Background/Aims: Inaccurate prediction of lymph node metastasis (LNM) may lead to unnecessary surgery following endoscopic resection of T1 colorectal cancer (CRC). We aimed to validate the usefulness of artificial intelligence (AI) models for predicting LNM in patients with T1 CRC. Methods: We analyzed the clinical data, laboratory results, pathological reports, and endoscopic findings of patients who underwent radical surgery for T1 CRC. We developed AI models to predict LNM using four algorithms: regularized logistic regression classifier (RLRC), random forest classifier (RFC), CatBoost classifier (CBC), and the voting classifier (VC). Four histological factors and four endoscopic findings were included to develop AI models. Areas under the receiver operating characteristics curves (AUROCs) were measured to distinguish AI model performance in accordance with the Japanese Society for Cancer of the Colon and Rectum guidelines. Results: Among 1,386 patients with T1 CRC, 173 patients (12.5%) had LNM. The AUROC values of the RLRC, RFC, CBC, and VC models for LNM prediction were significantly higher (0.673, 0.640, 0.679, and 0.677, respectively) than the 0.525 suggested in accordance with the Japanese Society for Cancer of the Colon and Rectum guidelines (vs RLRC, p<0.001; vs RFC, p=0.001; vs CBC, p<0.001; vs VC, p<0.001). The AUROC value was similar between T1 colon versus T1 rectal cancers (0.718 vs 0.615, p=0.700). The AUROC value was also similar between the initial endoscopic resection and initial surgery groups (0.581 vs 0.746, p=0.845). Conclusions AI models trained on the basis of endoscopic findings and pathological features performed well in predicting LNM in patients with T1 CRC regardless of tumor location and initial treatment method.

Background/Aims: Inaccurate prediction of lymph node metastasis (LNM) may lead to unnecessary surgery following endoscopic resection of T1 colorectal cancer (CRC). We aimed to validate the usefulness of artificial intelligence (AI) models for predicting LNM in patients with T1 CRC. Methods: We analyzed the clinical data, laboratory results, pathological reports, and endoscopic findings of patients who underwent radical surgery for T1 CRC. We developed AI models to predict LNM using four algorithms: regularized logistic regression classifier (RLRC), random forest classifier (RFC), CatBoost classifier (CBC), and the voting classifier (VC). Four histological factors and four endoscopic findings were included to develop AI models. Areas under the receiver operating characteristics curves (AUROCs) were measured to distinguish AI model performance in accordance with the Japanese Society for Cancer of the Colon and Rectum guidelines. Results: Among 1,386 patients with T1 CRC, 173 patients (12.5%) had LNM. The AUROC values of the RLRC, RFC, CBC, and VC models for LNM prediction were significantly higher (0.673, 0.640, 0.679, and 0.677, respectively) than the 0.525 suggested in accordance with the Japanese Society for Cancer of the Colon and Rectum guidelines (vs RLRC, p<0.001; vs RFC, p=0.001; vs CBC, p<0.001; vs VC, p<0.001). The AUROC value was similar between T1 colon versus T1 rectal cancers (0.718 vs 0.615, p=0.700). The AUROC value was also similar between the initial endoscopic resection and initial surgery groups (0.581 vs 0.746, p=0.845). Conclusions AI models trained on the basis of endoscopic findings and pathological features performed well in predicting LNM in patients with T1 CRC regardless of tumor location and initial treatment method.

Background/Aims: Inaccurate prediction of lymph node metastasis (LNM) may lead to unnecessary surgery following endoscopic resection of T1 colorectal cancer (CRC). We aimed to validate the usefulness of artificial intelligence (AI) models for predicting LNM in patients with T1 CRC. Methods: We analyzed the clinical data, laboratory results, pathological reports, and endoscopic findings of patients who underwent radical surgery for T1 CRC. We developed AI models to predict LNM using four algorithms: regularized logistic regression classifier (RLRC), random forest classifier (RFC), CatBoost classifier (CBC), and the voting classifier (VC). Four histological factors and four endoscopic findings were included to develop AI models. Areas under the receiver operating characteristics curves (AUROCs) were measured to distinguish AI model performance in accordance with the Japanese Society for Cancer of the Colon and Rectum guidelines. Results: Among 1,386 patients with T1 CRC, 173 patients (12.5%) had LNM. The AUROC values of the RLRC, RFC, CBC, and VC models for LNM prediction were significantly higher (0.673, 0.640, 0.679, and 0.677, respectively) than the 0.525 suggested in accordance with the Japanese Society for Cancer of the Colon and Rectum guidelines (vs RLRC, p<0.001; vs RFC, p=0.001; vs CBC, p<0.001; vs VC, p<0.001). The AUROC value was similar between T1 colon versus T1 rectal cancers (0.718 vs 0.615, p=0.700). The AUROC value was also similar between the initial endoscopic resection and initial surgery groups (0.581 vs 0.746, p=0.845). Conclusions AI models trained on the basis of endoscopic findings and pathological features performed well in predicting LNM in patients with T1 CRC regardless of tumor location and initial treatment method.

Background/Aims: Inaccurate prediction of lymph node metastasis (LNM) may lead to unnecessary surgery following endoscopic resection of T1 colorectal cancer (CRC). We aimed to validate the usefulness of artificial intelligence (AI) models for predicting LNM in patients with T1 CRC. Methods: We analyzed the clinical data, laboratory results, pathological reports, and endoscopic findings of patients who underwent radical surgery for T1 CRC. We developed AI models to predict LNM using four algorithms: regularized logistic regression classifier (RLRC), random forest classifier (RFC), CatBoost classifier (CBC), and the voting classifier (VC). Four histological factors and four endoscopic findings were included to develop AI models. Areas under the receiver operating characteristics curves (AUROCs) were measured to distinguish AI model performance in accordance with the Japanese Society for Cancer of the Colon and Rectum guidelines. Results: Among 1,386 patients with T1 CRC, 173 patients (12.5%) had LNM. The AUROC values of the RLRC, RFC, CBC, and VC models for LNM prediction were significantly higher (0.673, 0.640, 0.679, and 0.677, respectively) than the 0.525 suggested in accordance with the Japanese Society for Cancer of the Colon and Rectum guidelines (vs RLRC, p<0.001; vs RFC, p=0.001; vs CBC, p<0.001; vs VC, p<0.001). The AUROC value was similar between T1 colon versus T1 rectal cancers (0.718 vs 0.615, p=0.700). The AUROC value was also similar between the initial endoscopic resection and initial surgery groups (0.581 vs 0.746, p=0.845). Conclusions AI models trained on the basis of endoscopic findings and pathological features performed well in predicting LNM in patients with T1 CRC regardless of tumor location and initial treatment method.

Background: and purpose Whether pharmacologically altered high-density lipoprotein cholesterol (HDL-C) affects the risk of cardiovascular events is unknown. Recently, we have reported the Prevention of Cardiovascular Events in Asian Patients with Ischaemic Stroke at High Risk of Cerebral Haemorrhage (PICASSO) trial that demonstrated the non-inferiority of cilostazol to aspirin and superiority of probucol to non-probucol for cardiovascular prevention in ischemic stroke patients (clinicaltrials.gov: NCT01013532). We aimed to determine whether on-treatment HDL-C changes by cilostazol and probucol influence the treatment effect of each study medication during the PICASSO study. Methods: Of the 1,534 randomized patients, 1,373 (89.5%) with baseline cholesterol parameters were analyzed. Efficacy endpoint was the composite of stroke, myocardial infarction, and cardiovascular death. Cox proportional hazards regression analysis examined an interaction between the treatment effect and changes in HDL-C levels from randomization to 1 month for each study arm. Results: One-month post-randomization mean HDL-C level was significantly higher in the cilostazol group than in the aspirin group (1.08 mmol/L vs. 1.00 mmol/L, P<0.001). The mean HDL-C level was significantly lower in the probucol group than in the non-probucol group (0.86 mmol/L vs. 1.22 mmol/L, P<0.001). These trends persisted throughout the study. In both study arms, no significant interaction was observed between HDL-C changes and the assigned treatment regarding the risk of the efficacy endpoint. Conclusions Despite significant HDL-C changes, the effects of cilostazol and probucol treatment on the risk of cardiovascular events were insignificant. Pharmacologically altered HDL-C levels may not be reliable prognostic markers for cardiovascular risk.

Purpose: The purpose of this study was to identify the concordant or discordant genomic profiling between primary and matched metastatic tumors in patients with colorectal cancer (CRC) and to explore the clinical implication. Materials and Methods: Surgical samples of primary and matched metastatic tissues from 158 patients (335 samples) with CRC at Korea University Anam Hospital were evaluated using the Ion AmpliSeq Cancer Hotspot Panel. We compared genetic variants and classified them as concordant, primary-specific, and metastasis-specific variants. We used a combination of principal components analysis and clustering to find genomic groups. Kaplan-Meier curves were used to appraise survival between genomic groups. We used machine learning to confirm the correlation between genetic variants and metastatic sites. Results: A total of 282 types of deleterious non-synonymous variants were selected for analysis. Of a total of 897 variants, an average of 40% was discordant. Three genomic groups were yielded based on the genomic discrepancy patterns. Overall survival differed significantly between the genomic groups. The poorest group had the highest proportion of concordant KRAS G12V and additional metastasis-specific SMAD4. Correlation analysis between genetic variants and metastatic sites suggested that concordant KRAS mutations would have more disseminated metastases. Conclusion Driver gene mutations were mostly concordant; however, discordant or metastasis-specific mutations were present. Clinically, the concordant driver genetic changes with additional metastasis-specific variants can predict poor prognosis for patients with CRC.

Background: Hodgkin's lymphoma (HL) constitutes 10%–20% of all malignant lymphomas and has a high cure rate (5-year survival, around 90%). Recently, interest has increased concerning preventing secondary complications (secondary cancer, endocrine disorders) in long-term survivors. We aimed to study the epidemiologic features and therapeutic outcomes of HL in children, adolescents, and young adults in Korea. Methods: We performed a multicenter, retrospective study of 224 patients aged < 25 years diagnosed with HL at 22 participating institutes in Korea from January 2007 to August 2016. Results: A higher percentage of males was diagnosed at a younger age. Nodular sclerosis histopathological HL subtype was most common, followed by mixed cellularity subtype.Eighty-one (36.2%), 101 (45.1%), and 42 (18.8%) patients were classified into low, intermediate, and high-risk groups, respectively. Doxorubicin, bleomycin, vinblastine, dacarbazine was the most common protocol (n = 102, 45.5%). Event-free survival rate was 86.0% ± 2.4%, while five-year overall survival (OS) rate was 96.1% ± 1.4%: 98.7% ± 1.3%, 97.7% ± 1.6%, and 86.5% ± 5.6% in the low, intermediate, and high-risk groups, respectively (P = 0.021). Five-year OS was worse in patients with B-symptoms, stage IV disease, highrisk, splenic involvement, extra-nodal lymphoma, and elevated lactate dehydrogenase level.In multivariate analysis, B-symptoms and extra-nodal involvement were prognostic factors for poor OS. Late complications of endocrine disorders and secondary malignancy were observed in 17 and 6 patients, respectively. Conclusion This is the first study on the epidemiology and treatment outcomes of HL in children, adolescents, and young adults in Korea. Future prospective studies are indicated to develop therapies that minimize treatment toxicity while maximizing cure rates in children, adolescents, and young adults with HL.

Thermal ablation using radiofrequency is a new, minimally invasive modality employed as an alternative to surgery in patients with benign thyroid nodules and recurrent thyroid cancers. The Task Force Committee of the Korean Society of Thyroid Radiology (KSThR) developed recommendations for the optimal use of radiofrequency ablation for thyroid tumors in 2012. As new meaningful evidences have accumulated, KSThR decided to revise the guidelines. The revised guideline is based on a comprehensive analysis of the current literature and expert consensus.
Advisory Committees ; Catheter Ablation* ; Consensus ; Humans ; Thyroid Gland* ; Thyroid Neoplasms ; Thyroid Nodule ; Ultrasonography

Bioresorbable vascular scaffold (BRS) is an innovative device that provides structural support and drug release to prevent early recoil or restenosis, and then degrades into nontoxic compounds to avoid late complications related with metallic drug-eluting stents (DESs). BRS has several putative advantages. However, recent randomized trials and registry studies raised clinical concerns about the safety and efficacy of first generation BRS. In addition, the general guidance for the optimal practice with BRS has not been suggested due to limited long-term clinical data in Korea. To address the safety and efficacy of BRS, we reviewed the clinical evidence of BRS implantation, and suggested the appropriate criteria for patient and lesion selection, scaffold implantation technique, and management.
Coronary Disease ; Drug Liberation ; Drug-Eluting Stents ; Humans ; Korea ; Stents ; Thrombosis