial intelligence, machine learning (ML) has developed rapidly in the past decade. In particular, ML has become widely used in the medical field, and it has helped in the diagnosis and treatment of cancer. Different algorithms of ML have different roles in diagnosis, treatment, and prognosis. This article reviews recent research, explains the application of different ML models in HCC, and provides suggestions for follow-up research.

Prognostic value of LI-RADS category on MRI in patients with primary hepatic lymphoepithelioma-like carcinoma.

OBJECTIVES: To compare the clinical and MRI features of primary hepatic lymphoepithelioma-like carcinoma (LELC) categorized as LR-M or LR-4/5 using the Liver Imaging Reporting and Data System (LI-RADS) version 2018 and to determine the prognostic factors for recurrence-free survival (RFS). METHODS: In this retrospective study, 37 patients with surgically confirmed LELC were included. Two independent observers evaluated preoperative MRI features according to the LI-RADS version 2018. Clinical and imaging features were compared between two groups. RFS and the associated factors were evaluated using Cox proportional hazards regression analysis, Kaplan-Meier analysis, and log-rank test. RESULTS: In total, 37 patients (mean age, 58.5 ± 10.3 years) were evaluated. Sixteen (43.2%) LELCs were categorized as LR-M and twenty-one (56.8%) LELCs were categorized as LR-4/5. In the multivariate analysis, the LR-M category was an independent factor for RFS (HR 7.908, 95% CI 1.170-53.437; p = 0.033). RFS rates were significantly lower in patients with LR-M LELCs than in patients with LR-4/5 LELCs (5-year RFS rate, 43.8% vs.85.7%; p = 0.002). CONCLUSION: The LI-RADS category was significantly associated with postsurgical prognosis of LELC, with tumor categorized as LR-M having a worse RFS than those categorized as LR-4/5. KEY POINTS: • Lymphoepithelioma-like carcinoma patients categorized as LR-M have worse recurrence-free survival than those categorized as LR-4/5. • MRI-based LI-RADS categorization was an independent factor for postoperative prognosis of primary hepatic lymphoepithelioma-like carcinoma.

Analysis of mutations in DNA damage repair pathway gene in Chinese patients with hepatocellular carcinoma.

The incidence of hepatocellular carcinoma (HCC) has increased in these years. DNA damage repair (DDR) pathway is required in response to DNA damage Gene mutations in DDR pathway play an important role in different stages of tumorigenesis and development. Based on the importance of DDR pathway in precision therapy of multiple cancers, we analyzed DDR gene mutations in Chinese patients with HCC. The results showed that (tumor mutation burden) TMB was significantly higher in HCC patients who carried somatic mutations in DDR than in non-carriers, and TMB in patients with DS, MMR mutations and DDR genes mutations such as RAD50, MLH1, MSH2, CHEK2 was significantly higher than that in wild-type patients. Based on the results of next-generation sequencing (NGS) testing, we are trying to provide clues for targeted therapy and provide feasible basis for PD-1/PD-L1 immune checkpoint inhibitor therapy.

A new model based inflammatory index and tumor burden score (TBS) to predict the recurrence of hepatocellular carcinoma (HCC) after liver resection.

To establish a model based on inflammation index and tumor burden score (TBS) to predict recurrence of hepatocellular carcinoma (HCC) after liver resection. A retrospective study was performed on 217 patients who diagnosed HCC underwent liver resection at Xiangya Hospital Central South University from June 1, 2017 to June 1, 2019. According to the receiver operating characteristic (ROC) curve, the optimal cut-off value of inflammatory index and the TBS was determined by the Youden index. Prediction performance was compared by the area under the receiver operating characteristic curve (AUC). Cox regression analysis was used to determine the risk factors for the recurrence of HCC after liver resection. According to the independent risk factors of the patients, a prediction model for HCC was established based on inflammation index and tumor burden score (TBS).The prediction performance of the model was compared with single index (TBS group and NLR group) and traditional HCC stage models (TNM stage and BCLC stage). MLR = 0.39, NLR = 2.63, PLR = 134, SII = 428 and TBS = 8.06 are the optimal cut-off values. AUC of SII, PLR, NLR, MLR and TBS were 0.643, 0.642, 0.642, 0.618 and 0.724respectively. MVI (P = 0.005), satellite nodule (P = 0.017), BCLC B-C stage (P = 0.013), NLR > 2.63 (P = 0.013), TBS > 8.06 (P = 0.017) are independent risk factors for the recurrence of HCC after liver resection. According to this study, the optimal inflammatory index NLR combined with TBS was obtained. The AUC of NLR-TBS model was 0.762, not only better than NLR group (AUC = 0.630) and TBS group (AUC = 0.671), also better than traditional BCLC (AUC = 0.620) and TNM (AUC = 0.587) stage models. Interestingly, we found that NLR and TBS should be good prognostic factor for recurrence of HCC after liver resection. The NLR-TBS model based the best inflammatory index (NLR) and TBS have a better prediction performance and the prediction performance of NLR-TBS model not only better than NLR group and TBS group, but better than BCLC and TNM stage models.

TNF-α derived from M2 tumor-associated macrophages promotes epithelial-mesenchymal transition and cancer stemness through the Wnt/ß-catenin pathway in SMMC-7721 hepatocellular carcinoma cells.

M2-polarized tumor-associated macrophages (M2-TAMs) infiltrating the tumor microenvironment contribute to hepatocellular carcinoma (HCC) progression. It was reported that cancer cells undergoing EMT will acquire stemness characteristics. Here, the HCC SMMC-7721 cell line was co-cultured with M2-TAMs polarized from THP-1 cells in vitro. In in vivo studies, we used nude mice subcutaneous tumor model to test whether the growth of the tumor was affected by M2-TAMs. Subsequently, EMT, stemness and Wnt/ß-catenin pathway related markers were detected in cells and subcutaneous tumor tissues. TNF-α was also assessed in both the co-culture system supernatants and in nude mice serum. We found that SMMC-7721 underwent EMT and acquired stemness after co-culture with M2-TAMs, and resulted in larger tumor size following subcutaneous injection of SMMC-7721 suspended in M2-TAMs supernatants compared with SMMC-7721 alone. Enzyme linked immunosorbent assay showed that TNF-α expression was elevated in supernatants of M2-TAMs and positively correlated with tumor size in the serum of nude mice. Furthermore, we found that the Wnt/ß-catenin pathway was a downstream target of TNF-α and that the Wnt/ß-catenin inhibitor ICG-001 partially reversed EMT and attenuated cancer stemness. Our results indicate that TNF-α derived from M2-TAMs promote EMT and cancer stemness cells via the Wnt/ß-catenin pathway.