Purpose: In this study, we aimed to determine the clinicopathologic, radiologic, and molecular significance of the tumor invasiveness to further stratify the patients with high-grade (HG) upper tract urothelial carcinoma (UTUC) who can be treated less aggressively. Materials and Methods: Clinicopathologic and radiologic characteristics of 166 surgically resected HG UTUC (48 noninvasive, and 118 invasive) cases were evaluated. Six noninvasive UTUC cases with intratumoral tumor grade heterogeneity were selected for whole-exome sequencing (WES) to understand the underlying molecular pathophysiology. Barcode-tagging sequencing was done for validation of the target genes from WES data. Results: Patients with noninvasive UTUC showed no cancer-specific death with better cancer-specific survival (p < 0.001) and recurrence-free survival (p < 0.001) compared to the patients with invasive UTUC. Compared to the invasive UTUC, noninvasive UTUC was correlated to a low grade (LG) on the preoperative abdominal computed tomography (CT) grading system (p < 0.001), histologic intratumoral tumor grade heterogeneity (p=0.018), discrepancy in preoperative urine cytology diagnosis (p=0.018), and absence of urothelial carcinoma in situ (p < 0.001). WES of the heterogeneous components showed mutually shared HRAS and FGFR3 mutations shared between the HG and LG components. HRAS mutation was associated with the lower grade on preoperative abdominal CT and intratumoral tumor grade heterogeneity (p=0.045 and p < 0.001, respectively), whereas FGFR3 mutation was correlated to the absence of carcinoma in situ (p < 0.001). Conclusion According to our comprehensive analysis, HG noninvasive UTUC can be preoperatively suspected based on distinct preoperative radiologic, cytologic, histologic, and molecular features. Noninvasive HG UTUC shows excellent prognosis and thus should be treated less aggressively.

PURPOSE: Although gemcitabine plus cisplatin has been established as the standard first-line chemotherapy for patients with advanced biliary tract cancer (BTC), overall prognosis remains poor. We investigated the efficacy of a novel triplet combination of oxaliplatin, irinotecan, and S-1 (OIS) for advanced BTC. MATERIALS AND METHODS: Chemotherapy-naive patientswith histologically documented unresectable or metastatic BTC were eligible for this multicenter, single-arm phase II study. Patients received 65 mg/m2 oxaliplatin (day 1), 135 mg/m2 irinotecan (day 1), and 40 mg/m2 S-1 (twice a day, days 1-7) every 2 weeks. Primary endpoint was objective response rate. Targeted exome sequencing for biomarker analysis was performed using archival tissue. RESULTS: In total, 32 patients were enrolled between October 2015 and June 2016. Median age was 64 years (range, 40 to 76 years), with 24 (75%) male patients; 97% patients had metastatic or recurrent disease. Response rate was 50%, and median progression-free survival and overall survival (OS) were 6.8 months (95% confidence interval [CI], 4.8 to 8.8) and 12.5 months (95% CI, 7.0 to 18.0), respectively. The most common grade 3-4 adverse events were neutropenia (32%), diarrhea (6%), and peripheral neuropathy (6%). TP53 and KRAS mutations were the most frequent genomic alterations (42% and 32%, respectively), and KRAS mutations showed a marginal relationship with worse OS (p=0.07). CONCLUSION: OIS combination chemotherapy was feasible and associated with favorable efficacy outcomes as a first-line treatment in patients with advanced BTC. Randomized studies are needed to compare OIS with gemcitabine plus cisplatin.
Biliary Tract Neoplasms* ; Biliary Tract* ; Cholangiocarcinoma ; Cisplatin ; Diarrhea ; Disease-Free Survival ; Drug Therapy ; Drug Therapy, Combination ; Exome ; Humans ; Male ; Neutropenia ; Peripheral Nervous System Diseases ; Prognosis ; Triplets

Background: Bladder cancer is characterized by frequent mutations, which provide potential therapeutic targets for most patients. The effectiveness of emerging personalized therapies depends on an accurate molecular diagnosis, for which the accurate estimation of the neoplastic cell percentage (NCP) is a crucial initial step. However, the established method for determining the NCP, manual counting by a pathologist, is time-consuming and not easily executable. Methods: To address this, artificial intelligence (AI) models were developed to estimate the NCP using nine convolutional neural networks and the scanned images of 39 cases of urinary tract cancer. The performance of the AI models was compared to that of six pathologists for 119 cases in the validation cohort. The ground truth value was obtained through multiplexed immunofluorescence. The AI model was then applied to 41 cases in the application cohort that underwent next-generation sequencing testing, and its impact on the copy number variation (CNV) was analyzed. Results: Each AI model demonstrated high reliability, with intraclass correlation coefficients (ICCs) ranging from 0.82 to 0.88. These values were comparable or better to those of pathologists, whose ICCs ranged from 0.78 to 0.91 in urothelial carcinoma cases, both with and without divergent differentiation/ subtypes. After applying AI-driven NCP, 190 CNV (24.2%) were reclassified with 66 (8.4%) and 78 (9.9%) moved to amplification and loss, respectively, from neutral/minor CNV. The neutral/minor CNV proportion decreased by 6%. Conclusions These results suggest that AI models could assist human pathologists in repetitive and cumbersome NCP calculations.

Background: Bladder cancer is characterized by frequent mutations, which provide potential therapeutic targets for most patients. The effectiveness of emerging personalized therapies depends on an accurate molecular diagnosis, for which the accurate estimation of the neoplastic cell percentage (NCP) is a crucial initial step. However, the established method for determining the NCP, manual counting by a pathologist, is time-consuming and not easily executable. Methods: To address this, artificial intelligence (AI) models were developed to estimate the NCP using nine convolutional neural networks and the scanned images of 39 cases of urinary tract cancer. The performance of the AI models was compared to that of six pathologists for 119 cases in the validation cohort. The ground truth value was obtained through multiplexed immunofluorescence. The AI model was then applied to 41 cases in the application cohort that underwent next-generation sequencing testing, and its impact on the copy number variation (CNV) was analyzed. Results: Each AI model demonstrated high reliability, with intraclass correlation coefficients (ICCs) ranging from 0.82 to 0.88. These values were comparable or better to those of pathologists, whose ICCs ranged from 0.78 to 0.91 in urothelial carcinoma cases, both with and without divergent differentiation/ subtypes. After applying AI-driven NCP, 190 CNV (24.2%) were reclassified with 66 (8.4%) and 78 (9.9%) moved to amplification and loss, respectively, from neutral/minor CNV. The neutral/minor CNV proportion decreased by 6%. Conclusions These results suggest that AI models could assist human pathologists in repetitive and cumbersome NCP calculations.

Background: Bladder cancer is characterized by frequent mutations, which provide potential therapeutic targets for most patients. The effectiveness of emerging personalized therapies depends on an accurate molecular diagnosis, for which the accurate estimation of the neoplastic cell percentage (NCP) is a crucial initial step. However, the established method for determining the NCP, manual counting by a pathologist, is time-consuming and not easily executable. Methods: To address this, artificial intelligence (AI) models were developed to estimate the NCP using nine convolutional neural networks and the scanned images of 39 cases of urinary tract cancer. The performance of the AI models was compared to that of six pathologists for 119 cases in the validation cohort. The ground truth value was obtained through multiplexed immunofluorescence. The AI model was then applied to 41 cases in the application cohort that underwent next-generation sequencing testing, and its impact on the copy number variation (CNV) was analyzed. Results: Each AI model demonstrated high reliability, with intraclass correlation coefficients (ICCs) ranging from 0.82 to 0.88. These values were comparable or better to those of pathologists, whose ICCs ranged from 0.78 to 0.91 in urothelial carcinoma cases, both with and without divergent differentiation/ subtypes. After applying AI-driven NCP, 190 CNV (24.2%) were reclassified with 66 (8.4%) and 78 (9.9%) moved to amplification and loss, respectively, from neutral/minor CNV. The neutral/minor CNV proportion decreased by 6%. Conclusions These results suggest that AI models could assist human pathologists in repetitive and cumbersome NCP calculations.

Background: Bladder cancer is characterized by frequent mutations, which provide potential therapeutic targets for most patients. The effectiveness of emerging personalized therapies depends on an accurate molecular diagnosis, for which the accurate estimation of the neoplastic cell percentage (NCP) is a crucial initial step. However, the established method for determining the NCP, manual counting by a pathologist, is time-consuming and not easily executable. Methods: To address this, artificial intelligence (AI) models were developed to estimate the NCP using nine convolutional neural networks and the scanned images of 39 cases of urinary tract cancer. The performance of the AI models was compared to that of six pathologists for 119 cases in the validation cohort. The ground truth value was obtained through multiplexed immunofluorescence. The AI model was then applied to 41 cases in the application cohort that underwent next-generation sequencing testing, and its impact on the copy number variation (CNV) was analyzed. Results: Each AI model demonstrated high reliability, with intraclass correlation coefficients (ICCs) ranging from 0.82 to 0.88. These values were comparable or better to those of pathologists, whose ICCs ranged from 0.78 to 0.91 in urothelial carcinoma cases, both with and without divergent differentiation/ subtypes. After applying AI-driven NCP, 190 CNV (24.2%) were reclassified with 66 (8.4%) and 78 (9.9%) moved to amplification and loss, respectively, from neutral/minor CNV. The neutral/minor CNV proportion decreased by 6%. Conclusions These results suggest that AI models could assist human pathologists in repetitive and cumbersome NCP calculations.