Springer 2005
Journal of Neuro-Oncology (2005) 74: 59–63 DOI 10.1007/s11060-004-5949-8
Clinical Study
Gliosarcomas: analysis of 11 cases do two subtypes exist? Maurizio Salvati, Emanuela Caroli1, Antonino Raco2, Felice Giangaspero, Roberto Delfini2 and Luigi Ferrante1 Department of Neurological Sciences – Neurosurgery, INM Neuromed IRCCS, Pozzilli (Is), Italy; 1Department of Neurological Sciences – Neurosurgery, Policlinico S. Andrea, University of Rome ‘La Sapienza’, Italy; 2Department of Neurological Sciences – Neurosurgery, Policlinico Umberto I, University of Rome ‘La Sapienza’, Italy
Key words: chemotherapy, gliosarcoma, prognosis, radiotherapy, survival Summary There are conflicting reports regarding gliosarcomas. The goal of this study is to examine clinical, radiological, surgical and therapeutic aspects of 11 patients with gliosarcoma. Between 1993 and 2001, 11 patients with cerebral gliosarcoma were treated at our Institute. Ten patients underwent surgery and one patient had stereotactic biopsy. Four patients received whole brain radiotherapy with 60Co, five underwent radiotherapy with LINAC extended 2 cm beyond the edema margins. One patient refused any additional treatment after surgery and one patient was not treated postoperatively for poor clinical conditions (KPS 40). Chemotherapy (temozolomide) was administered to four patients. Four patients had a prevalence of sarcomatous component that corresponded to surgical and radiological aspects similar to meningioma while six patients showed a prevalence of gliomatous component and radiological and surgical aspects similar to those of glioblastomas. Surgical resection was total in six and subtotal in four patients. Patients with prevalent sarcomatous component showed median survival time more prolonged than patients with prevalent gliomatous component (71 ± 6 weeks vs. 63 ± 6; P ¼ 0.0417). Moreover, the survival rate differed in relation to the therapy: patients treated with multimodality therapy (surgery, radiotherapy and chemotherapy) had a longer survival time than patients treated in single or bimodality. Despite prognosis of gliosarcomas remains poor, a multidisciplinary approach (surgery, radiotherapy and chemotherapy) seems to be associated with slight more prolonged survival times.
Introduction Gliosarcoma is a rare malignant primary brain tumor characterized by a biphasic tissue pattern with areas of glial and mesenchymal differentiation. Recent genetic studies suggest a monoclonal origin of gliosarcoma and an evolution of the sarcomatous component due to acquisition of a mesenchymal phenotype in a highly malignant astrocytic tumor [1–5]. Genetic aberrations, clinical features, and prognosis are similar to those of multiform glioblastoma [1,2,4,6,7]. Most of the literature concerning gliosarcoma consists of case reports or focuses on histopathological aspects [8–14]. The purpose of this report is to analyze clinical, radiological, surgical, therapeutic and prognostic aspects of 11 patients with gliosarcoma.
Materials and methods Between 1993 and 2000, 11 patients were treated to our Department for cerebral gliosarcoma. The diagnosis was made by histopathological examination in all cases. Diagnosis of gliosarcoma was based on the following histological criteria defined by Meis in 1991 [10]: (a) the tumour must be bimorphic, that is composed of two morphologically distinct malignant cells population; (b) one component must be astrocytic with areas of
necrosis, fulfilling the criteria for glioblastoma; (c) the sarcomatous component must resemble a spindle cell sarcoma; and (d) a minimum of one confluent sarcomatous area must fill one medium power field [10· objective with 10· eyepiece). Moreover, trichromic stain and immunohistochemistry for GFAP, and reticulin stain were performed to distinguish the glial from the sarcomatous component of the neoplasia [4,15]. Two cases, in addition to glial and sarcomatous elements, contained cartilaginous and osteoid tissue. The diagnosis of a gliosarcoma was confirmed in all 11 specimens upon neuropathological re-examination for this study. Preoperative CT was performed in all patients and MRI in 9 patients. Early postoperative MRI was performed in five patients. In all patients serial postoperative MRI (9 patients) or CT (two patients) were performed. Surgical removal was estimated as total if residual tumor was £10%, subtotal if residual tumor was £50%. Total tumor removal was documented by early postoperative MRI (12 h after operation) in five cases. The degree of tumor resection was based only on the neurosurgeon’s intraoperative impression in five patients (one total and four subtotal). In some patients gliosarcoma showed radiological and surgical aspects mimicking a meningioma, that is hyperdense signal and marked contrast enhancement on
60 CT and firm and well-demarcated limits at surgery. In other patients, gliosarcoma showed radiological and surgical aspects mimicking a glioma. In the latter event, gliosarcoma appears as a dyshomogeneous mass with diffuse or ring-shaped enhancement on CT/MRI and diffusely infiltrating with no cleavage plane and necrotic areas at surgery. All but one patient were proposed to receive postoperative radiotherapy. The average time between surgery and the initiation of radiotherapy was 20 days. Four patients received whole brain radiotherapy with 60Co in 5 weeks, five underwent 60 Gy radiotherapy with LINAC extended 2 cm beyond the edema margins. One patient refused any additional treatment after surgery, and one patient was not treated postoperatively for poor clinical conditions (KPS 40). Chemotherapy was administered to four patients. The only drug used was temozolomide. Chemotherapy was delivered 200 mg/m2/day for 5 days every 28 days until disease progression. Survival functions were estimated by Kaplan–Meier method using the package SPSS.10.
Results Nine patients were male and two female. Their ages ranged from 39 to 81 years (median 52 years). Location of the gliosarcoma was temporal in six patients (55%), frontal in two (18%), parietal in two (18%) and occipital in one (9%). CT scans showed hyperdense lesions in four patients and hypodense mass in seven. In contrastenhanced CT scans, the four hyperdense lesions showed marked contrast enhancement, the seven hypodense lesions showed ring-shaped enhancement. At MRI most lesions (seven cases) appeared as a inhomogeneous mass and all sholwed diffuse or ring-shaped enhancement. Two lesions showed a broad-dural based involvement and a diffuse and inhomogeneous enhancement. Marked perilesional edema was a constant radiological feature. Surgical resection was reported as gross total in six patients and subtotal in four. One patient had stereotactic biopsy. In the five patients who had early postoperative MRI, surgeon’s report correlated with the amount of tumor resection demonstrated on postoperative imaging. At operation four tumours were firm, dural-based and had well demarcated margins so resembled a meningioma. The other six tumors were necrotic and infiltrating, resembling glioblastoma. Eight patients had a preoperative KPS of 90, two patients 80 and one 40. Postoperative KPS was increased to
100 in 5 patients, decreased of 10 points in one, and remained stable in the remainder. The histological examination was performed on five sections for each case. The pathological specimens revealed a sarcomatous component inferior or equal to 25% in seven patients (group 1), and more than 50% in four patients (group 2). Gliosarcomas with prevalence of sarcomatous component [4] (group 2) had surgical and radiological aspects similar to a meningioma while gliosarcomas with prevalence of gliomatous component [6] (group 1) showed radiological and surgical aspects similar to those of glioblastomas. Median survival times for patients of group 2 were significantly longer than that of the groups 1 (71 ± 6 weeks vs. 63 ± 6; P ¼ 0.0417) (Table 1 and Figure 1). The survival rate differed also in relation to the therapy (Table 2): • Two patients (mean age 80 years) had no additional treatment after surgery and had a survival rate of 11 and 21 weeks, respectively. • Nine patients treated with postoperative radiotherapy had a median survival time of 66 ± 1 weeks (CI 95% 63,69). • Five of these patients (mean age 49.6 years) had radiation therapy alone after surgery, (one LINAC 60 -Gy and four whole brain 60Co in five weeks). Their median survival time was 63 weeks. Two of these had subtotal surgical removal and the remainder had total removal. • Four patients (mean age 49.7 years) received radiotherapy with LINAC (60Gy) followed by chemotherapy with temozolomide. One of these patients had subtotal surgical removal. Three patients had gross macroscopic removal of the lesion. The median survival time of these four patients was 69.5 weeks. There is no significant difference in terms of survival between patients treated with 60Co and LINAC. The median recurrence or time to progression of the gliosarcoma with gliomatous predominance was 53.5 weeks vs. 62 weeks in gliosarcoma with sarcomatous predominance (Table 2). Eight patients died of recurrent or disease progression from 11 to 75 weeks after diagnosis, their median survival time was 62.5 week. The remaining three patients underwent a second surgical resection for tumor recurrence 56, 44 and 41 weeks after the first operation. Two of these three patients had a total removal at first operation and one subtotal removal. The resection at second operation was total in two patients and subtotal in one. These
Table 1. Survival time (weeks) after diagnosis of gliosarcoma Parameter
Histological & surgical finding
SE = standard error. CI = confidence interval.
Gliomatous prevalence Sarcomatous prevalence
n
Mean ± S.E. (95% CI) Median ± S.E. (95% CI)
Statistical significance
6
57 63 74 71
P = 0.0417
4
± ± ± ±
7 6 5 6
(42.71) (52.74) (64.83) (58.84)
61
Figure 1. Survival functions.
patients presented a second recurrence at 5, 37 and 26 weeks from second operation, respectively. One patient was not operated on because tumour recurrence involved midline structures and died one months after diagnosis of recurrence. The remainder patients were reoperated on and died of tumor recurrence with involvement of the midline structures at 4 and 5 weeks, respectively. The median survival time in patients reoperated on was 68 weeks. Patients with sarcoma predominance recurred later (mean 59.7 weeks) than those with glioma predominance (mean 47 weeks).
Discussion Gliosarcoma is a morphologically defined glioblastoma variant, originally described in 1895 by Stroebe [16]. Since then the histopathogenesis of the gliosarcoma is a matter of controversy. Whereas morphological studies suggested an evolution of the sarcomatous component from microvascular proliferation within a highly malignant glioblastoma [17,18], recent genetic studies revealed a common origin of the glial and sarcomatous component from glial cells [1–5,19]. The concept of a
monoclonal origin is strongly supported by recent genetic analyses which show identical genetic aberrations in both gliomatous and sarcomatous components [1–5]. Biernat et al. [1] demonstrated the presence of identical p53 mutations in the two tumour areas. Homozygous p16 deletion, PTEN mutation and co-amplification of MDM2 and CDK4 are other gene alterations detected both in gliomatous and sarcomatous component of gliosarcoma [19]. The reported incidence of gliosarcomas varies between 2% and 10% of all malignant gliomas [4,9,11,20,21]. We found 11 cases of gliosarcomas from a series of 500 patients with malignant glioma treated in the same period. The 2.2% incidence at our institution is similar to prior published reports. Gliosarcomas usually occur in the fifth – sixth decade of life and have a slight male preponderance [1,4,9,11–13,22–30]. This is consistent with our data (M/F ratio 11/2 and median age 52 years). Multiple reports in the literature describe the higher tendency of gliosarcomas to metastasise to extracranial and intraaxial sites than that of glioblastoma [11,31]. This tendency is reported to be as high as 15% in some studies [6,31], but extraneural metastases were not found in our eleven patients. In spite of multimodality therapy, the prognosis of patients affected by gliosarcoma remains poor. The median survival has been reported to range from 6 to 14.8 months [4,7,9,11–13,27,32,33]. Median survival in our patients was 16.2 months. Longer survival times up to 6 years have only occasionally been reported [7,33–35]. It has been noticed that prevalent sarcomatous component of gliosarcomas corresponds to radiological and surgical aspects suggestive of meningioma. Patients affected by gliosarcoma with the above mentioned characteristics seem to have a slightly more prolonged survival times [7,12,32,36]. This seems to be consistent with our results. We found a significant statistical correlation between radiological and surgical features of gliosarcoma and survival (P ¼ 0.0417) (Table 1 – Figure 1). However, we admit that the small number of cases which, of course, is not surprising considering the fact that these are very rare tumors,
Table 2. Case
Age/Sex
KPS
Location – size
Surgical resection
Surgical & histological findings
RT
CHT
Recurrence (Weeks)
Survival (Weeks)
1 2 3 4 5 6 7 8 9 10 11
54/M 52/M 58/M 45/F 39/M 49/F 44/M 58/M 48/M 81/M 79/M
90 90 90 90 90 80 90 90 90 40 80
F – 3 cm P – 4 cm T – 4.5 cm F – 5.5 cm T – 4 cm P – 4 cm T – 5 cm T – 5.5 cm T – 3.8 cm O – 6.5 cm T – 3cm
T T ST T ST T T T ST SB ST
S S G G G S G S G / G
YES YES YESa YES YES YESa YESa YESa YESa NO NO
NO NO NO NO NO YES YES YES YES NO NO
Yes · 1 (59) Yes · 1 (71) Progression (51) Yes · 3 (56,5,4) Progression (60) Yes · 3 (44,37,5) Yes (59) Yes (65) Progression · 2 (41,26) Progression Progression (19)
62 75 56 65 63 86 66 71 68 11 21
T = total removal; SB = stereotactc biopsy; ST = subtotal removal; S = sarcomatous component prevalent and surgical aspect of meningioma; G = gliomatous component prevalent and surgical aspect of glioblastoma; RT = radiotherapy; Size = max diameter; CHT = chemotherapy; a = LINAC.
62 may cause possible selection bias in groups. Anyway, it has been well documented that extent of survival is correlated with completeness of surgical resection, so those gliosarcomas that resemble meningiomas both in location and in terms of demarcation from the brain are amenable to radical surgical resection. On the other hand, the gliosarcomas that resemble glioblastomas are intrinsic to the brain and more difficult to remove radically. This may be the main explanation for the survival time difference found in our study. In fact, all four sarcomatous tumors of our series underwent total resection compared to only two of six gliomatous tumors. Age and KPS have no influenced the difference of survival time of the two groups of our patients. We found also that the recurrence and progression time of gliosarcoma with predominant gliomatous component were inferior (53.5 weeks) to those of the gliosarcoma with predominant sarcomatous component (62 weeks). Several authors report significant biological similarities in the behaviour of gliosarcoma and glioblastoma [4,8,10], then the same treatment should be applied for these two kinds of tumor. On the basis of this observation, we have treated patients with gliosarcoma as those with glioblastoma. Total surgical removal was possible in six patients and subtotal in four. We have adopted two postoperative therapeutic modalities over the years: radiation therapy (60Co or LINAC) and, more recently, radiotherapy (LINAC) plus chemotherapy with temozolomide (Table 2). Our choice of temozolomide over other agents is due to its high tolerability and promising results seen in patients with malignant gliomas. The benefits of chemotherapy are still not clear because the studies reported in the literature refer to different schedules of treatment [8,10,24]. In addition, one should keep in mind that chemotherapic agents that are active against soft tissue sarcoma might be effective as well for gliosarcoma, especially for those one with prevalent sarcomatous component. Chemotherapy appear to have an important role for soft tissue sarcoma [37,38]. However, it is unknown if gliosarcomas respond to sarcoma-based chemotherapy because the vast majority of the studies emphasizes the significant similarity of gliosarcoma with glioblastoma [4,8,10]. Temozolomide has been clinically investigated in different cancers including advanced soft tissue sarcoma [39]. However, more data are required before temozolomide can be judged in this disease because the prognosis for patients who received temozolomide in this study was grave [39]. A variety of chemotherapeutic agents have been used in cerebral gliosarcomas such as nitrosoureas (BCNU, CCNU, PCNU) [8,10], misonidazole [10], dacarbazine [10], mythramycin [11], and ametophterin [11]. However, only Morantz et al. [11] comment on the effect of chemotherapy on outcome. In this study of 24 patients with gliosarcoma, chemotherapy was used in adjunct to radiotherapy in nine cases. Five patients were given mithramycin, two BCNU, one adriamycin, and one intracarotid amethopterin. The authors found a modest increase in survival in gliosarcoma patients with additional chemotherapy (36 weeks) compared with
radiation therapy alone (33 weeks). Meis et al. [10] in a comparative study between multiform glioblastoma and gliosarcomas (26 cases) found that none of the treatment regimens, which included various combination of radiation therapy and chemotherapy (BCNU, dacarbazine, semustine, misonidazole), improved survival of gliosarcoma over glioblastoma being 8.3 months vs. 9.6 months. Similarly, Galanis et al. [8] supported the conclusion that patients with gliosarcoma should be treated identically to those with glioblastoma. These authors treated all patients with external radiation and nitrosurea-based chemotherapy (carmustine, PCNU and dibromodulcitol) and reported a median survival time of 28 weeks. In our small series, chemotherapy seems to have added beneficial effect to the radiotherapy, both in patients with gliosarcoma resembling meningioma and in those patients affected by gliosarcoma with prevalent gliomatous component (median survival 69.5 weeks vs. 63 weeks in patients with only postoperative radiotherapy) (Table 2). The median survival rate of our patients is substantially higher than that previously reported for any other chemo-radiotherapy regimen. This should prompt to further investigate in further studies the role of temozolomide in treatment of gliosarcomas.
Conclusion Our current treatment policy of gliosarcoma consists of surgical removal as extended as possible, radiotherapy extended 2 cm beyond the edema margins (LINAC) and chemotherapy with temozolomide. Despite the fact our series is small, we can suggest that patients treated with this protocol may have a more prolonged survival times than patients treated with surgery and radiotherapy. Also, our study corroborates the assumption of the existence of two subtypes of gliosarcoma that present different histological, radiological, surgical characteristics and prognosis.
References 1. Biernat W, Aguzzi A, Sure U, Grant JW, Kleihues P, Hegi ME: Identical mutations of the p53 tumor suppressor gene in the gliomatous and the sarcomatous components of gliosarcomas suggest a common origin from glial cells. J Neuropathol Exp Neurol 54: 651–656, 1995 2. Boerman RH, Anderl K, Herath J, Borrel T, Johnson N, Schaeffer-Klein J, Kirchhof A, Raap AK, Scheitauer BW, Jenkins RB: The glial and mesenchymal elements of gliosarcomas share similar genetic alterations. J Neuropathol Exp Neurol 55: 973– 981, 1996 3. Jones H, Steart PV, Weller RO: Spindle-cell glioblastoma or gliosarcoma? Neuropathol Appl Neurobiol 17: 177–187, 1991 4. Meis JM, Ho KL, Nelson JS: Gliosarcoma: a histologic and immunohistochemical reaffirmation. Mod Pathol 3: 19–24, 1990 5. Paulus W, Bayas A, Ott G, Roggendorf W: Interphase cytogenetics of glioblastoma and gliosarcoma. Acta Neuropathol (Berl) 88: 420–425, 1994 6. Cerame MA, Guthikonda M, Kohli CM: Extraneural metastases in gliosarcoma: a case report and review of the literature. Neurosurgery 17: 413–418, 1985
63 7. Maiuri F, Stella L, Benvenuti D, Giamundo A, Pettinato G: Cerebral gliosarcomas: correlation of computed tomographic findings, surgical aspect, pathological features, and prognosis. Neurosurgery 26: 261–267, 1990 8. Galanis E, Buckner JC, Dinapoli RP, Scheitauer BW, Jenkins RB, Hua Wang C, O’Fallon JR, Farr G: Clinical outcome of gliosarcoma compared with glioblastoma multiforme: North Central Cancer Treatment Group results. J Neurosurg 89: 425– 430, 1998 9. Lutterbach J, Guttenberger R, Pagenstecher A: Gliosarcoma: a clinical study. Radiother Oncol 61: 57–64, 2001 10. Meis JM, Martz KL, Nelson JS: Mixed glioblastoma multiforme and sarcoma. A clinicopathologic study of 26 radiation therapy oncology group cases. Cancer 7: 2342–2349, 1991 11. Morantz RA, Feigin I, Ransohoff J III: Clinical and pathological study of 24 cases of gliosarcoma. J Neurosurg 45: 398–408, 1976 12. Parekh HC, O’Donovan DG, Sharma RR, Keogh AJ: Primary cerebral gliosarcoma: report of 17 cases. Br J Neurosurg 9: 171– 178, 1995 13. Perry JR, Ang LC, Bilbao JM, Muller PJ: Clinicopathologic features of primary and postirradiation cerebral gliosarcoma. Cancer 75: 2910–2918, 1995 14. Sarkar C, Sharma MC, Sudha K, Gaikwad S, Varma A: A clinicopathological study of 29 cases of gliosarcoma with special reference to two unique variants. Indian J Med Res 106: 229–235, 1997 15. Kleihues P, Burger PC, Scheithauer BW: Histological Typing of Tumours of the Central Nervous System. Springer, Berlin Heidelberg, New York-Tokyo, 1993 16. Stroebe H: Uber entstehung und bau der hirngliome. Beitr Pathol Anat 18: 405–486, 1895 17. Feigin I, Allen LB, Lipkin L, Groos SW: The endotelial hyperplasia of the cerebral blood vessels within brain tumors and its sarcomatous transformation. Cancer 11: 264–277, 1958 18. Haddad SF, Moore SA, Schelper RL, Goeken JA: Smooth muscle can comprise the sarcomatous component of gliosarcomas. J Neuropathol Exp Neurol 51: 493–498, 1992 19. Reis RM, Konu-Lebleblicioglu D, Lopes JM, Kleihues P, Ohgaki H: Genetic profile of gliosarcomas. Am J Pathol 156: 425–432, 2000 20. Codd MB, Kurland LT: Descriptive epidemiology of primary intracranial neoplasms. Prog Exp Tumor Res 29: 1–11, 1985 21. Feigin IH, Gross SW: Sarcoma arising in glioblastoma of the brain. Am J Pathol 31: 6333–6353, 1995 22. Beute BJ, Fobben ES, Hubschmann O, Zablow A, Eanelli T, Solitare GB: Cerebellar gliosarcoma: report of a probable radiation-induced neoplasm. AJNR Am J Neuroradiol 12: 554– 556, 1991 23. Biernat W, Hegi M, Aguzzi A, Kleihues P: Gliosarcoma. In: Kleihues P, Cavenee WK (eds) Pathology and Genetics of Tumours of the Nervous System. IARC, Lyon, 1997, pp. 27–28 24. Classen J, Hoffmann W, Kortmann RD, Lehr A, Meyermann R, Palmbach M, Bamberg M: Gliosarcoma–case report and review of the literature. Acta Oncol 36: 771–774, 1997
25. Garret P: Glioblastoma and fibroblastoma of the brain with extracranial metastases. Cancer 11: 888–894, 1958 26. Ho KL: Histogenesis of sarcomatous component of the gliosarcoma: an ultrastructural study. Acta Neuropathol (Berl) 8: 178– 188, 1990 27. Khan EM, Pandey R, Gupta S, Mittal P: Gliosarcomas: pathological spectrum. Indian J Cancer 3: 118–123, 1994 28. Slowik F, Jellinger K, Gaszo L, Fischer J: Gliosarcomas: histological, immunohistochemical, ultrastructural, and tissue culture studies. Acta Neuropathol (Berl) 67: 201–210, 1985 29. Tamura T, Nakamura S, Shirata K, Sato K, Tsubokawa T, Sawada T: Cerebellar malignant glioma after radiation therapy for suprasellar germinoma-case report. Neurol Med Chir (Tokyo) 29: 223–229, 1989 30. Wargotz ES, Sidawy MK, Jannotta FS: Thorotrast-associated gliosarcoma. Including comments on thorotrast use and review of sequelae with particular reference to lesions of the central nervous system. Cancer 62: 58–66, 1988 31. Smith DR, Hardman JM, Earle KM: Contiguous glioblastoma multiforme and fibrosarcoma with extracranial metastasis. Cancer 24: 270–276, 1969 32. Cervoni L, Celli P: Cerebral gliosarcoma: prognostic factors. Neurosurg Rev 19: 93–96, 1996 33. Schiffer D, Giordana MT, Mauro A, Migheli A: GFAP, FVIII/ Rag, laminin, and fibronectin in gliosarcomas: an immunohistochemical study. Acta Neuropathol 63: 108–116, 1984 34. Lee YY, Castillo M, Nauert C, Moser RP: Computed tomography of gliosarcoma. AJNR 6: 527–531, 1985 35. Winkler PA, Buttner A, Tomezzoli A, Weis S: Histologically repeatedly confirmed gliosarcoma with long survival: review of the literature and report of a case. Acta Neurochir (Wien) 142: 91–95, 2000 36. Dwyer KW, Naul LG, Hise JH: Gliosarcoma: MR features. J Comput Assist Tomogr 20: 719–723, 1996 37. Raney RB Jr, Tefft M, Newton WA et al.: Improved prognosis with intensive treatment of children with cranial soft tissue sarcomas arising in nonorbital parameningeal sites: a report from the intergroup rhabdomyosarcoma study. Cancer 59: 147–155, 1987 38. Schutte J, Kellner R, Seeber S: Ifosfamide in the treatment of soft-tissue sarcomas: experience at the West Germany Tumour Center, Essen. Cancer Chemother Pharmacol 31: (Suppl 2) 194– 198, 1993 39. Woll PJ, Crowther D, Johnson PWM et al.: Temozolamide in adult patients with advanced soft tissue sarcoma: a phase II study of the EORTC soft tissue and bone sarcoma group. Eur J Cancer 35: 410–412, 1999
Address for offprints: Emanuela Caroli, Via Meropia 85, 00147 Rome, Italy; Tel.: +39-06-80345487; Fax: +39-06-44888057; E-mail:
[email protected]
