| BackgroudCervical cancer is the third most common cancer in women worldwide, resulting in an estimated 275,000 deaths in 2008. Numerous studies have demonstrated that a higher CIN grade correlates with a greater risk that lesions will progress to invasive cervical cancer. Liquid-based cytology (LBC) was proved to be an alternative to conventional cytology with improved specimen adequacy and sensitivity in detecting cervical abnormalities. Usually, most of high-grade squamous intraepithelial lesions (HSIL) and squamous cell carcinomas (SCC) on LBC slides do not pose diagnostic difficulties. However, It is sometimes difficult to reliably distinguish SCC from HSIL due to low cellularity, lack of tumor diathesis and overlapping tissue fragments. Cell blocks have been recognized as a useful adjunct in the diagnosis of nongynecologic cytology. Cell blocks were found to provide good tissue architecture and multiple sections for performing ancillary tests such as immunohistochemistry, in situ hybridization and ploidy analysis.Cervical cytologic diagnoses for unequivocal squamous epithelial abnormalities include low-grade squamous intraepithelial lesions (LSIL), high-grade squamous intraepithelial lesions (HSIL), and squamous cell carcinoma (SCC).This classification scheme provides standardized terminology and guidance for clinical management. Despite the clear guidelines for interpretation of Pap tests, many specimens contain cells that cannot be easily placed into the accepted schema. The diagnostic dilemma of overestimating LSIL or underestimating HSIL is made more difficult by the fact that the Pap test specimen is one of the most litigated of all pathologic specimens. To reduce the discordance rate between cytology diagnosis and follow-up biopsies, a variety of objective and reliable diagnostic tools, including immunocytochemistry, have been investigated.p16INK4a (p16), a tumor suppressor protein that regulates cell proliferation by inhibiting cell cycle G1 progression, has been demonstrated to be strongly overexpressed in almost all high-grade precursor lesions and invasive cancers of the cervix uteri. In cervical cytology, p16 also has been used as a surrogate marker for the presence of HSIL or more advanced lesions. Because HSIL and SCC consistently express p16, immunostaining for this marker has no practical utility in distinguishing SCC from HSIL.Insulin-like growth factor Ⅱ messenger RNA (mRNA)-binding protein 3 (IMP3) is a member of the insulin-like growth factor mRNA binding protein family.IMP3 is an oncofetal protein and is expressed in developing tissues during embryogenesis but not in adult human normal cells.IMP3 expression promotes cell adhesion and invadopodia formation, thereby suggesting that IMP3 plays a pivotal role in the migration and invasion found during cancer development and cancer progression. Previous studies have demonstrated that IMP3 is overexpressed in various malignancies.Data regarding the expression of IMP3 in cervical squamous cell carcinoma (SCC) are scarce. A recent study has found that IMP3 is not expressed in normal tissue and CIN Ⅰ but can be detected in CIN Ⅱ, CIN Ⅲ, and SCC. However, the prognostic relevance of IMP3 expression in cervical SCC has not yet been assessed.We characterize the morphological features of gynecological samples, and compare the morphological features of HSIL and SCC sections, and then we investigate the diagnostic value of IMP3 expression in cell blocks from residual LBC and to determine whether p16 and IMP3 can serve as a molecular biomarkers to reduce the discordance rate between LBC and follow-up biopsies. We examined IMP3 expression in cervical carcinoma cell lines and determined the correlation of IMP3 expression. with clinicopathological parameters and patient outcome. In addition, we examined the staining pattern of IMP3 in CIN Ⅰ-Ⅲ and SCC to study its role in the development of these lesions.Three parts were investigated in this paper.Objective1.To develop technique to prepare cell block from LBC.2. To evaluate the diagnostic value and compare morphological features of cell block sections of high-grade squamous intraepithelial lesions (HSIL) and squamous cell carcinomas (SCC).3. To evaluate the utility of p16INK4a (p16) and IMP3 staining of LBC specimens to increase the concordance rate.4. The correlation between IMP3 expression and cervical cancer progression and prognosis was assessed by immunohistochemistry.Methods1. Morphological Features of Cell Blocks Prepared from Residual Liqui-PREP Samples Can Distinguish between High-Grade Squamous Intraepithelial Lesions and Squamous Cell Carcinomas.Liqui-PREP cleaning solution was poured into cell block. The slide was stained with the Papanicolaou stain. All slides were classified according to the 2001 Bethesda System. The following morphological features were as-sessed on the slides from each case.2. Combined detection of p16INK4a and IMP3 can increase the concordance rate between cervical cytologic and histologic diagnosisProblematic cases were selected to prepare cell blocks from residual Liqui-PREP samples. For the purposes of the current study, all LBC slides corresponding to the cell blocks were reviewed and cytological diagnoses (according to the 2001 Bethesda System) were confirmed or reassigned based on a consensus reached by two pathologists. All cervical samples were poured into cell block and detected with antibody against IMP3 and p16 by immunocytochemistry.3. IMP3 expression is associated with poor survival in cervical squamous cell carcinomaIMP3 expression was detected in normal cervical tissue, CIN and SCC by Immunohistochemical staining, in addition, IMP3 expression were detected in cervical carcinoma cell lines from mRNA and protein level by Real-time PCR and Western blot. The relationship between IMP3 expression and clinicopathological characteristics of patients and survival analysis were investigated.4. Statistical analysisThe SPSS software 13.0 was used for the statistical analysis. The χ2 test was used to assess the association between categorical variables. Single factor survival analysis was used by log-Rank and Kaplan-Meier curve, Multivariate analysis of prognosis by Cox model. Real-time PCR was assessed by using the one-way ANOVA. Differences are considered statistically significant at p< 0.05.Results1. Morphological Features of Cell Blocks Prepared from Residual Liqui-PREP Samples Can Distinguish between High-Grade Squamous Intraepithelial Lesions and Squamous Cell Carcinomas.Ninety-three percent (40/43) of cell block diagnoses were consistent with histological diagnoses, which was better than the cytological diagnoses (81.4%;35/43). The mean cell block size was 0.54 cm (range,0.3-1.0cm) for HSIL and 0.84 cm (range,0.4-1.4 cm) for SCC (P<0.0001). Cell clusters were present in 70.8%(17/24) of HSIL and 100%(19/19) of SCC(P<0.0001). No epithelial-stromal interface, stromal invasion or tumor necrosis was observed on HSIL cell block sections, which is in contrast to the 84.2%(16/19),68.4%(13/19) and 42.1%(8/19) on SCC cell blocks, respectively (P<0.0001).2. Combined detection of p16INK4a and IMP3 can increase the concordance rate between cervical cytologic and histologic diagnosisA total of 98 cell blocks with biopsy results, including 37 low-grade squamous intraepithelial lesions (LSIL),36 high-grade squamous intraepithelial lesions (HSIL), and 25 squamous cell carcinomas (SCC), were selected for the immunocytochemical analysis of p16 and IMP3. The LBC diagnoses corresponded with histological diagnoses for 59.5%(22/37),63.9%(23/36), and 88.0%(22/25) of LSIL, HSIL, and SCC lesions, respectively. We found a high frequency of p16 positivity in HSIL (72.2%) and SCC (100%), but not LSIL (29.7%). IMP3 was frequently expressed in SCC (84.0%), but rarely in LSIL (8.1%) and HSIL (25.0%). Cervical intraepithelial neoplasia 1 (CIN1) was negative for both p16 and IMP3, CIN2/3 tended to be positive for p16 and negative for IMP3, and SCC was positive for both p16 and IMP3. IMP3 is a useful diagnostic immunomarker that can be used to identify SCC and the combination of p16/IMP3 expression was found to improve the discrepant results between cytologic and histologic diagnoses. The staining patterns of p16 and IMP3 were divided into the following three categories:p16-/IMP3-, p16+/IMP3-, or p16+/IMP3+. Our data showed that CIN1, CIN2/CIN3, and SCC lesions all displayed distinct staining patterns of p16 and IMP3. SCC was frequently p16+/IMP3+, CIN2/3 was frequently p16+/IMP3-, and p16-/IMP3-was more frequently seen in CIN1.3. IMP3 expression is associated with poor survival in cervical squamous cell carcinoma(1) IMP3 expression in normal cervix, CIN and cervical SCCImmunohistochemical staining demonstrated that IMP3 was absent from all 40 normal cervix samples. In CIN specimens, IMP3 was expressed in nine of 204 cases (4.4%). Within the CIN cases, differential expression of IMP3 depended on the grade of the lesion. IMP3 expression was observed in 0 of 62 (0%) CIN Ⅰ,0 of 38 (0%) CIN Ⅱ, and nine of 104 (8.7%) CIN Ⅲ specimens. Of the 96 SCC cases, IMP3 expression was detected in 54 cases (56.3%). Significant difference in IMP3 expression existed between the all of the groups tested (p<0.001).Weak IMP3 expression was seen in eight of the nine (88.9%) CIN Ⅲ cases that were scored as IMP3 positive. In these cases, IMP3 staining was localized to the glandular involvement area. IMP3 expression was observed in the center of glandular involvement area, but was absent in the basal layer. Among the IMP3-positive SCC cases, IMP3 staining intensity was weak, moderate and strong in 15 (27.8%),20 (37.0%) and 19 (35.2%) specimens, respectively. The percentage of positive cells were scored using the following five categories:0 (1%-10% of cells stained),1 (11%-25%),2 (26%-50%),3 (51%-75%) and,4 (76%-100%). In SCC cases,11 (20.3%) samples were scored as 1,9 (16.7%) samples were scored as 2,9 (16.7%) samples were scored as 3, and 25 (46.3%) sample were scored as 4.SCC cases were further divided into a diffuse-positive group (88.9%,48 of 54 cases) and a central-negative group (11.1%, six of 54 cases), based on IMP3 staining. In the central-negative group, IMP3 staining was localized in the periphery of the tumor nests, with no IMP3 staining seen in the centers. IMP3 staining was localized in all the layers of the tumor nests in diffuse-positive group. Histologically, the central-negative group was characterized by a nesting growth pattern. Keratinization and intercellular bridges were evident in the periphery of the tumor nests, but keratin pearls were rarely observed. Less well-differentiated cells and mitotic figures are usually seen in in the centers of the nests.(2) Relationship between IMP3 expression and clinicopathological characteristicsClinicopathological features of each case was recorded, and its relationship to IMP3 expression was determined. IMP3 expression was significantly related to age (p<0.001), FIGO stage (p<0.001), and lymph node metastasis (p=0.00\). In contrast, little relationship was seen between IMP3 expression and histological grade (p=0.125), invasive interstitial depth (p=0.242), tumor size (p=0.298) and lympho-vascular space invasion (p-0.459).(3) Prognostic significance of IMP3 expressionIn order to determine the value of IMP3 as a prognostic indicator, follow-up studies were performed on patients from which clinical samples were obtained. Sixty patients were followed up for up to 60 months. The median survival time of patients in this study was 58.13 months (range,9-87). During the observation period,18 patients (30.0%) died.The Kaplan-Meier method was used to demonstrate that patients with IMP3-positive specimens had a poorer survival rate than those patients with IMP3-negative specimens (log-rankp=0.005).(4) IMP3 protein and mRNA expression in cervical cancer cell linesThe expression of IMP3 mRNA, as measured by the relative mRNA level, was determined in SiHa and HeLa cell lines, or normal cervix tissue. The fold increase of IMP3 mRNA was substantially increased in both SiHa (2.29 ± 0.14, p<0.001) and HeLa (1.33 ± 0.05, p<0.05), as compared to normal cervix tissue (1 ± 0.05). Furthermore, IMP3 expression was significantly higher in SiHa cell lines, as compared to HeLa cell lines (p<0.001). Western blot analysis further that IMP3 protein was expressed in SiHa and HeLa cell lines, whereas little IMP3 expression was observed in normal cervix.Conclusion1. Cell blocks the diagnostic accuracy of liquid-based cytology. The presence of stromal invasion, epithelial-stromal interface and tumor necrosis on cell block sections may be useful for accurate SCC diagnosis2. IMP3 is a useful diagnostic immunomarker that can be used to identify SCC and the combination of p16/IMP3 expression was found to improve the discrepant results between cytologic and histologic diagnoses.3. IMP3 expression is important in the progression from CIN Ⅲ to SCC and is a significant prognostic indicator of SCC.4. SCC cases were further divided into a diffuse-positive group and a central-negative group based on IMP3 staining and the unilty of sunset remain further research. |
PDF Full Text Request