Comparison of genomic profiling of circulating tumor DNA in pancreaticobiliary malignancies in plasma and bile.

BACKGROUND: Obtaining sufficient pancreaticobiliary tumor tissue for genomic profiling has limitations. Liquid biopsies using plasma do not provide sufficient sensitivity. Thus, this study aimed to determine the effectiveness of liquid biopsy between bile and plasma for identifying oncogenic and drug-matched mutations. METHODS: This study created a panel of 60 significantly mutated genes specific to pancreaticobiliary cancer (PBCA) and used it for genomic analysis of 212 deoxyribonucleic acid (DNA) samples (87 bile supernatant, 87 bile precipitate, and 38 plasma) from 87 patients with PBCA. The quantity of extracted DNA from bile and plasma was compared, as were genomic profiles of 38 pairs of bile and plasma from 38 patients with PBCA. Finally, we investigated 87 bile and 38 plasma for the ability to detect druggable mutations. RESULTS: The amount of DNA was significantly lower in plasma than in bile (p < .001). Oncogenic mutations were identified in 21 of 38 (55%) patients in bile and nine (24%) in plasma samples (p = .005). Bile was significantly more sensitive than plasma in identifying druggable mutations (p = .032). The authors detected 23 drug-matched mutations in combined bile and plasma, including five ERBB2, four ATM, three BRAF, three BRCA2, three NF1, two PIK3CA, one BRCA1, one IDH1, and one PALB2. CONCLUSIONS: Liquid biopsy using bile may be useful in searching for therapeutic agents, and using the obtained genomic information may improve the prognoses of patients with PBCA. PLAIN LANGUAGE SUMMARY: Genomic profiling of formalin-fixed paraffin-embedded tissues may provide actionable targets for molecular and immuno-oncological treatment. However, most pancreaticobiliary malignancies are unresectable and formalin-fixed paraffin-embedded tissues cannot be obtained. Although comprehensive genomic profiling tests using plasma have been used in recent years, the utility of those using bile is not clear. Our study revealed that bile identified more drug-matched mutations than plasma in advanced pancreaticobiliary cancer patients. Bile may help widen the patient population benefiting from targeted drugs.

Genomic profiling amplifies the utility of endoscopic ultrasound-guided fine needle biopsy by identifying clinically applicable druggable mutations in pancreatic cancer.

BACKGROUND: Genomic profiling of tumors is available, but whether the small fragment obtained via endoscopic ultrasound-guided fine needle biopsy (EUS-FNB) is sufficient for these examinations is unknown. Here we investigated whether EUS-FNB specimens are suitable for genomic profiling to identify oncogenic and drug-matched mutations. METHODS: We constructed a pancreatobiliary cancer panel for targeted panel sequencing that covered 60 significantly mutated genes and compared the results with those of whole-exome sequencing (WES). In total, 20 and 53 formalin-fixed paraffin-embedded tissues obtained via surgery and EUS-FNB were analyzed, respectively. First, we examined the DNA quality and genomic profiles of 20 paired samples from 20 malignant lesions obtained via surgery and EUS-FNB. We then tested 33 samples obtained via EUS-FNB from 24 malignant and 9 benign lesions for the discrimination of malignancy. Finally, we explored drug-matched mutations from EUS-FNB specimens. RESULTS: Although the DNA quantity obtained via surgery was higher than that obtained via EUS-FNB (P = 0.017), the DNA quality and mean depth were equivalent (P = 0.441 and P = 0.251). Panel sequencing of EUS-FNB specimens identified more oncogenic mutations than WES (90 % vs. 50 %). Furthermore, the number of oncogenic mutations did not differ between EUS-FNB and surgically resected specimens. Genomic profiling of EUS-FNB specimens enabled the discrimination of malignancy with 98 % accuracy. Of 44 malignant lesions, drug-matched alterations were identified in 14 % (6/44) of malignant lesions. CONCLUSION: EUS-FNB specimens can be widely utilized for diagnostic purposes, discrimination of malignancy, and detection of drug-matched mutations for the treatment of pancreatic cancer.

Detection of actionable mutations in cytological specimens obtained by endoscopic ultrasound-guided fine needle aspiration with rapid onsite evaluation in pancreatic cancer.

BACKGROUND: It is not clear whether archived cytological specimens (ACSs) obtained with endoscopic ultrasound-guided fine needle aspiration (EUS-FNA) with rapid onsite evaluation (ROSE) can be used for genomic profiling of tumors. We used ACSs to perform genomic analysis of specimens to identify oncogenic and druggable mutations. METHODS: A panel of 60 significantly mutated genes specific to pancreatobiliary cancer was created and used for genomic analysis of 113 specimens of 44 formalin-fixed paraffin-embedded (FFPE) tissues and 69 ACSs obtained by EUS-FNA with ROSE were included. The quantity and quality of DNA extracted from FFPE tissues and ACSs were compared. We also compared DNA from spray and touch ACSs. Next, genomic profiles were compared. We also evaluated detection of target gene mutations in each specimen. RESULTS: The amount of DNA in FFPE tissues was greater than in ACSs (P = 0.014), but the quality of DNA was comparable (P = 0.378). There was no quantitative or qualitative difference between spray and touch ACSs (P = 0.154 and P = 0.734, respectively). Oncogenic mutations were shared at 82 % in FFPE tissues and ACSs and 82 % in spray and touch ACSs. The sensitivity of genomic analysis in ACSs was 97 % (67 of 69), which was comparable to that of cytology (62 of 69, 90 %; P = 0.165), and was significantly higher than that of histology (32/44, 73 %; P < 0.001). Drug-matched mutations were identified in five of the 44 lesions (11 %). CONCLUSION: Genomic analysis of ACSs is useful in the prognosis of pancreatic cancer because detection of driver mutations is similar to detection in FFPE tissues.

Impaired Value of 99m Tc-GSA Scintigraphy as an Independent Risk Factor for Posthepatectomy Liver Failure in Patients with Hepatocellular Carcinoma.

BACKGROUND: Posthepatectomy liver failure (PHLF) was recently defined with the corresponding recommendations as follows: grade A, no change in clinical management; grade B, clinical management with noninvasive treatment; and grade C, clinical management with invasive treatment. In this study, we identified the risk factors for grade B and C PHLF in patients with hepatocellular carcinoma (HCC). METHODS: Of 339 HCC patients who underwent curative hepatic resection, 218 were included for analysis. The LHL15 index (uptake ratio of the liver to that of the liver and heart at 15 min) was measured by 99m Tc-GSA (99m technetium-labelled galactosyl human serum albumin); remnant LHL15 was calculated as LHL15 × [1 - (resected liver weight - tumor volume)/whole liver volume without tumor]. RESULTS: A total of 163 patients were classified as having no PHLF, whereas 17, 37, and 1 patient had PHLF grade A, B, and C, respectively. There were significant differences in indocyanine green R15, serum albumin, prothrombin time, Child-Pugh classification, LHL15 and remnant LHL15 between patients with grades B/C PHLF and patients with grade A or no PHLF. Only remnant LHL15 was identified as an independent risk factor for grades B/C PHLF (p = 0.023), with a cut-off value of 0.755. CONCLUSIONS: Remnant LHL15 was an independent risk factor for grades B/C PHLF. Patients with impaired remnant LHL15 value of <0.755 should be carefully monitored for PHLF.

Splenectomy attenuates murine liver fibrosis with hypersplenism stimulating hepatic accumulation of Ly-6C(lo) macrophages.

BACKGROUND & AIMS: Splenectomy in cirrhotic patients has been reported to improve liver function; however the underlying mechanism remains obscure. In the present study, we investigated the mechanism using a murine model, which represents well the compensated liver cirrhosis. METHODS: C57BL/6 male mice were allowed to drink water including thioacetamide (TAA: 300 mg/L) ad libitum for 32 weeks. After splenectomy at 32 weeks, mice were sacrificed on days one, seven, and 28, respectively, while TAA-administration was continued. Perioperative changes in peripheral blood and liver tissues were analyzed. RESULTS: TAA treatment of mice for 32 weeks reproducibly achieved advanced liver fibrosis with splenomegaly, thrombocytopenia, and leukocytopenia. After splenectomy, liver fibrosis was attenuated, and macrophages/monocytes were significantly increased in peripheral blood, as well as in the liver. Progenitor-like cells expressing CK-19, EpCAM, or CD-133 appeared in the liver after TAA treatment, and gradually disappeared after splenectomy. Macrophages/monocytes accumulated in the liver, most of which were negative for Ly-6C, were adjacent to the hepatic progenitor-like cells, and quantitative RT-PCR indicated increased canonical Wnt and decreased Notch signals. As a result, a significant amount of ß-catenin accumulated in the progenitor-like cells. Moreover, relatively small Ki67-positive hepatic cells were significantly increased. Protein expression of MMP-9, to which Ly-6G-positive neutrophils contributed, was also increased in the liver after splenectomy. CONCLUSIONS: The hepatic accumulation of macrophages/monocytes, most of which are Ly-6C(lo), the reduction of fibrosis, and the gradual disappearance of hepatic progenitor-like cells possibly play significant roles in the tissue remodeling process in cirrhotic livers after splenectomy.

Hemodynamic analysis and treatment of an enlarging extrahepatic portal aneurysm: report of a case.

Aneurysms in the portal venous system are relatively rare. We report the case of an extrahepatic portal venous aneurysm, detected incidentally by ultrasonography. The patient, a 75-year-old woman, was initially observed over 18 months, during which time, the aneurysm grew from 36 mm × 32 mm to 51 mm × 37 mm in size, without symptoms. Hemodynamic analysis employing computational flow dynamics technique showed obvious turbulence in the aneurysm, and the wall shear stress (WSS) against that part of the aneurysmal wall was greater than in other sites. To prevent complications such as spontaneous rupture and portal vein thrombosis, the aneurysm was resected, with reconstruction of the portal trunk. While careful follow-up is sufficient for most portal venous aneurysms, its enlargement could indicate possible spontaneous rupture. The increased WSS against part of the aneurysmal wall most likely accounts for the aneurysm enlargement in this case.

Primary adenocarcinoma of the minor duodenal papilla with mass-forming chronic pancreatitis: report of a case.

A 70-year-old male was treated for gastric ulcers. Follow-up upper gastrointestinal endoscopy revealed an irregular, elevated tumor in the second portion of the duodenum. Upon pathological inspection of a biopsy specimen, a diagnosis of adenocarcinoma was made, and the patient was admitted to our hospital. Computed tomography showed an irregular mass in the pancreatic head and dilatation of the main pancreatic duct and bile duct. Pancreatic head carcinoma with infiltration of the duodenum was diagnosed, and pylorus-preserving pancreaticoduodenectomy was performed. A histopathological examination of the resected specimen showed moderately differentiated adenocarcinoma in the minor duodenal papilla and chronic pancreatitis in the pancreatic head. Therefore, primary adenocarcinoma of the minor duodenal papilla with mass-forming chronic pancreatitis was diagnosed. Currently, the patient is alive without recurrence 17 months after the surgery. Primary adenocarcinoma of the minor duodenal papilla is extremely rare. We herein report this case, and also provide a review of the literature.

Nationwide survey in Japan regarding splenectomy/partial splenic embolization for interferon treatment targeting hepatitis C virus-related chronic liver disease in patients with low platelet count.

AIM: In chronic liver disease associated with hepatitis C virus (HCV), a low platelet count is a major obstacle in carrying out interferon (IFN) treatment. We used a questionnaire to clarify the extent to which splenectomy/partial splenic embolization (PSE) is performed before IFN treatment, as well as the efficacy and complications thereof. METHODS: Two questionnaires were distributed to 413 medical institutes in Japan specializing in the treatment of liver diseases, and responses were obtained from 204 institutes. Furthermore, a more detailed questionnaire was completed by 10 institutes that experienced cases of death. RESULTS: In patients with HCV genotype 1b and a high viral load (HCV1b/High), the sustained viral response (SVR) rate was 28% for the splenectomy group and 22% for the PSE group, with no significant difference between these groups. In patients that were not HCV1b/High, the SVR rate was higher in those that underwent splenectomy (71%) compared to the PSE group (56%; P = 0.025). There were cases of death in seven of 799 splenectomy cases (0.89%) and four of 474 PSE cases (0.84%). Infectious diseases were involved in nine of 11 cases of death, with a peculiar patient background of Child-Pugh B (6/10) and an age of 60 years or greater (7/11). CONCLUSION: The application of splenectomy/PSE before IFN treatment should be avoided in patients with poor residual hepatic function and/or elderly patients. In HCV1b/High patients, splenectomy/PSE should be performed only after selecting those in which IFN treatment should be highly effective.

Prediction of Postoperative Hepatic Failure after Liver Resection for Hepatocellular Carcinoma: Significance of the Aspartate Aminotransferase-to-Platelet Ratio Index.

BACKGROUND/AIMS: The serum aspartate aminotransferase-to-platelet ratio index (APRI) is a biomarker for hepatic fibrosis. The relationship between the APRI and postoperative hepatic failure is unclear. METHODOLOGY: The risk factors for postoperative hepatic failure and the APRI were evaluated in 457 patients who underwent liver resection for HCC. RESULTS: Nineteen patients (4.2%) experienced postoperative hepatic failure and five (1.1%) died. An increased APRI (p = 0.039), increased total bilirubin (p = 0.044), longer operation (p = 0.035) and increased intraoperative blood loss (p = 0.028) were independent risk factors in the multivariate analysis. Incidence of postoperative hepatic failure in patients with an APRI ≥ 1.57 (13/127, 10%) was significantly higher than in patients with an APRI < 1.57 (6/330,1.8%, p = 0.0002). Moreover, incidence of hepatic failure in high APRI cases with both an operation ≥ 500 min and intraoperative blood loss ≥ 1L (6/33 (18.1%)) tended to be higher than in those with lower values (7/94 (7.4%), p = 0.051). CONCLUSIONS: Increased APRI (≥ 1.57) may be a preoperative predictor of postoperative hepatic failure. Meticulous surgery with shorter operations and reduced blood loss may reduce the incidence of postoperative hepatic failure, even in patients with a high APRI.

Non-B non-C Hepatocellular Carcinoma: the Risk Factors for Recurrence and the Types of Recurrence Following Hepatic Resection.

BACKGROUND/AIMS: Risk factors for recurrence and types of recurrence following hepatic resection for non-B non-C hepatitis hepatocellular carcinoma (NBC-HCC) have not yet been established. METHODOLOGY: The clinicopathological data of 76 patients with NBC-HCC were retrospectively reviewed. Risk factors for postoperative recurrence were analyzed using univariate and multivariate analyses. In addition, types of intrahepatic recurrence were investigated. RESULTS: Of the 76 patients, 38 (50%) developed recurrence during the follow-up period, with disease-free survival rates at 1/3/5 years of 72%/46%/40%, respectively. Of the 38 patients with recurrence, 36 (95%) were found to have recurrence within three years after surgery. Of the 38 patients, 34 exhibited intrahe patic recurrence. In multivariate analysis, Child-Pugh B (p = 0.009) and microscopic vascular invasion (MVI) (p = 0.002) were independent risk factors for postoperative recurrence. Based on our definitions, of the 34 patients with intrahepatic recurrence, recurrence at the stump was present in one patient, multicentric recurrence in 11 patients and intrahepatic metastasis in 22 patients. CONCLUSIONS: Child-Pugh B and MVI are independent risk factors for the postoperative recurrence. Although most recurrences occurred within three years after hepatic resection, incidence of multicentric recurrence is not negligible. Preventing recurrence according to types of recurrence is therefore considered to be essential.

Development of a molecular barcode detection system for pancreaticobiliary malignancies and comparison with next-generation sequencing.

BACKGROUND: Obtaining sufficient tumor tissue for genomic profiling is challenging in pancreaticobiliary cancer (PBCA). We determined the utility of molecular barcoding (MB) of liquid biopsies (bile, duodenal fluid, and plasma) for highly sensitive genomic diagnosis and detection of druggable mutations for PBCA. METHODS: Two in-house panels of 60 genes (non-MB panel) and 21 genes using MB (MB panel) were used for the genomic analysis of 112 DNA samples from 20 PBCA patients. We measured the yield of DNA and compared the genomic profiles of liquid samples obtained using the non-MB panel and the MB panel. The utility of the panels in detecting druggable mutations was investigated. RESULTS: A significantly greater amount of DNA was obtained from bile supernatants and precipitates compared to tumor samples (P < 0.001 and P = 0.001, respectively). The number of mutations per patient was significantly higher using the MB panel than using the non-MB panel (2.8 vs. 1.3, P = 0.002). Tumor-derived mutations were detected more frequently using the MB panel than the non-MB panel (P = 0.023). Five drug-matched mutations were detected in liquid samples. CONCLUSIONS: Liquid biopsy with MB may have utility in providing genomic information for the prognosis of patients with PBCA.

Liquid biopsy of wash samples obtained via endoscopic ultrasound-guided fine-needle biopsy: Comparison with liquid biopsy of plasma in pancreatic cancer.

OBJECTIVES: Pancreatic cancer (PC) has a poor prognosis and limited treatment options. Liquid biopsy, which analyzes circulating tumor DNA (ctDNA) in blood, holds promise for precision medicine; however, low ctDNA detection rates pose challenges. This study aimed to investigate the utility of wash samples obtained via endoscopic ultrasound-guided fine-needle biopsy (EUS-FNB) as a liquid biopsy for PC. METHODS: A total of 166 samples (42 formalin-fixed paraffin-embedded [FFPE] tissues, 80 wash samples, and 44 plasma samples) were collected from 48 patients with PC for genomic analysis. DNA was extracted and quantified, and 60 significantly mutated genes were sequenced. The genomic profiles of FFPE tissues, wash samples, and plasma samples were compared. Finally, the ability to detect druggable mutations in 80 wash samples and 44 plasma samples was investigated. RESULTS: The amount of DNA was significantly lower in plasma samples than in wash samples. Genomic analysis revealed a higher detection rate of oncogenic mutations in FFPE tissues (98%) and wash samples (96%) than in plasma samples (18%) and a comparable detection rate in FFPE tissues and wash samples. Tumor-derived oncogenic mutations were detected more frequently in wash samples than in plasma samples. Furthermore, the oncogenic mutations detection rate remained high in wash samples at all PC stages but low in plasma samples even at advanced PC stages. Using wash samples was more sensitive than plasma samples for identifying oncogenic and druggable mutations. CONCLUSIONS: The wash sample obtained via EUS-FNB is an ideal specimen for use as a liquid biopsy for PC.

Regional hepatic regeneration after liver resection correlates well with preceding changes in the regional portal circulation in humans.

BACKGROUND AND AIMS: While portal hemodynamics largely affects the liver regeneration after partial hepatectomy, whether the remnant liver homogeneously regenerates is unclear, especially in humans. We hypothesized that change in flow distribution varies in each remnant portal branch after liver resection in humans and the liver consequently regenerates heterogeneously. METHODS: Twenty-two patients who underwent anatomical hepatic resection preserving intact drainage veins were analyzed. Based on perioperative contrast-enhanced computed tomography, the regional hepatic regeneration in each segment was analyzed using a region growing software. The perioperative change in the distribution of blood flow in each portal branch was assessed using the computational flow dynamics technique. The correlation between the change in the portal flow distribution and the later regional hepatic regeneration was investigated. RESULTS: The distribution of portal blood flow in each remnant branch largely changed at 2 weeks (71-389 %). Each remnant segment also heterogeneously regenerated at 3 months (85-204 %). Meanwhile, a good correlation between the regional regeneration rate at 3 months and the relative change in the flow distribution in each circulating portal branch at 2 weeks was detected in each patient (r = 0.74-0.99). CONCLUSIONS: After partial hepatectomy, the change in blood flow varies in each remnant portal branch and the liver heterogeneously regenerates in humans. The good correlation between the earlier change in the portal flow distribution and the later regional hepatic regeneration strongly suggests that the portal venous flow most likely regulates the non-uniform liver regeneration after hepatic resection in humans.

Safety of hepatic resection for hepatocellular carcinoma in obese patients with cirrhosis.

PURPOSE: This study aimed at investigating the safety of hepatic resection for hepatocellular carcinoma (HCC) in obese patients with cirrhosis in Japan. METHODS: We reviewed the clinical records of 202 patients with liver cirrhosis, who underwent hepatic resection for HCC between January, 2001 and August, 2011. The patients were divided into three groups according to their body mass index (BMI): the normal body weight (BMI < 24.9 kg/m(2)), obese class I (BMI 25.0-29.9 kg/m(2)), and obese class II (BMI ≥ 30 kg/m(2)) groups. We compared the patient backgrounds, intraoperative factors, and postoperative complications among the three groups. RESULTS: The normal body weight, obese class I, and obese class II groups comprised 138 (68.3 %), 55 (27.2 %), and 9 (4.5 %) patients, respectively. The incidence of non-B non-C cirrhosis was higher in the obese class II group (22 %) than in the normal body weight group (14 %, p = 0.034). Intraoperative blood loss tended to be higher in the obese class II patients than in the other two groups. Postoperative complications and mortality did not differ significantly among the three groups. According to multivariate analysis, obesity was not a risk factor for postoperative complications (Clavien-Dindo classification Grade III or higher) or mortality. CONCLUSION: Hepatic resection for HCC can be performed safely in obese patients with cirrhosis.

Fascin, a novel marker of human hepatic stellate cells, may regulate their proliferation, migration, and collagen gene expression through the FAK-PI3K-Akt pathway.

Fascin is a component of actin bundles and may regulate various cellular events. The expression and function of fascin in human hepatic stellate cells (HSCs) has remained largely uncharacterized. Fascin expression in human liver tissue was studied using immunohistochemistry. To identify cells expressing fascin, double immunofluorescent staining with vimentin, α-smooth muscle actin (α-SMA), or fibulin-2 was performed and analyzed with confocal microscopy. In culture experiments, fascin expression and the phosphorylation of focal adhesion kinase (FAK) and Akt in LX-2 cells, a cell line of human HSCs, were investigated using western blot. Specific siRNAs were used to reduce the expression of fascin in LX-2 cells. Proliferation and migration were assayed with a CyQuant assay kit and a Matrigel-coated culture insert system, respectively. Levels of matrix metalloproteinase (MMP)-2 and collagen mRNAs were examined using quantitative RT-PCR. Immunohistochemistry revealed the expression of fascin along sinusoids and overlapping with vimentin and α-SMA in both non-fibrotic and fibrotic liver tissue, but it was almost absent in periportal myofibroblastic cells and did not colocalize with fibulin-2, a marker of portal myofibroblasts. In addition, fascin immunoreactivity was almost undetectable in septa of fibrotic human liver tissue. The expression of fascin in LX-2 cells was confirmed using western blot. Two different specific siRNAs against fascin significantly reduced the number of viable LX-2 cells to 65% compared with control cultures and downregulated the mRNAs levels of types I and III collagen and MMP-2 to 62%, 65%, and 70% of control levels, respectively. This condition also reduced the migration activity of LX-2 cells to 46% of control cells and the phosphorylation level of both FAK and Akt. Fascin may be an excellent novel marker of human HSCs that distinguishes HSCs from periportal myofibroblasts. Fascin may regulate functions of human HSCs through the FAK-phosphoinositide 3-kinase-Akt pathway.

Primary epithelioid hemangioendothelioma of the retroperitoneum: report of a case.

We herein report the case of a 48-year-old Japanese female with retroperitoneal epithelioid hemangioendothelioma (EHE), a rare malignant vascular tumor of intermediate grade. She was referred to our hospital because a retroperitoneal tumor was found during a medical checkup, in which strong accumulation of (18)F-fluorodeoxyglucose (FDG) was observed by (18)F-FDG-positron emission tomography (PET). A histological examination of the resected tumor revealed that it consisted of large epithelioid cells with vesicular nuclei, and clear cells with vacuolated cytoplasm and intracytoplasmic lumina. These cells expressed CD31 and vimentin, and the final pathological diagnosis was EHE. Postoperative surveillance with FDG-PET revealed distant metastasis in Virchow's lymph node 7 months after the operation. After dissection of the metastatic lymph node, the patient has been free from recurrence for 13 months. Close follow-up with FDG-PET seemed to be useful for surveillance of the recurrence of this tumor with unpredictable behavior, making an early treatment for the recurrent lesions possible.

Detection of actionable mutations in archived cytological bile specimens.

BACKGROUND/PURPOSE: There are limitations in obtaining sufficient pancreaticobiliary tumor tissue for genomic profiling of tumors. Herein, we investigated whether archived cytological specimen (ACS) is suitable for genomic profiling to identify oncogenic and drug-matched mutations. METHODS: We constructed a pancreaticobiliary cancer panel for targeted sequencing covering 60 significantly mutated genes (280 220 bp). Eighty DNA samples (19 formalin-fixed paraffin-embedded [FFPE] tissues and 61 ACS) from 44 patients with pancreaticobiliary disease were analyzed. We compared genomic profiles of 19 FFPE and 29 ACS from 19 patients with malignancies (Validation Cohort). We tested 32 ACS from 25 patients (15 malignant and 10 benign) for the ability to discriminate between malignant and benign disorders (Testing Cohort). We explored whether actionable and drug-matched mutations (Validation and Testing Cohorts) could be identified from ACS. RESULTS: Oncogenic mutations observed in ACS were identical to those identified in FFPE specimens (76% consistency). Genomic profiling using only ACS discriminated between malignant and benign disorders with 93% accuracy, 91% sensitivity, and 100% specificity. Actionable and drug-matched mutations were identified in 74% and 32% of ACS, respectively, and in 79% and 21% in FFPE samples, respectively. CONCLUSION: Archived cytological specimen may be used to discriminate malignancy and to detect drug-matched mutations in patients with advanced pancreaticobiliary cancer.

Cytoglobin-expressing cells in the splenic cords contribute to splenic fibrosis in cirrhotic patients.

BACKGROUND AND AIM: Among several noninvasive evaluation methods of portal hypertension (PH), the measurement of spleen stiffness is a reliable method for predicting esophageal variceal bleeding; however, the underlying mechanisms for increased stiffness remain unclear. We attempted to elucidate the pathological changes to the spleen and the underlying mechanisms in patients with PH. METHODS: Histological examination was performed using splenic tissues from 42 patients with PH who underwent laparoscopic splenectomy, and the results were compared with those from patients without PH. RESULTS: In addition to splenic sinus congestion, diffuse fibrosis was detected in the splenic cords in the red pulp of patients with PH. The degree of the fibrosis was well correlated with severity in thrombocytopenia and splenomegaly. Cells expressing α-smooth muscle actin dramatically increased in the splenic cord. Cytoglobin (Cygb) expression was detected in human splenic cords as reported in animal reticular cells, and fluorescent double immunostaining revealed that these cells expressed α-smooth muscle actin in patients with PH, suggesting transformation of Cygb-expressing cells to myofibroblastic cells. Expression levels of nicotinamide adenine dinucleotide phosphate oxidase (NOX) 2, nitrotyrosine, and transforming growth factor-ß were markedly upregulated in the red pulp of patients with PH, implying a significant role of oxidative stress in the mechanism for splenic fibrosis. CONCLUSION: Splenic fibrosis progresses along with advancement of PH. Cygb-expressing cells in the splenic cord possibly participate in this process through mechanisms including oxidative stress.

Minimally invasive laparoscopic liver resection: 3D MDCT simulation for preoperative planning.

BACKGROUND/PURPOSE: Laparoscopic liver resection has not gained wide acceptance compared with other laparoscopic procedures. We evaluated the impact of simulated surgery using data from multidetector CT scanning on planning for laparoscopic hepatectomy. METHODS: The hepatectomy simulation system was programmed to perform three-dimensional reconstruction of the vasculature and to calculate the liver resection volume and surgical margin. In 35 patients undergoing laparoscopic hepatectomy or laparoscopy-assisted hepatectomy, the liver resection volume and margin were estimated by simulation preoperatively. Then, the estimated values were compared with the actual resected liver weight and margin. RESULTS: Three-dimensional reconstruction allowed stereoscopic identification of the tumor-bearing portal vein and draining vein. The predicted liver resection volume and margin both showed a significant correlation with the actual values: the mean difference was 21 mL (P < 0.0001) and 1.3 mm (P < 0.01), respectively. Preoperative planning based on simulated resection facilitated laparoscopic mobilization of the liver and mini-laparotomy resection of a large tumor located in the upper right lobe. CONCLUSIONS: Three-dimensional simulation of hepatectomy facilitated intraoperative identification of the vascular anatomy, and accurately predicted the resected liver volume and surgical margin. This simulation method should contribute to preoperative planning for safe and curative laparoscopic hepatectomy.

A case with hepatic portal vein gas who required delayed elective surgery.

INTRODUCTION: Hepatic portal venous gas (HPVG) is believed to be an indication for emergent surgery because it is associated with high mortality rate. However, the recent increase in the use of modern abdominal computed tomography (CT) has resulted in the detection of HPVG in more benign conditions. Therefore, the decision-making process whether we chose emergent surgery or conservative treatment without surgery is important for the patients with HPVG. CASE PRESENTATION: An 84-year-old male was referred to our hospital due to the sudden onset of abdominal pain and massive hepatic portal vein gas on emergent CT. The Acute Physiology and Chronic Health Evaluation (APACHE) II Score was calculated as 17; slightly elevated comparing with the other cases who were successfully treated without surgery. Although the PHVG was remained at follow up CT on the next day after the onset, the symptoms were improved. We selected conservative treatment without emergent surgery and he discharged on 9th day after the onset. However, he was suffered from right lower abdominal pain and vomiting and admitted our hospital on 23th day. He developed ischemic intestinal stenosis and underwent a surgery of partial resection of ileum. CONCLUSIONS: The clinical finding of this case showing subtle differences from cases who were successfully treated without surgery. We hope this report will help physician's decision-making process for HPVG.