Splenic vein tumor thrombosis is a major prognostic factor in distal pancreatic adenocarcinoma.

BACKGROUND: The prognostic value of splenic vessel involvement in distal pancreatic adenocarcinoma remains controversial. The aim of the study was to assess its prognostic relevance in a large multicenter cohort. METHODS: Patients who underwent pancreatosplenectomy for distal pancreatic adenocarcinoma were identified from 5 pancreatic surgical centers. A pathology review of the surgical specimens was performed to assess splenic vessel involvement, defined as invasion of the vessel's adventitia or deeper, and confirm the presence of splenic vein tumor thrombosis. Prognostic factors associated with overall and relapse-free survival were evaluated. RESULTS: 149 patients underwent upfront surgery. Splenic vascular involvement was observed in 69 of them (46.3%). A parietal infiltration of the splenic artery or splenic vein was observed in 26 (17.5%) and 49 patients (32.8%), respectively. A pathologic tumor thrombosis of the splenic vein was identified in 22 patients (14.8%) and associated with larger tumors (>20 mm) (P = .023), more perineural (P = .017), and lymphovascular (P = .002) invasion, and more positive lymph node (P = .001). After a median follow-up of 50.8 months (95% confidence interval: 44.3-57.3), the cumulative 5-year overall and relapse-free survival were 46.2% and 33%, respectively. In multivariate analysis, in addition to lymph node metastasis (hazard ratio = 1.8; 95% confidence interval [1.1-3.1]; P = .023) and perineural invasion (hazard ratio = 3.5; 95% confidence interval [1.3-9.7]; P = .016), presence of splenic vein tumor thrombosis was the only splenic vascular involvement that affected independently the overall survival (HR = 2.3; 95% confidence interval [ 1.3-4.3]; P = .006). CONCLUSION: In resectable distal pancreatic adenocarcinoma, a pathologic tumor thrombosis of the splenic vein is an independent prognostic factor of overall survival. To define the perioperative oncological strategy, a preoperative evaluation of splenic vessel involvement and thrombosis is needed.

[Eosinophilic inflammation of the digestive tract]. / Inflammation à éosinophiles du tube digestif.

Eosinophilic inflammation of the digestive tract is an inflammatory disease characterized by extensive infiltration of eosinophils into the gastrointestinal tract. It can be either a primary disorder of the digestive tract or be secondary to another cause of tissue eosinophilia. Primary disorders include eosinophilic esophagitis (OE) and eosinophilic gastroenteritis (GEEo). These are 2 rare pathologies considered to be diseases related to a Th2-mediated food allergy. The role of the pathologist is twofold: (1) he must make the diagnosis of tissue esosinophilia and propose the various causes, knowing that a secondary cause is the most frequently observed; (2) identify the abnormal number of polymorphonuclear eosinophils, which implies knowing the normal distribution of eosinophils in the different digestive segments. To carry the diagnosis of EO, the threshold of polymorphonuclear eosinophils must be ≥ 15/fields × 400. There is no predefined threshold concerning the other segments of the digestive tract to carry the diagnosis of GEEO. In addition, to make the diagnosis of primary digestive tissue eosinophilia, the patient must be symptomatic with histological evidence of eosinophilia and have ruled out all secondary causes. The main differential diagnosis of OE is gastroesophageal reflux disease. The differential diagnoses of GEEo are multiple, including primarily drugs and parasitic infections.

The Appendix Orchestrates T-Cell Mediated Immunosurveillance in Colitis-Associated Cancer.

BACKGROUND & AIMS: Although appendectomy may reduce colorectal inflammation in patients with ulcerative colitis (UC), this surgical procedure has been suggested to be associated with an increased risk of colitis-associated cancer (CAC). Our aim was to explore the mechanism underlying the appendectomy-associated increased risk of CAC. METHODS: Five-week-old male BALB/c mice underwent appendectomy, appendicitis induction, or sham laparotomy. They were then exposed to azoxymethane/dextran sodium sulfate (AOM/DSS) to induce CAC. Mice were killed 12 weeks later, and colons were taken for pathological analysis and immunohistochemistry (CD3 and CD8 staining). Human colonic tumors from 21 patients with UC who underwent surgical resection for CAC were immunophenotyped and stratified according to appendectomy status. RESULTS: Whereas appendectomy significantly reduced colitis severity and increased CAC number, appendicitis induction without appendectomy led to opposite results. Intratumor CD3+ and CD8+ T-cell densities were lower after appendectomy and higher after appendicitis induction compared with the sham laparotomy group. Blocking lymphocyte trafficking to the colon with the anti-α4ß7 integrin antibody or a sphingosine-1-phosphate receptor agonist suppressed the inducing effect of the appendectomy on tumors' number and on CD3+/CD8+ intratumoral density. CD8+ or CD3+ T cells isolated from inflammatory neo-appendix and intravenously injected into AOM/DSS-treated recipient mice increased CD3+/CD8+ T-cell tumor infiltration and decreased tumor number. In UC patients with a history of appendectomy, intratumor CD3+ and CD8+ T-cell densities were decreased compared with UC patients without history of appendectomy. CONCLUSIONS: In UC, appendectomy could suppress a major site of T-cell priming, resulting in a less efficient CAC immunosurveillance.

Pathology of Cholangiocarcinomas.

Cholangiocarcinomas (CCA) are heterogeneous tumors that arise from epithelial cells of the biliary tract. They represent the second primary liver malignancy, after hepatocellular carcinoma. Recent epidemiological data show an increased incidence of intrahepatic CCA without any identified causes. According to their location on the biliary tract, intrahepatic, perihilar (p) and distal (d) CCA can be individualized. Intrahepatic CCA (iCCA) are subdivided into small duct type iCCA and large duct type iCCA, according to the level or size of the biliary duct affected. These two subgroups are characterized by distinct risk factors, gross aspect, histopathological and molecular features, and therapeutic management. The role of biopsy in iCCA is to confirm the diagnosis and to eliminate various differential diagnostics, in particular, metastases. In p/d CCA, biopsy requires more invasive approaches, and tissue samples are difficult to obtain, leading to a high rate of false negatives. In this review, we will discuss the different classifications of CCA (anatomical and macroscopic). We will describe the various microscopic and phenotypic subtypes of CCA. Finally, we will deal with their mode of extension, the role of biopsy and pre-neoplastic lesions.

Zinc Finger E-Box Binding Homeobox 1 Promotes Cholangiocarcinoma Progression Through Tumor Dedifferentiation and Tumor-Stroma Paracrine Signaling.

BACKGROUND AND AIMS: Zinc finger E-box binding homeobox 1 (ZEB1) is a transcription factor that promotes metastatic and stem cell features, which has been associated with poor prognosis in cholangiocarcinoma (CCA), a desmoplastic cancer enriched in cancer-associated fibroblasts (CAFs). We aimed to define ZEB1 regulatory functions in malignant and stromal compartments of CCA. APPROACH AND RESULTS: Bioinformatic and immunohistochemical analyses were performed to determine correlations between ZEB1 and markers of progressiveness in human intrahepatic CCA (iCCA). Gain-of-function and loss-of-function models were generated in CCA cells and liver myofibroblasts as a model of CAFs. Conditioned media (CM) was used to unravel tumor-stroma interplay. In vivo experiments were performed using a xenograft CCA model. ZEB1 expression in tumor cells of human iCCA was associated with undifferentiated tumor and vascular invasion. In vitro, ZEB1 promoted epithelial-mesenchymal transition and stemness in tumor cells, leading to cell migration and spheroid formation. In vivo, ZEB1-overexpressing CCA cells formed larger tumors with more abundant stroma. Expression of cellular communication network factor 2 (CCN2, encoding connective tissue growth factor [CTGF]) was increased in tumor cells from ZEB1-overexpressing xenografts and correlated with ZEB1 expression in human tumors. In vitro, CM from ZEB1-overexpressing tumor cells or recombinant CTGF induced myofibroblast proliferation. ZEB1 was also expressed by CAFs in human CCA, and its expression correlated with CCN2 in myofibroblasts and CCA stroma. In mice, cotransplantation of CCA cells with ZEB1-depleted myofibroblasts reduced CCA progressiveness compared to CCA cells/ZEB1-expressing myofibroblasts. Furthermore, ZEB1 controls the expression of paracrine signals (i.e., HGF and IL6) in tumor cells and myofibroblasts. CONCLUSIONS: ZEB1 plays a key role in CCA progression by regulating tumor cell-CAF crosstalk, leading to tumor dedifferentiation and CAF activation.

Smooth muscular layer: A new helpful criterion to reclassify tumor deposits into metastatic lymph nodes in patients with colo-rectal adenocarcinoma.

CONTEXT: The origin of tumor deposit in colorectal cancer is still unknown, and currently there is no single morphological feature to distinguish a metastatic lymph node from a tumor deposit. Histologically, the normal lymph node capsule and trabeculae contain a smooth muscular layer, which when present in extramural deposits would strongly suggest their lymph node origin. OBJECTIVE: We analyze the value of the smooth muscular layer criterion in reclassifying tumor deposit into metastatic lymph node. DESIGN: A total of 458 colo-rectal carcinomas surgical specimens treated or not by neoadjuvant (radio)chemotherapy were retrospectively included. Harvested tumor deposits were analyzed by Hematoxylin and Eosin and elastin staining on 10 consecutive serial sections and by α- smooth muscle actin immunostaining. RESULTS: A total of 129 tumor deposits were identified. 77 (60%) tumor deposits were reclassified into metastatic lymph node, of which 63 (49%) presented a smooth muscular layer on the initial Hematein Eosin staining and/or after serial tissue sections, confirmed by positive α-smooth muscle actin immunostaining in 43 out of 45 cases (90%). Fourteen (18%) additional tumor deposits were reclassified into metastatic lymph node by the appearance of lymphoid tissue after serial sections. CONCLUSIONS: The presence of a smooth muscular layer in a presumable tumor deposit is helpful in pointing out its lymph node origin in patients with colo-rectal carcinomas. This criterion could improve the inter-observer agreement of tumor deposit identification, allowing accurate nodal staging and better assessment of patient's prognosis.

CARMN-NOTCH2 fusion transcript drives high NOTCH2 expression in glomus tumors of the upper digestive tract.

Glomus tumors (GTs) are perivascular tumors mostly occurring in the distal extremities. Rare cases arise in the digestive tract and may be misdiagnosed with neuroendocrine or gastrointestinal stromal tumors. We aimed to specify the features of GT of the upper digestive tract. Clinical, histological, phenotypic, and molecular features of 16 digestive GTs were analyzed, of whom two underwent whole exome and RNA sequencing to search for gene alterations. RNA-sequencing disclosed a t(1:5)(p13;q32) translocation, which resulted in the fusion of CARMN and NOTCH2 in two GTs. The fusion gene encoded a protein sequence corresponding to the NOTCH2 intracellular domain that functions as transcription factor. These finding was supported by high expression of genes targeted by NOTCH. The CARMN-NOTCH2 translocation was detected in 14 out of 16 (88%) GTs of the upper digestive tract; but in only in two out of six cutaneous GTs (33%). Most digestive GT arose from the stomach (n = 13), and the others from duodenal (2) or oesophagous (1). Nuclear expression of NOTCH2 was detected in the 14 cases containing the fusion transcripts. The CARMN-NOTCH2 fusion transcript may contribute to activation of the NOTCH2 pathway in GT and drive tumor development. The high frequency of this translocation in GT of the upper digestive track suggest that detection of nuclear NOTCH2 expression may be useful diagnostic biomarker of these tumors.

Prognostic value of desmoplastic stroma in intrahepatic cholangiocarcinoma.

Intrahepatic cholangiocarcinomas (iCCs) are primary tumors of the liver characterized by the presence of a desmoplastic stroma. While tumor stroma may have a protective or a pejorative value depending on the type of malignant disease, the precise role of the desmoplastic stroma in iCC remains poorly understood. The aim of the present study was to evaluate the prognostic value of stromal compartment in iCC through a multiparametric morphological analysis. Forty-nine surgically resected iCCs were included. For all cases, tumor paraffin blocks of iCCs were selected for stromal morphological characterization through quantitative and qualitative approaches using immunohistochemistry and second-harmonic generation imaging. Intratumor heterogeneity was also evaluated in regards with the different stromal features. High proportionated stromal area (PSA) (defined by stromal to tumor area ratio) was inversely correlated with vascular invasion (62.5% vs 95.7%, p = 0.006) and positively correlated with well-differentiated grade (60% vs 12.5%, p = 0.001). Patients with high PSA had a better disease-free survival (DFS) than patients with low stromal area (60% vs 10%, p = 0.077). Low activated stroma index (defined by cancer-associated fibroblasts number to stromal area ratio) was associated with a better DFS (60% vs 10%, p = 0.05). High collagen reticulation index (CRI), defined as the number of collagen fiber branches within the entire length of the collagen network, was associated with a poorer overall survival (42% vs NR, p = 0.026). Furthermore, we showed that CRI was also an homogeneous marker throughout the tumor. Based on morphological features, desmoplastic stroma seems to exert a protective effect in patients with iCC. Stromal collagen reticulation may provide additional clinically relevant information. In addition, these data support the potential value to evaluate CRI in biopsy specimen.

[Clinicopathologic features and postoperative outcome of the radiation-induced enteritis: a retrospective study of 41 patients]. / Caractéristiques clinicopathologiques et résultats postopératoires de l'entérite radio-induite : une étude rétrospective de 41 patients.

INTRODUCTION: Surgery is required in patients with symptoms of chronic radiation-induced enteritis (CRE) resistant to medical therapy. The study aimed to correlate histopathologic features of CRE to the clinical presentation and the postoperative recurrence. MATERIAL AND METHOD: All patients with small bowel resection performed for CRE between 2006 and 2017 were studied. Histological data were retrospectively correlated to initial clinical data and to postoperative recurrence of CRE (occlusion, need for parenteral nutrition) observed during a median follow-up of 32 months. RESULTS: Forty-one patients were studied (39 women and 2 men, median age 62 yo at time of radiation for pelvic cancer, 80% gynecologic). Median time to surgery after radiation was 3 years. Ileocaecal resections (80% of patients) removed 60cm (median length). Histologically, a diffuse obliterative arteriopathy was present in 24 (59%) patients, highly associated to amyotrophy, villous atrophy and ulceration observed in 66, 63 and 34% of patients respectively (P<.05). Diffuse arteriopathy was uncorrelated with patient's age and vascular risk factor (tobacco, diabetes, hypertension, dyslipidemia). Median time to surgery after radiation was longer in patients presenting with obliterative arteriopathy (13 years vs. 2.6 years, P=0.0002). During follow-up, half of the patients had a recurrence of CRE, uncorrelated to the arteriopathy. CONCLUSION: Radiation-induced enteritis requiring late surgery after radiation presented histologically with a diffuse obliterative arteriopathy and ischemic features. In our center, half of the patients were cured by surgery. The arterial injury was not a risk factor for postoperative recurrence.

Microscopic intramural extension of rectal cancer after neoadjuvant chemoradiation: A meta-analysis based on individual patient data.

OBJECTIVE: In selected rectal cancer patients with residual local disease following neoadjuvant chemoradiation (CRT) and the preference of an organ preservation pathway, additional treatment with dose escalation by endoluminal radiotherapy (RT) may ultimately result in a clinical complete response. To date, the widespread introduction of selective endoluminal radiation techniques is hampered by a lack of evidence-based guidelines that describe the radiation treatment volume in relation to the residual tumor mass. In order to convert an incomplete response into a complete one with additional treatment such as dose-escalation with endoluminal RT from a theoretical perspective, it seems important to treat all remaining microscopic tumor cells after CRT. In this setting, residual tumor extension beneath normal appearing mucosa (microscopic intramural spread - MIS) becomes relevant for accurate tumor volume and margin estimation. With the goal of providing evidence-based guidelines that define an appropriate treatment volume and patient selection, we present results from a meta-analysis based on individual patient data of studies that have assessed the extent or range of MIS of rectal cancers after neoadjuvant CRT. This meta-analysis should provide an estimate of the residual tumor volume/extension that needs to be targeted by any additional radiation therapy boost in order to achieve complete tumor eradication after initial incomplete or near-complete response following standard CRT. METHODS AND MATERIALS: A PubMed search was performed. Additional articles were selected based on identification from reference lists. Papers were eligible when reporting MIS in patients who were treated by total mesorectal excision or local excision/transanal endoscopic microsurgery (TEM) after neo-adjuvant long-course CRT. The mean MIS was calculated for the entire group along with the 70th until 95th percentiles. Additional exploratory subgroup analyses were performed. RESULTS: Individual patient data from 349 patients with residual disease from five studies were analyzed. 80% of tumors showed no MIS. In order to appropriately treat MIS in 95% of rectal cancer patients after CRT, a margin of 5.5 mm around the macroscopic tumor would suffice. An exploratory subgroup analysis showed that T-stage after CRT (ypT) and time interval between neoadjuvant CRT and surgery are significant factors predicting the extent of MIS (p < 0.001.) The group of ypT1 had the smallest MIS, followed by the ypT3-4 group, while the ypT2 group had the largest MIS (p < 0.001). Regarding time interval between CRT and surgery, a statistically significant difference was seen when comparing the three time-interval groups (less than 8 weeks, 8-12 weeks, and more than 12 weeks), where waiting more than 12 weeks after CRT resulted in the largest MIS (p < 0.0001). CONCLUSION: Based on this meta-analysis, in order to treat the MIS for 95% of rectal cancer patients after CRT, a Clinical Target Volume (CTV) margin of 5.5 mm from the lateral most edge of the macroscopic tumor would suffice. 80% of tumors showed no MIS and would not require an extra CTV margin for treatment. These findings support the feasibility of localized radiotherapy boosts for dose-escalation to improve response among patients with incomplete response after standard CRT and can also be applied in the surgical setting.

Is Routine Splenectomy Justified for All Left-Sided Pancreatic Cancers? Histological Reappraisal of Splenic Hilar Lymphadenectomy.

BACKGROUND: Although splenectomy is recommended during resection for left-sided resectable pancreatic ductal adenocarcinoma (PDAC) to perform lymphadenectomy of station 10 (splenic hilum), no level I evidence justifies this procedure. This study aims to evaluate the rate of lymph node (LN) and contiguous involvement of the splenic hilum in resectable distal PDAC. METHODS: We retrospectively reviewed all patients who underwent splenopancreatectomy for PDAC in the past 10 years. Station 10 LN were routinely isolated, and all corresponding microscopic slides were reinterpreted by a pathologist. The computed tomography (CT) results of patients with tumoral involvement of the spleen or splenic hilum by contiguity (TISOSH) and ≤ 10 mm between the tumor and spleen on pathology were blindly reviewed by two radiologists to evaluate CT for diagnosis of TISOSH. RESULTS: We included 110 consecutive patients, including 104 with analyzable station 10 LN. The tumor was N+ in 58 (53%) patients. The median number of LN identified at station 10 was 2.0 ± 3.0. No station 10 LNs were detected in 42 (40%) patients. No patients had tumor-positive LN at station 10. TISOSH was found in nine (8%) patients, and was significantly associated with tail location (p = 0.001), tumor size (p = 0.005), and multivisceral involvement (p = 0.015). For diagnosis of TISOSH, the sensitivity and specificity of CT were respectively 89% and 95% for radiologist 1 and 89% and 100% for radiologist 2. CONCLUSIONS: Splenic preservation during resection of distal PDAC may be an option in selected patients with body tumors and no suspected splenic or splenic hilum involvement on preoperative CT.

Is magnetic resonance imaging useful for the management of patients with rectal villous adenoma? A study of 45 consecutive patients treated by transanal endoscopic microsurgery.

PURPOSE: Very few data are available about the clinical relevance of magnetic resonance (MR) imaging in preoperative evaluation of rectal villous adenoma. The aim is to evaluate the impact of MR imaging for the surgical management of rectal villous adenoma treated by transanal endoscopic microsurgery (TEM). METHODS: All patients with histologically proven rectal villous tumours operated by TEM who had a preoperative MR imaging between 2009 and 2017 were retrospectively reviewed. All patients underwent TEM because preoperative evaluation suggested systematically usT0 or usT1 tumour. Pathological stage was blindly compared to preoperative MR imaging (location according to the anal verge and the peritoneal reflection, amount of circumferential involvement, tumour size and staging) and preoperative transrectal ultrasonography (TRUS) results. RESULTS: Forty-five patients were included (24 men, mean age 65 ± 8 years) with TRUS data available only in 37. Pathologic results were pT0-pTis in 32, pT1 in 10 and pT2 in 3. TRUS diagnosed correctly 36/37 lesions (97%) and understaged one pT2 tumour. A significant correlation between TRUS and pathologic results was noted (r = 0.99; p = 0.01). MR imaging diagnosed correctly 19/42 pTis-T1 and 1/3 pT2 tumours (46%). Overstaging by MR imaging was noted in 25 cases (54%). No correlation between MR imaging and pathologic results was noted (r = 0.7; p = 0.3). CONCLUSION: Preoperative evaluation of rectal villous adenoma is overstaged by MRI in more than half of the patients. This study suggests that the indication of local excision by TEM for rectal villous adenoma should be based on TRUS rather than on MRI.

Small bowel adenocarcinoma complicating Crohn's disease: a single-centre experience emphasizing the importance of screening for dysplasia.

Small bowel adenocarcinoma (SBA) complicating Crohn's disease (CD) is rare and generally found incidentally on surgical specimens. We report our experience in CD-associated SBA observed this last decade in a tertiary referral centre in order to update its incidence, clinical presentation and pathological features. All SBAs diagnosed in patients who underwent surgery for CD between 2006 and 2016 were retrospectively included. Clinico-pathological characteristics were reviewed, and follow-up was updated. SBA was diagnosed in 9 (1.7%) of 522 patients who underwent SB resection(s) after a median CD duration of 15 years [0-32]. The median age at diagnosis was 46 years. Seven (78%) patients had obstructive symptoms refractory to medical treatment. Pre-operative biopsy revealed neoplasia in five (56%) patients (dysplasia in three and SBA in two) justifying the surgery. Two (29%) of the seven patients with imaging had features suggestive of cancer. In all specimens, SBA developed in active ileitis with adjacent dysplasia. Stage I low-grade tubulo-glandular adenocarcinoma was observed in 33% of patients. Stage IV high-grade adenocarcinoma was observed in 56% of patients, and mucinous/signet ring cell differentiation predominated in 44% of patients. Molecular analysis showed no BRAF mutation, a KRAS mutation in one case and a microsatellite instability phenotype suggestive of Lynch syndrome in one case. After a median follow-up of 24 months [7-82], four (44%) patients died with advanced stage IV SBA. This surgical series confirms that CD-associated SBA is rare with an incidence of 1.7%. Adjacent dysplasia was present in all specimens and was identified before surgery in all patients who benefit from ileal biopsies. This strengthens the importance of screening all longstanding CD by endoscopy if surgery is not considered.

Epithelial-mesenchymal transition in cholangiocarcinoma: From clinical evidence to regulatory networks.

Cholangiocarcinoma (CCA) is an aggressive tumor with a poor prognosis due to its late clinical presentation and the lack of effective non-surgical therapies. Unfortunately, most of the patients are not eligible for curative surgery owing to the presence of metastases at the time of diagnosis. Therefore, it is important to understand the steps leading to cell dissemination in patients with CCA. To metastasize from the primary site, cancer cells must acquire migratory and invasive properties by a cell plasticity-promoting phenomenon known as epithelial-mesenchymal transition (EMT). EMT is a reversible dynamic process by which epithelial cells gradually adopt structural and functional characteristics of mesenchymal cells, and has lately become a centre of attention in the field of metastatic dissemination. In the present review, we aim to provide an extensive overview of the current clinical data and the prognostic value of different EMT markers that have been analysed in CCA. We summarize all the regulatory networks implicated in EMT from the membrane receptors to the main EMT-inducing transcription factors (SNAIL, TWIST and ZEB). Furthermore, since a tumor is a complex structure not exclusively formed by tumor cells, we also address the prominent role of the main cell types of the desmoplastic stroma that characterizes CCA in the regulation of EMT. Finally, we discuss the therapeutic considerations and difficulties faced to develop an effective anti-EMT treatment due to the redundancies and bypasses among the pathways regulating EMT.

Distal intramural and tumor spread in the mesorectum after neoadjuvant radiochemotherapy in rectal cancer: about 124 consecutive patients.

This observational prospective study aimed to assess the distribution of intramural and mesorectal tumor spread in mid/low rectal cancer after neoadjuvant radiochemotherapy. Distribution of mesorectal metastatic lymph nodes (MLNs) and mesorectal extranodal cancer tissue (EX), according to the tumor location, were analyzed. Distal intramural tumor spread was also performed. A total of 1676 LNs, 135 MLNs, and 69 EX were detected on 124 consecutive surgical specimens. Forty-two patients (34%) had MLNs. Six patients (4.8%) were classified as ypN1c. Distal viable cancer spread was observed in 3 patients (2.4%), all with mid rectal carcinoma. Two patients (1.6%) presented distal direct intramural extension less than 1 cm; and 1 (0.8%), with EX localized no more than 2 cm from the lower edge of the tumor. MLNs (76%) and EX (94%) were preferentially localized in the peritumoral area and in the first 3 cm just above the tumor. No viable distal intramural or mesorectal spread was observed in low rectal carcinoma. Distal intramural and mesorectal cancer spread is a rare event after neoadjuvant RCT. These results suggest that the 1-cm distal margin recommended in patients with low rectal carcinoma could be reduced with insurance to obtain a negative distal margin. The knowledge of preferential localization of MLNs and EX would help the pathologist to improve patient's lymph node staging.

Local excision of low rectal cancer treated by chemoradiotherapy: is it safe for all patients with suspicion of complete tumor response?

PURPOSE: The purpose of this study is to assess if local excision (LE) could be proposed if suspicion of complete tumor response (CR) after neoadjuvant chemoradiotherapy (CRT) for low rectal cancer (LRC) and this despite a potential risk of nodes (N+) or other tumor deposits (OTD) left in place. The aim was to assess in patients with LRC treated by CRT: (a) pathologic results of LE and total mesorectal excision (TME) in case of preoperative suspicion of CR and (b) the risk of N+ or OTD on TME if ypT0-Tis-T1 tumor. PATIENTS: Among 202 patients with LRC after CRT, 33 (16 %) with suspicion of CR underwent LE (n = 20) because of comorbidities and/or indication of definitive stoma or TME (n = 13). Pathologic examination of LE and TME specimens and oncological outcomes were assessed. Furthermore, 40/202 patients with pathologic CR on TME specimen (ypT0-Tis-T1) were assessed for possible N+ or OTD. RESULTS: In the 33 patients with suspicion of CR: (a) after LE, tumor was ypT0-Tis-T1 in only 15/20 cases (75 %); (b) after TME, tumor was ypT0-Tis-T1 in only 7/13 cases (54 %). Among 40 patients with ypT0-Tis-T1 tumor on TME specimen, 4 (10 %) presented N+ and/or OTD. CONCLUSION: In LRC with suspicion of CR after CRT, LE deserves a word of caution: 25 % of patients have in fact ypT2-T3 tumors. Furthermore, in patients with ypT0-Tis or T1 on TME specimen, a 10 % risk of N+ and/or ODT is observed. Thus, patient with suspicion of CR after CRT and treated by LE is exposed to a possible incomplete oncologic treatment.

Loss of ezrin in human intrahepatic cholangiocarcinoma is associated with ectopic expression of E-cadherin.

AIMS: Ezrin connects proteins from the plasma membrane to the subcortical cytoskeleton, and contributes to epithelial integrity by interacting with the cell-cell adhesion molecule E-cadherin. In the liver, ezrin is restricted to cholangiocytes, where it regulates biliary secretory functions. During carcinogenesis, ezrin expression is impaired and associated with enhancement of cell migratory activity in cancer cells; therefore, we aimed to analyse ezrin in cholangiocarcinogenesis. METHODS AND RESULTS: Ezrin expression was evaluated by immunohistochemistry on tissue microarrays from 94 surgical specimens of intrahepatic cholangiocarcinoma (CCA), and correlated with clinicopathological factors and E-cadherin expression. Ezrin function was also analysed in human CCA cell lines. In CCA, ezrin was negative/weakly expressed in 49 cases (52%) and moderately/strongly expressed in 45 cases (48%), mostly in cell cytoplasm. The negative/weak expression of ezrin was more frequent in peripheral than in perihilar CCA (P = 0.002), and was associated with high tumour size (P = 0.001), low mucus secretion (P = 0.042), the presence of satellite nodules (P = 0.024), and ectopic cytoplasmic expression of E-cadherin (P = 0.005). In vitro, silencing of ezrin in CCA cells caused internalization of E-cadherin and favoured cell migration. CONCLUSIONS: Ezrin is down-regulated during cholangiocarcinogenesis, and its loss results in a more aggressive phenotype.