p53 immunohistochemical positivity as a prognostic marker in intracranial tumours

Prognostic Implication of Clinical and Pathologic Features in Patients with Glioblastoma Multiforme Treated with Concomitant Radiation plus Temozolomide

Aims and background

Glioblastoma multiforme is the most common and most malignant primary brain tumor in adults. The current standard of care for glioblastoma is surgical resection to the extent feasible, followed by adjuvant radiotherapy plus temozolomide, given concomitantly with and after radiotherapy. This report is a prospective observational study of 43 cases treated in the Department of Radiotherapy, University of Rome La Sapienza, Italy. We examine the relationship between pathologic features and objective response rate in adult patients treated with concomitant radiation plus temozolomide to identify clinical, neuroradiologic, pathologic, and molecular factors with prognostic significance.

Methods

Forty-three consecutive patients (24 males and 19 females), ages 15-77 years (median, 57) with newly diagnosed glioblastoma multiforme, were included in this trial between 2002 and 2004 at our department. All patients were treated with surgery (complete resection in 81%, incomplete in 19%) followed by concurrent temozolomide (75 mg/m2/day) and radiotherapy (median tumor dose, 60 Gy), followed by temozolomide, 200 mg/m2/day for 5 consecutive days every 28 days. Neurologic evaluations were performed monthly and cranial magnetic resonance bimonthly. We analyzed age, clinical manifestations at diagnosis, seizures, Karnofsky performance score, tumor location, extent of resection, proliferation index (Ki-67 expression), p53, platelet-derived growth factor and epidermal growth factor receptor immunohistochemical expression as prognostic factors in the patients. The Kaplan-Meier statistical method and logrank test were used to assess correlation with survival.

Results

Fourteen patients (32%) manifested clinical and neuroradiographic evidence of tumor progression within 6 months of surgery. In contrast, 5 patients (12%) showed no disease progression for 18 months from the beginning of treatment. Median overall survival was 19 months. Multivariate analysis revealed that an age of 60 years or older (P <0.03), a postoperative performance score ≤70 (P = 0.04), the nontotal tumor resection (P = 0.03), tumor size >4 cm (P = 0.01) and proliferation index overexpression (P = 0.001) were associated with the worst prognosis. p53, PDGF and EGFR overexpression were not significant prognostic factors associated with survival.

Conclusions

The results suggest that analysis of prognostic markers in glioblastoma multiforme is complex. In addition to previously recognized prognostic variables such as age and Karnofsky performance score, tumor size, total resection and proliferation index overexpression were identified as predictors of survival in a series of patients with glioblastoma multiforme.

Age-dependent prognostic effects of EGFR/p53 alterations in glioblastoma: study on a prospective cohort of 140 uniformly treated adult patients

AIMS

To assess the prognostic influence of EGFR amplification/overexpression, p53 immunoreactivity and their age-dependent prognostic effects in a large prospective cohort of uniformly treated adult patients with newly diagnosed glioblastoma.

METHODS

Tumours from a uniformly treated prospective cohort of adult patients with newly diagnosed glioblastoma (n=140) were examined for EGFR amplification by fluorescence in situ hybridisation and EGFR/p53 expression by immunohistochemistry. Statistical methods were employed to assess the degree of association between EGFR amplification/overexpression and p53 immunopositivity. Survival analyses were performed by employing Cox proportional hazard models to assess the independent prognostic value of EGFR/p53 alterations and test the propensity for risk with age by assessing their interaction with patient age.

RESULTS

A strong positive correlation between EGFR amplification and EGFR overexpression (rho=0.5157; p<0.0001; CI 0.3783 to 0.6309) and a negative association of EGFR amplification (rho=-0.3417; p<0.0001; CI -0.4842 to -0.1816) and EGFR overexpression (rho=-0.3095; p<0.001; CI -0.4561 to -0.1465) with p53 immunopositivity was observed. Only patient age (HR: 1.029; p=0.004; CI 1.009 to 1.049) was associated with shorter survival by univariate Cox regression analysis. Multivariable Cox proportional hazards models revealed a statistically significant interaction between EGFR overexpression and age to be associated with shorter survival (HR: 1.001; p<0.0001; CI 1.000 to 1.002), thus predicting a higher hazard with increasing age. No age interaction of EGFR amplification status (HR: 1.001; p=0.642; CI 0.995 to 1.008) and p53 immunopositivity (HR: 1.000; p=0.841; CI 0.999 to 1.001) was noted in this cohort.

CONCLUSIONS

The prognostic value of EGFR overexpression is age-dependent, and there is a propensity for a higher hazard with increasing patient age. Identifying such groups of patients with more aggressive disease becomes mandatory, since they would benefit from intense therapeutic protocols targeting EGFR.

Worse outcome in primary glioblastoma multiforme with concurrent epidermal growth factor receptor and p53 alteration

Primary glioblastoma multiforme (GBM), in contrast with secondary GBM, has been associated with the presence of EGFR amplification and absence of p53 mutation. In this study, we analyzed relevant molecular and clinical variables in 194 primary GBMs and tested them for survival analysis. Although most of the tumors showed a mutually exclusive pattern, concurrent alterations of EGFR and p53 were detected. Survival analysis of CDK4 amplification revealed a highly significant association with a worse clinical outcome (P = .01), whereas MDM2, CDK6, PTEN, and p21 were not associated with patient survival. Multivariate analysis including the significant clinical and molecular variables revealed CDK4 amplification, age, and radiotherapy to be markers with independent prognostic value. In addition, the primary GBM tumors showing simultaneous EGFR and p53 alterations were significantly associated with worse survival (P < .01). These results highlight the prognostic value of CDK4 amplification and of simultaneous EGFR-p53 alterations in the clinical outcome of patients with primary GBM.

Mutant, wild type, or overall p53 expression: freedom from clinical progression in tumours of astrocytic lineage

Abnormalities of the p53 tumor-suppressor gene are found in a significant proportion of astrocytic brain tumours. We studied tumour specimens from 74 patients evaluated over 20 years at the Massachusetts General Hospital, where clinical outcome could be determined and sufficient pathologic material was available for immunostaining. p53 expression studies employed an affinity-purified p53 monoclonal antibody, whose specificity was verified in absorption studies and, in a minority of cases, a second antibody recognising a different epitope of p53. Significant overexpression of p53 protein was found in 48% of the 74 tumours included in this series and high levels of expression were associated with higher mortality from astrocytic tumours (P<0.001, log rank). Multivariate analyses revealed that immunohistochemically detected p53 was an independent marker of shortened progression-free and overall actuarial survival in patients with astrocytic tumours, suggesting that increased expression of p53 plays an important role in the pathobiology of these tumours. In a subset of 36 cases, coding regions of the p53 gene were completely sequenced via SSCP and direct DNA sequencing, revealing that overexpression of p53 protein is not always associated with point mutations in conserved exons of the p53 gene. Finally, we confirmed p53 protein expression in early-passage human glioma cell lines of known p53 mutational status and immunostaining scores. Although grade continues to be the strongest prognostic variable, the use of p53 staining as a prognostic indicator, in contrast to mutational DNA analyses, may be a useful adjunct in identifying patients at higher risk of treatment failure.

Prognostic value of Mdm2, p53 and p16 in patients with astrocytomas

Surgical cure of gliomas infiltrating into the brain is practically impossible and their clinical course is primarily determined by the biological behavior of the tumor cell. The purpose of this study was to analyze retrospectively prognostic input of p53, Mouse double minute-2 (Mdm2) and p16 in 103 uniformly treated patients with astrocytic tumors. The expression of these molecules was measured by immunohistochemical procedure. Prognostic evaluation was performed with the multivariate proportional hazards model. The follow-up period lasted 19 (5-122) months for the survivors. We observed that 66% of gliomas showed mutated p53, while only 17% overexpressed Mdm2, the p53-regulatory molecule. Besides, almost 50% of gliomas lost p16 immunopositivity. Only p53 labeling showed a positive correlation with the grade of malignancy, according with the WHO classification. The association between mutated p53 and histological grade remained when prognostic variables were considered in a multivariate analysis. No association between p53 status and overall survival was found. On the other hand, Mdm2 overexpression and, unexpectedly, p16 immunopositivity were associated with a shorter survival in an univariate analysis. However, Cox-regression analysis showed that only Mdm2 in female patients was an independent prognostic factor, associated with shorter survival. In conclusion, our results suggest that Mdm2 could be a relevant marker in determining the evolution of glioma patients and could provide a more objective way to classify astrocytomas.

Prognostic significance of annexin VII expression in glioblastomas multiforme in humans

Object. Glioblastoma multiforme (GBM) is the most common and lethal primary brain tumor in adults. It is nearly uniformly fatal, with a median survival time of approximately 1 year, despite modern treatment modalities. Nevertheless, a range of survival times exists around this median. Efforts to understand why some patients live longer or shorter than the average may provide insight into the biology of these neoplasms. The annexin VII (ANX7) gene is located on the human chromosome 10q21, a site long hypothesized to harbor tumor suppressor genes associated with prostate and other cancers. To test whether ANX7 expression might be a predictor for GBMs, we examined ANX7 expression, p53 accumulation, and the MIB-1 labeling index in a retrospective series of 99 GBMs.

Methods. In all 99 cases, the patient's age, Karnofsky Performance Scale (KPS) score before surgery, extent of surgery, tumor location, and immunohistochemical features were analyzed using univariate and multivariate analyses to identify whether any significance exists among ANX7 expression, p53 accumulation, the MIB-1 labeling index, and survival time. Kaplan—Meier analyses demonstrated that a higher KPS score before surgery (< 0.0001), total tumor excision (p = 0.0072), young patient age (p = 0.03), and ANX7 expression (p = 0.0006) correlated with longer survival. Multivariate Cox regression analyses demonstrated that ANX7 expression was the strongest predictor of outcome (p < 0.0001), independent of all other variables. In addition, ANX7 expression correlated with higher MIB-1 immunostaining, but did not correlate with p53 accumulation. Moreover, a significant positive correlation was observed between p53 and MIB-1 staining.

Conclusions. These findings indicate that a higher KPS score before surgery, total tumor excision, young patient age, and ANX7 expression correlate with longer survival in patients with GBMs. Multivariate Cox regression analyses demonstrated that ANX7 expression was the strongest predictor of outcome (p < 0.0001) and was independent of all other variables.

p53 in nonneoplastic central nervous system lesions: an immunohistochemical and genetic sequencing study

OBJECTIVE

Immunostaining for p53 commonly is considered a marker of neoplasia. Previous studies of nonneoplastic processes have yielded conflicting results.

METHODS

To test the assumption that p53 immunoreactivity indicates neoplasia, we examined 60 formalin-fixed, paraffin-embedded biopsies of nonneoplastic central nervous system lesions, including gliosis (n = 12), infarction (n = 9), demyelinating disease (n = 23), progressive multifocal leukoencephalopathy (n = 11), and herpes simplex virus encephalitis (n = 5). Diffuse astrocytomas (n = 50) of World Health Organization Grades 2 to 4 also were studied, as were six control autopsy brains. The avidin-biotin-peroxidase complex method was used with commercially available monoclonal antisera to both p53 (clone DO7; Dako, Carpinteria, CA) and mdm2 (Dako), a protein known to stabilize p53. Two samples of each nonneoplastic lesion also were subjected to deoxyribonucleic acid isolation, amplification, and sequencing of exons 5 to 8 of TP53.

RESULTS

Although it was low level in most instances, p53 immunoreactivity was noted in all but normal control samples. In reactive lesions, staining was largely observed in astrocytes and histiocytes. Scant oligodendroglia also were labeled in demyelinating disease. The progressive multifocal leukoencephalopathy samples revealed exceptionally strong staining in astrocytes and infected oligodendrocytes. Staining also was noted in occasional endothelial cells and neurons, and in rare lymphocytes. Immunoreactivity for mdm2, studied only in nonneoplastic lesions, was moderate to strong in all cases and limited to reactive astrocytes and histiocytes. No TP53 mutations were noted in the nonneoplastic lesions studied. To some extent, all astrocytomas exhibited p53 immunopositivity, particularly high-grade lesions.

CONCLUSION

p53 immunoreactivity is not limited to astrocytomas, but it can be observed in lesions that often are mistaken for glioma. No TP53 mutations accompany p53 expression in nonneoplastic lesions, and mdm2 may be responsible for persistence of p53 expression in these processes.

Immunohistochemical markers for prognosis of cerebral glioblastomas

Glioblastoma is the commonest neuroectodermal tumor and the most malignant in the range of cerebral astrocytic gliomas. The prognostic utility of various biological markers for glioblastomas has been broadly tested but the results obtained are regarded as controversial. In the present study, 302 glioblastoma specimens were studied to evaluate a possible association between clinical outcome and expression of some immunohistochemical variables. Furthermore, tumors examined were subdivided on the three cytological subsets--small-cell (SGB), pleomorphic-cell (PGB) and gemistocytic (GGB). Immunohistochemical variables differed between various subsets: the number of p53-positive tumors was found to be prevailed among the PGB, whereas the number of tumors with EGFR and mdm2 positivity was significantly greater in SGB. GGB contained significantly lowest mean proliferating cell nuclear antigen (PCNA) labeling index (LI), greater number of p21ras positive cases, and higher mean apoptotic index (AI). Survival time in patients with SGB, EGFR and mdm2-positivity and PCNA LI >40% was found to be significantly shorter, whereas presence of p21ras and AI >0.5% were associated with prolonged survival. Multivariate analysis revealed that survival time is associated with SGB, EGFR-positivity, and AI (p = 0.0023, p = 0.0035 and p = 0.0029 respectively). We conclude that although some immunohistochemical variables were found to be significant for glioblastoma outcome, they appear to be closely related to biology of single cytological subsets. Furthermore, these variables exhibited no prognostic value when they were analyzed within each cytological subset separately. Therefore, the glioblastoma subdivision on three cytological subsets proposed by us is carrying some element of rationality but, undoubtedly, requires further prospective studies.

Immunohistochemical markers for prognosis of anaplastic astrocytomas

Among the entire spectrum of astrocytic neoplasms, just anaplastic astrocytoma (or grade III astrocytoma) appears to be a more enigmatic tumor entity with vague criteria for pathological diagnosis, unclear biological behavior and diverse clinical outcome. Attempts have been made to identify biological markers that would be useful in prediction of prognosis of anaplastic astrocytomas but the results obtained are controversial. In the present study, survival data on 63 patients with anaplastic astrocytoma were studied to evaluate a possible association between clinical outcome and expression of some immunohistochemical variables. Both the progression-free (PFS) and overall (OS) survival times were significantly reduced for patients older than 45 years, for anaplastic astrocytomas containing multiple mitoses, for Ki-67 LI > 5%, for cyclin A LI > 4% and for PTEN-negative tumors. We found no differences in survival times in patients with or without p53 immunoreactivity and also in cases with different values of p16 and p27 immunostaining. Multivariate analysis revealed that risk of tumor progression and death is independently associated with tumors containing multiple mitoses and for PTEN-negative tumors. According to the data from the CART modeling, tumors were subdivided based on the three following subsets: (1) Anaplastic astrocytomas with solitary mitosis. (2) Anaplastic astrocytomas with multiple mitoses and PTEN positivity. (3) Anaplastic astrocytomas with multiple mitoses and PTEN negativity. Thus, the results obtained reveal the advantage of combined approach including evaluation of routine histological parameters and immunohistochemical variables for further clinical subdivision of anaplastic astrocytomas.

The prognostic value of tumor markers in patients with glioblastoma multiforme: analysis of 32 patients and review of the literature

Although several studies have examined brain tumor markers for prognostic value, few investigations have stratified analysis based on specific histologic grade. The objective of this study was to evaluate a single histologic grade of glioma, the grade IV glioma or glioblastoma (World Health Organization Classification), with a comprehensive panel of tumor markers in an attempt to identify those with prognostic significance. Tumor samples from a cohort of patients with glioblastoma multiforme (n = 32) were examined for tumor markers, DNA analysis, and clinical variables in an attempt to determine a 'profile' for this tumor. We used univariate and multivariate statistical analysis to determine the prognostic value of tumor cell ploidy, percent S-phase, DNA index, p53, and Ki-67 labeling index, as well as the variables of gender, race, age, location of tumor, history of chemotherapy, and primary versus recurrent tumor. Two additional tumor markers, multidrug resistance gene 1 and glutathione-S-transferase subtype pi, were included in the sample testing, but were not analyzed statistically. Univariate analysis indicated that increasing age had a strong association with decreased survival. Female gender, increasing Ki-67, no chemotherapy before sample collection, and primary glioblastoma showed some association with decreased survival in the univariate model. The univariate results indicated that race, side of tumor, ploidy, S-phase, DNA index, and p53 had no prognostic value. Multivariate modeling demonstrated that age, gender, and Ki-67 were the strongest factors associated with survival. The relevant literature is reviewed.

Analysis of 2 antiapoptotic factors in gliomas: bcl-2 overexpression and p53 mutations

BACKGROUND

p53 mutations and immunoreactivity have been described in human gliomas. During the past few years, some authors have found bcl-2 overexpression in astrocytomas, although their correlation with histological grade is a matter of disagreement. A relation between bcl-2 overexpression and p53 immunoreactivity has also been suggested.

OBJECTIVES

To analyze the frequency of presentation of bcl-2 and p53, their clinicopathologic implications, and their possible coexpression.

METHODS

We studied p53 and bcl-2 with immunohistochemical and molecular methods in 61 gliomas (including 21 astrocytomas, 9 anaplastic astrocytomas, 29 glioblastomas, 1 oligodendroglioma, and 1 mixed glioma).

RESULTS

We discovered a high level of bcl-2 overexpression (57%). Overexpression of bcl-2 can be an early event in gliomas tumorigenesis, although no correlation was found with any of the clinicopathologic parameters studied. p53 mutations were present in a small proportion of gliomas (17%). p53 immunoreactivity was present in 34 cases (57%), and it was related to histological grade and a supratentorial location. A high percentage of tumors (26 cases, 42%) presented p53 immunoreactivity without p53 mutations.

CONCLUSIONS

Since there was no relation between bcl-2 overexpression and p53 mutations or p53 immunoreactivity, both factors may not act together in the genesis and evolution of gliomas.

Distinct radiochemotherapy protocols differentially influence cellular proliferation and expression of p53 and Bcl-2 in glioblastoma multiforme relapses in vivo

Several protocols for the adjuvant treatment of glioblastoma multiforme (GBM) are currently being evaluated. In this context, little is known about the influence of radiochemotherapy on apoptosis and the expression of apoptosis-related proteins in vivo. We have analyzed the incidence of apoptosis using in situ nick translation (ISNT) and expression of Ki-67 (MIB- 1), p53 (DO-1 and DO-7), Bcl-2 and transglutaminase II (TGase II) by immunohistochemistry in 41 patients with GBM and their matched relapses. Sixteen patients received radiochemotherapy, 18 irradiation and 7 no treatment. Radiochemotherapy resulted in an increase in Bcl-2+ cells (p = 0.013). Irradiation caused the reduction of MIB-1+ (p = 0.0015), DO-7+ (p = 0.0043) and the increase of Bcl-2+ cells (p = 0.016). We calculated a positive correlation between high TGase II scores in patients preceding radiochemotherapy (p = 0.0186) and no treatment (p = 0.0158), low ISNT scores (p = 0.0018) and high DO-1 scores (p = 0.0233) in patients preceding irradiation and short time to progression. These data show that distinct postsurgical radiochemotherapy protocols differentially alter cellular proliferation and expression of p53 and Bcl-2 in GBM relapses. Furthermore, we show that ISNT, DO-I and TGase II labeling scores are therapy-specific predictors of time to progression in GBM patients.

Radiation and concomitant weekly administration of paclitaxel in patients with glioblastoma multiforme. A phase II study

The present study was conducted to evaluate the activity and toxicity profile of radiation (RT) and concomitant chemotherapy in patients with glioblastoma multiforme (GBM). Thirty-nine patients were treated postoperatively with RT and concomitant administration of paclitaxel. Cranial irradiation was initiated 2-3 weeks postoperatively and was administered in 2.0 fractions, one fraction per day, for 5 consecutive days per week, to a total of 60 Gy. Paclitaxel was delivered at a dose of 100 mg/m2 over 3-h once weekly for 6 weeks. Thirty-three patients received all 6 cycles of paclitaxel according to the protocol. Totally, 217 cycles were delivered all of them at full dose. The median relative dose intensity of paclitaxel was 1 (range 0.88-1.1). Three (7.5%) patients achieved complete and 9 (23%) partial response, while 12 (30.5%) patients demonstrated stabilization of the disease. Side effects from combined chemoradiotherapy were mainly mild. Grade III toxicity included infection (7.5%) and alopecia (5%). Median time to progression was 6 (range 0.9-27) months and median survival 10.7 (range 0.9-39.5+) months. The present study has clearly shown that 100 mg/m2 of paclitaxel in 1-h infusion weekly can be safely given concomitantly with RT in patients with GBM with manageable toxicity. However, the efficacy of this combined modality treatment does not appear to be superior to that of RT alone.

Accumulation of mutant p53(V143A) modulates the growth, clonogenicity, and radiochemosensitivity of malignant glioma cells independent of endogenous p53 status

Alterations of the p53 gene have been attributed a major role in the development and resistance to therapy of several human cancers. Accumulation of p53 in tumor cells may result from mutations associated with prolonged half-life or from stabilization of wild-type p53 by different mechanisms. To address the role of p53 accumulation in the response of malignant glioma cells to radiochemotherapy, we expressed the p53 mutant p53(V143A) in five human malignant glioma cell lines with different genetic and functional p53 status. Accumulation of p53(V143A) modulated proliferation in three and clonogenicity in four of five cell lines without a clear pattern with regard to their endogenous p53 status. p53(V143A) inhibited the camptothecin-induced accumulation of p21(WAF1/CIP1) in cell lines with p53 functional wild-type activity, but not in cell lines lacking p53 activity, consistent with a transdominant-negative effect of p53(V143A). Irradiation induced a moderate G2/M arrest in all cell lines, irrespective of the p53 status, that was unaffected by p53(V143A). Radiosensitivity as well as sensitivity to BCNU, teniposide (VM26), topotecan, vincristine, Taxol, and cisplatin both in cytotoxic cell death and in clonogenic cell death was unchanged in p53(V143A)-transfected cells with few exceptions. These data do not support the hypothesis that accumulation of mutant p53 is a major determinant of the response to adjuvant radiochemotherapy in human malignant glioma cells.

Prognostic value of tumour associated antigen immunoreactivity and apoptosis in cerebral glioblastomas: an analysis of 168 cases

AIMS

To evaluate a possible association between clinical outcome in patients with glioblastoma and expression of some immunohistochemical variables and apoptosis.

METHODS

168 selected patients with cerebral glioblastomas were studied retrospectively. Tumour specimens were examined immunohistochemically with antibodies to proliferating cell nuclear antigen (PCNA), p53, bcl-2, and epidermal growth factor receptor (EGFR) to detect the intracellular receptor domain. Apoptosis was detected by in situ end labelling. Multivariate analysis was performed using the Cox proportional hazard model.

RESULTS

On univariate analysis the PCNA labelling index, immunoexpression of EGFR, and the apoptotic index were significantly related to glioblastoma outcome. Survival time was reduced as PCNA labelling index increased and apoptotic index decreased (p = 0.0073 and p = 0.00031, respectively). Survival time in patients with EGFR positive tumours was found to be reduced (p = 0.00024). Multivariate analysis showed independent prognostic value for the EGFR positivity and apoptotic index only (p = 0.0053 and p = 0.0039, respectively). There was no association between clinical outcome of glioblastoma and p53 or bcl-2 immunostaining.

CONCLUSIONS

EGFR immunoreactivity and apoptotic index were found to be useful for assessing prognosis of individual glioblastomas but it seems unlikely that p53 and bcl-2 immunohistochemistry will be of value in determining survival in such patients.

Analysis of proliferation markers and p53 expression in gliomas of astrocytic origin: relationships and prognostic value

Consecutive paraffin sections of 105 astrocytomas and 15 oligoastrocytomas were examined for expression of p53, MIB-1 (Ki-67), and proliferating cell nuclear antigen (PCNA). The tumors had been examined previously for genetic abnormalities and by flow cytometry. Regardless of the tumor's stage and grade and the patient's age and gender, p53 expression was found in 40% of tumors. Although p53 expression was associated with a loss on chromosome 17p and was more frequent in aneuploid tumors, it had no association with survival time. The MIB-1 and PCNA labeling indices increased with increasing tumor grade but showed no association with other clinicopathological parameters. In individual tumors, there was poor concordance between any of the variables (MIB-1, PCNA, and p53). Results for p53 and MIB-1 were similar for both astrocytomas and oligoastrocytomas. The MIB-1 and PCNA values appeared to have prognostic utility in univariate analysis but not after adjusting for patient age and tumor grade. The poor concordance between MIB-1 and PCNA in individual tumors indicates that any one means of assessing proliferative potential in gliomas may not be reliable.

MDM2 and p53 expression in gliomas: a multivariate survival analysis including proliferation markers and epidermal growth factor receptor

p53 and the murine double minute 2 (MDM2) oncoprotein expression was evaluated in paraffin-embedded tissue from 61 patients with central nervous system gliomas (53 astrocytomas and eight oligodendrogliomas) and related to proliferation-associated markers [i.e. proliferating cell nuclear antigen (PCNA), Ki-67 and nuclear organizer regions (NORs)] and epidermal growth factor receptor (EGFR). We used the monoclonal antibodies PC-10, MIB-1, DO-1, 1B1O and EGFR 113 and the colloid silver nitrate (AgNOR) technique. MDM2 and p53 were co-expressed in 28% of cases. A p53-positive/MDM2-negative phenotype was observed in 15% and a p53-negative/MDM2-positive phenotype in 20% of cases. There was a positive correlation of p53 and MDM2 expression with grade and proliferation indices. Univariate analysis in the group of diffuse astrocytomas showed that older age, high histological grade, high PCNA labelling index (LI) and high AgNOR score were associated with reduced overall survival (P < 0.05). p53 LI, Ki-67 LI, AgNOR score, tumour location and grade influenced disease-free survival (P < 0.05), whereas the only parameters affecting post-relapse survival were histological grade and Ki-67 LI (P < 0.1). Multivariate analysis revealed that age, radiotherapy, PCNA LI and p53 LI were the independent predictors of overall survival. p53 LI, Ki-67 LI, MDM2 LI, EGFR LI, grade and type of therapy were independent predictors of disease-free survival, and grade was the only independent predictor of post-relapse survival. Our results indicate that p53 LI and MDM2 LI, EGFR expression as well as proliferation markers (PCNA and Ki-67) are useful indicators of overall and disease-free survival in diffuse astrocytoma patients.