Klaus Kitz

Defining the Role of Stereotactic Radiosurgery Versus Microsurgery in the Treatment of Single Brain Metastases

Summary Stereotactic radiosurgery (RS) and surgery have proved to be effective treatment modalities for brain metastasis. We followed 133 patients whose treatment for intracranial disease was either RS or a single surgical resection at the University of Vienna from August 1992 through October 1996. All patients who received additional Whole Brain Radiotherapy were included. This was a retrospective, case-control study comparing these treatment modalities. Sixty-seven patients were treated by RS and 66 patients were treated by microsurgery. The median size of the treated lesions for RS patients was 7800 mm3, and 12500 mm3 for microsurgery patients, respectively. The median dose delivered to the tumour margin for RS patients was 17 gray. The median survival for patients after RS was 12 months, and 9 months for patients after microsurgery. This difference was not statistically significant (p=0.19). Comparison of local tumour control, defined as absence of regrowth of a treated lesion, showed that tumours following RS had a preferred local control rate (p<0.05). Univariate and multivariate analysis showed that this fact was due to a greater response rate of “radioresistant” metastasis to RS (p<0.005). Postradiosurgical complications included the onset of peritumoural oedema (n=5) and radiation necrosis (n=1). Two patients after microsurgery experienced local wound infection. One postoperative death occurred due to pulmonary embolism in this group. On the basis of our data we conclude that RS and microsurgery combined with Whole Brain Radiotherapy are comparable modalities in treating single brain metastasis. Concerning morbidity and local tumour control, in particular in cases of “radioresistant” primary tumours, RS is superior. Therefore we advocate RS except for cases of large tumours (>3 cm in maximum diameter) and for those with mass effect.

Proliferation activity in pituitary adenomas: measurement by monoclonal antibody Ki-67

The monoclonal antibody (MAb) Ki-67 detects a nuclear antigen expressed by proliferating cells during the entire cell cycle. In contrast to conventional histological techniques, the use of MAb Ki-67 on frozen sections or cytological smear preparations allows direct determination of the growth rate of tumors routinely. Sixty-two pituitary adenomas were investigated by use of the MAb Ki-67 in a two-step avidin-biotin-peroxidase complex technique. The proliferation activity ranged from 0.1 to 2.8%. There was no significant difference between the proliferation and hormonal state of the adenomas. Adenomas for which there was histological evidence of dural infiltration, however, showed a statistically significant higher proliferation activity (P less than 0.05) compared to noninvasive adenomas.

Monocyte subpopulations in human gliomas: expression of Fc and complement receptors and correlation with tumor proliferation

SummaryCryostat sections of 12 gliomas and of 3 peritumoral brain tissue samples were investigated for mononuclear cell infiltration by immunohistochemistry, concentrating on cells expressing monocyte/macrophage markers. Only low numbers of T cells were detected in the tumors, whereas in average 20%–30% of all cells present in the samples were recognized by various macrophage markers. These cells carried surface epitopes with known function, like Fc-γ (Fcg) and complement receptors. Microglial cells, in comparison to typical debris laden macrophages, were only recognized by a restricted panel of macrophages markers (anti-Fcg receptors 1, 2, 3, complement receptor CR3, HLA DR, common leucocyte antigen CD45 and the monocyte marker RM3/1). In peritumoral tissue mainly dendritic, microglia-like cells were present, which revealed decreased expression of antigens CD4, RM3/1 and Fcg receptors in comparison to those in gliomas. A significant positive correlation was found between the number of RM3/1 or CR3 (CD11b)-positive cells and the proliferation rate of the tumors as documented by the number of bromodeoxyuridine-positive or Ki-67+ cells.

Local tumor control and morbidity after one to three fractions of stereotactic external beam irradiation for uveal melanoma

BACKGROUND AND PURPOSE To evaluate prospectively local tumor control and morbidity after 1-3 fractions of stereotactic external beam irradiation (SEBI) in patients with uveal melanoma, unsuitable for ruthenium-106 brachytherapy or local resection. MATERIAL AND METHODS This phase I/II study includes 62 selected patients with uveal melanoma. The mean initial tumor height was 7.8+/-2.8 mm. With the Leskell gamma knife SEBI, 41 patients (66%) were irradiated with two equal fractions of 35, 30 or 25 Gy/fraction, 14 patients (22%) were treated with three fractions of 15 Gy each, and seven patients (11%) with small tumor volumes below 400 mm(3) were treated with one fraction of 45 Gy. The mean total dose was 54+/-8 Gy. The minimal follow-up period was 12 months, and the median follow-up was 28.3 months. Data on radiation-induced side-effects were analyzed with the Cox proportional hazards model for possible risk factors. RESULTS Local tumor control was achieved in 98% and tumor height reduction in 97%. The mean relative tumor volume reductions were 44, 60 and 72% after 12, 24 and 36 months, respectively. Seven patients developed metastases (11%). Secondary enucleation was performed in eight eyes (13%). Morbidity was significant in tumors exceeding 8 mm in initial height; it was comparable and acceptable in those smaller. In the stepwise multiple Cox model, tumor localization, height and volume, planning target volume (PTV), total dose and patient age were identified as the strongest risk factors for radiation-induced lens opacities, secondary glaucoma, uveitis, eyelash loss and exudative retinal detachment. In this model, the high-dose volume irradiated with more than 10 Gy/fraction was the strongest risk factor for radiation-induced uveitis. CONCLUSIONS Stereotactic external photon beam irradiation and a total dose of 45-70 Gy delivered in one to three fractions are highly effective at achieving local tumor control in uveal melanoma. Further clinical studies using smaller fraction doses, and consequent smaller high-dose volumes, are justified to optimize dose and fractionation. Fractionated stereotactic irradiation has a challenging potential as an eye-preserving treatment in uveal melanoma.

Gamma knife radiosurgery of skull base meningiomas.

Summary Background. The standart surgical treatment of meningiomas is total resection of the tumour. The complete removal of skull base meningiomas can be difficult because of the proximity of cranial nerves. Stereotactic radiosurgery (SRS) is an effective therapy, either for adjuvant treatment in case of subtotal or partial tumour resection, or as solitary treatment in asymptomatic meningiomas. Method. Between September 1992 and October 1995, SRS using the Leksell Gamma Knife was performed on 46 patients (f:m=35:15), ranging in age from 35 to 81 years, with skull base meningiomas at the Neurosurgical Department of the University of Vienna. According to the indication of gamma knife radiosurgery (GKRS) the patients (n=46) were devided into two subgroups. Group I (combined procedure: subtotal resection followed by GKRS as a planned procedure or because of a recurrent meningioma), group II (GKRS as the primary treatment). Histological examination of tumour tissue was available for 31 patients (67%) after surgery covering 25 benign (81%) and 6 malignant (19%) meningioma subtypes. Findings. The overall tumour control rate after a mean follow-up period of 48 months (ranging from 36 to 76 months) was 96% (97.5% in benign and 83% in malignant meningiomas). Group I displayed a 96.7% tumour control rate, followed by group II with 93.3% respectively. Neurological follow-up showed an improvement in 33%, stable clinical course in 58% and a persistant deterioration of clinical symtoms in 9%. Remarkable neurological improvement after GKRS was observed in group II (47%), whereas in group I (26%) the amelioration of symptoms was less pronounced. Interpretation. GKRS in meningiomas is a safe and effective treatment. A good tumour control and low morbidity rate was achieved in both groups (I, II) of our series, either as a primary or adjunctive therapeutic approach. The planned combination of microsurgery and GKRS extends the therapeutic spectrum in the treatment of meningiomas. Reduction of tumour volume, increasing the distance to the optical pathways and the knowledge of the actual growing tendency by histological evaluation of the tumour minimises the risk of morbidity and local regrowth. Small and sharply demarcated tumours are in general ideal candidates for single high dose-GKRS, even after failed surgery and radiation therapy, and in special cases also in larger tumour sizes with an adapted/reduced margine dose.

In situ analysis of cell kinetics in human brain tumors

SummaryA newly developed in vitro labeling method with bromodeoxyuridine (BrdU) identifies S phase cells in situ in freshly obtained surgical tissue of human brain tumors which is subsequently fixed and embedded in paraffin for BrdU immunovisualization. For the first time, the BrdU labeling index (LI) is successfully compared here with the LI obtained by immunostaining of frozen sections of the same tumors with monoclonal antibody Ki-67 which identifies all proliferating cells, i.e., the growth fraction. LIs were counted in at least five different areas with high density of labeled cells; at least 1,000 cells were counted. In 13 metastatic tumors, Ki-67 LI was 8.3%–62.6%, and BrdU LI was 5.1%–28.0%. In 18 gliomas, Ki-67 LI was 1.4%–19.3%, and BrdU LI was 0.2%–11.6%. In 7 meningiomas, Ki-67 LI was 0.3%–3.0%, and BrdU LI was 0%–2.0%. Statistical comparison of Ki-67 and BrdU LIs by linear regression analysis revealed a highly significant correlation: BrdU LI=0.99+0.34 Ki-67 LI (r=0.92,P<0.001). A significant heterogeneity of proliferation patterns may occur within one sample from area to area, as well as between different samples of the same tumor, especially in gliomas; thus, some subjective influence on LIs by arbitrary sampling and selection could occur in quantitative evaluation of in situ cell kinetics of human brain tumors. This study indicates that our in vitro BrdU-labeling method allows the in situ identification of S phase cells in excellently preserved fixed tumor tissue which is well suited for further histological examination. This method compares favorably with Ki-67 labeling of frozen sections and might emerge as a powerful new tool for the routine study of cell proliferation in surgical specimens of human brain tumors.

Gamma-Knife Radiosurgery for Brain Metastases of Renal Cell Carcinoma: Results in 23 Patients

Summary From Jan. 1993 to Sept. 1995 23 patients suffering from brain metastases from renal cell carcinoma were treated with the Leksell Gamma Knife at the University of Vienna. At the time of diagnosis 13 patients had single and 10 patients presented with multiple metastatic lesions with a total of 44 metastases in MRI scans. Median tumour volume was 5500 cmm (range 100–24000 cmm). Predominant neurological symptoms and signs were different forms of hemiparesis, focal and generalized seizures, cognitive deficit, headache, dizziness, ataxia and CN XII paresis. Fourteen patients received Gamma Knife Radiosurgery (GKRS) with a median dose of 22 Gy (range 8–30 Gy) at the tumour margin. Nine patients underwent a combined treatment of a radiosurgical boost with a median dose of 18 Gy (range 10–22 Gy) at the tumour margin followed by Whole Brain Radiotherapy (total dose 30 Gy/2 weeks). In 20 patients tumour volume reduction up to 30% of the primary tumour volume was found after 4 weeks, evaluated on CT or MRI. A total remission was seen in 4 cases 3 months after GKRS. We achieved a local tumour control of 96%. Rapid neurological improvement after GKRS was seen in 17 patients. The median survival time was 11 months; the one-year actual survival in this unselected group was 48%. Five long term survivors were still alive, 18 patients had subsequently died, 15 of them of general tumour progression. GKRS induces a significant tumour remission accompanied by rapid neurological improvement and therefore provides the opportunity for extended high quality survival. Neither local tumour control was improved nor CNS relapse free survival was prolonged significantly by additional WBRT.

Acoustic Neurinomas with Macrocysts Treated with Gamma Knife Radiosurgery

Six cases of acoustic neurinomas with macrocystic components are presented. In three cases the cystic portion was within the tumor, while in the other three, the cyst was peritumoral, in the form of a cul-de-sac within the arachnoid, in other words it was not a true tumor cyst. The six tumors are from a series of 74 acoustic neurinomas treated by radiosurgery with a minimum follow-up of 18 months. In all cases, enlargement of the associated cyst was observed as early as 4 months after radiosurgery. Clinical signs and symptoms such as facial weakness, trigeminal symptoms, vertigo and dizziness and coordination disorders developed between 4 and 8 months. In three cases (two intramural cysts and one combined peri- and intramural cyst), subacute microsurgery was performed to treat the progression of neurological symptoms. One case had spontaneous rupture of an intramural cyst, one case of a peritumoral cyst, after progression showed a slow spontaneous size decrease after 2 years, and one case is still under observation. In the reported series, the dose at the tumor margin ranged between 11 and 17 Gy (mean 13.8 +/- 2.5 [SD] Gy) and the maximal dose between 24 and 40 Gy (mean 30.6 +/- 6.2 Gy). In view of the findings in this study, one should perhaps be cautious in advising radiosurgery for this subgroup of acoustic tumors.

Argyrophilic nucleolar organizer region proteins (Ag-NORs) in human brain tumors: relations with grade of malignany and proliferation indices

SummaryProliferation indices and mean number of silver-stained nucleolar organizer region-associated proteins (Ag-NORs) are compared in 65 brain tumors, including 34 gliomas, 8 meningiomas, 17 metastatic tumors, and 6 other tumors. Immunocytochemical investigations include labeling with the monoclonal antibody Ki-67 which identifies the whole growth fraction, and with a monoclonal antibody against bromodeoxyuridine (BrdUrd) which detects cells in the S phase of the cell cycle after in vitro incubation with BrdUrd. When all types of tumors are collectively considered, mean numbers of Ag-NORs did not correlate with Ki-67 and Brd-Urd labeling indices (LIs) and mitotic index. Among tumor subtypes, only meningiomas showed significant correlations between Ag-NOR counts, LIs, and malignancy. Mean number of Ag-NORs did not correlate with proliferation indices and tumor grade in low-grade and high-grade gliomas. However, recurrent high-grade gliomas showed a tendency to higher Ag-NOR counts than primary tumors. This study indicates that counting of Ag-NORs in paraffin sections is of limited value in tumor neuropathology. Correlations found in meningeal tumors should be substantiated in larger series.

Prognostic factor analysis for multiple brain metastases after gamma knife radiosurgery: results in 97 patients.

Stereotactic radiosurgery (SR) is being used with increasing frequency in the treatment of brain metastases. This study provides data from a clinical experience with radiosurgery in the treatment of cases with multiple metastases and identifies parameters that may be useful in the proper selection and therapy of these patients. From January 1993 to April 1997, 97 patients (43 women and 54 men; median age 58 years) suffering from multiple brain metastases (median 3; range 2–4) in MRI scans, received SR with the Gamma Knife. The median dose at the tumor margin was 20 Gy (range 17–30 Gy). Median tumor volume was 3900 cmm (range 100–10 000). Different forms of hemiparesis, focal and generalized seizures, cognitive deficit, headache, dizziness and ataxia had been the predominant neurological symptoms. Major histologies included lung carcinoma (44%), breast cancer (21%), renal cell carcinoma (10%), colorectal cancer (8%), and melanoma (7%).The median survival time was 6 months after SR. The actual one-year survival rate was 26%. In univariate and multivariate analysis, a higher Karnofsky performance rating and absence of extracranial metastases had a significantly positive effect on survival. Local tumor control was achieved in 94% of the patients. Complications included the onset of peritumoral edema (n=5) and necrosis (n=1).SR induces a significant tumor remission accompanied by neurological improvement and, therefore, provides the opportunity for prolonged high quality survival. We conclude that radiosurgical treatment of multiple brain metastases leads to an equivalent rate of survival when compared to the historic experience of patients treated with whole brain radiotherapy. Patients presenting initially with a higher Karnofsky performance rating and without extracranial metastases had a median survival time of nine months. Each such case should therefore be evaluated based on these factors to determine an optimal treatment regimen.