RESUMEN
AIMS: Lymph node metastasis (LNM) has not been found in more than 85% of patients with early invasive colorectal adenocarcinoma (T1-CRAC) who undergo surgery after therapeutic endoscopy due to the risk for LNM. Better histological risk assessment for LNM of endoscopically resected T1-CRAC is important to avoid unnecessary additional surgery. METHODS AND RESULTS: We evaluated cancer gland rupture (CGR), i.e. cancer glands with a discontinuous epithelial lining, at the invasive front, as a potential risk factor for LNM by histological examination of differentiated T1-CRAC from 217 patients who underwent surgery with or without therapeutic endoscopy. CGR was represented by C-shaped neoplastic glands with a variable inflammatory or stromal reaction, and was occasionally accompanied by mucus lake or abscess formation. CGR was observed in 168 (77%) cases, including all 20 cases with LNM, and the odds ratio of LNM was higher for CGR than for deep invasion (depth of submucosal invasion ≥1000 µm). All cases with LNM were found among 148 cases with deep invasion and positive CGR, whereas no LNM was detected in 29 cases with deep invasion and negative CGR, regardless of vascular invasion or tumour budding. In the 148 cases, LNM was detected in 18 (19%) of 93 cases with positive vascular invasion or high-grade tumour budding, and in two (4%) of 55 cases without either. CONCLUSIONS: Our findings suggest that CGR is an easily applied and objective histological finding for predicting LNM that could be useful for assessing the risk for LNM of endoscopically resected T1-CRAC with deep invasion.
Asunto(s)
Adenocarcinoma/patología , Algoritmos , Neoplasias Colorrectales/patología , Metástasis Linfática/patología , Invasividad Neoplásica/patología , Adenocarcinoma/cirugía , Anciano , Neoplasias Colorrectales/cirugía , Endoscopía del Sistema Digestivo , Femenino , Humanos , Masculino , Persona de Mediana Edad , Factores de RiesgoRESUMEN
BACKGROUND: Sarcoidosis is caused by Th1-type immune responses to unknown agents, and is linked to the infectious agent Propionibacterium acnes. Many strains of P. acnes isolated from sarcoid lesions cause intracellular infection and autophagy may contribute to the pathogenesis of sarcoidosis. We examined whether P. acnes induces autophagy. METHODS: Three cell lines from macrophages (Raw264.7), mesenchymal cells (MEF), and epithelial cells (HeLa) were infected by viable or heat-killed P. acnes (clinical isolate from sarcoid lymph node) at a multiplicity of infection (MOI) of 100 or 1000 for 1 h. Extracellular bacteria were killed by washing and culturing infected cells with antibiotics. Samples were examined by colony assay, electron-microscopy, and fluorescence-microscopy with anti-LC3 and anti-LAMP1 antibodies. Autophagy-deficient (Atg5-/-) MEF cells were also used. RESULTS: Small and large (≥5 µm in diameter) LC3-positive vacuoles containing few or many P. acnes cells (LC3-positive P. acnes) were frequently found in the three cell lines when infected by viable P. acnes at MOI 1000. LC3-positive large vacuoles were mostly LAMP1-positive. A few small LC3-positive/LAMP1-negative vacuoles were consistently observed in some infected cells for 24 h postinfection. The number of LC3-positive P. acnes was decreased at MOI 100 and completely abolished when heat-killed P. acnes was used. LC3-positive P. acnes was not found in autophagy-deficient Atg5-/- cells where the rate of infection was 25.3 and 17.6 times greater than that in wild-type Atg5+/+ cells at 48 h postinfection at MOI 100 and 1000, respectively. Electron-microscopic examination revealed bacterial cells surrounded mostly by a single-membrane including the large vacuoles and sometimes a double or multi-layered membrane, with occasional undigested bacterial cells in ruptured late endosomes or in the cytoplasm. CONCLUSION: Autophagy was induced by intracellular P. acnes infection and contributed to intracellular bacterial killing as an additional host defense mechanism to endocytosis or phagocytosis.