Letter: Sorafenib hepatotoxicity may be enhanced during treatment of advanced hepatocellular carcinoma in HIV-infected patients

A novel machine learning model based on sparse structure learning with adaptive graph regularization for predicting drug side effects

Drug side effects are closely related to the success and failure of drug development. Here we present a novel machine learning method for side effect prediction. The proposed method treats side effect prediction as a multi-label learning problem and uses sparse structure learning to model the relationships between side effects. Additionally, the proposed method adopts the adaptive graph regularization strategy to explore the local structure in drug data and fuse multiple types of drug features. An alternating optimization algorithm is proposed to solve the optimization problem. We collected chemical structures and biological pathway features of drugs as the inputs of our method to predict drug side effects. The results of the cross-validation experiment showed that our method could significantly improve the prediction performance compared to the other state-of-the-art methods. Besides, our model is highly interpretable. It could learn the drug neighbourhood relationships, side effect relationships, and drug features related to side effects. We systematically validated the information extracted by the model with independent data. Some prediction results could also be supported by literature reports. The proposed method could be applied to integrate both chemical and biological data to predict side effects and helps improve drug safety.

Multikinase inhibitor-induced liver injury in patients with cancer: A review for clinicians

BACKGROUND

Multikinase inhibitors (MKI) are targeted molecular agents that have revolutionized cancer management. However, there is a paucity of data concerning MKI-related liver injury risk and clinical guidelines for the management of liver toxicity in patients receiving MKI for cancer are scarce.

DESIGN

We conducted a PubMed search of articles in English published from January 2000 to December 2018 related to hepatotoxicity of the 29 FDA-approved MKIs at doses used in clinical practice. The search terms were the international non-proprietary name of each agent cross-referenced with «hepatotoxicity», «hepatitis», «hepatic adverse event», or «liver failure», and «phase II clinical trial», «phase III clinical trial», or «case report».

RESULTS

Following this search, 140 relevant studies and 99 case reports were considered. Although asymptomatic elevation of aminotransferase levels has been frequently observed in MKI clinical trials, clinically significant hepatotoxicity is a rare event. In most cases, the interval between treatment initiation and the onset of liver injury is between one week and two months. Liver toxicity is often hepatocellular and less frequently mixed. Life-threatening MKI-induced hepatic injury has been described, involving fulminant liver failure or death. Starting from existing data, a description of MKI-related liver events, grading of hepatotoxicity risk, and recommendations for management are also given for various MKI molecules.

CONCLUSION

All MKIs can potentially cause liver injury, which is sometimes irreversible. As there is still no strategy available to prevent MKI-related hepatotoxicity, early detection remains crucial. The surveillance of liver function during treatment may help in the early detection of hepatotoxicity. Furthermore, the exclusion of potential causes of hepatic injury is essential to avoid unnecessary MKI withdrawal.

Survival of patients treated with sorafenib for hepatocellular carcinoma recurrence after liver transplantation: a systematic review and meta-analysis

BACKGROUND

Data on survival and safety of sorafenib for hepatocellular carcinoma recurrence after liver transplant are still equivocal.

AIM

We performed a meta-analysis of published studies, with the aim of estimating the 1-year rates of survival, analysing the variability in survival rates and, finally, identifying the factors associated with a longer survival.

METHODS

Data from 8 of the 17 selected studies were pooled, while the other 9 were excluded because survival rates were missing. All included studies were retrospective.

RESULTS

Overall, the 1-year survival ranged from 18% to 90%. Tumour progression was the main cause of death. The second cause was bleeding, reported only in patients undergoing m-Tor inhibitor therapy. The pooled estimate of 1-year survival was 63%. There was a significant heterogeneity among studies (P < 0.0001). Among the 34 variables assessed by univariate meta-regression, 5 were associated with an increase in the 1-year survival rate: (1) male gender (P = 0.001); (2) Time to progression (P = 0.038); and adverse drug events, divided in (3) gastrointestinal (P = 0.038), (4) cardiovascular (P = 0.029), and (5) dermatological (P = 0.014).

CONCLUSIONS

Additional data from multicentre prospective studies are required to clearly determine if sorafenib is a safe and acceptable treatment in hepatocellular carcinoma recurrence after liver transplant. Nevertheless, its association with m-Tor inhibitors should be discouraged.

Hepatocellular Carcinoma and Liver Transplantation: State of the Art

Hepatocellular carcinoma (HCC) is an aggressive tumor that often occurs in chronic liver disease and cirrhosis. The incidence of HCC is growing worldwide. With respect to any other available treatment for liver cancer, liver transplantation (LT) has the highest potential to cure. LT allows for removal at once of both the tumor ("seed") and the damaged-hepatic tissue ("soil") where cancerogenesis and chronic liver disorders have progressed together. The Milan criteria (MC) have been applied worldwide to select patients with HCC for LT, yielding a 4-year survival rate of 75%. These criteria represent the benchmark for patient selection and are the basis for comparison with any other suggested criteria. However, MC are often considered to be too restrictive, and recent data show that between 25% and 50% of patients with HCC are currently transplanted beyond conventional indications. Consequently, any unrestricted expansion of selection criteria will increase the need for donor organs, lengthen waiting periods, increase drop-out rates, and impair outcomes on intention-to-treat analysis. Management of HCC recurrence after LT is challenging. There are a few reports available regarding the safety and efficacy of sorafenib for HCC recurrence after LT, but the data are heterogeneous. A multi-center prospective randomized controlled trial comparing placebo with sorafenib is advised. Alternatively, a meta-analysis of patient survival with sorafenib for HCC recurrence after LT could be helpful to characterize the therapeutic benefit and safety of sorafenib. Here, we review the use of LT for HCC, with particular emphasis on the selection criteria for transplantation in patients with HCC and management of HCC recurrence after LT.