Deep Learning-Based Survival Analysis Identified Associations Between Molecular Subtype and Optimal Adjuvant Treatment of Patients With Gastric Cancer.

PURPOSE: Gastric cancer (GC) is the third-leading cause of cancer-related deaths. Several pivotal clinical trials of adjuvant treatments were performed during the previous decade; however, the optimal regimen for adjuvant treatment of GC remains controversial. PATIENTS AND METHODS: We developed a novel deep learning-based survival model (survival recurrent network [SRN]) in patients with GC by including all available clinical and pathologic data and treatment regimens. This model uses time-sequential data only in the training step, and upon being trained, it receives the initial data from the first visit and then sequentially predicts the outcome at each time point until it reaches 5 years. In total, 1,190 patients from three cohorts (the Asian Cancer Research Group cohort, n = 300; the fluorouracil, leucovorin, and radiotherapy cohort, n = 432; and the Adjuvant Chemoradiation Therapy in Stomach Cancer cohort, n = 458) were included in the analysis. In addition, we added Asian Cancer Research Group molecular classifications into the prediction model. SRN simulated the sequential learning process of clinicians in the outpatient clinic using a recurrent neural network and time-sequential outcome data. RESULTS: The mean area under the receiver operating characteristics curve was 0.92 ± 0.049 at the fifth year. The SRN demonstrated that GC with a mesenchymal subtype should elicit a more risk-adapted postoperative treatment strategy as a result of its high recurrence rate. In addition, the SRN found that GCs with microsatellite instability and GCs of the papillary type exhibited significantly more favorable survival outcomes after capecitabine plus cisplatin chemotherapy alone. CONCLUSION: Our SRN predicted survival at a high rate, reaching 92% at postoperative year 5. Our findings suggest that SRN-based clinical trials or risk-adapted adjuvant trials could be considered for patients with GC to investigate more individualized adjuvant treatments after curative gastrectomy.

Effect of DA-9701 on Gastric Motor Function Assessed by Magnetic Resonance Imaging in Healthy Volunteers: A Randomized, Double-Blind, Placebo-Controlled Trial.

BACKGROUND: Improving gastric accommodation and gastric emptying is an attractive physiological treatment target in patients with functional dyspepsia (FD). We evaluated the effect of DA-9701, a new drug for FD, on gastric motor function after a meal in healthy volunteers using magnetic resonance imaging (MRI). METHODS: Forty healthy volunteers were randomly allocated to receive either DA-9701 or placebo. After 5 days of treatment, subjects underwent gastric MRI (60 min before and 15, 30, 45, 60, 90, and 120 min after a liquid test meal). Gastric volume was measured through 3-dimensional reconstruction from MRI data. We analyzed 4 outcome variables including changes in total gastric volume (TGV), proximal TGV, and proximal to distal TGV ratio after a meal and gastric emptying rates after adjusting values at the pre-test meal. RESULTS: Changes in TGV and proximal TGV after a meal did not differ between the DA-9701 and placebo groups (difference between groups -25.9 mL, 95% confidence interval [CI] -54.0 to 2.3 mL, P = 0.070 and -2.9 mL, 95% CI -30.3 to 24.5 mL, P = 0.832, respectively). However, pre-treatment with DA-9701 increased postprandial proximal to distal TGV ratio more than placebo (difference between groups 0.93, 95% CI 0.08 to 1.79, P = 0.034). In addition, pre-treatment with DA-9701 significantly increased gastric emptying as compared with placebo (mean difference between groups 3.41%, 95% CI 0.54% to 6.29%, P = 0.021, by mixed model for repeated measures). CONCLUSION: Our results suggested that DA-9701 enhances gastric emptying and does not significantly affect gastric accommodation in healthy volunteers. Further studies to confirm whether DA-9701 enhances these gastric motor functions in patients with FD are warranted. TRIAL REGISTRATION: ClinicalTrials.gov NCT02091635.

Characterization of brivanib therapy response in hepatocellular carcinoma xenografts using ¹H HR-MAS spectroscopy and histopathology.

Angiogenesis inhibition is an attractive therapeutic strategy in the management of solid tumors. Vascular endothelial growth factor (VEGF) and fibroblast growth factor (FGF) are key factors in growth and neovascularization of hepatocellular carcinoma (HCC). Brivanib is a novel, orally available dual tyrosine kinase inhibitor that selectively targets the key angiogenesis receptors VEGF­R2, FGF­R1 and FGF­R2. Recently, high­resolution magic angle spinning magnetic resonance spectroscopy (HR­MAS MRS) has provided the opportunity to investigate more detailed metabolic profiles from intact tissue specimens that are correlated with histopathology and is thus, a promising tool for monitoring changes induced by treatment. In the present study, 1H HR­MAS MRS and immunohistochemistry were used to investigate the antitumor efficacy of brivanib in HCC xenograft models. Tumor growth was significantly suppressed in brivanib­treated mice compared with the controls and treatment was associated with the inhibition of angiogenesis, increased apoptosis and inhibition of cell proliferation. Furthermore, HR­MAS techniques showed altered metabolic profiles between the two groups. HR­MAS spectra demonstrated a significant decrease in choline metabolite levels in the treated groups, concurrent with decreased cell proliferation and increased apoptosis. The results showed that 1H HR­MAS MRS provides quantitative metabolite information that may be used to analyze the efficacy of brivanib treatment in Hep3B tumor xenografts. Thus, the HR­MAS MRS technique may be a complementary method to support histopathological results and increase its potential for use in the clinic.

Radiofrequency ablation for viable hepatocellular carcinoma around retained iodized oil after transcatheter arterial chemoembolization: usefulness of biplane fluoroscopy plus ultrasound guidance.

OBJECTIVE: To assess the technical feasibility and local efficacy of biplane fluoroscopy plus US-guided percutaneous radiofrequency ablation (RFA) for viable hepatocellular carcinoma (HCC) around retained iodized oil after transcatheter arterial chemoembolization (TACE). MATERIALS AND METHODS: Our prospective study was approved by our institutional review board and informed consent was obtained from all participating patients. For patients with viable HCC around retained iodized oil after TACE, biplane fluoroscopy plus US-guided RFA was performed. We evaluated the rate of technical success and major complications on a post-RFA CT examination and local tumor progression with a follow-up CT. RESULTS: Among 40 consecutive patients, 19 were excluded due to one of the following reasons: poorly visible HCC on fluoroscopy (n = 13), high risk location (n = 2), RFA performed under monoplane fluoroscopy and US guidance (n = 2), and poorly identifiable new HCCs on US (n = 2). The remaining 21 patients with 21 viable HCCs were included. The size of total tumors ranged from 1.4 to 5.0 cm (mean: 3.2 cm) in the longest diameter. Technical success was achieved for all 21 HCCs, and major complications were observed in none of the patients. During the follow-up period (mean, 20.3 months; range, 6.5-29.9 months), local tumor progression was found in two patients (2/21, 9.5%). Distant intrahepatic metastasis developed in 76.2% (16/21) of patients. CONCLUSION: When retained iodized oil around the tumor after TACE hampers the targeting of the viable tumor for RFA, biplane fluoroscopy plus US-guided RFA may be performed owing to its technical feasibility and effective treatment for viable HCCs.

Pattern and chronological change of iodized oil retention in radiofrequency ablation-induced vascular injury area: differentiation from iodized oil retention in recurrent hepatocellular carcinoma on computed tomography.

PURPOSE: The purpose of this study was to evaluate the pattern and the chronological change of iodized oil retention in a radiofrequency ablation (RFA)-induced vascular injury area as compared with that in recurrent hepatocellular carcinoma (HCC). MATERIALS AND METHODS: Informed consents from all the patients and approval from the institutional review board were obtained. The computed tomographic scans of 226 patients who underwent both RFA and subsequent transcatheter arterial chemoembolization for recurrence were retrospectively reviewed. The RFA-induced vascular injury area that did not have HCC recurrence in it were examined and assessed whether iodized oil was retained after transcatheter arterial chemoembolization. We evaluated the incidence, the shape, and the duration of the iodized oil retention in the RFA-induced vascular injury area and compared them with those found in recurrent HCC from the same patient. The rate of misinterpretation was calculated. RESULTS: Of 59 RFA-induced vascular injury areas that were within the range of segmental embolization, 33 RFA-induced vascular injury areas (55.9%) in 31 patients retained iodized oil. The shapes of the iodized oil in the RFA-induced vascular injury area were diffuse-dense (n = 13), diffuse-faint (n = 8), nodular-dense (n = 8), or nodular-faint (n = 4). In 29 (93.5%) of 31 patients, at least one recurrent HCC retained iodized oil, and they were all nodular-dense in shape. The duration of the retention in RFA-induced vascular injury area (mean [SD], 3.3 [3.2] months) was significantly shorter than that in recurrent tumors (21.2 [12.5] months, P < 0.001). In 21.2% (7/33) of the cases, the iodized oil density in the RFA-induced vascular injury area was clinically misinterpreted as recurrent HCC. CONCLUSIONS: The RFA-induced vascular injury area frequently retains iodized oil, which could mimic recurrent HCC. However, we determined that iodized oil in the RFA-induced vascular injury area differs from that in HCC in variety in shape and shorter duration of retention.

Indeterminate small, low-attenuating hepatocellular nodules on helical CT in patients with chronic liver disease: 2-year follow-up.

To evaluate the clinical significance of indeterminate, small, low-attenuating nodular lesions on helical dynamic computed tomography (CT) in chronic liver diseases, CT images were reviewed retrospectively in 281 patients. Indeterminate, low-attenuating nodular lesions less than 20 mm in diameter were followed for 24 to 35 months. Of 127 nodules in 73 patients, 21 nodules turned out to be hepatocellular carcinomas (16%), 25 nodules (20%) became larger, while the remaining 81 nodules (64%) remained unchanged or disappeared. Some low-attenuating nodules larger than 10 mm in diameter may develop into hepatocellular carcinoma.

Liver abscess after percutaneous radiofrequency ablation for hepatocellular carcinomas: frequency and risk factors.

OBJECTIVE: The purpose of this study was to clarify the frequency and risk factors of liver abscess formation after percutaneous radiofrequency ablation in patients with hepatocellular carcinoma. MATERIALS AND METHODS: Over a 4-year period, 603 patients with 831 hepatocellular carcinomas measuring 5 cm or less in maximum diameter who underwent a total of 751 percutaneous radiofrequency ablation procedures were enrolled in this study. We retrospectively reviewed the medical records and analyzed the overall frequency of liver abscess, risk factors for abscess, and clinical features of the patients. The relationships between liver abscess and potential risk factors were analyzed using either generalized estimating equations or multiple logistic regression analysis. RESULTS: Liver abscess developed in 14 tumors of 13 patients after 13 (13/751 [1.7%]) ablation procedures. Generalized estimating equations and multiple logistic regression analysis of various potential risk factors revealed that preexisting biliary abnormality prone to ascending biliary infection (p = 0.0088), tumor with retention of iodized oil from previous transcatheter arterial chemoembolization (p = 0.040), and treatment with an internally cooled electrode system (p = 0.016) were associated with a significant risk of liver abscess formation. No patient died of liver abscess, and all successfully recovered from liver abscess with parenteral antibiotics and percutaneous clearance of pus. CONCLUSION: Although liver abscess formation was infrequent in patients who underwent percutaneous radiofrequency ablation for hepatocellular carcinoma, the patients with significant risk factors-preexisting biliary abnormality prone to ascending biliary infection, tumor with retention of iodized oil, and treatment with an internally cooled electrode system-for liver abscess formation should be closely monitored after treatment.