Search

Background & Aims: Hepatocellular carcinoma (HCC) is characterized by a high mortality rate. The Liver Imaging Reporting and Data System (LI-RADS) results in a considerable number of indeterminate observations, rendering an accurate diagnosis difficult. Methods: We developed four deep learning models for diagnosing HCC on computed tomography (CT) via a training–validation–testing approach. Thin-slice triphasic CT liver images and relevant clinical information were collected and processed for deep learning. HCC was diagnosed and verified via a 12-month clinical composite reference standard. CT observations among at-risk patients were annotated using LI-RADS. Diagnostic performance was assessed by internal validation and independent external testing. We conducted sensitivity analyses of different subgroups, deep learning explainability evaluation, and misclassification analysis. Results: From 2,832 patients and 4,305 CT observations, the best-performing model was Spatio-Temporal 3D Convolution Network (ST3DCN), achieving area under receiver-operating-characteristic curves (AUCs) of 0.919 (95% CI, 0.903–0.935) and 0.901 (95% CI, 0.879–0.924) at the observation (n = 1,077) and patient (n = 685) levels, respectively during internal validation, compared with 0.839 (95% CI, 0.814–0.864) and 0.822 (95% CI, 0.790–0.853), respectively for standard of care radiological interpretation. The negative predictive values of ST3DCN were 0.966 (95% CI, 0.954–0.979) and 0.951 (95% CI, 0.931–0.971), respectively. The observation-level AUCs among at-risk patients, 2–5-cm observations, and singular portovenous phase analysis of ST3DCN were 0.899 (95% CI, 0.874–0.924), 0.872 (95% CI, 0.838–0.909) and 0.912 (95% CI, 0.895–0.929), respectively. In external testing (551/717 patients/observations), the AUC of ST3DCN was 0.901 (95% CI, 0.877–0.924), which was non-inferior to radiological interpretation (AUC 0.900; 95% CI, 0.877–-923). Conclusions: ST3DCN achieved strong, robust performance for accurate HCC diagnosis on CT. Thus, deep learning can expedite and improve the process of diagnosing HCC. Impact and implications:: The clinical applicability of deep learning in HCC diagnosis is potentially huge, especially considering the expected increase in the incidence and mortality of HCC worldwide. Early diagnosis through deep learning can lead to earlier definitive management, particularly for at-risk patients. The model can be broadly deployed for patients undergoing a triphasic contrast CT scan of the liver to reduce the currently high mortality rate of HCC.

Background: Afaitnib has shown anti-tumor activity against metastatic EGFR-mutated NSCLC after prior failure to first generation EGFR-TKI and chemotherapy. We prospectively evaluated the efficacy and safety of afatinib in Chinese patients who previously failed first-generation TKI and chemotherapy under a compassionate use program (CUP) and compared to the erlotinib cohort.Methods: Patients who suffered from metastatic EGFR-mutated NSCLC previously responsive to first-generation TKI and chemotherapy received afatinib until progression, loss of clinical benefits or intolerable toxicity. Treatment response, survival and safety were evaluated and compared to the erlotinib cohort.Results: Twenty-five and 28 patients received afatinib and erlotinib respectively. More patients in the afatinib group had worse performance status (ECOG 2) than the erlotinib group (p = 0.008). After a median follow-up of 12.1 months, afatinib demonstrated comparable objective response rate (ORR) (20.0% vs. 7.1%, p = 0.17) but significantly higher disease control rate (DCR) (68.0% vs. 39.3%, p = 0.04) compared to erlotinib. Median progression-free survival (PFS) (4.1 months [95% CI, 2.7-5.5 months] vs. 3.3 months [95% CI, 2.2-4.3 months], p = 0.97) and overall survival (OS) were not different between the two groups (10.3 months [95% CI, 7.5-13.0 months] vs. 10.8 months [95% CI, 7.4-14.2 months], p = 0.51). Multivariate analyses revealed that age ≤ 70 years and time to progression (TTP) ≥ 18 months for the 1st TKI therapy were prognostic of PFS (p = 0.006 and p = 0.008 respectively). Afatinib caused less rash (60.0% vs. 67.9%, p = 0.04) but more diarrhea (60.0% vs. 10.7%, p = 0.002) compared to erlotinib.Conclusion: Afatinib produced encouraging clinical efficacy as 2nd TKI therapy with manageable safety profiles in our Chinese patients after failure to another TKI and systemic chemotherapy. This study was registered at ClinicalTrials.gov (NCT02625168) on 3rd December 2015. [ABSTRACT FROM AUTHOR]

Introduction: Current guidelines discourage the use of direct-acting antiviral (DAA) containing protease-inhibitor (PI) in advanced HCV cirrhosis. We aimed to compare the real-world tolerability of PI vs. non-PI DAA regimens in this population. Methods: We identified advanced cirrhosis patients treated with DAA from the REAL-C registry. The primary outcome was significant worsening or improvement in CPT or MELD scores following DAA treatment. Results: From the REAL-C registry of 15,837 patients, we included 1077 advanced HCV cirrhosis patients from 27 sites. 42% received PI-based DAA. Compared to non-PI group, the PI group was older, had higher MELD and higher percentage with kidney disease. Inverse probability of treatment weighting (IPTW; matching on age, sex, history of clinical decompensation, MELD, platelet, albumin, Asia site, Asian ethnicity, hypertension, hemoglobin, genotype, liver cancer, ribavirin) was used to balance the two groups. In the IPTW-matched cohorts, the PI and non-PI groups had similar SVR12 (92.9% vs. 90.7%, p = 0.30), similar percentages of significant worsening in CTP or MELD scores at posttreatment week 12 and 24 (23.9% vs. 13.1%, p = 0.07 and 16.5% vs. 14.6%, p = 0.77), and similar frequency of new HCC, decompensating event, and death by posttreatment week 24. In multivariable analysis, PI-based DAA was not associated with significant worsening (adjusted odds ratio = 0.82, 95% CI 0.38–1.77). Conclusion: Tolerability and treatment outcomes were not significantly different in advanced HCV cirrhosis treated with PI-based (vs. non-PI) DAA up to CTP-B or MELD score of 15. Safety of PI-based DAA in those with CTP-C or MELD beyond 15 awaits further data. [ABSTRACT FROM AUTHOR]

Background: Despite HCV cure, patients remain at risk for HCC, but risk factor data for HCC following SVR are limited for Asian patients. Methods: To address this gap, we analyzed 5814 patients (5646 SVR, 168 non-SVR) from the Real-World Evidence from the Asia Liver Consortium for HCV (REAL-C) who did not have HCC or a history of HCC at baseline (pre-DAA treatment) and did not develop HCC within 6 months of baseline. To assess the effect of SVR on HCC incidence, we used 1:4 propensity score matching [(PSM), age, sex, baseline cirrhosis, and baseline AFP] to balance the SVR and non-SVR groups. Results: In the PSM cohort (160 non-SVR and 612 SVR), the HCC incidence rate per 100 person years was higher in the non-SVR compared to the SVR group (5.26 vs. 1.94, p < 0.001). Achieving SVR was independently associated with decreased HCC risk (adjusted HR [aHR]: 0.41, p = 0.002). Next, we stratified the SVR cohort of 5646 patients to cirrhotic and noncirrhotic subgroups. Among cirrhotic SVR patients, aged ≥ 60, having an albumin bilirubin grade (ALBI) of 2 or 3 (aHR: 2.5, p < 0.001), and baseline AFP ≥ 10 ng/mL (aHR: 1.6, p = 0.001) were associated with higher HCC risk, while among the non-cirrhotic SVR group, only baseline AFP ≥ 10 ng/mL was significant (aHR: 4.26, p = 0.005). Conclusions: Achieving SVR decreases HCC risk; however, among East Asians, patients with elevated pretreatment AFP remained at risk. Pretreatment AFP, an easily obtained serum marker, may provide both prognostic and surveillance value for HCC in East Asian patients who obtained SVR. [ABSTRACT FROM AUTHOR]

A 49-year-old man with pharyngeal dysphagia after encephalitis is presented in this case study. Sixteen months earlier the patient experienced a sudden severe fever which resulted in encephalitis, leading to liquid dysphagia. Despite receiving an initial treatment of swallowing therapy, the patient's liquid dysphagia did not improve. Functional neuromuscular electrical stimulation, which is a new treatment method, was then applied to the patient. The patient showed improvement in the pharyngeal phase of swallowing. Clinical and treatment observations are reported. [ABSTRACT FROM AUTHOR]

The aim of this work was to study the psychometric properties of the Chinese version of the Swallow Quality-of-Life Questionnaire (CSWAL-QOL) validated in the Hong Kong Chinese-speaking population. With convenience sampling, a cross-sectional survey was launched to evaluate the validity and reliability of the CSWAL-QOL. One hundred subjects with swallowing problems were recruited to evaluate the construct validity and internal consistency, and 20 subjects were recruited for the test-retest reliability. Construct validity was validated through factor analysis (both exploratory and confirmatory) and a correlation study between the CSWAL-QOL and the World Health Organization Quality-of-Life Questionnaire-abbreviated version [WHOQOL-BREF (HK)]. Reliability was estimated using tests of internal consistency and test-retest reliability. The psychometric properties of the CSWAL-QOL were found to be largely similar to those of the SWAL-QOL, except the three-item eating desire scale of the CSWAL-QOL, which showed insignificant results in the validity and reliability tests. The CSWAL-QOL is the first validated Chinese version of the SWAL-QOL in Hong Kong. It is a clinically valid and reliable tool for assessing the quality of life in dysphagic Chinese patients in Hong Kong, regardless of the causes of dysphagia. [ABSTRACT FROM AUTHOR]