Osteopontin expression is a valuable marker for prediction of short-term recurrence in WHO grade I benign meningiomas

J Neurooncol (2010) 100:217 223 DOI 10.1007/s11060-010-0164-2 CLINICAL STUDY - PATIENT STUDY Osteopontin expression is a valuable marker for prediction of short-term recurrence in WHO grade I benign meningiomas Kuan-Yin Tseng Min-Huey Chung Huey-Kang Sytwu Horng-Mo Lee Kuan-Yu Chen Chen Chang Chih-Kung Lin Che-Hung Yen Jia-Hong Chen Gu-Jiun Lin Hsin-I Ma Yi-Shian Yeh Da-Tong Ju Ming-Ying Liu Dueng-Yuan Hueng Received: 6 December 2009 / Accepted: 15 March 2010 / Published online: 29 April 2010 Ó Springer Science+Business Media, LLC. 2010 Abstract Prediction of recurrence remains a challenge in histopathological benign/grade I tumors. Osteopontin (OPN) plays important roles in tumorigenesis, invasion, and metastasis of several human cancers. In this study, we investigated OPN protein expression by evaluating the differences between recurrent and non-recurrent histologically benign meningiomas. Thirty-two patients were enrolled, and 23 benign non-recurrent meningiomas and 9 benign recurrent meningiomas were followed for a mean of 34 months after complete surgical resection (Simpson grades I and II). Cytoplasmic OPN staining was evaluated K.-Y. Tseng K.-Y. Chen H.-I. Ma D.-T. Ju (&) M.-Y. Liu (&) D.-Y. Hueng (&) Department of Neurological Surgery, Tri-Service General Hospital, National Defense Medical Center 325, Sec. 2, Cheng-Kung Road, Neihu 114, Taipei, Taiwan, ROC e-mail: hondy2195@yahoo.com.tw D.-T. Ju e-mail: wxyz670628@yahoo.com.tw M.-Y. Liu e-mail: tsghns01@ndmctsgh.edu.tw M.-H. Chung Graduate Institute of Nursing, College of Nursing, Taipei Medical University, Taipei, Taiwan, ROC H.-K. Sytwu H.-I. Ma D.-Y. Hueng Graduate Institute of Medical Sciences, National Defense Medical Center, Taipei, Taiwan, ROC H.-M. Lee Department of Medical Laboratory Sciences and Biotechnology, Graduate Institute of Medical Sciences, College of Medicine, Taipei Medical University, Taipei, Taiwan, ROC C. Chang Institute of Biomedical Science, Academia Sinica, Nankang, Taipei, Taiwan, ROC by means of immunohistochemistry (IHC) score and by use of the Allred-8-unit system. We examined clinical biological data, their relationship with tumor recurrence, and the expression of OPN. Our results showed that meningioma recurrence correlated significantly with OPN IHC score (P = 0.001). An OPN Allred score between 0 and 3 was associated with a recurrence-free time of more than 25 months. In comparison, an OPN Allred score from 4 to 8 was indicative of a shorter average recurrence-free time. We concluded that OPN IHC score may play a role in prediction of the recurrence of the grade I meningiomas. Moreover, determination of the OPN Allred score is a C.-K. Lin Department of Pathology, Tri-Service General Hospital, National Defense Medical Center, Taipei, Taiwan, ROC C.-H. Yen Department of Neurology, Tri-Service General Hospital, National Defense Medical Center, Taipei, Taiwan, ROC J.-H. Chen Division of Hemato-Oncology, Department of Internal Medicine, Tri-Service General Hospital, National Defense Medical Center, Taipei, Taiwan, ROC H.-K. Sytwu G.-J. Lin Graduate Institute of Life Sciences, National Defense Medical Center, Taipei, Taiwan, ROC Y.-S. Yeh Department of Neurosurgery, Shuang Ho Hospital, Taipei Medical University, Center of Excellence for Cancer Research, Taipei Medical University, Taipei, Taiwan, ROC H.-M. Lee Institute of Pharmaceutical Science and Technology, Central Taiwan University of Science and Technology, Taichung, Taiwan, ROC

218 J Neurooncol (2010) 100:217 223 reliable, quantitative tool for predicting recurrence-free time in benign meningioma patients. Keywords Osteopontin Immunohistochemical staining Allred score Meningioma Recurrence Introduction Meningiomas are the commonest tumors of the central nervous system and are believed to arise from specialized arachnoid cap cells. The most relevant factor for progression-free survival (PFS) is their histopathological grade and surgical resection. However, apparently complete tumor resection and removal of the involved dura and bone do not always prevent the recurrence even of histologically benign (WHO grade I) meningiomas [1]. In addition, borderline cases exit in all grading schemes and there still remains a group of cases that show aggressive behavior despite histologically benign meningiomas and as many as 7 20% of these are known to recur [2]. Osteopontin (OPN) is an integrin-binding protein which has been shown to be involved in tumor progression [3 5]by regulation of apoptosis, proliferation, adhesion, migration, invasion, metastasis, and angiogenesis [3]. Overexpression of OPN has been correlated with poor clinical outcome [6 9]. It has been implicated as a candidate target for cancer therapy [5]. Although positive staining of OPN in the psammoma body has been reported [10], its correlation with clinical recurrence of meningiomas is still largely unknown. The objective of this study was to investigate OPN expression in intracranial WHO grade I benign meningiomas, and to assess its correlation with clinical and pathological characteristics. Materials and methods Patients This study was approved by the institutional review board of our hospital and followed the tenets of the Declaration of Helsinki. We collected specimens from 32 patients with benign meningiomas (12 men and 20 women; age range, 20 80 years; mean age, 56 years) with gross total tumor resection between 2000 and 2008. All patients included in the study submitted to image examinations such as computerized tomography (CT) or magnetic resonance images (MRI) pre and post-surgery. Tumor volumes were obtained, approximately, using the spheroid formula (V = 4/3p 9 abc), where a and b correspond to the maximum perpendicular diameters obtained from the sum of axial images where the tumor is actually present, multiplied by slices thickness. Paraffin-embedded tumor tissues were obtained from the Department of Pathology, Tri-Service General Hospital. All cases were grade I meningiomas with a mitosis index of less than 4 mitotic figure/10 fields, according to the WHO 2000 classification. Absence of any residual tumor, resection, or extensive coagulation of dural attachment, were regarded to be essential findings for inclusion in the study. During a mean follow-up time of 34 months (range 5 65 months), nine of the 32 patients (28%) experienced recurrence. The criteria for diagnosing recurrence were the demonstration during regular follow-up, and MRI and/or CT evidence of tumor appearance after complete resection. Tissue microarray (TMA) construction The TMA was conducted at Array Biotechnology. Hematoxylin and eosin (H&E)-stained sections of the selected tumor cases were examined and representative areas of the solid tumor were identified by a pathologist for sampling. Thirty-two meningiomas were selected for TMA placement. Two 2-mm cores per donor block were transferred to a recipient block tissue microarrayer. Three-micron-thick sections were cut from the recipient block and transferred to glass slides using an adhesive tape transfer system for ultraviolet cross linkage. Immunohistochemical (IHC) staining Anti-OPN mouse monoclonal antibody (mab53, prepared against recombinant glutathione S-transferase human OPN fusion protein, dilution at 1:1000) and anti-ki-67 monoclonal antibody (mab) (mouse mab obtained from Dako, Denmark; dilution 1:100) were applied to 3-lm sections from TMA, using the avidin biotin peroxidase complex method, following the manufacture s instructions. In brief, the immunostaining was performed manually at room temperature. Endogenous peroxidase and nonspecific background staining were blocked by incubating slides with 3% aqueous hydrogen peroxide for 10 min. After washing with PBS for 5 min, slides were blocked with normal serum for 20 min, followed by incubation with the anti-opn primary antibody, at the given dilution, for 60 min. After rinsing with PBS for 5 min, sections were incubated with a biotinylated secondary antibody for 20 min. After washing with PBS for 5 min, slides were incubated with avidin biotin complex for 30 min and washed again. Chromogen was developed with 10 mg 3,3 0 -diaminobenzidine tetrahydrochloride diluted in 12 ml Tris buffer for 2 min. All samples were lightly counterstained with Mayer s hematoxylin for 30 s before dehydration and mounting. A section of colon cancer that was previously proven to be OPN positive [13] was used as a positive control. Mouse IgG was used as the negative control

J Neurooncol (2010) 100:217 223 219 serum. Cytoplasmic immunostaining intensity was assessed for OPN on a scale of 0 (no staining), 1, 2, 3, and 4 (strongest). Also, the percentage of immunostaining intensity in each cell was calculated. The OPN IHC score, obtained by multiplying the percentage of cells by its corresponding immunostaining intensity, ranged from 0 to 400. The OPN Allred-8-unit system (OPN Allred score) [13] was an alternative method for verifying OPN protein expression across a wide range of tumor histologies semiquantitatively. We determined the tumor epithelial cells proportion score and intensity score. The proportion score included the fraction of positively stained tumor cells and was: 0 = none; 1 = \1/100th; 2 = 1/100th to 1/10th; 3 = 1/10th to 1/3; 4 = 1/3 to 2/3; 5 = [2/3. The estimated average staining intensity of the positive tumor cells was expressed as follows; 0 = none; 1 = weak; 2 = intermediate; 3 = strong. For the statistical analysis an OPN Allred score of 0 3 was considered low, and an OPN Allred score of 4 8 was considered high. All tissue sections of Ki-67 staining were examined at high power magnification (9400). The number of cells stained positively with anti-ki-67 antibody and the total number of tumor cells was counted in several representative fields containing more than 1000 cells and their ratio was expressed as the Ki-67 labeling index (LI) %. Statistical analysis The basic comparisons between the non-recurrent and the recurrent groups were undertaken by primary examination of the demographic characteristics and the meningiomas therapeutic issues using independent t tests for the continuous variables and v 2 tests for the categorical variables, with no consideration of the follow-up period. The recurrent group included those patients who had meningiomas at the end of the follow-up again. As a result of the retrospective design of the study setting, we applied the survival analyses to the influencing factors generated by reviewing literature and patients charts, until recurrence of meningiomas as the endpoint. The duration of the follow-up, in months, was defined as the period from the date of the operation to the date of recurrence from meningiomas (the event in survival analysis), or the end of the follow-up period. The specific issues of major concern in this study were OPN IHC score, OPN Allred score, Ki-67, and perifocal edema factors potentially predicting the prognosis of recurrence of cancer. Hazard ratios (HRs) (with 95% confidence intervals; CI) were estimated for the relative risks of the prediction factors. Univariate Cox regressions were then used to estimate the HRs for these factors compared with the occurrences of meningioma. Thereafter, the age and sex of the patient, the type of meningioma, location, volume, symptom, and calcification were regarded as potential confounders in the multivariate analyses. SPSS (Chicago, Illinois, USA) 15.0 software for Windows was used to perform all of the statistical analyses in this study, with the significance level (a value) being set at 0.05. Results Clinicopathological characteristics and analysis of OPN expression The pathological diagnosis comprises 15 meningotheliomatous, eight fibrous, six transitional, and three psammomatous types. There were 14 tumors (43%) located in the cerebral convexity, seven in the parasagittal region (28%), three in the tentorial region (9%), two falx tumors (6%), and six skull base tumors (18%). The average tumor volume was 295.23 cm 3, ranging from 26.3 cm 3 to a maximum of 717.5 cm 3. Peritumoral brain edema of 2 cm or larger was found in 40.6% of the benign meningiomas. Immunostaining for OPN in cytoplasmic components of different pathological subtypes of recurrent and non-recurrent meningiomas is shown in Fig. 1. Moreover, Table 1 lists results from quantitative analysis of OPN IHC scores for patients of the two different groups with recurrent and non-recurrent meningiomas after initial complete surgery. The mean immunostaining score of OPN was 30.87 (SD = 46.16) for the group with non-recurrent meningiomas, whereas it was 192.22 (SD = 83.03) for recurrent benign meningiomas. Comparing the OPN IHC score between benign non-recurrent and recurrent meningiomas showed a significant difference (P = 0.001). The mean Ki-67 LI of benign nonrecurrent and benign recurrent meningiomas is given in Table 1, and was shown to be statistically significant by using the unpaired t-test (P = 0.002). Univariate analyses, multivariate analysis, and Cox regressions We used the univariate analyses by Cox regressions to compare each variable for prognosis of benign meningiomas, as shown in Table 1. The follow-up time was also considered as a variable in the model. Cases of OPN IHC score, perifocal edema, and Ki-67 revealed significant correlation with recurrence in meningiomas. Moreover, the outcomes of the multivariate analyses presented in Table 1 show that after controlling for potential confounders, the OPN IHC score (HR = 1.02; 95% CI:1.00 1.03) and the Ki-67 LI (HR = 1.41; 95% CI: 1.05 1.89) were identified as statistically significant factors. To study the correlation between OPN expression and tumor recurrence, the OPN Allred score was compared with the tumoral recurrence-free time by Kaplan Meier analysis. As shown in Fig. 2, the estimated 25-month

220 J Neurooncol (2010) 100:217 223 Fig. 1 Comparison of OPN in different types of grade I recurrent and non-recurrent meningiomas. a. Low expression of OPN is noted in non-recurrent meningotheliomatous meningioma (original magnification 9100). b A tumor with high degree of staining intensity was evidenced by OPN staining in a recurrent meningotheliomatous meningioma (original magnification 9100). c An intense negative immunoreaction for OPN was present in a non-recurrent fibrous meningioma (original magnification 9100). d OPN staining in a recurrent fibrous meningioma (original magnification 9100). e OPN negative immunoreaction was evident in a non-recurrent transitional meningioma (original magnification 9100). f A strongly positive OPN immunoreaction was present in a recurrent transitional meningioma. g A positive OPN immunoreaction was only present in psammoma bodies in a non-recurrent psammoma type meningioma, but cytoplasmic immunostaining intensity was weak (original magnification 9100). h A positive OPN immunoreaction was present in both psammoma bodies and cytoplasm in recurrent psammoma type meningioma (original magnification 9100) recurrence-free time in Group A (OPN Allred score between 0 and 3) and Group B (OPN Allred score between 4 and 8), were 100% and 18.2%, respectively. The cumulative non-recurrence rate for patients with OPN Allred scores above the cut-off value (the Allred score [3) was significantly lower than that of the patients with OPN Allred score below the cut-off value. Discussion Meningiomas are benign indolent neoplasms occurring in middle age, with a subset of them behaving aggressively and having a higher tendency to recur [2]. Types of surgical resection and histological data are major predictors of tumor recurrence. In the literature, estimated five-year

J Neurooncol (2010) 100:217 223 221 Table 1 Statistical correlations between recurrence and the clinicopathological data analyzed throughout independent t test and v 2 tests, and univariate and multivariate analyses Variables Non-recurrence Recurrence Independent t test and v 2 tests P value Univariate analyses P value Multivariate analysis P value Age, mean (SD), years 58.00 (16.19) 52.67 (15.73) 0.41 0.38 0.824 Gender 0.19 0.25 0.290 Female 16 (69.6) 4 (44.4) Male 7 (30.4) 5 (55.6) Pathological diagnosis 0.45 Meningotheliomatous 11 (47.8) 4 (44.4) Fibrous type 6 (26.1) 2 (22.2) Transitional type 3 (13.0) 3 (33.3) Psammomatous type 3 (13.0) 0 Location 0.12 Convexity 7 (30.4) 7 (77.8) Parasagittal 7 (30.4) 0 Tentorial 2 (8.7) 1 (11.1) Falcine 2 (8.7) 0 Skull base 5 (21.7) 1 (11.1) Volume (cm 3 ), mean (SD) 300.82 (205.27) 280.98 (189.27) 0.80 0.88 0.746 Symptom 0.25 Seizure 2 (8.7) 2 (22.2) Motor deficit 10 (43.5) 1 (11.1) Dizziness 6 (26.1) 2 (22.2) Headache 5 (21.7) 4 (44.4) Calcification 0.87 No 16 (69.9) 6 (66.7) Yes 7 (30.4) 3 (33.3) Osteopontin IHC score, mean (SD) 30.87 (46.16) 192.22 (83.03) 0.001* 0.001* 0.001* Ki-67 (LI), mean (SD) 3.43 (3.80) 9.00 (5.27) 0.002* 0.002* 0.002* Perifocal edema 0.001* 0.007* 0.059* No 18 (78.3) 1 (11.1) Yes 5 (21.7) 8 (88.9) SD, standard deviation; IHC, Immunohistochemical; LI, labeling index * Statistical significance, P \ 0.05 recurrence after the gross total resection of histologically benign meningiomas varies from 0 to 32% [11]. There may be some tumors that are equivocal. Meanwhile, the WHO classification does not provide more precise quantitative indicators enabling assessment according to numerical scoring systems, especially considering a finding of fewer than four mitoses of importance [12]. So far, there is no single histological result enabling prediction of recurrence among benign meningiomas (WHO grade I) [2]. Therefore, biomarkers are needed to aid prediction of recurrence, and searching for candidate targets for therapy in meningiomas. In this study, we assessed the OPN protein by immunohistochemistry in intracranial benign meningiomas (WHO grade I). We used tissue microarray technology, which enables OPN protein to be determined at reasonable cost and has the advantage of minimizing experimental variations. Our findings indicated a significant correlation between high cytoplasmic OPN IHC score and recurrence of WHO grade I benign meningiomas. Conversely, low OPN staining was identified in benign non-recurrent meningiomas. This result demonstrated that the OPN IHC score was statistically significantly different between benign non-recurrent tumors and benign recurrent meningiomas. Our observation regarding OPN overexpression in recurrent benign meningiomas was supported by previous studies on the role of OPN in invasiveness, carcinogenesis,

222 J Neurooncol (2010) 100:217 223 Fig. 2 Kaplan Meier meningioma-specific recurrence-free curve for 32 patients classified as group A (Allred score between 0 and 3) and group B (Allred between 4 and 8) according to OPN Allred score. The recurrence rate for patients with OPN Allred score above the cut-off value (Allred score [3) was significantly worse than that for patients with OPN Allred score below the cut-off value (*, P \ 0.05) and metastasis [6, 9, 14, 15]. In addition, our study revealed the importance of osteopontin in early recurrence of meningiomas in accordance with the role of osteopontin in early recurrence of hepatocellular carcinoma [14]. Furthermore, in our study, the statistically significant correlation of recurrent versus non-recurrent meningiomas between OPN and peritumor edema size was also noted. These findings are consistent with literature reports that vascular endothelial growth factor may induce expression of OPN in endothelial cells and further cause angiogenesis, vascular permeability, and peritumor edema formation [16, 17]. Together, these findings support the idea that OPN and peritumoral edema are preferentially expressed by tumor cells with invasive properties, suggesting they may be predictors for tumor recurrence. In efforts to develop a system whereby quantitative indicators can be used to predict the recurrence-free time of meningioma patients, we used the Allred 8-unit system. We found that for patients in group A (Allred score between 0 and 3) 25-month incidence of benign meningioma recurrence was zero (0%). Group B (Allred score between 4 and 8) had a higher 25-month recurrence rate (81.8%), as shown in Fig. 2. We therefore propose that an Allred score exceeding 3 could be used as a threshold values for predicting short-term recurrence in benign meningioma patients. To compare, side by side, the results of OPN with other published positive findings associated with meningioma recurrence, we also assessed peritumor brain edema and Ki-67 staining. Our finding further supported a statistical significant correlation between peritumoral brain edema size and tumoral recurrence. A few studies have assumed that tumor cortical invasion was the cause of peritumoral edema, resulting in a greater incidence of tumoral recurrence [18]. Simultaneously, those tumors presented elevated expression of factors related to tumoral invasion, for example matrix metalloproteinases and angiogenic factors [19]. In this fashion, we believe that peritumoral edema formation of meningiomas may be related to their invading potential and interaction with the adjacent cortex, resulting in hematoencephalic barrier breakage. However, in multivariate analysis, peritumoral brain edema size was not statistically significantly different among patients with recurrence compared to nonrecurrence. Although the reason is not well known, it is suggested that the usefulness of peritumoral brain edema size in establishing prognosis in benign meningioma is affected by the initial location, mainly in the skull base, where tumoral contact occurs with the brain stem, which will not occur following the same formation mechanisms of convexity meningiomas or be a predictor of tumor invasive potential [18]. In the literature, proliferative activity, Ki-67 LI, has also become a potential tool in these tumor subsets to predict recurrence [20]. In our univariate analysis it was demonstrated that Ki-67 LI was statistically significantly different between the benign non-recurrent and benign recurrent meningiomas. The hypothetical reason for this could be that in the process of transformation of the tumor grade, the tumor cells are biologically active, which is evidenced by the tumor s innate proliferative potential, and are yet to show the histological features of atypical meningioma [2]. However, the literature contains conflicting results. Karja et al. [21] reported that mean Ki-67 expression was not significantly different, being 3.3 ± 0.4 for the nonrecurrent meningioma and 3.9 ± 0.5 for the recurrent meningioma with similar Dako IHC antibody (used in ours also). Bruna et al. [22] found Ki-67 LI was an independent predictor of both tumor recurrence and overall survival in patients with atypical meningioma and patients with anaplastic meningioma. These variations between studies organized by different institutes might in part be because of clinical diversity between the selected patients pool and grade of meningiomas, therefore bias might exist especially in retrospective studies [23]. In conclusion, examinations for meningiomas using OPN IHC score and OPN Allred score are very important for prediction of future short-term recurrence in WHO grade I meningiomas. Meningiomas with an OPN Allred score [3 should be followed up meticulously, even when they are totally resected or histologically classified as being in the benign category. However, a large long-term prospective study will help to confirm these results and their positive predictability.

J Neurooncol (2010) 100:217 223 223 Acknowledgments This study was supported by grants from Tri- Service General Hospital TSGH-C97-4-S05, TSGH-C99-069, TSGH- C99-072, TSGH-C99-073, TSGH-C99-074, TSGH-C99-149, TSGH- C99-133, and TSGH-C99-032; National Science Council grants NSC94-3112-B-016-004 and NSC95-3112-B-016-002, and Teh-Tzer Study Group for Human Medical Research Foundation (B971113), and in part by C.Y. Foundation for Advancement of Education, Sciences and Medicine, and grant (DOH99-TD-B-111-003) from Taiwan Department of Health for the Center of Excellence for Clinical Trial and Research in Neuroscience (Jia-Wei Lin), and NSC97-2314-B- 038-046-MY2 (Ming-Huey Chung). References 1. Maiuri F, De Caro Mdel B, Esposito F, Cappabianca P, Strazzullo V, Pettinato G, de Divitiis E (2007) Recurrences of meningiomas: predictive value of pathological features and hormonal and growth factors. J Neurooncol 82:63 68 2. Vankalakunti M, Vasishta RK, Das Radotra B, Khosla VK (2007) MIB-1 immunolabeling: a valuable marker in prediction of benign recurring meningiomas. Neuropathology 27:407 412 3. Chakraborty G, Jain S, Behera R, Ahmed M, Sharma P, Kumar V, Kundu GC (2006) The multifaceted roles of osteopontin in cell signaling, tumor progression and angiogenesis. Curr Mol Med 6:819 830 4. Denhardt DT, Noda M (1998) Osteopontin expression and function: role in bone remodeling. J Cell Biochem Suppl 30 31:92 102 5. Weber GF (2001) The metastasis gene osteopontin: a candidate target for cancer therapy. Biochim Biophys Acta 1552:61 85 6. Furger KA, Menon RK, Tuck AB, Bramwell VH, Chambers AF (2001) The functional and clinical roles of osteopontin in cancer and metastasis. Curr Mol Med 1:621 632 7. Tuck AB, Chambers AF (2001) The role of osteopontin in breast cancer: clinical and experimental studies. J Mammary Gland Biol Neoplasia 6:419 429 8. Tuck AB, Elliott BE, Hota C, Tremblay E, Chambers AF (2000) Osteopontin-induced, integrin-dependent migration of human mammary epithelial cells involves activation of the hepatocyte growth factor receptor (Met). J Cell Biochem 78:465 475 9. Toy H, Yavas O, Eren O, Genc M, Yavas C (2009) Correlation between osteopontin protein expression and histological grade of astrocytomas. Pathol Oncol Res 15:203 207 10. Tunio GM, Hirota S, Nomura S, Kitamura Y (1998) Possible relation of osteopontin to development of psammoma bodies in human papillary thyroid cancer. Arch Pathol Lab Med 122: 1087 1090 11. Ho DM, Hsu CY, Ting LT, Chiang H (2002) Histopathology and MIB-1 labeling index predicted recurrence of meningiomas: a proposal of diagnostic criteria for patients with atypical meningioma. Cancer 94:1538 1547 12. Takahashi JA, Ueba T, Hashimoto N, Nakashima Y, Katsuki N (2004) The combination of mitotic and Ki-67 indices as a useful method for predicting short-term recurrence of meningiomas. Surg Neurol 61: 149 155; discussion 155 146 13. Coppola D, Szabo M, Boulware D, Muraca P, Alsarraj M, Chambers AF, Yeatman TJ (2004) Correlation of osteopontin protein expression and pathological stage across a wide variety of tumor histologies. Clin Cancer Res 10:184 190 14. Pan HW, Ou YH, Peng SY, Liu SH, Lai PL, Lee PH, Sheu JC, Chen CL, Hsu HC (2003) Overexpression of osteopontin is associated with intrahepatic metastasis, early recurrence, and poorer prognosis of surgically resected hepatocellular carcinoma. Cancer 98:119 127 15. Jan HJ, Lee CC, Shih YL, Hueng DY, Ma HI, Lai JH, Wei HW, Lee HM (2010) Osteopontin regulates human glioma cell invasiveness and tumor growth in mice. Neuro Oncol 12:58 70 16. Bellahcene A, Castronovo V, Ogbureke KU, Fisher LW, Fedarko NS (2008) Small integrin-binding ligand N-linked glycoproteins (SIBLINGs): multifunctional proteins in cancer. Nat Rev Cancer 8:212 226 17. Shijubo N, Uede T, Kon S, Nagata M, Abe S (2000) Vascular endothelial growth factor and osteopontin in tumor biology. Crit Rev Oncog 11:135 146 18. Mantle RE, Lach B, Delgado MR, Baeesa S, Belanger G (1999) Predicting the probability of meningioma recurrence based on the quantity of peritumoral brain edema on computerized tomography scanning. J Neurosurg 91:375 383 19. Paek SH, Kim CY, Kim YY, Park IA, Kim MS, Kim DG, Jung HW (2002) Correlation of clinical and biological parameters with peritumoral edema in meningioma. J Neurooncol 60:235 245 20. Hsu DW, Efird JT, Hedley-Whyte ET (1998) MIB-1 (Ki-67) index and transforming growth factor-alpha (TGF alpha) immunoreactivity are significant prognostic predictors for meningiomas. Neuropathol Appl Neurobiol 24:441 452 21. Karja V, Sandell PJ, Kauppinen T, Alafuzoff I (2010) Does protein expression predict recurrence of benign World Health Organization grade I meningioma? Hum Pathol 41:199 207 22. Bruna J, Brell M, Ferrer I, Gimenez-Bonafe P, Tortosa A (2007) Ki-67 proliferative index predicts clinical outcome in patients with atypical or anaplastic meningioma. Neuropathology 27: 114 120 23. Keizman D, Issakov J, Meller I, Maimon N, Ish-Shalom M, Sher O, Merimsky O (2009) Expression and significance of EGFR in malignant peripheral nerve sheath tumor. J Neurooncol 94: 383 388