Que, Florencia G.

Spontaneous hemorrhage from hepatic tumors is an uncommon but serious complication. Recently, interventional radiologic (IR) techniques are being used increasingly in the management of these patients. We report our 16-year experience in managing spontaneous hemorrhage from liver tumors. Twenty-six consecutive patients were diagnosed with spontaneous liver hemorrhage between 1995 and 2011. Initial management was operative in eight, IR in six, and supportive in 12 patients. Of those managed operatively, five were segmentectomies; one hemihepatectomy; one wedge resection; and one packing who later died from coagulopathy. In the IR patients, seven had an angiographic embolization; two required reembolization; one underwent resection of a hepatic adenoma 21 days after angiographic embolization. The malignant lesions included hepatocellular carcinoma (n = 6), angiosarcoma (n = 1), metastatic squamous cell carcinoma (n = 1), metastatic leiomyosarcoma (n = 1), nonsquamous cell carcinoma (n = 1), or metastatic angiosarcoma (n = 1). Benign diseases included hepatic adenoma (n = 5), end-stage liver disease (n = 1), and polycystic liver (n = 1). Spontaneous hemorrhage from the liver occurs evenly from benign or malignant causes, one-third of which are primary liver disease. If the patients presents emergently, angiographic embolization may control the bleeding and allow for elective resection once the sequelae of bleeding have resolved.

Hepatocytes do not express Bcl-2, a repressor of apoptosis. In contrast, cholangiocytes, which are in direct contact with bile, do express Bcl-2. Because cholestasis results in the retention of bile within hepatocytes, we reasoned cholestasis may induce hepatocellular expression of Bcl-2. Thus our aim was to determine whether hepatocytes express Bcl-2 or alter expression of other Bcl-2 family members in cholestasis using the bile duct-ligated (BDL) rat as a model of cholestasis. De novo Bcl-2 expression was observed in hepatocytes of BDL rats assessed by reverse transcriptase-polymerase chain reaction and immunoblot analysis. Immunohistochemistry demonstrated that Bcl-2 expression in hepatocytes was greater in periportal hepatocytes than pericentral hepatocytes. Expression of Bcl-x (an antiapoptotic Bcl-2 family protein) was not altered by bile duct ligation, whereas expression of Bax (a proapoptotic Bcl-2 family protein) increased slightly as determined by Northern and Western blot analyses. Bcl-2-positive hepatocytes isolated from BDL rats were resistant to induction of apoptosis by 50 microM glycochenodeoxycholate. Our results demonstrate, for the first time, expression of Bcl-2 by hepatocytes during cholestasis. We suggest that hepatocellular expression of Bcl-2 during cholestasis is an adaptive phenomenon to resist apoptosis by toxic bile salts.

Cholangiocytes are the target of a group of liver diseases termed the cholangiopathies that include conditions characterized by periductal inflammation and cholangiocyte apoptosis. Because inflammation is associated with oxidative stress, we developed the hypothesis that cholangiocytes exposed to oxidative stress will be depleted of endogenous cytoprotective molecules, leading to cholangiocyte apoptosis. To begin to test this hypothesis, we explored the relationships among glutathione (GSH) depletion, expression of Bcl-2 (a protooncogene that inhibits apoptosis), and apoptosis in a nonmalignant human cholangiocyte cell line. Monolayers of human bile duct epithelial cells, derived from normal liver and immortalized by SV40 transformation, were depleted of GSH using buthionine sulfoximine (BSO). Bcl-2 expression was assessed by quantitative immunoblot analysis, and apoptosis quantified by fluorescence microscopy using the DNA binding dye 4′,6′-diamidino-2-phenylindole. Bcl-2 message was assessed by RNase protection assay, and Bcl-2 protein synthesis and half-life by pulse-chase analysis. Exposure of human cholangiocytes in culture to BSO reduced GSH levels by 93 ± 3% ( P < 0.01). In addition, treatment of cholangiocytes with BSO reduced Bcl-2 levels by 87 ± 2% ( P < 0.01) and was associated with a time-dependent increase in the number of cells undergoing apoptosis; ∼11 ± 1% of cultured cells demonstrated morphological changes of apoptosis by 72 h compared with 1.5 ± 0.1% in untreated cholangiocytes ( P < 0.01). Maintenance of GSH levels by addition of glutathione ethyl ester in the presence of BSO blocked the BSO-associated increase in apoptosis in BSO-treated cholangiocytes and also prevented the decrease in Bcl-2 protein. BSO treatment of cholangiocytes did not change steady-state levels of bcl-2 mRNA or Bcl-2 protein synthesis. However, Bcl-2 protein half-life decreased 57% in BSO-treated vs. untreated cells. Our results using a human cholangiocyte cell line demonstrate that reduction in the cellular levels of an antioxidant such as GSH results in increased degradation of Bcl-2 protein and an increase in apoptosis. These data provide a mechanistic link between the consequences of oxidative stress and cholangiocyte apoptosis, an observation that may be important in the pathogenesis of the inflammatory cholangiopathies.

Abstract Background: Pseudomyxoma peritonei (PMP) is a rare abdominal malignancy. We hypothesized that next generation exomic sequencing would identify recurrent mutations that may have prognostic or therapeutic implications. Methods and Materials: Ten patients were selected based on availability of tissue and adequate follow-up. Mutations were tested for in 236 cancer-related genes in formalin-fixed paraffin-embedded (FFPE) slides using next-generation exomic sequencing (Foundation Medicine, Cambridge, MA). Immunohistochemistry (IHC) was used to detect MCL1 expression. Results: Detectable mutations were found in 80% of patients. Seventy percent harbored a KRAS mutation, most commonly involving codon 12 (G12D in four cases, and one case each of G12V, G12A and A59T). Four GNAS mutations (R201H/R201C substitutions) were also detected. MCL1 and JUN were concurrently amplified in three patients. One patient with MCL1 and JUN amplification had concurrent amplification of MYC and NFKBIA. ZNF703 was amplified in one patient. Patients with MCL1 amplification were also found to express MCL1 by IHC, but MCL1 expression was also detected in some patients without amplification. Conclusions: We are first to report MCL1 and JUN co-amplification in PMP. Expression of MCL1 may not be completely dependent upon amplification. The prognostic and therapeutic implications of these recurrent mutational events are the subject of ongoing investigation. Citation Information: Mol Cancer Ther 2013;12(11 Suppl):A135. Citation Format: Aaron S. Mansfield, Terence Sio, Travis E. Grotz, Rondell P. Graham, Julian R. Molina, Florencia G. Que, Robert C. Miller. Concurrent MCL1 and JUN amplification in pseudomyxoma peritonei. [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2013 Oct 19-23; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2013;12(11 Suppl):Abstract nr A135.

Abstract Background Outcomes for patients with hepatocellular carcinoma (HCC) without cirrhosis and factors associated with disease progression remain unclear. The goals of this single-institution study were to define the outcomes for such patients, and to determine factors associated with survival and disease progression. Methods This was a retrospective review of consecutive patients with HCC without cirrhosis who underwent hepatic resection between 1985 and 2003. Survival was estimated by the Kaplan–Meier method and risk factors were identified by Cox proportional hazards models. Results A total of 143 patients were enrolled, of whom 29·4 per cent had identifiable risk factors for chronic liver disease. Major resection (at least three segments) was undertaken in 63·6 per cent of patients. The operative mortality rate was 3·5 per cent. Median disease-free survival was 2·4 years. Multivariable analysis revealed presence of multiple tumours as the only independent predictor of tumour recurrence. Median overall survival was 3·3 years. Factors independently associated with decreased overall survival were multiple tumours, high histological grade, perioperative transfusion, male sex and age at least 66 years. Conclusion Patients with HCC but without cirrhosis have acceptable outcomes after resection. Specific risk factors for the development of HCC in these patients have yet to be defined.

Abstract Hepatic resection is the only treatment for patients with colorectal cancer metastatic to the liver that has resulted in long-term survival. This apparent efficacy of hepatectomy has prompted efforts to expand the surgical approach for disease progression within the liver. A review of personal experience and of the literature was performed in an attempt to define the role of surgery for disease progression. Twenty-one patients who underwent hepatic resection between 1983 and 1991 for isolated disease progression in the liver were retrospectively reviewed. The median follow-up of patients still alive was 1·7 years (range 4 months to 4·5 years). The median survival from the date of repeat hepatic resection was 3·4 years with an estimated actuarial survival rate of 43 per cent at 4 years. These patients experienced no significant morbidity and the mortality rate was 5 per cent. Hepatic resection of metastatic colorectal carcinoma can produce long-term survival without prohibitive risk. These findings support an aggressive surgical approach for metastatic progression in the liver from colorectal carcinoma.