Reproducible and clinically meaningful differential diagnosis is possible between lobular endocervical glandular hyperplasia and 'adenoma malignum' based on common histopathological criteria
Department of Pathology II, National Defense Medical College, Tokorozawa, Japan. Pathology International
(Impact Factor: 1.69).
07/2005; 55(7):412-8. DOI: 10.1111/j.1440-1827.2005.01846.x
The aim of the present study was to determine if the differential diagnosis between lobular endocervical glandular hyperplasia (LEGH) and minimal deviation adenocarcinoma (MDA), or 'adenoma malignum', is reproducible when clear criteria for these two lesions are given. A total of 44 proliferative endocervical glandular lesions were collected, for which differential diagnosis from MDA was considered to be necessary. Seven observers independently classified these 44 lesions into LEGH, LEGH with adenocarcinoma in situ (AIS), MDA, or common cervical adenocarcinoma, according to the following criteria: LEGH was non-invasive proliferation of endocervical glandular cells without any obvious adenocarcinoma component. MDA was very well-differentiated endocervical-type mucinous adenocarcinoma composed mostly of LEGH-looking glands but containing the component of obviously invasive adenocarcinoma. LEGH with AIS was defined as continuous coexistence of LEGH and AIS. Among these four diagnostic categories, the interobserver agreement level was substantial (kappa = 0.618). The level increased to almost perfect (kappa = 0.928) between the group of non-invasive lesions consisting of LEGH and LEGH with AIS and the other group of invasive lesions comprising MDA and common adenocarcinoma. When the modal diagnosis was adopted as the final diagnosis for individual lesions, the 5 year survival rate of patients after surgery was 100% for the non-invasive lesions but only 54% for the invasive lesions (P < 0.01). It is clearly shown that reproducible differential diagnosis is possible between LEGH, LEGH with AIS, and MDA and that such a differentiation is clinically meaningful.
Available from: Shin Nishio
- "Before LEGH became an established clinical entity, adenocarcinoma with LEGH components might have been included in MDA, because the LEGH component was believed to constitute malignant glands. In our previous studies, both true MDA and adenocarcinoma with LEGH components were included in MDA  . True MDA, extremely well-differentiated mucinous adenocarcinoma, is composed mainly of well-formed glands resembling LEGH. "
[Show abstract] [Hide abstract]
ABSTRACT: Lobular endocervical glandular hyperplasia (LEGH) is usually assumed to be a benign tumor-like lesion of the glands of the uterine cervix. However, LEGH has been associated with obvious cervical adenocarcinoma. The clinicopathological significance of coexistence of LEGH with adenocarcinoma remains unclear. We microscopically examined the presence or absence of LEGH components in 95 stage Ib cervical adenocarcinomas. Gastric mucin was detected with the use of clone HIK1083. Associations of the coexistence of LEGH components with clinicopathological variables were analyzed. LEGH components were present in 16 cases (16.8%). Gastric mucin was positive in all 16 LEGH components, as compared with only 6 of the 95 adenocarcinoma components. Of the 16 adenocarcinomas with LEGH components, 15 were well-differentiated mucinous adenocarcinomas, and one was poorly differentiated adenocarcinoma. The mortality rate of tumor recurrence was 25% (4 of 16) in patients whose tumors had LEGH components, and 21.5% (17 of 79) in those whose tumors had no LEGH components. There was no significant difference in survival. Early cervical adenocarcinoma was relatively frequently associated with LEGH components. LEGH may be one of the factors related to the development of cervical adenocarcinoma, but adenocarcinoma with LEGH components does not necessarily develop into a highly aggressive "adenoma malignum."
[Show abstract] [Hide abstract]
ABSTRACT: Endocervical-type mucinous adenocarcinoma (ECA) of the uterine cervix is defined as a tumor composed of cells resembling those of the endocervical glands, but recent studies have demonstrated that a minority of ECAs displays a gastric immunophenotype. The aim of this study was to assess the significance of the gastric phenotype. Fifty-three cases of mucinous adenocarcinoma of the uterine cervix (37 FIGO stage IB, 4 stage IIA, and 12 stage IIB) were reviewed and reevaluated using a newly established morphologic criteria for distinguishing gastric type adenocarcinoma, which was defined as a tumor showing clear and/or pale eosinophilic and voluminous cytoplasm, with distinct cell borders. The results were correlated with gastric immunophenotype, determined by HIK1083 and MUC6 immunostaining, and patient outcome. Following the current World Health Organization scheme (2003), 47 tumors (89%) were classified as ECA, 1 (2%) as intestinal type, 1 (2%) as mixed endocervical and intestinal type, and 4 (8%) as minimal deviation adenocarcinoma. Twelve of 47 (26%) ECAs and all 4 minimal deviation adenocarcinomas, reclassified as gastric type using the novel criteria, were frequently positive for HIK1083 with a rate of 75% (12/16), whereas only 11% (4/37) of nongastric tumors were positive. There was no significant difference in MUC6 reactivity between gastric and nongastric type tumors (31%, 5/16 vs. 16%, 6/37; P=0.4). Patients with gastric-type adenocarcinomas had a significantly decreased 5-year disease-specific survival rate (30 vs. 77%; P<0.0001), and the gastric type morphology was related to a significant risk for disease recurrence compared with the nongastric type (P=0.001; HR, 4.5; 95% confidence interval, 1.42-14.2). HIK1083-positivity was also related to decreased 5-year disease-specific survival rate (38% vs. 74%; P<0.005). Mucinous adenocarcinoma of the uterine cervix with gastric immunophenotype can be a distinct morphologic variant showing an aggressive clinical course.
[Show abstract] [Hide abstract]
ABSTRACT: The aim of this study was to investigate differences in the process of carcinogenesis between adenocarcinoma coexistent with LEGH and conventional adenocarcinoma. And we intend to describe appropriate treatment plans for LEGH in this study.
Using the surgical pathology files of patients who visited the University of Yamanashi Hospital, Yamanashi Central Hospital and Kofu Municipal Hospital between 1996 and 2005, pathological diagnoses were reevaluated based on criteria for the diagnosis of LEGH by Nucci et al. As for the cases including adenocarcinoma with LEGH: (a) we created a map showing position of the LEGH component and adenocarcinoma component and squamo-columnar junction (SCJ) in HE-stained specimens, (b) immunohistochemical staining was performed using antibodies to CEA, HIK1083 and p53, and (c) detection of HPV DNA was performed using PCR and in situ hybridization (ISH).
Endocervical adenocarcinoma was observed coexistent with LEGH in 5 cases (19.2%). (a) LEGH was located in a remote place from the SCJ. Sizes of lesions in the 5 cases ranged from 18 to 35 mm in width and 7 to 16 mm in depth. (b) HIK1083 was diffusely immunopositive in the cytoplasm of LEGH component and focal immunopositive in 4 cases with adenocarcinoma component. Immunopositivity for CEA was seen in the cytoplasm of adenocarcinoma component in 4 cases. Immunopositivity for p53 was seen in adenocarcinoma component nuclei in 2 cases. (c) HPV DNA was not detected using PCR and ISH in either LEGH or adenocarcinoma components.
The present study suggests that clear differences exist in the process of carcinogenesis between adenocarcinoma associated with LEGH and conventional adenocarcinoma. LEGH may represent a precursor of cervical adenocarcinoma independent of HPV infection. As LEGH displays characteristics of precancerous mucinous adenocarcinoma, surgical treatment should be considered for LEGH growing beyond a certain size.
Data provided are for informational purposes only. Although carefully collected, accuracy cannot be guaranteed. The impact factor represents a rough estimation of the journal's impact factor and does not reflect the actual current impact factor. Publisher conditions are provided by RoMEO. Differing provisions from the publisher's actual policy or licence agreement may be applicable.