Development of a Deep Learning System for Intra-Operative Identification of Cancer Metastases.

OBJECTIVE: The aim of this study was to develop and test a prototype of a deep learning surgical guidance system (CASL) that can intra-operative identify peritoneal surface metastases on routine laparoscopy images. BACKGROUND: For a number of cancer patients, operative resection with curative intent can end up in early recurrence of the cancer. Surgeons misidentifying visible peritoneal surface metastases is likely a common reason. METHODS: CASL was developed and tested using staging laparoscopy images recorded from 132 patients with histologically-confirmed adenocarcinoma involving the gastrointestinal tract. The data included images depicting 4287 visible peritoneal surface lesions and 3650 image patches of 365 biopsied peritoneal surface lesions. The prototype's diagnostic performance was compared to results from a national survey evaluating 111 oncologic surgeons in a simulated clinical environment. RESULTS: In a simulated environment, surgeons' accuracy of correctly recommending a biopsy for metastases while omitting a biopsy for benign lesions was only 52%. In this environment, the prototype of a deep learning surgical guidance system demonstrated improved performance in identifying peritoneal surface metastases compared to oncologic surgeons with an area under the receiver operating characteristic curve of 0.69 (oncologic surgeon) versus 0.78 (CASL) versus 0.79 (human-computer combined). A proposed model would have improved the identification of metastases by 5% while reducing the number of unnecessary biopsies by 28% compared to current standard practice. CONCLUSIONS: Our findings demonstrate a pathway for an artificial intelligence system for intra-operative identification of peritoneal surface metastases, but still requires additional development and future validation in a multi-institutional clinical setting.

Fine-Needle Aspiration of Epithelial-Myoepithelial Carcinoma of the Parotid Gland With Prominent Adenoid Cystic Carcinoma-Like Cribriform Features: Avoiding a Diagnostic Pitfall.

OBJECTIVES: Epithelial-myoepithelial carcinoma (EMC) is a rare salivary gland malignancy associated with an overall good prognosis. Fine-needle aspiration (FNA) typically shows a dual population of myoepithelial and ductal cells. Rarely, acellular matrix globules are present, raising a differential diagnosis of adenoid cystic carcinoma (AdCC), a more aggressive salivary gland malignancy associated with a poor long-term prognosis. We report an FNA case of EMC containing a predominant pattern of AdCC-like spherical globules. METHODS: We compare features of an unusual case of EMC with those of cribriform AdCC to arrive at morphologic clues to the correct diagnosis. RESULTS: Distinguishing features of EMC on FNA include (1) a prominent population of myoepithelial cells vs the predominance of basaloid cells in AdCC and (2) cohesive matrix globules with a peripheral rim of pale-staining basement membrane material compared with the dyscohesive matrix globules of AdCC. Immunochemical markers (S100, CD117, and MyB) are also useful. CONCLUSIONS: Although EMC and AdCC can both contain spherical matrix globules, close evaluation of the cytomorphology of the globules and their relationship to surrounding cells provides a clue to distinguish the two neoplasms.

Evaluation of guanylyl cyclase C lymph node status for colon cancer staging and prognosis.

PURPOSE: The prognostic significance of guanylyl cyclase C (GCC) gene expression in lymph nodes (LNs) was evaluated in patients with stage II colon cancer who were not treated with adjuvant chemotherapy. We report a planned analysis performed on 241 patients. METHODS: GCC mRNA was quantified by RT-qPCR using formalin-fixed LN tissues from patients with untreated stage II colon cancer who were diagnosed from 1999-2006 with at least ten LNs examined and blinded to clinical outcomes. Lymph node ratio (LNR) is the number of GCC-positive nodes divided by total number of informative LNs. Risk categories of low (0-0.1) and high (>0.1) for LNR were chosen by significance using Cox regression models. The data were tested for association with time to recurrence. RESULTS: Twenty-nine patients (12%) had a disease recurrence or cancer death. The LNR significantly predicted higher recurrence risk for 84 patients (34.9%) classified as high risk (hazard ratio (HR), 2.38; P=0.02). The estimated 5-year recurrence rates were 10% and 27% for the low- and high-risk groups, respectively. After adjusting for age, T stage, number of nodes assessed, and MMR status, a significant association remained (HR, 2.61; P=0.02). In a subset of patients (n=181) with T3 tumor, ≥12 nodes examined and negative margins, a significant association between the GCC LNR and recurrence risk also was observed (HR, 5.06; P=0.003). CONCLUSIONS: Our preliminary results suggest that detection of GCC mRNA in LNs is associated with risk of disease recurrence in patients with untreated stage II colon cancer. A larger validation study is ongoing.

Quantifying gastric Helicobacter pylori infection: a comparison of quantitative culture, urease breath testing, and histology.

Although there are several methods to detect Helicobacter pylori infection, there is no simple validated test to quantify the density of infection, which is believed to play a major role in the pathogenesis of H. pylori-associated gastritis and peptic ulceration. The aim of this study was to evaluate and compare noninvasive and invasive tests for assessing the level of H. pylori infection so as to facilitate the development and clinical testing of new antibiotic treatments. Healthy volunteers (n=323) were screened for H. pylori infection by serology and, if positive (n=86), invited to undergo (13)C urea breath testing (UBT) (n=55). An increase of >2.4 parts per thousand (13)CO(2) at 15 min compared to baseline was considered a positive test. Total cumulative urease activity (mumol) at 60 min was also calculated. UBT-positive subjects underwent endoscopy and nine biopsies were obtained from defined sites for quantitative culture and histological grading using the modified Sydney System. A total of 19 subjects were studied, 4 of whom underwent repeat testing. All subjects were positive for H. pylori by serology, UBT, culture, and histology. The increase in (13)CO(2) at 30 min correlated with the total cumulative urease activity at 60 min (r (2)=0.92, P< 0.0001). Bacterial counts (log cfu/biopsy; mean+/-SD) were 3.9+/-0.5, 3.9+/-0.4, and 3.9+/-0.6 at the lesser curve antrum, greater curve antrum, and corpus, respectively. There was no significant correlation between UBT results and bacterial counts at any biopsy site. Nor was there any significant correlation between the histology grading and either the UBT or the bacterial counts at any site. This study indicates that there is little correlation among the three methods used to measure bacterial burden in H. pylori infection. Thus, decrements in bacterial numbers during single-agent therapy cannot be measured reliably by UBT and therefore cannot be used to evaluate the potential efficacy of novel agents to treat gastric H. pylori infection.

Native liver xanthogranulomatous cholangiopathy in primary sclerosing cholangitis: impact on posttransplant outcome.

A retrospective analysis of 51 primary sclerosing cholangitis (PSC) patients who underwent liver transplant (LT) identified 16 with xanthogranulomatous cholangiopathy (XGC) at the native liver hilum. Pre-LT clinical and laboratory data and post-LT course and outcome of patients with XGC were compared with the 35 PSC patients without XGC. The XGC and non-XGC groups were similar with respect to age and laboratory data at the time of LT. Pre-LT cholecystectomy was performed in 44% versus 26% and biliary bypass procedure in 38% versus 26% of patients with and without XGC, respectively (P = NS). Peri-operative complications resulted in six (38%) deaths or retransplantation within 60 days of LT in the XGC group compared with 4 (11%) in the non-XGC group (P =.05). Patient survival at 60 and 100 days post-LT was better in the non-XGC group (P =.01). The causes of death or retransplantation within 60 days post-LT in the patients with XGC included primary nongraft function (1), uncontrolled bleeding (3), and sepsis (2), while in the non-XGC group these were uncontrolled bleeding (2), sepsis (1), and primary nongraft function (1). Mean graft survival +/- SD was 1,081 +/- 1,584 days in patients with XGC versus 2,149 +/- 1,679 days in patients without XGC. The presence of XGC in the native liver hilum of PSC patients undergoing LT was associated with a higher rate of early post-LT mortality or retransplantation. In conclusion, no pre-LT clinical features or laboratory tests were identified that predicted the presence of XGC in PSC patients.

Liver transplantation for primary sclerosing cholangitis: a long-term clinicopathologic study.

The course and outcome of patients after liver transplantation (LT) for primary sclerosing cholangitis (PSC) are still debated. Our purpose is to define retrospectively, the post-LT clinicopathologic findings seen in 51 PSC patients with a follow-up of 2 to 14 years. Of the total 51 patients, 16 with native liver hilar xanthogranulomatous cholangiopathy (XGC) had median graft and patient survival of 573 and 835 days, respectively compared with 2489 and 2794 days, respectively, in 35 patients without XGC. Perioperative complications resulted in 9 early deaths (day 0 to 52). Of the remaining 42 patients, 6 had recurrent PSC (R-PSC) with typical histologic and cholangiographic findings, 12 had autoimmune liver disease-not otherwise specified with histology of autoimmune hepatitis/overlap syndrome, 3 had chronic rejection, 4 had ischemic cholangiopathy, and 17 had no recurrence. The presence of inflammatory bowel disease, total ischemia time of > or =11 hours, recipient-donor ABO and HLA Class I and II matches, and the type of immunosuppression did not affect the post-LT outcome. Recipient-donor gender mismatch was more common in R-PSC than in the nonrecurrent group (P=0.045). Post-LT malignancies were significantly more common in the nonrecurrent cases compared with all others combined (P=0.031) and caused deaths in 4. The majority of deaths (11/13) in other groups were due to sepsis complicating graft dysfunction. In conclusion, allograft autoimmune liver disease was seen in 18 (43%) of 42 long-term post-LT PSC patients, with progression in 5 of 18 patients. Features of PSC were seen in 6 (33%) of 18. Native liver XGC negatively impacted post-LT graft and patient survival. Increased incidence of malignancies in the nonrecurrent group may reflect overimmunosuppression in those patients.

Nodular pulmonary amyloidosis following liver transplantation: report of a case and review of the literature.

Nodular pulmonary lesions seen in liver transplant recipients have a broad differential diagnosis including both infectious and noninfectious etiologies. Here, we report the first case of nodular pulmonary amyloidosis, an uncommon and benign localized form of amyloidosis occurring after orthotopic liver transplantation for end-stage primary biliary cirrhosis.

Xanthogranulomatous choledochitis: a previously undescribed mass lesion of the hepatobiliary and ampullary region.

Solid liver and pancreatic masses are commonly neoplastic in nature; however, inflammatory lesions mimicking carcinoma are at times encountered in these sites. We report two cases of previously undescribed inflammatory mass lesions of the liver and pancreas that originated in the biliary tract. Detailed clinical and histologic evaluations were performed in two patients who underwent right partial hepatic lobectomy and Whipple's resection for presumed hepatic and pancreatic neoplasms. In case 1, with a remote history of cholecystectomy and recent extraction of a stone from the common bile duct, a liver mass in segment 6 was discovered incidentally. In case 2, a periampullary pancreatic mass was diagnosed radiographically following papillotomy and stent insertion for stricture and biliary calculous disease. The histologic findings in both cases were similar, localized around a part of the biliary tract, and consisted of inspissated bile, acute and chronic inflammation, abundant lipid-laden macrophages, fibrosis, and giant cell reaction. No neoplasm was identified. On the basis of the close resemblance of these features to those seen in xanthogranulomatous cholecystitis, the lesions seen here were termed xanthogranulomatous choledochitis. In conclusion, xanthogranulomatous choledochitis is a benign inflammatory process involving the biliary tract that can form a mass lesion within the liver or pancreas and thus mimic a neoplasm. Extensive sampling of the lesion is required to rule out an underlying neoplastic process. In our patients a propensity to form lithogenic bile and a prior history of biliary tract operative procedure were present.

Liver transplantation for alcoholic liver disease with or without hepatitis C.

Alcoholic cirrhosis with or without hepatitis C is a common indication for liver transplantation (LT). Comparative post-LT data for the 2 groups are not available. Our aim is to compare the clinicopathologic features of patients with alcoholic liver disease (ETOH) and ETOH/HCV at the time of and after LT and to determine the impact of concomitant hepatitis C virus (HCV) on ETOH patients undergoing LT. A comparative clinical and pathologic analysis at LT and after LT was performed for 56 patients with ETOH and 32 patients with ETOH/HCV. All 88 had cirrhosis at LT. Other native liver features for ETOH and ETOH/HCV, respectively, were: >2+ inflammation 50/56 and 26/32, Mallory's hyalin 12/56 and 6/32, steatosis 9/56 and 7/32, large cell dysplasia 12/56 and 6/32, hepatoma 4/56 and 4/32, iron deposition 24/56 and 12/32; none was statistically significant. The post-LT findings for ETOH and ETOH/HCV were as follows: 1-year survival 93% and 97%; alive 36/56 (419-4,348 days) and 27/32 (488-5,516 days); deaths 20/56 and 5/32; ETOH recurrence 5/56 (all alive) and 3/32 (1 dead); post-LT HCV 4/56 (acquired) and 22/32 (recurrent). Native liver histology in ETOH and ETOH/HCV patients was similar. Post-LT HCV recurrence was common (69%) in ETOH/HCV patients but resulted in death in only 6%. Post-LT ETOH recurrence was uncommon (9%) and progression to liver failure was rare (1.1%). Post-LT outcome for ETOH was excellent, and concomitant HCV did not affect survival.

Angiographic outcome after intracoronary X-Sizer helical atherectomy and thrombectomy: first use in humans.

The objective of this study was to evaluate the early angiographic outcome in the first human subjects who underwent intracoronary atherectomy and thrombectomy using the X-Sizer helical cutting and aspiration system. Percutaneous coronary interventions in patients with thrombo-occlusive disease or friable degenerative saphenous vein grafts are associated with considerable periprocedural morbidity and mortality, predominantly related to microscopic distal embolization. X-Sizer catheter system is a novel atherectomy and thrombectomy device that consists of a helix cutter connected to a handheld motor drive unit and a vacuum collection chamber for aspiration of excised atheroma, thrombus, and debris. Quantitative coronary angiography was obtained in 14 patients before and after X-Sizer extraction atherectomy with adjunctive balloon angioplasty and stenting. Thirteen native coronary arteries and one saphenous vein graft were treated. Mean preprocedural reference vessel diameter was 3.06 +/- 0.66 mm. There were 71.4% AHA/ACC type B2 and C lesions. Preprocedural thrombus was present in nine patients and total occlusion in 64% of cases. Minimal luminal diameter was increased from 0.29 +/- 0.47 mm to 1.32 +/- 0.64 mm, a gain of 1.04 +/- 0.69 mm after atherectomy. Final total gain was 1.47 +/- 0.61 mm. Mean diameter stenosis was reduced from 89.3% to a final residual stenosis of 14.4%. Postatherectomy distal embolization occurred in one patient who had heavy preprocedural thrombus burden. No episodes of perforation, distal coronary spasm, abrupt closure, or slow/no-reflow occurred. The angiographic analysis of the first cohort of human subjects suggests that X-Sizer helical atherectomy is a feasible method of removing occlusive tissue or thrombus in coronary artery disease with a low angiographic complication rate. A large-scale randomized phase II clinical trial is underway to determine the ultimate safety and efficacy of this device in thrombo-occlusive native coronary arteries and saphenous vein grafts.