A158 FINDINGS FROM A MULTIDISCIPLINARY DATABASE FOR PATIENTS WITH BARRETT'S ESOPHAGUS AND EARLY ESOPHAGEAL ADENOCARCINOMA

Background

Previously, Barrett’s esophagus (BE) with high grade dysplasia (HGD) or neoplasia was treated with esophagectomy; however, recent guidelines support the use of endoscopic mucosal resection (EMR) for T1a esophageal adenocarcinoma (EAC) and potentially T1b EAC. Long term data for outcomes from EMR are lacking and these treatments are often provided with minimal collaboration between gastroenterologists and thoracic surgeons.

Purpose

Our primary aim was to describe the findings from our multidisciplinary database of patients with Barrett’s esophagus and EAC undergoing endoscopic treatment. Secondary aims were to compare the overall survival and recurrence-free survival of patients undergoing endoscopic resection to those undergoing esophagectomy.

Method

For the endoscopic resection cohort, a combined retrospective and prospective database was created containing demographic, clinical, and oncologic variables for patients undergoing endoscopic resection for early stage EAC from 2009 to 2021. For the esophagectomy cohort, a pre-existing retrospective database including patients undergoing esophagectomy for esophageal cancer from 2012 to 2018 at a single institution was used. A multivariate cox proportional hazards model was developed for recurrence-free survival and overall survival using a hypothesis driven approach. A kaplan-meier (KM) curve with associated log-rank test was created to evaluate recurrence-free survival and overall survival stratified by treatment modality.

Result(s)

A total of 108 patients were included in the analysis (73 EMR, 35 esophagectomy). Baseline characteristics including age, sex, and co-morbidities were similar among the two groups. KM curves stratified by treatment modality are provided in Figure 1. Esophagectomy was associated with greater DFS on univariate log-rank test (p = 0.0127), but no difference in OS (p = 0.9306). There was no significant difference between esophagectomy and endoscopic resection in the cox-model for OS (HR 1.03, 95% CI 0.45-2.32, p = 0.914). Endoscopic resection was associated with increased hazards of disease recurrence in the cox model for DFS (HR 2.56, 95% CI 1.1-6.0, p = 0.032). In the logistic regression model, high grade disease (OR 5.43, 95% CI 1.1 – 26.1, p = 0.035) and submucosal involvement (OR 7.8, 1.9-31.4, p = 0.004) were identified as significant predictors of positive margin necessitating esophagectomy after initial endoscopic resection. All patients who experienced a positive margin after endoscopic therapy were evaluated by a thoracic surgeon and proceeded to esophagectomy.

Conclusion(s)

In this largely retrospective analysis, our multidisciplinary approach was shown to be highly efficacious in the treatment of BE with EAC. Through optimizing collaboration between thoracic surgeons and gastroenterologists, patients receive the best therapeutic approach for their unique condition, taking into account oncologic factors and clinical comorbidities.

Please acknowledge all funding agencies by checking the applicable boxes below

None

Disclosure of Interest

None Declared

Transient expression of GAP-43 within the hippocampus after global brain ischemia in rat

Neuroanatomical methods have been used to study selective vulnerability after global brain ischemia. A consistent pattern of ischemic neuronal damage is found in the rodent hippocampus with loss of CA1 neurons and of some cells in the hilus of the dentate gyrus. Very little is known about plastic changes that would be expected in ischemia-resistant areas such as CA3 neurons and granule cells. Neuronal plasticity after lesions may be indicated by changes in labeling with antibodies to the growth-associated protein 43 (GAP-43). Expression of GAP-43 as a marker for neuronal plasticity was studied here in the hippocampus after global brain ischemia. Halothane-anesthetized rats were subjected to 20 min of transient forebrain ischemia using four-vessel occlusion. In situ hybridization was used to study GAP-43 mRNA at 1, 3, 6, and 12 h and at 1, 3, and 7 days after ischemia. Immunostaining was carried out with two different antibodies to GAP-43 in brains which were perfusion-fixed after 1, 2, 4, and 7/8 days. In the control hippocampus, GAP-43 mRNA was localized to CA1-CA3 and the hilus. Moderate increases in cellular signals were seen in hilar cells and granule cells early after ischemia, and some changes occurred in CA3 at late stages. Hybridization was lost in CA1 due to cell death. With immunostaining, GAP-43 was not seen in the cytoplasm of neurons, whereas dense labeling occurred in a differentiated pattern in the axonal and dendritic layers. At 1 day after ischemia, neurons in the hilus of the dentate gyrus and in the stratum pyramidale and lucidum of CA3 showed strong cytoplasmic labeling for GAP-43. Few cells were labeled in these regions at 2 days, and none at later stages. Pyramidal cells in CA1 and CA3 areas and granule cells were never labeled. These studies demonstrate a transient expression of GAP-43 mRNA and protein in a subset of vulnerable neurons after transient brain ischemia. The cytoplasmic localization in hilar neurons could be due to increased synthesis of GAP-43 or to changes in axoplasmic transport. It is suggested that axonal damage occurs in hilar cells which stimulates GAP-43 expression. The increased production of trophic factors after ischemia in granule cells could also cause plastic changes in hilar cells. Since hilar neurons are in a strategic position to control the excitability of the dentate area, increased expression of GAP-43 may indicate an important pathophysiological process. In seizure experiments, strong expression of GAP-43 mRNA in granule cells was associated with abnormal mossy fiber sprouting and development of chronic epilepsy. The relevance of the minor GAP-43 mRNA upregulation after ischemia must be considered. The changes in CA3 neurons at several days after ischemia might represent a plastic response to a loss of CA1 neurons.

C-fos mediates antipsychotic-induced neurotensin gene expression in the rodent striatum

The ubiquitous inducibility of the immediate-early gene c-fos in the central nervous system has led to the search for downstream genes which are regulated by its product, Fos. Recent evidence suggests that c-fos induction by a single injection of the classical antipsychotic haloperidol may contribute to the subsequent increase in neurotensin gene expression in the rodent striatum. Consistent with this proposal, in the present study haloperidol-induced Fos-like immunoreactivity and neurotensin/neuromedin N messenger RNA were found to be expressed by the same population of striatal neurons. Moreover, inhibition of haloperidol-induced c-fos expression by intrastriatal injection of antisense phosphorothioate oligodeoxynucleotides complimentary either to bases 109-126 or 127-144 of c-fos attenuated the subsequent increase in neurotensin/neuromedin N messenger RNA. However, injection of a sense phosphorothioate oligodeoxynucleotide corresponding to bases 127-144 of c-fos did not reduce haloperidol-induced c-fos or neurotensin/neuromedin N expression. Furthermore, constitutive expression of Jun-like immunoreactivity in the striatum was not reduced by either the sense or antisense phosphorothioate oligodeoxynucleotides. Similarly, the sense and antisense phosphorothioate oligodeoxynucleotide failed to reduce proenkephalin messenger RNA, which is located in the same striatal neurons that express haloperidol-induced neurotensin/neuromedin N messenger RNA, which is located in the same striatal neurons that express haloperidol-induced neurotensin/neuromedin N messenger RNA. Lastly, haloperidol-induced increases in nerve growth factor I-A-, JunB- and FosB-like immunoreactivity and fosB messenger RNA were not decreased by intrastriatal injection of either the sense or antisense phosphorothioate oligodeoxynucleotides. These results indicate that the antisense phosphorothioate oligodeoxynucleotides attenuated haloperidol-induced neurotensin/neuromedin N expression by selectively reducing c-fos expression and emphasize the potential importance of immediate-early gene induction in the mechanism of action of this antipsychotic drug.

Colonic neoplasms following ureterosigmoidostomy

Attention is drawn to the occurrence of colonic tumors at the site of ureterosigmoidostomy. A case of carcinoma is described, the literature is reviewed and regular long-term followup is emphasized.