Prognosis Of Astrocytomas Research Articles

Expression analyses of 27 DNA repair genes in astrocytoma by TaqMan low-density array

DNA repair systems act to maintain genome integrity in the face of replication errors, environmental insults, and the cumulative effects of age. The mRNA expressions of 27 genes of the DNA repair system as well as their correlation with the clinical characteristics were studied in human astrocytoma. We applied TaqMan low-density array to investigate the mRNA expressions of 27 DNA repair genes in 40 astrocytoma tissues (10 of grade II, 10 of grade III, and 20 of grade IV, according to the WHO Grading System). And the normal brain tissues from 10 non-astrocytoma patients were collected as the control. In addition, correlation of their mRNA levels with clinical characteristics was also analyzed. We found that the expression of the 13 genes were significantly ( P < 0.01) down-regulated in grade II, III, IV of astrocytoma compared to normal brain tissues, including ERCC1, ERCC2, ERCC3, ERCC4, MGMT, MLH1, MLH3, NTHL1, OGG1, RAD50, SMUG1, XRCC4 and XRCC5. Meanwhile, we found that the expression of MSH2, MSH6, NUDT1 and XRCC3 were only significantly lower in grade II and III of astrocytoma, and the expression of MRE11A and MUS81 were only significantly lower in grade III and IV. But the expression of MPG, MSH3, MUTHY and RAD51 were not changed in any grade of astrocytoma. Furthermore, we found that the decrease expression of eight genes was significantly ( P < 0.05) associated with a poor prognosis, including ERCC3, ERCC4, MLH3, MRE11A, NTHL1, RAD50, XRCC4 and XRCC5. We suggest that TaqMan low-density array is an effective multivariate technique to examine the expression of DNA repair genes in astrocytomas, which can be applied to identify tumor-specific genes. We also suggest that the down-regulation of some DNA repair genes may be associated with pathogenesis and poor prognosis of astrocytoma.

Role of MIB1 in Predicting Survival in Patients with Glioblastomas

Histologic immunomarkers of cell cycle proteins have been utilized for prognosis in high-grade astrocytic tumors. One such marker, MIB1, an antibody immunoreactive throughout the cell cycle, is predictive of more aggressive disease and poorer prognosis in astrocytomas. An independent role of MIB1 analysis for survival prediction and clinical management within histologic grades has not been clearly proven. This study retrospectively evaluated MIB1 reactivity in tissue samples from 116 patients with glioblastomas on initial medical presentation. Clinical variables considered included gender, age, Karnofsky Performance Scores (KPS), extent of surgical resection, adjuvant radiation and survival. Univariate and multivariate analyses were used to correlate these variables with MIB1 staining. MIB1 staining does not predict overall survival or response to adjuvant therapy as an independent risk factor. MIB1 labeling does not predict patient survival as an independent variable and does not predict response to additional therapies. Patient survival with glioblastoma was predicted by KPS, age, extent of resection and use of adjuvant radiotherapy.

Defects in Cell Growth Regulation by C18:0-Ceramide and Longevity Assurance Gene 1 in Human Head and Neck Squamous Cell Carcinomas

In this study, endogenous long chain ceramides were measured in 32 human head and neck squamous cell carcinoma (HNSCC) and 10 nonsquamous head and neck carcinoma tumor tissues, as compared with adjacent noncancerous tissues, by liquid chromatography/mass spectroscopy. Interestingly, only one specific ceramide, C(18:0)-ceramide, was selectively down-regulated in the majority of HNSCC tumor tissues. On the other hand, in nonsquamous tumor tissues, this selectivity for C18-ceramide was not detected. These data suggested the hypotheses that decreased levels of C18-ceramide might impart a growth advantage to HNSCC cells and that increased generation of C18-ceramide may be involved in the inhibition of growth. These roles were examined by reconstitution of C18-ceramide at physiologically relevant concentrations in UM-SCC-22A cells (squamous cell carcinoma of hypopharynx) via overexpression of mammalian upstream regulator of growth and differentiation factor 1 (mUOG1), a mouse homologue of longevity assurance gene 1 (mLAG1), which has been shown to specifically induce the generation of C18-ceramide. Liquid chromatography/mass spectroscopy analysis showed that overexpression of the mLAG1/mUOG1 resulted in increased levels of only C(18:0)-ceramide by approximately 2-fold, i.e. concentrations similar to those of normal head and neck tissues. Importantly, increased generation of C18-ceramide by mLAG1/mUOG1 inhibited cell growth (approximately 70-80%), which mechanistically involved the modulation of telomerase activity and induction of apoptotic cell death by mitochondrial dysfunction. In conclusion, this study demonstrates, for the first time, a biological role for LAG1 and C18-ceramide in the regulation of growth of HNSCC.

Evaluation of molecular markers in low-grade diffuse astrocytomas: loss of p16 and retinoblastoma protein expression is associated with short survival.

Diffuse astrocytomas have a median survival of 6-8 years. However, in a minority of cases that are histologically low grade, progression is rapid, leading to death within 2 years. Loss of p16, retinoblastoma protein, and deleted-in-colon-carcinoma protein expression, and monosomy of chromosome 10 have been shown to occur in malignant astrocytic tumors. We have investigated the prognostic value of expression of these markers, using techniques applicable in many histopathology laboratories, in diffuse astrocytomas that are histologically low grade. Paraffin sections from 71 diffuse, supratentorial, low-grade astrocytomas, from patients with at least 8-year survival data, were immunostained with antibodies to p16, deleted-in-colon-carcinoma protein, p53, Ki67, and retinoblastoma protein. In situ hybridization with a digoxigenin-labeled probe to chromosome 10 was used to assess chromosomal loss. In most cases there was immunostaining of virtually all tumor cell nuclei with antibodies to p16 and retinoblastoma protein. Three of the 68 tumors in which assessment of p16 was possible included discrete foci with lack of detectable immunoreactivity in tumor cells. The three patients concerned had a significantly shortened median survival (1.1 years vs 4.4 years in those without loss of p16; p <0.01). In six of the 61 cases where assessment of retinoblastoma protein was possible, <70% of tumor cell nuclei showed immunoreactivity. These six patients had a shorter survival (4.0 years) than had the remaining patients (5.4 years), although this difference was not statistically significant. The tumor from one of these patients included areas where only 36% of tumor cells showed retinoblastoma protein immunoreactivity, and this patient survived only 1.5 years. Tumors showing loss of both p16 and retinoblastoma were not seen. p53 and deleted-in-colon-carcinoma protein expression was highly variable and did not correlate with survival. Tumors with monosomy for chromosome 10 were not identified. Both polyploidy and the Ki67 labeling index were significantly associated with the p53 labeling index but not with survival. Focal loss of p16 or retinoblastoma protein is demonstrable in approximately 5% and 10% of diffuse low-grade diffuse astrocytomas, respectively. Tumors with focal loss of immunoreactivity for these proteins are associated with shorter survival than those without, suggesting that immunohistochemistry for p16 and retinoblastoma protein may be a useful adjunct to other methods for assessing the prognosis of astrocytomas.

High cyclin E/low p27Kip1 expression is associated with poor prognosis in astrocytomas

Cyclin E and p27Kip1 are co-regulators of the G1- to S-phase transition and closely related to tumor behavior. The purpose of this study was to examine expression of cyclin E and p27Kip1 in astrocytomas and to evaluate the relationships between expression of these cell-cycle regulators and prognosis of patients with astrocytoma. Tissue samples from 130 astrocytomas (WHO grade 1 n = 5, grade 2 n = 23, grade 3 n = 64, grade 4 n = 38) were examined immunohistochemically for cyclin E and p27Kip1 expression. Patient charts were reviewed for clinical presentation, and survival was followed. The cyclin E labeling index (LI) tended to increase with tumor grade (Kruskal-Wallis, P = 0.0104). For patients with primary astrocytomas, the 50% survival times for the low cyclin E LI (< 5%) group and the high cyclin E LI (> or = 5%) group were 53.7 months and 19.8 months. In combined analysis of cyclin E and p27Kip1 expression, the low cyclin E/high p27Kip1 LI (> or = 50%) group had the best survival (50% survival time: 103.2 months), the low cyclin E/low p27Kip1 LI (> or = 50%) and the high cyclin E/high p27Kip1 LI groups moderate survival (24.1 and 27.5 months), and the high cyclin E/low p27Kip1 LI group the worst survival (13.1 months). Multivariate analysis identified the combined factor, high cyclin E/low p27Kip1, as a novel independent prognostic factor for survival time (P = 0.0037, relative risk = 2.4). This study suggested that combined analysis of cyclin E and p27Kip1 expression was considered to be potentially useful in predicting the prognosis of patients with astrocytoma.

The p53 tumor suppressor gene and pediatric brain tumors.

Mutations of the p53 tumor suppressor gene are common in astrocytomas that occur in adults but are rarely found in astrocytomas or medulloblastomas of children. This difference may contribute to the overall better prognosis of astrocytomas in children and potentially make pediatric astrocytomas and medulloblastomas more sensitive to antitumor therapy.