Deep Learning Classification and Quantification of Pejorative and Nonpejorative Architectures in Resected Hepatocellular Carcinoma from Digital Histopathologic Images.

Liver resection is one of the best treatments for small hepatocellular carcinoma (HCC), but post-resection recurrence is frequent. Biotherapies have emerged as an efficient adjuvant treatment, making the identification of patients at high risk of recurrence critical. Microvascular invasion (mVI), poor differentiation, pejorative macrotrabecular architectures, and vessels encapsulating tumor clusters architectures are the most accurate histologic predictors of recurrence, but their evaluation is time-consuming and imperfect. Herein, a supervised deep learning-based approach with ResNet34 on 680 whole slide images (WSIs) from 107 liver resection specimens was used to build an algorithm for the identification and quantification of these pejorative architectures. This model achieved an accuracy of 0.864 at patch level and 0.823 at WSI level. To assess its robustness, it was validated on an external cohort of 29 HCCs from another hospital, with an accuracy of 0.787 at WSI level, affirming its generalization capabilities. Moreover, the largest connected areas of the pejorative architectures extracted from the model were positively correlated to the presence of mVI and the number of tumor emboli. These results suggest that the identification of pejorative architectures could be an efficient surrogate of mVI and have a strong predictive value for the risk of recurrence. This study is the first step in the construction of a composite predictive algorithm for early post-resection recurrence of HCC, including artificial intelligence-based features.

Evaluation of the interest to combine a CD4 Th1-inducer cancer vaccine derived from telomerase and atezolizumab plus bevacizumab in unresectable hepatocellular carcinoma: a randomized non-comparative phase II study (TERTIO - PRODIGE 82).

BACKGROUND: Several cancer immunotherapies that target the PD-L1/PD-1 pathway show promising clinical activity in patients with hepatocellular carcinoma (HCC). However, the standard of care in first-line treatment with atezolizumab (anti-PD-L1 therapy) in combination with bevacizumab is associated with a limited objective response rate. Telomerase reverse transcriptase (TERT) activation meets the criteria of oncogenic addiction in HCC and could be actionable therapeutic target and a relevant tumor antigen. Therefore we hypothesized that combining anti-PD-1/PD-L1 therapy with an anti-telomerase vaccine might be an attractive therapy in HCC. UCPVax is a therapeutic cancer vaccine composed of two separate peptides derived from telomerase (human TERT). UCPVax has been evaluated in a multicenter phase I/II study in non-small cell lung cancers and has demonstrated to be safe and immunogenic, and is under evaluation in combination with atezolizumab in a phase II clinical trial in tumors where telomerase reactivation contributes to an oncogene addiction (HPV+ cancers). The aim of the TERTIO study is to determine the clinical interest and immunological efficacy of a treatment combining the CD4 helper T-inducer cancer anti-telomerase vaccine (UCPVax) with atezolizumab and bevacizumab in unresectable HCC in a multicenter randomized phase II study. METHODS: Patients with locally advanced, metastatic or unresectable HCC who have not previously received systemic anti-cancer treatment are eligible. The primary end point is the objective response rate at 6 months. Patients will be allocated to a treatment arm with a randomization 2:1. In both arms, patients will receive atezolizumab at fixed dose of 1200 mg IV infusion and bevacizumab at fixed dose of 15 mg/kg IV infusion, every 3 weeks, according to the standard of care. In the experimental arm, these treatments will be combined with the UCPVax vaccine at 0.5 mg subcutaneously. DISCUSSION: Combining anti-PD-1/PD-L1 therapy with an anti-telomerase vaccine gains serious consideration in HCC, in order to extend the clinical efficacy of anti-PD-1/PD-L1. Indeed, anti-cancer vaccines can induce tumor-specific T cell expansion and activation and therefore restore the cancer-immunity cycle in patients lacking pre-existing anti-tumor responses. Thus, there is a strong rational to combine immune checkpoint blockade therapy and anticancer vaccine (UCPVax) in order to activate antitumor T cell immunity and bypass the immunosuppression in the tumor microenvironment in HCC. This pivotal proof of concept study will evaluate the efficacy and safety of the combination of a CD4 Th1-inducer cancer vaccine derived from telomerase (UCPVax) and atezolizumab plus bevacizumab in unresectable HCC, as well as confirming their synergic mechanism, and settling the basis for a new combination for future clinical trials. TRIAL REGISTRATION: NCT05528952.

Indocyanine green fluorescence patterns of hepatocellular carcinoma correlate with pathological and molecular features.

BACKGROUND: Intraoperative Indocyanine Green Dye (ICG) routinely used in hepatobiliary surgery identifies different fluorescent patterns of hepatocellular carcinoma (HCC), a highly heterogeneous cancer. We aimed to correlate these patterns with gene mutations and extensive pathological features beyond the well-known tumor differentiation. METHODS: Between February 2017 and December 2019, 21 HCC in 16 consecutive patients who underwent intraoperative ICG fluorescence imaging were included. Pathological review was performed by one pathologist blinded to fluorescence features. Random forest machine learning algorithm correlated pathological features of the tumor, peritumoral and non-tumoral liver, and gene mutations from a 28 gene-panel with rim and intra-lesion fluorescence. RESULTS: Three HCC had negative intra-lesion and rim-like emission, 7 HCC had homogeneous pattern and 11 heterogeneous patterns in whom 3 with rim-like emission. Rim emission was associated with peritumoral vascular changes, lower differentiation and lower serum AFP level. Homogeneous intra-lesion fluorescence was associated with lower necrosis rate, thinner capsule, absence of peritumoral liver changes, and higher serum AFP level. Heterogeneous HCC without rim harbored lesser TP53 and ARID1A mutations. CONCLUSION: Tumoral and peri-tumoral fluorescence classification of HCC yielded a possible intraoperative pathological and molecular characterization. These preliminary observations could lead to intraoperative refinement in surgical strategy.

Nivolumab versus sorafenib in advanced hepatocellular carcinoma (CheckMate 459): a randomised, multicentre, open-label, phase 3 trial.

BACKGROUND: Single-agent nivolumab showed durable responses, manageable safety, and promising survival in patients with advanced hepatocellular carcinoma in the phase 1-2 CheckMate 040 study. We aimed to investigate nivolumab monotherapy compared with sorafenib monotherapy in the first-line setting for patients with advanced hepatocellular carcinoma. METHODS: In this randomised, open-label, phase 3 trial done at medical centres across 22 countries and territories in Asia, Australasia, Europe, and North America, patients at least 18 years old with histologically confirmed advanced hepatocellular carcinoma not eligible for, or whose disease had progressed after, surgery or locoregional treatment; with no previous systemic therapy for hepatocellular carcinoma, with Child-Pugh class A and Eastern Cooperative Oncology Group performance status score of 0 or 1, and regardless of viral hepatitis status were randomly assigned (1:1) via an interactive voice response system to receive nivolumab (240 mg intravenously every 2 weeks) or sorafenib (400 mg orally twice daily) until disease progression or unacceptable toxicity. The primary endpoint was overall survival assessed in the intention-to-treat population. Safety was assessed in all patients who received at least one dose of study drug. This completed trial is registered with ClinicalTrials.gov, NCT02576509. FINDINGS: Between Jan 11, 2016, and May 24, 2017, 743 patients were randomly assigned to treatment (nivolumab, n=371; sorafenib, n=372). At the primary analysis, the median follow-up for overall survival was 15·2 months (IQR 5·7-28·0) for the nivolumab group and 13·4 months (5·7-25·9) in the sorafenib group. Median overall survival was 16·4 months (95% CI 13·9-18·4) with nivolumab and 14·7 months (11·9-17·2) with sorafenib (hazard ratio 0·85 [95% CI 0·72-1·02]; p=0·075; minimum follow-up 22·8 months); the protocol-defined significance level of p=0·0419 was not reached. The most common grade 3 or worse treatment-related adverse events were palmar-plantar erythrodysaesthesia (1 [<1%] of 367 patients in the nivolumab group vs 52 [14%] of patients in the sorafenib group), aspartate aminotransferase increase (22 [6%] vs 13 [4%]), and hypertension (0 vs 26 [7%]). Serious treatment-related adverse events were reported in 43 (12%) patients receiving nivolumab and 39 (11%) patients receiving sorafenib. Four deaths in the nivolumab group and one death in the sorafenib group were assessed as treatment related. INTERPRETATION: First-line nivolumab treatment did not significantly improve overall survival compared with sorafenib, but clinical activity and a favourable safety profile were observed in patients with advanced hepatocellular carcinoma. Thus, nivolumab might be considered a therapeutic option for patients in whom tyrosine kinase inhibitors and antiangiogenic drugs are contraindicated or have substantial risks. FUNDING: Bristol Myers Squibb in collaboration with Ono Pharmaceutical.

Number of hepatocellular carcinoma nodules in patients listed for liver transplantation within alpha-fetoprotein score: a new prognostic risk factor.

In France, the listing for liver transplantation (LT) for hepatocellular carcinoma (HCC) requires an AFP score ≤2. This study evaluates whether the number of nodules assessed immediately before LT has a prognostic value among patients already listed within AFP score. Among 143 recipients transplanted with an AFP score ≤2 between 2013 and 2017 in our center, the number of nodules was considered at listing on the waiting list and at last imaging before LT. We compared the overall survival (OS) and disease-free survival (DFS) post-LT of patients with ≤3 and >3 nodules (current classification), and aimed to propose a new criteria to exclude patients on list at high risk of recurrence. The 3-year OS of patients with ≤3 HCC vs. >3 HCC at listing was of 90.3% vs. 67.3%, respectively (P = 0.04). At last imaging, eight listed patients presented ≥5 HCC nodules and had a significantly lower OS than <5 nodules patients (5-year OS: 24.4% vs. 78.1%; P = 0.01). Although the current AFP score offers satisfactory outcomes, we highlight the poorer outcomes when ≥5 nodules persist or appear after listing. A modification of the AFP score is mandatory to consider exclusion of high-risk patients already listed for LT program.

High trough levels of everolimus combined to sorafenib improve patients survival after hepatocellular carcinoma recurrence in liver transplant recipients.

Recurrence of hepatocellular carcinoma (HCC) following liver transplantation (LT) occurs in 10%-20% of patients transplanted for HCC. The treatment of HCC recurrence after LT remains a challenge. Consecutive patients who underwent LT for HCC between 2005 and 2015 at our center were recruited. Characteristics of patients with recurrence, modalities of treatment and outcome were collected retrospectively. Patient survival was analyzed according to HCC recurrence therapeutic strategy. Among 306 transplanted patients, 43 patients (14.1%) developed recurrence with a median survival time after recurrence of 10.9 months (95%CI: 6.6-18.6). Survival of patients treated with Sorafenib (SOR) and everolimus (EVL) (n = 19) was significantly better than that of the group treated with other strategies (n = 24) (P = 0.001). Multivariable analysis demonstrated that SOR plus EVL therapy and absence of dissemination at diagnosis of recurrence were independent predictive factors of prolonged survival after recurrence. Among the patients who treated with EVL, survival of patients with controlled EVL blood trough levels ≥5 ng/ml was significantly better compared to those with EVL trough levels <5 ng/ml (P = 0.021). Combination therapy of sorafenib and everolimus was an independent predictor for better survival after HCC recurrence. Patients with controlled everolimus trough level ≥5 ng/ml might get the best survival benefit.

Liver Resection for Early Hepatocellular Carcinoma: Preoperative Predictors of Non Transplantable Recurrence and Implications for Treatment Allocation.

BACKGROUND AND AIMS: LR and LT are the standard curative options for early HCC. LT provides best long-term survival but is limited by organ shortage. LR, readily available, is hampered by high recurrence rates. Salvage liver transplantation is an efficient treatment of recurrences within criteria. The aim of the study was to identify preoperative predictors of non transplantable recurrence (NTR) to improve patient selection for upfront LR or LT at initial diagnosis. STUDY DESIGN: Consecutive LR for transplantable HCC between 2000 and 2015 were studied. A prediction model for NTR based on preoperative variables was developed using sub-distribution hazard ratio after multiple imputation and internal validation by bootstrapping. Model performance was evaluated by the concordance index after correction for optimism. RESULTS: A total of 148 patients were included. Five-year overall survival and recurrence free survival were 73.6% and 29.3%, respectively (median follow-up 45.8 months). Recurrence rate was 54.8%. NTR rate was 38.2%. Preoperative model for NTR identified >1 nodule [sub-distribution hazard ratio 2.35 95% confidence interval (CI) 1.35-4.09], AFP >100 ng/mL (2.14 95% CI 1.17-3.93), and F4 fibrosis (1.93 95% CI 1.03-3.62). The apparent concordance index of the model was 0.664 after correction for optimism. In the presence of 0, 1, and ≥2 factors, NTR rates were 2.6%, 22.7%, and 40.9%, respectively. The number of prognostic factors was significantly associated with the pattern of recurrence (P = 0.001) and 5-year recurrence free survival (P < 0.001). CONCLUSIONS: Cirrhosis, >1 nodule, and AFP >100 ng/mL were identified as preoperative predictors of NTR. In the presence of 2 factors or more upfront transplantation should be probably preferred to resection in regard of organ availability. Other patients are good candidates for LR and salvage liver transplantation should be encouraged in eligible patients with recurrence.

Hepatocellular Carcinomas With Mutational Activation of Beta-Catenin Require Choline and Can Be Detected by Positron Emission Tomography.

BACKGROUND & AIMS: In one-third of hepatocellular carcinomas (HCCs), cancer cells have mutations that activate ß-catenin pathway. These cells have alterations in glutamine, bile, and lipid metabolism. We investigated whether positron emission tomography (PET) imaging allows identification of altered metabolic pathways that might be targeted therapeutically. METHODS: We studied mice with activation of ß-catenin in liver (Apcko-liv mice) and male C57Bl/6 mice given injections of diethylnitrosamine, which each develop HCCs. Mice were fed a conventional or a methionine- and choline-deficient diet or a choline-deficient (CD) diet. Choline uptake and metabolism in HCCs were analyzed by micro-PET imaging of mice; livers were collected and analyzed by histologic, metabolomic, messenger RNA quantification, and RNA-sequencing analyses. Fifty-two patients with HCC underwent PET imaging with 18F-fluorodeoxyglucose, followed by 18F-fluorocholine tracer metabolites. Human HCC specimens were analyzed by immunohistochemistry, quantitative polymerase chain reaction, and DNA sequencing. We used hepatocytes and mouse tumor explants for studies of incorporation of radiolabeled choline into phospholipids and its contribution to DNA methylation. We analyzed HCC progression in mice fed a CD diet. RESULTS: Livers and tumors from Apcko-liv mice had increased uptake of dietary choline, which contributes to phospholipid formation and DNA methylation in hepatocytes. In patients and in mice, HCCs with activated ß-catenin were positive in 18F-fluorocholine PET, but not 18F-fluorodeoxyglucose PET, and they overexpressed the choline transporter organic cation transporter 3. The HCC cells from Apcko-liv mice incorporated radiolabeled methyl groups of choline into phospholipids and DNA. In Apcko-liv mice, the methionine- and choline-deficient diet reduced proliferation and DNA hypermethylation of hepatocytes and HCC cells, and the CD diet reduced long-term progression of tumors. CONCLUSIONS: In mice and humans, HCCs with mutations that activate ß-catenin are characterized by increased uptake of a fluorocholine tracer, but not 18F-fluorodeoxyglucose, revealed by PET. The increased uptake of choline by HCCs promotes phospholipid formation, DNA hypermethylation, and hepatocyte proliferation. In mice, the CD diet reverses these effects and promotes regression of HCCs that overexpress ß-catenin.

Pharmacokinetics and safety of olaparib in patients with advanced solid tumours and mild or moderate hepatic impairment.

AIMS: Olaparib, a potent oral poly(ADP-ribose) polymerase inhibitor, is partially hepatically cleared. We investigated the pharmacokinetics (PK) and safety of olaparib in patients with mild or moderate hepatic impairment to provide dosing recommendations. METHODS: This Phase I open-label study assessed the PK, safety and tolerability of single doses of olaparib 300-mg tablets in patients with advanced solid tumours. Patients had normal hepatic function (NHF), or mild (MiHI; Child-Pugh class A) or moderate (MoHI; Child-Pugh class B) hepatic impairment. Blood was collected for PK assessments for 96 hours. Patients could continue taking olaparib 300 mg twice daily for long-term safety assessment. RESULTS: Thirty-one patients received ≥1 dose of olaparib and 30 were included in the PK assessment. Patients with MiHI had an area under the curve geometric least-squares mean (GLSmean) ratio of 1.15 (90% confidence interval 0.72, 1.83) and a GLSmean maximum plasma concentration ratio of 1.13 (0.82, 1.56) vs those with NHF. In patients with MoHI, GLSmean ratio for area under the curve was 1.08 (0.66, 1.74) and for maximum plasma concentration was 0.87 (0.63, 1.22) vs those with NHF. For patients with mild or moderate hepatic impairment, no new safety signals were detected. CONCLUSION: Patients with MiHI or MoHI had no clinically significant changes in exposure to olaparib compared with patients with NHF. The safety profile of olaparib did not differ from a clinically relevant extent between cohorts. No olaparib tablet or capsule dose reductions are required for patients with MiHI or MoHI.

Impact on Oncological Outcomes and Intent-to-Treat Survival of Resection Margin for Transplantable Hepatocellular Carcinoma in All-Comers and in Patients with Cirrhosis: A Multicenter Study.

BACKGROUND: The outcomes of liver resection (LR) with a narrow margin in patients with transplantable hepatocellular carcinoma (HCC) have not been studied. The aim was to assess whether narrow margin following up-front LR impacts the incidence, timing, pattern, and transplantability of tumor recurrence in patients with initially transplantable HCC. METHODS: All initially transplantable HCC patients undergoing hepatectomy with either narrow (<10 mm) or wide (≥10 mm) margins from 2007 to 2016 at four Western university centers were compared in terms of recurrence, transplantability of recurrence, recurrence-free survival (RFS), and intention-to-treat overall survival (ITT-OS). Independent predictors of non-transplantability of recurrence were assessed. RESULTS: This study included 187 patients (narrow group, n = 107 vs. wide group, n = 80). Recurrence was significantly more frequent in the narrow margin group (44% vs. 26%; p = 0.01) with a shorter RFS (p = 0.03). The transplantability of recurrence and ITT-OS were, however, not different between the two groups. The presence of satellite nodules on the resected specimens emerged as the sole independent predictor of non-transplantability of tumor recurrence. The stratification of the analysis according to the presence of cirrhosis achieved essentially the same results as in the whole study population. CONCLUSIONS: Narrow margin was associated with a higher tumor recurrence rate and a shorter RFS for patients with initially transplantable HCC. However, transplantability of recurrence and long-term ITT-OS were not impaired.

Impact of combined selective internal radiation therapy and sorafenib on survival in advanced hepatocellular carcinoma.

BACKGROUND & AIMS: Sorafenib is the recommended treatment for patients with advanced hepatocellular carcinoma (HCC). We aimed to compare the efficacy and safety of a combination of sorafenib and selective internal radiation therapy (SIRT) - with yttrium-90 (90Y) resin microspheres - to sorafenib alone in patients with advanced HCC. METHODS: SORAMIC is a randomised controlled trial comprising diagnostic, local ablation and palliative cohorts. Based on diagnostic study results, patients were assigned to local ablation or palliative cohorts. In the palliative cohort, patients not eligible for TACE were randomised 11:10 to SIRT plus sorafenib (SIRT + sorafenib) or sorafenib alone. The primary endpoint was overall survival (OS; Kaplan-Meier analysis) in the intention-to-treat (ITT) population. RESULTS: In the ITT cohort, 216 patients were randomised to SIRT + sorafenib and 208 to sorafenib alone. Median OS was 12.1 months in the SIRT + sorafenib arm, and 11.4 months in the sorafenib arm (hazard ratio [HR] 1.01; 95% CI 0.81-1.25; p = 0.9529). Median OS in the per protocol population was 14.0 months in the SIRT + sorafenib arm (n = 114), and 11.1 months in the sorafenib arm (n = 174; HR 0.86; p = 0.2515). Subgroup analyses of the per protocol population indicated a survival benefit of SIRT + sorafenib for patients without cirrhosis (HR 0.46; 0.25-0.86; p = 0.02); cirrhosis of non-alcoholic aetiology (HR 0.63; p = 0.012); or patients ≤65 years old (HR 0.65; p = 0.05). Adverse events (AEs) of Common Terminology Criteria for AE Grades 3-4 were reported in 103/159 (64.8%) patients who received SIRT + sorafenib, 106/197 (53.8%) patients who received sorafenib alone (p = 0.04), and 8/24 (33.3%) patients who only received SIRT. CONCLUSION: Addition of SIRT to sorafenib did not result in a significant improvement in OS compared with sorafenib alone. Subgroup analyses led to hypothesis-generating results that will support the design of future studies. LAY SUMMARY: Sorafenib given orally is the recommended treatment for patients with advanced hepatocellular carcinoma (HCC). In selective internal radiation therapy (SIRT), also known as radioembolisation, microscopic, radioactive resin or glass spheres are introduced into the blood vessels that feed the tumours in the liver. This study found that the addition of SIRT with 90yttrium-loaded resin microspheres to sorafenib treatment in people with advanced HCC did not significantly improve overall survival compared with sorafenib treatment alone. However, the results give an indication of how future studies using this combination therapy in people with advanced HCC could be designed. STUDY REGISTRATION: EudraCT 2009-012576-27, NCT0112 6645.

Outcomes of sequential treatment with sorafenib followed by regorafenib for HCC: Additional analyses from the phase III RESORCE trial.

BACKGROUND & AIMS: The RESORCE trial showed that regorafenib improves overall survival (OS) in patients with hepatocellular carcinoma progressing during sorafenib treatment (hazard ratio [HR] 0.62, 95% confidence interval [CI] 0.50-0.78; p <0.0001). This exploratory analysis describes outcomes of sequential treatment with sorafenib followed by regorafenib. METHODS: In RESORCE, 573 patients were randomized 2:1 to regorafenib 160 mg/day or placebo for 3 weeks on/1 week off. Efficacy and safety were evaluated by last sorafenib dose. The time from the start of sorafenib to death was assessed. Time to progression (TTP) in RESORCE was analyzed by TTP during prior sorafenib treatment. RESULTS: HRs (regorafenib/placebo) for OS by last sorafenib dose were similar (0.67 for 800 mg/day; 0.68 for <800 mg/day). Rates of grade 3, 4, and 5 adverse events with regorafenib by last sorafenib dose (800 mg/day vs. <800 mg/day) were 52%, 11%, and 15% vs. 60%, 10%, and 12%, respectively. Median times (95% CI) from the start of sorafenib to death were 26.0 months (22.6-28.1) for regorafenib and 19.2 months (16.3-22.8) for placebo. Median time from the start of sorafenib to progression on sorafenib was 7.2 months for the regorafenib arm and 7.1 months for the placebo arm. An analysis of TTP in RESORCE in subgroups defined by TTP during prior sorafenib in quartiles (Q) showed HRs (regorafenib/placebo; 95% CI) of 0.66 (0.45-0.96; Q1); 0.26 (0.17-0.40; Q2); 0.40 (0.27-0.60; Q3); and 0.54 (0.36-0.81; Q4). CONCLUSIONS: These exploratory analyses show that regorafenib conferred a clinical benefit regardless of the last sorafenib dose or TTP on prior sorafenib. Rates of adverse events were generally similar regardless of the last sorafenib dose. LAY SUMMARY: This analysis examined characteristics and outcomes of patients with hepatocellular carcinoma who were treated with regorafenib after they had disease progression during sorafenib treatment. Regorafenib provided clinical benefit to patients regardless of the pace of their disease progression during prior sorafenib treatment and regardless of their last sorafenib dose. The sequence of sorafenib followed by regorafenib for hepatocellular carcinoma may extend survival beyond what has been previously reported. ClinicalTrials.gov NCT01774344.

Regorafenib for patients with hepatocellular carcinoma who progressed on sorafenib treatment (RESORCE): a randomised, double-blind, placebo-controlled, phase 3 trial.

BACKGROUND: There are no systemic treatments for patients with hepatocellular carcinoma (HCC) whose disease progresses during sorafenib treatment. We aimed to assess the efficacy and safety of regorafenib in patients with HCC who have progressed during sorafenib treatment. METHODS: In this randomised, double-blind, parallel-group, phase 3 trial done at 152 sites in 21 countries, adults with HCC who tolerated sorafenib (≥400 mg/day for ≥20 of last 28 days of treatment), progressed on sorafenib, and had Child-Pugh A liver function were enrolled. Participants were randomly assigned (2:1) by a computer-generated randomisation list and interactive voice response system and stratified by geographical region, Eastern Cooperative Oncology Group performance status, macrovascular invasion, extrahepatic disease, and α-fetoprotein level to best supportive care plus oral regorafenib 160 mg or placebo once daily during weeks 1-3 of each 4-week cycle. Investigators, patients, and the funder were masked to treatment assignment. The primary endpoint was overall survival (defined as time from randomisation to death due to any cause) and analysed by intention to treat. This trial is registered with ClinicalTrials.gov, number NCT01774344. FINDINGS: Between May 14, 2013, and Dec 31, 2015, 843 patients were screened, of whom 573 were enrolled and randomised (379 to regorafenib and 194 to placebo; population for efficacy analyses), and 567 initiated treatment (374 received regorafenib and 193 received placebo; population for safety analyses). Regorafenib improved overall survival with a hazard ratio of 0·63 (95% CI 0·50-0·79; one-sided p<0·0001); median survival was 10·6 months (95% CI 9·1-12·1) for regorafenib versus 7·8 months (6·3-8·8) for placebo. Adverse events were reported in all regorafenib recipients (374 [100%] of 374) and 179 (93%) of 193 placebo recipients. The most common clinically relevant grade 3 or 4 treatment-emergent events were hypertension (57 patients [15%] in the regorafenib group vs nine patients [5%] in the placebo group), hand-foot skin reaction (47 patients [13%] vs one [1%]), fatigue (34 patients [9%] vs nine patients [5%]), and diarrhoea (12 patients [3%] vs no patients). Of the 88 deaths (grade 5 adverse events) reported during the study (50 patients [13%] assigned to regorafenib and 38 [20%] assigned to placebo), seven (2%) were considered by the investigator to be related to study drug in the regorafenib group and two (1%) in the placebo group, including two patients (1%) with hepatic failure in the placebo group. INTERPRETATION: Regorafenib is the only systemic treatment shown to provide survival benefit in HCC patients progressing on sorafenib treatment. Future trials should explore combinations of regorafenib with other systemic agents and third-line treatments for patients who fail or who do not tolerate the sequence of sorafenib and regorafenib. FUNDING: Bayer.

Growth factor receptor binding protein 14 inhibition triggers insulin-induced mouse hepatocyte proliferation and is associated with hepatocellular carcinoma.

Metabolic diseases such as obesity and type 2 diabetes are recognized as independent risk factors for hepatocellular carcinoma (HCC). Hyperinsulinemia, a hallmark of these pathologies, is suspected to be involved in HCC development. The molecular adapter growth factor receptor binding protein 14 (Grb14) is an inhibitor of insulin receptor catalytic activity, highly expressed in the liver. To study its involvement in hepatocyte proliferation, we specifically inhibited its liver expression using a short hairpin RNA strategy in mice. Enhanced insulin signaling upon Grb14 inhibition was accompanied by a transient induction of S-phase entrance by quiescent hepatocytes, indicating that Grb14 is a potent repressor of cell division. The proliferation of Grb14-deficient hepatocytes was cell-autonomous as it was also observed in primary cell cultures. Combined Grb14 down-regulation and insulin signaling blockade using pharmacological approaches as well as genetic mouse models demonstrated that Grb14 inhibition-mediated hepatocyte division involved insulin receptor activation and was mediated by the mechanistic target of rapamycin complex 1-S6K pathway and the transcription factor E2F1. In order to determine a potential dysregulation in GRB14 gene expression in human pathophysiology, a collection of 85 human HCCs was investigated. This revealed a highly significant and frequent decrease in GRB14 expression in hepatic tumors when compared to adjacent nontumoral parenchyma, with 60% of the tumors exhibiting a reduced Grb14 mRNA level. CONCLUSION: Our study establishes Grb14 as a physiological repressor of insulin mitogenic action in the liver and further supports that dysregulation of insulin signaling is associated with HCC. (Hepatology 2017;65:1352-1368).

Somatostatin receptors in resected hepatocellular carcinoma: status and correlation with markers of poor prognosis.

AIMS: To investigate the status of somatostatin receptors (SSTRs) in resected hepatocellular carcinoma (HCC). METHODS AND RESULTS: Transcript and protein levels of SSTR2, SSTR3 and SSTR5 were investigated, with real-time polymerase chain reaction (PCR) and manual and automated immunohistochemistry (IHC), in 53 resected HCCs and paired non-tumour tissues. SSTR1, SSTR4, SSTR5TMD4 and SSTR5TMD5 were analysed with real-time PCR. SSTR3 and SSTR5 transcripts were expressed in ~25% of HCCs, but not in adjacent non-tumour tissues. SSTR1 and SSTR2 transcripts were overexpressed in 42% and 32% of HCCs, respectively. SSTR4, SSTR5TMD4 and SSTR5TMD5 were not detected. Membrane staining for SSTR2 was detected in 38% of HCCs, whereas SSTR5 protein was detectable in only 11% of HCCs. SSTR3 protein was detected in the majority of HCCs and adjacent non-tumour liver tissues, but membrane staining was <20% of that in HCCs. The results obtained with the two IHC methods were highly correlated (P < 0.0001). Statistical analyses also showed a positive correlation between SSTR2 membrane staining and cytokeratin 19 expression (P = 0.04), serum α-fetoprotein level (P = 0.002), and poor differentiation (P = 0.05). CONCLUSIONS: Membrane SSTR2 is detected reliably in HCCs by IHC, and is a potential therapeutic target, as it is coexpressed with markers of poor prognosis.

Liver biopsy for diagnosis of presumed benign hepatocellular lesions lacking magnetic resonance imaging diagnostic features of focal nodular hyperplasia.

BACKGROUND & AIMS: The contribution of liver biopsy for the diagnosis of presumed benign hepatocellular lesions lacking the diagnostic features of focal nodular hyperplasia (FNH) on magnetic resonance imaging (MRI) is unknown. We evaluated liver biopsy and MRI performances in this setting. METHODS: Magnetic resonance imaging and slides of liver biopsies performed for a presumed benign hepatocellular lesion (2006-2013) without the typical features of FNH on MRI were blindly reviewed (n = 45). Eighteen lesions were surgically removed and also analyzed. The final diagnosis was the diagnosis established after surgery or on the biopsy in the absence of surgery. RESULTS: The final diagnosis was FNH (n = 19), hepatocellular adenoma (HCA, n = 15), hepatocellular carcinoma (n = 3) and indefinite (n = 4). Four lesions corresponded to non hepatocellular lesions. FNH, HNF1A mutated and inflammatory HCA were diagnosed accurately on the biopsy in 95%, 67% and 100% of the cases respectively. Diagnostic performance of liver biopsy for HNF1A mutated HCA was lower because of the lack of non-tumoral tissue. Diagnosis based on morphological analysis was certain and correct in 27 cases. Immunostaining allowed a definite diagnosis in 12 additionnal cases. Radiological diagnosis was in agreement with the histological diagnosis in 75.6% of the cases, with a very high sensitivity (97%) and specificity (100%) for the diagnosis of HNF1A mutated HCA. CONCLUSIONS: Liver biopsy has a good diagnostic performance particularly for FNH and inflammatory HCA, and sampling of non-lesional tissue is highly recommended. A biopsy does not seem necessary if H-HCA is diagnosed on MRI.

Early and resectable HCC: Definition and validation of a subgroup of patients who could avoid liver transplantation.

BACKGROUND: Liver transplantation (LT) remains the best curative option for early hepatocellular carcinoma (HCC) but is limited by the ongoing graft shortage. The present study aimed at defining the population in which primary liver resection (LR) could represent the best alternative to LT. METHODS: An exploration set of 357 HCC patients (LR n = 221 and LT n = 136) operated between 2000-2012 was used in order to identify factors associated with survival following LR and define a good prognosis (GP) group for which LR may challenge the results of upfront LT. These factors were validated in an external validation set of 565 HCC patients operated at another center (LR n = 287 LR and LT n = 278). RESULTS: In the exploration set, factors associated with survival on multivariate analysis were a solitary lesion, a diameter <50 mm, a well-moderately differentiated lesion, the absence of microvascular invasion, and preoperative AST level <2N. Thirty-nine patients (18%) displayed all these criteria and constituted the GP patients. Overall survivals at 1, 3, and 5 years did not significantly differ between GP resected patients, and the in Milan transplanted patients (93, 80.4, and 80.4% vs. 86.9, 82, and 78.8%, P = 0.79). In the validation cohort, patients with GP factors of survival still displayed better overall survivals than those without (P = 0.036) but also displayed better survivals than in Milan HCC transplanted patients (P = 0.005). CONCLUSION: In a group of early HCC patients gathering all factors of GP, primary LR achieves at least similar survival as upfront LT and should be the approach of choice.

Gemcitabine and oxaliplatin with or without cetuximab in advanced biliary-tract cancer (BINGO): a randomised, open-label, non-comparative phase 2 trial.

BACKGROUND: Gemcitabine plus a platinum-based agent (eg, cisplatin or oxaliplatin) is the standard of care for advanced biliary cancers. We investigated the addition of cetuximab to chemotherapy in patients with advanced biliary cancers. METHODS: In this non-comparative, open-label, randomised phase 2 trial, we recruited patients with locally advanced (non-resectable) or metastatic cholangiocarcinoma, gallbladder carcinoma, or ampullary carcinoma and a WHO performance status of 0 or 1 from 18 hospitals across France and Germany. Eligible patients were randomly assigned (1:1) centrally with a minimisation procedure to first-line treatment with gemcitabine (1000 mg/m(2)) and oxaliplatin (100 mg/m(2)) with or without cetuximab (500 mg/m(2)), repeated every 2 weeks until disease progression or unacceptable toxicity. Randomisation was stratified by centre, primary site of disease, disease stage, and previous treatment with curative intent or adjuvant therapy. Investigators who assessed treatment response were not masked to group assignment. The primary endpoint was the proportion of patients who were progression-free at 4 months, analysed by intention to treat. This study is registered with ClinicalTrials.gov, number NCT00552149. FINDINGS: Between Oct 10, 2007, and Dec 18, 2009, 76 patients were assigned to chemotherapy plus cetuximab and 74 to chemotherapy alone. 48 (63%; 95% CI 52-74) patients assigned to chemotherapy plus cetuximab and 40 (54%; 43-65) assigned to chemotherapy alone were progression-free at 4 months. Median progression-free survival was 6·1 months (95% CI 5·1-7·6) in the chemotherapy plus cetuximab group and 5·5 months (3·7-6·6) in the chemotherapy alone group. Median overall survival was 11·0 months (9·1-13·7) in the chemotherapy plus cetuximab group and 12·4 months (8·6-16·0) in the chemotherapy alone group. The most common grade 3-4 adverse events were peripheral neuropathy (in 18 [24%] of 76 patients who received chemotherapy plus cetuximab vs ten [15%] of 68 who received chemotherapy alone), neutropenia (17 [22%] vs 11 [16%]), and increased aminotransferase concentrations (17 [22%] vs ten [15%]). 70 serious adverse events were reported in 39 (51%) of 76 patients who received chemotherapy plus cetuximab (34 events in 19 [25%] patients were treatment-related), whereas 41 serious adverse events were reported in 25 (35%) of 71 patients who received chemotherapy alone (20 events in 12 [17%] patients were treatment-related). One patient died of atypical pneumonia related to treatment in the chemotherapy alone group. INTERPRETATION: The addition of cetuximab to gemcitabine and oxaliplatin did not seem to enhance the activity of chemotherapy in patients with advanced biliary cancer, although it was well tolerated. Gemcitabine and platinum-based combination should remain the standard treatment option. FUNDING: Institut National du Cancer, Merck Serono.

Genotype-phenotype relationships in the low-phospholipid-associated cholelithiasis syndrome: a study of 156 consecutive patients.

UNLABELLED: The low-phospholipid-associated cholelithiasis syndrome (LPAC; OMIM 171060) is a peculiar form of intrahepatic cholelithiasis occurring in young adults, associated with ABCB4/MDR3 gene sequence variations. Our aim was to determine the genotype-phenotype relationships in 156 consecutive patients with the criteria of LPAC syndrome. A variant was detected in 79 (61 missense and 18 truncating sequence variants), 63 being monoallelic. The clinical features (age at onset, high prevalence in women, frequency and severity of acute and chronic complications, intrahepatic cholestasis of pregnancy [ICP]) were similar in the patients with or without ABCB4 gene sequence variation. Truncating variations were associated with an earlier onset of symptoms both in women and men. Acute and chronic biliary complications were variant-independent. Half of the women who had pregnancy developed ICP. The frequency of ICP and fetal complications were similar in patients with missense and truncating variants. CONCLUSION: The LPAC syndrome is more frequent in women and highly associated with ICP. Half of the patients harbored missense or truncating variants of the ABCB4 gene. The characteristics of the patients without detectable variant are similar to those with variant, indicating that yet unexplored regions of the ABCB4 and other genes may be involved.

[Metabolic syndrome, non alcoholic hepatic steatopathy and hepatocellular carcinoma: so dangerous liaisons]. / Syndrome métabolique, stéatopathie hépatique non alcoolique (NASH) et carcinome hépatocellulaire: des liaisons bien dangereuses....

Obesity and the metabolic syndrome are fast-growing epidemics associated with an increased risk for many types of cancer. In the liver, inflammatory and angiogenic changes due to insulin resistance and fatty liver disease are associated with an increased incidence of liver cancer. Regardless of underlying liver disease, cirrhosis remains the most important risk factor for hepatocellular carcinoma (HCC) although cases of HCC arising without cirrhosis do not exclude the possibility of a direct carcinogenesis secondary to non-alcoholic fatty liver disease. Moreover, metabolic syndrome and its different features may also increase the risk of HCC in the setting of chronic liver diseases of other causes, such as viral hepatitis or alcohol abuse. Taking into account all these data, it is necessary to better determine the risk of developing HCC in patients with metabolic syndrome to improve the screening guidelines and develop prophylactic treatments in this setting.