Chemotherapy of pediatric brain-stem tumors

Cervicomedullary glioblastoma: A report of two cases with review of literature

Background

Cervicomedullary glioblastoma is an extremely rare clinical entity and the principles of its management are not well understood.

Case Description

We report two cases of cervicomedullary glioblastoma in young patients aged 12 and 30 years with contrasting clinical presentation and outcomes. The 12-year-old child had rapid onset bulbar symptoms, with frank infiltration of the medulla due to which the patient succumbed within 4 weeks of surgery. The 30-year-old adult had a relatively slow disease onset and progression and made a good neurological recovery without disease progression at 16 months after surgery. To the best of our knowledge, we also report only the second adult patient in the literature with a dorsally exophytic cervicomedullary glioblastoma. Difficulties in diagnosis and management are discussed with a review of the pertinent literature.

Conclusion

The overall outcome depends on the rapid progression and severity of preoperative symptoms and the degree of tumor infiltration noted in imaging and during surgery.

Neuroradiology of Brain Stem and Cervicomedullary tumours

Brain stem and cervicomedullary tumours are typical of paediatric age, 80% of them occurring in patients under 18 years of age, and comprising 10–15% of all childhood and adolescent brain tumours, as well as 20–25% of infratentorial locations.

They are characteristically pontine tumours (60% of the cases), but they commonly extend to involve the medulla, midbrain and cerebellum. Although most brain stem tumours are low grade gliomas, their prognosis is extremely severe (no more than 20% of patients are alive 3 years after diagnosis and the 5-year survival rate is 5%) because both the typical infiltrating nature and the neuro-anatomical location usually make them surgically unresectable. Surgery is generally limited to biopsy, partial decompression, or excission of the exophytic components, because of the extremely severe functional sequelae of even minor resections. Thus, the mainstay of therapy has been based on irradiation alone or combined with chemotherapy, doses of 5000–5500 cGy being usually adequate for tumour shrinkage or remission, even if recurrence is common after 10–15 months.

At present, it is very important to establish reliable, homogeneous, objective, and reproducible diagnostic criteria for the identification of patient subsets with predictable histology, prognosis and possible therapeutic management, in some cases histology, site and relationship of tumour enabling total or subtotal resection with a lower operative risk. Since its introduction, magnetic resonance imaging (MRI) has appeared the procedure of choice for the neuroradiologic study of the brain stem and brain stem tumours, enabling a more precise definition of their margins, a correct assessment of intrinsic and exophytic components, as well as a satisfactory characterization of pathologic tissue. MRI studies should include good quality T1-, PD- and T2-weighted images and T1-weighted images after gadolinium i.v., T2-weighted sagittal images being required for complete evaluation of tumour extent. Computed tomography is still superior in the identification of calcifications and acute intratumoral haemorrhage; it is rapidly performed, thus representing the first choice procedure in emergency, i.e. the diagnosis of hydrocephalus. Epstein has proposed the most widely accepted classification system of brain stem tumours, essentially based on neuroradiological findings, surgical and stereotactic biopsy and histology generally resulting in understaging. This classification system separates intrinsic (diffuse, focal, cervico-medullary), exophytic (anterolateral into cerebellopontine angle, posterolateral into brachium pontis, posterior into fourth ventricle) and cerebrospinal fluid seeding (positive cytology or myelography) tumors.

More recently, Barkovich – based on a multicentric study of some of the most important paediatric neurosurgery and neuro-oncology centres of the United States – has clearly defined the neuroradiological parameters which must be considered for an objective and reproducible assessment of brain stem gliomas, in order to identify patient subsets characterized by predictable histology, prognosis and possible therapeutic management. We agree with him, emphasizing that the evaluation of brain stem tumours must include a careful interpretation of all MRI findings (tumour site and origin; dimensions/degree of brain stem enlargement; tumour caudo-cranial and transverse extension; exophytic components; tumour characteristics as defined by MRI signal intensity; cysts, haemorrhage, necrosis, calcifications; ventricular dimensions and hydrocephalus; leptomeninengeal seeding) that can help in the definition of the following tumour subsets: diffuse pontine tumours, medullary tumours, cervicomedullary tumours, focal brain stem tumours.

With regard to the neuroradiological follow-up, in patients undergoing surgery (for biopsy decompression or less frequently for radicality), the role of the neuroradiologist is similar to that in other fields of neurosurgery, and concerns the quantitation of the extent of the resection and the identification of possible parenchymal injuries or postoperative haemorrhage, always keeping in mind the negative effects of postoperative reactive phenomena and the blood-brain disruption 24–48 hours to 30–40 days after surgery. However, the major contribution of the neuroradiologist is the objective evaluation of irradiation effects and recently of combined irradiation and chemotherapy. In our experience, the efficacy of irradiation is well evaluated by MRI only 3 or 6 months after the end of treatment, even if a clinical improvement is possible after 30–60 days. However, tumour shrinkage is rarely drastic, and its disappearance, as well as that of signal alterations, is exceptional. Qualitative modifications occurring within the tumour are more difficult to interpret, because of the appearance of cystic or pseudocystic areas, markedly increased enhancement, and small areas of haemorrhage may be related both to irradiation-induced regressive modifications and disease progression.

In conclusion, MRI represents the gold standard in the evaluation of brain stem and cervicomedullary tumors, always enabling a precise definition of tumour site and extent, and in most cases the diagnosis of nature, thus allowing the identification of patients who can undergo radical microsurgery. MRI follow-up controls the extent of resection and the effect of combined irradiation and chemotherapy; disease progression is evidenced and spinal seeding can be diagnosed.

However, the differentiation between tumour recurrence and irradiation-induced injury may be difficult if only based on morphological data. These limitations of MRI will probably be reduced by the advances in ultra-fast MRI technology and 18F-fluoro-deoxy-glucose positron emission tomography which supplies in vivo metabolic and functional information.

Clinico-radiological profile and nuances in the management of cervicomedullary junction intramedullary tumors

BACKGROUND

Cervicomedullary junction (CMJ) intramedullary tumors comprise of tumors that often pose a surgical challenge even in the present era. Though classified under brainstem glioma CMJ tumors are well amenable for surgical resection and have a good outcome. Various factors are involved in the outcome of these patients following surgery and a proper pre-operative assessment is often required to reduce the morbidity and mortality.

MATERIALS AND METHODS

Patients admitted in the Department of Neurosurgery with a diagnosis of CMJ intramedullary tumors from January 2001 to January 2010 were included in the study. Patients were analyzed retrospectively regarding their symptomatology, clinical findings, radiology and outcome after surgery. All patients underwent pre-operative magnetic resonance imaging (MRI) and post-operatively all were managed in the neurosurgery intensive care unit for days to weeks or as dictated by the clinical condition of the patient.

RESULTS

A total of 32 patients were included in the present study. The number of males was 21 (65.6%) and females were 11 (34.4%) respectively. The mean age of presentation was 22.97 ± 9.8 years and the mean duration of pre-operative symptoms was 13.3 ± 12.9 months. The tumor had extension from the CMJ into the cervical region in 17 (53.1%) and into the medullary region in 14 (43.8%) patients. Tumor decompression was done in 9 (28.1%) patients and gross near total excision done in 23 (71.87%) patients.

CONCLUSIONS

Cervicomedullary tumors are a subset of tumors quite distinct from the usual brainstem tumors. Patients having predominant cervical involvement present early and have less post-operative deficits. Those with predominantly more medullary involvement present late, hence have a much more morbid outcome. Though closely related to vital neural structures, surgery forms the mainstay of treatment. Adequate pre-operative planning and preparation of the patient along with intense post-operative monitoring and ventilatory assistance as and when required helps in a good surgical outcome.

Magnetic resonance imaging compared with biopsy in the diagnosis of brainstem diseases of childhood: a multicenter review

OBJECT

Data analysis was performed in a multicenter study to evaluate magnetic resonance (MR) imaging for classification of brain tumors, prognosis, and prediction of tumor histological diagnosis.

METHODS

The clinical, MR imaging, and histological findings obtained in 142 pediatric cases of brainstem disease were assessed in a multicenter study performed as a blinded review. Clinical data were available in 142 cases, histopathological findings in 126, and MR images in 131. The subgroup of cases involving brainstem gliomas (78 cases) was analyzed separately. Images that met criteria for evaluation were reviewed in random order by three experienced observers who were initially blinded to clinical data as well as histopathological diagnoses. Subsequently, the images were randomized again and provided to the observers for review together with the clinical symptoms of the specific patients. The three observers were able to correctly identify lesions as tumors or nontumorous disease on MR images in 99, 96, and 95% of cases, resulting in an overall sensitivity of 0.94, a specificity of 0.43, a positive predictive value of 0.96, and a negative predictive value of 0.45. Awareness of clinical symptoms did not change the results.

CONCLUSIONS

Based on 14 imaging criteria together with the patient's clinical history and symptoms, laboratory data (results of cerebrospinal fluid analysis as well as infectious and immunological parameters), and imaging follow up, a diagnosis of brainstem tumor, as opposed to demyelination, encephalitis, or granuloma, could generally be made. Given these findings, there is only rarely a need for biopsy, and in those patients in whom it is considered, the potential costs and benefits must be carefully assessed on a case-by-case basis.

Biopsy of brain stem glioma using motor-evoked potential mapping by direct peduncular stimulation and individual adjuvant therapy. Case report

A 23-year-old man presented with a brain stem glioma manifesting as a 6-month history of right hemiparesis and diplopia. Serial magnetic resonance imaging showed an intrinsic diffuse brain stem glioma that gradually localized to the left cerebral peduncle after initial adjuvant therapy. Surgery was performed through a left subtemporal transtentorial approach under motor-evoked potential (MEP) mapping by direct peduncular stimulation. The lateral aspect of the midbrain was exposed, a train of five bipolar 25 mA pulses was applied, and MEPs recorded from the extremities. MEPs were only recorded from the left extremities even with left cerebral peduncular stimulation. Partial resection of the tumor was safely performed, with slight temporary neurological worsening. The histological diagnosis was anaplastic astrocytoma. Individual adjuvant therapy based on the results of real-time reverse transcription-polymerase chain reaction of O6-methylguanine-deoxyribonucleic acid methyltransferase achieved an almost complete tumor response. Surgery under pyramidal tract mapping and intensive postoperative adjuvant therapy resulted in a good outcome despite the presence of a generally intractable intrinsic brain stem glioma.

Phase II study of antineoplaston A10 and AS2-1 in patients with recurrent diffuse intrinsic brain stem glioma: a preliminary report

OBJECTIVE

A phase II study of antineoplaston A10 and AS2-1 was conducted to evaluate the antineoplastic activity in patients with recurrent diffuse intrinsic brain stem glioma.

PATIENTS AND METHODS

This report describes the results of treatment of the first 12 patients admitted to the study. Patients received escalating doses of antineoplaston A10 and AS2-1 by intravenous bolus injections. The median duration of treatment was 6 months and the average dosage of antineoplaston A10 was 11.3 g/kg/day and of antineoplaston AS2-1 0.4 g/kg/day. Responses were assessed by gadolinium-enhanced magnetic resonance imaging of the head.

RESULTS

Of ten evaluable patients, complete response was determined in two cases (20%), partial response in three (30%), stable disease in three (30%) and progressive disease in two (20%). Survival at 2 years was 33.3%. Currently, of all 12 patients, two (17%) were alive and tumour free for over 5 years since initial diagnosis; one was alive for more than 5 years, and another for more than 4 years from the start of treatment. Only mild and moderate toxicities were observed, which included three cases of skin allergy, two cases of anaemia, fever and hypernatraemia, and single cases of agranulocytosis, hypoglycaemia, numbness, tiredness, myalgia and vomiting.

CONCLUSION

The results of this study compared favourably with the responses of patients treated with radiation therapy and chemotherapy. The study continues with accrual of additional patients.

Prolonged convection-enhanced delivery into the rat brainstem

OBJECTIVE

Prolonged convection-enhanced delivery was used in an attempt to achieve large volumes of distribution (V(d)) in the rat brainstem. Clinical assessment and histological analysis were performed to establish the safety of this approach.

METHODS

For evaluation of V(d,), 10 rats underwent stereotactic cannula placement into the brainstem. Five rats underwent a 24-hour infusion (volume of infusion [V(i)], 200 microl), and 5 rats underwent a 7-day infusion (V(i), 2 ml) of fluorescein isothiocyanate-dextran. Serial brainstem sections were imaged with ultraviolet illumination, and V(d) was assessed. For assessment of clinical tolerance, 30 additional rats underwent chronic infusions of an isotonic saline solution into the brainstem. Serial neurological examinations were performed, followed by histological analysis after the animals' death.

RESULTS

No animal demonstrated clinically recognized neurological deficits. Foci of organizing necrosis were limited to the site of infusion and cannula tract. V(d) increased linearly with increasing V(i) (range, 24.8-130.6 mm(3)). Maximal cross sectional area of fluorescence and craniocaudal extent of fluorescence increased with increasing V(i). Fluorescence was detected throughout the entire brainstem beyond the compact area of highly concentrated tracer.

CONCLUSION

Prolonged convection-enhanced delivery can be applied safely in the rat brainstem with no recognized limitations of V(d) and minimal histological changes beyond the site of infusion. Chronic brainstem infusions may enhance the potential application of convection-enhanced delivery for therapeutic purposes in treating diffuse pontine gliomas.

Effect of hyperosmolar mannitol on convection-enhanced delivery into the rat brain stem

INTRODUCTION

Convection-enhanced delivery (CED) can safely achieve high local infusate concentrations within the rat brain stem with predictable distribution volumes. The authors investigated the effects of co-infusion or systemic administration of hyperosmolar mannitol on distribution parameters for infusions into the rat brain stem.

METHODS

Fifteen rats underwent stereotactic cannula placement into the pontine nucleus oralis (PnO) followed by infusions at a constant rate to a total volume of 1 microl. Five rats underwent infusion of fluorescein isothiocyanate (FITC)-dextran diluted in 20% mannitol. Five rats received an intraperitoneal injection of 20% mannitol 10 min prior to infusion of FITC-dextran diluted in isotonic saline. As a control group, 5 rats underwent infusion of FITC-dextran diluted in isotonic saline without mannitol administration. Serial brain sections were imaged using confocal microscopy with ultraviolet illumination, and distribution volume (Vd) was calculated by computer image analysis. Histologic analysis was performed on adjacent sections.

RESULTS

Volumes of distribution were not significantly increased by co-infusion of mannitol directly into the brain stem or by systemic mannitol administration compared to infusion without mannitol. Similarly, mannitol administration by either means failed to significantly alter maximal cross-sectional area or cranio-caudal extent of fluorescence. No animal demonstrated a postoperative neurological deficit or histologic evidence of tissue disruption.

CONCLUSIONS

Neither systemic administration nor co-infusion of hyperosmolar mannitol significantly affects distribution parameters for CED infusions into the rat brain stem.

Convection-enhanced delivery into the rat brainstem

OBJECT

Convection-enhanced delivery (CED) can be used safely to achieve high local infusate concentrations within the brain and spinal cord. The use of CED in the brainstem has not been previously reported and may offer an alternative method for treating diffuse pontine gliomas. In the present study the authors tested CED within the rat brainstem to assess its safety and establish distribution parameters.

METHODS

Eighteen rats underwent stereotactic cannula placement into the pontine nucleus oralis without subsequent infusions. Twenty rats underwent stereotactic cannula placement followed by infusion of fluorescein isothiocyanate (FITC)-dextran at a constant rate (0.1 microl/minute) until various total volumes of infusion (V(i)s) were reached: 0.5, 1, 2, and 4 microl. Additional rats underwent FITC-dextran infusion (V, 4 microl) and were observed for 48 hours (five animals) or 14 days (five animals). Serial (20-microm thick) brain sections were imaged using confocal microscopy with ultraviolet illumination, and the volume of distribution (Vd) was calculated using computer image analysis. Histological analysis was performed on adjacent sections. No animal exhibited a postoperative neurological deficit, and there was no histological evidence of tissue disruption. The Vd increased linearly (range 15.4-55.8 mm3) along with increasing Vi, with statistically significant correlations for all groups that were compared (p < 0.022). The Va/Vi ratio ranged from 14 to 30.9. The maximum cross-sectional area of fluorescence (range 9.8-20.9 mm2) and the craniocaudal extent of fluorescence (range 2.8-5.1 mm) increased with increasing Vi.

CONCLUSIONS

Convection-enhanced delivery can be safely applied to the rat brainstem with substantial and predictable V(d)s. This study provides the basis for investigating delivery of various candidate agents for the treatment of diffuse pontine gliomas.

Radiotherapy followed by high dose busulfan and thiotepa: a prospective assessment of high dose chemotherapy in children with diffuse pontine gliomas

BACKGROUND

The role of high dose chemotherapy (HDC) in patients with pediatric brain tumors currently is ill-defined. The purpose of this pilot study was to assess the feasibility and the benefit of HDC after radiotherapy in a group of children with newly diagnosed diffuse pontine gliomas.

METHODS

Patients eligible for study were ages 3-18 years with diffuse intrinsic tumors arising in the pons, who were not treated previously with radiotherapy or chemotherapy. Histologic confirmation was not mandatory, provided clinical findings and magnetic resonance imaging were typical. Patients were given focal radiotherapy followed 2-3 months later by HDC. Busulfan (150 mg/m(2) on Days 8, 7, 6, and 5) and thiotepa (300 mg/m(2) on Days 4, 3, and 2) were administered prior to autologous bone marrow transplantation. Survival was the endpoint, and the statistical procedure was based on sequential subgroup analysis.

RESULTS

Thirty-six patients were entered on to the study, 12 of whom underwent stereotactic biopsy or open surgery at the time of diagnosis. One patient eventually was excluded due to inappropriate eligibility criteria. All 35 eligible patients received irradiation. Early progression (9 patients) and parental refusal (2 patients) precluded the use of HDC in 11 patients. Three patients died of HDC-related complications. All 21 patients who survived HDC eventually died of disease progression. The median survival time was 10 months for the study group. The median survival time in the subgroup of patients who received HDC was 10 months (range, 3-26 months). Statistical analysis did not suggest any evidence of survival benefit.

CONCLUSIONS

For patients with diffuse pontine gliomas, survival using this aggressive treatment modality does not appear to be any better than that reported for conventional radiotherapy.

High dose chemotherapy with autologous bone marrow rescue for children with diffuse pontine brain stem tumors. Children's Cancer Group

PURPOSE

Diffuse pontine tumors are highly lethal, and there are few long-term survivors with the standard treatment of external beam irradiation. We investigated the effectiveness of high-dose thiotepa and etoposide-based chemotherapy regimens with autologous bone marrow rescue (ABMR) in children with pontine tumors.

PATIENTS AND METHODS

Sixteen children with diffuse pontine tumors were treated. Ten had resistant or recurrent tumors. All ten had previously received irradiation; five had also received chemotherapy and one, beta-interferon. Three high-dose chemotherapy regimens were employed. Six patients received three days of thiotepa (300 mg/m2/day) and etoposide (250-500 mg/m2/day) (TE); two received three days of carmustine (BCNU) (200 mg/m2/day divided every 12 hours) followed by TE (BTE); and two received three days of carboplatin (500 mg/m2/day) followed by TE (CTE). Six other patients had newly-diagnosed tumors and had not received any prior treatment. They all received the BTE regimen and subsequently were treated with hyperfractionated irradiation (7200-7800 cGy) beginning approximately six weeks post-ABMR.

RESULTS

There were two toxic deaths (13%), both in previously treated patients, due to multiorgan system failure and Candida septicemia in one case each. Median survival of the patients with resistant or recurrent disease was 4.7 months (range 0.1-18.7) from time of ABMR. Median survival of the newly-diagnosed patients was 11.4 months (range 7.6-17.1) from the time of ABMR.

CONCLUSION

High-dose chemotherapy utilizing these regimens followed by ABMR did not appear to prolong survival compared to conventional therapy in these children with pontine tumors. Alternative strategies need to be developed for this highly lethal disease.

Intra-arterial chemotherapy for brain stem glioma: report of four cases

Four patients with pontine gliomas were treated with radiation and intra-arterial chemotherapy using ACNU, cisplatin, and/or carboplatin. All underwent biopsy and were histologically proven to have gliomas. Even though three were benign histologically, all appeared malignant on neuroimaging. Two patients had complete remissions; the others had partial responses; three patients died, 36, 14, and 15 months after the primary diagnosis. One patient, now in complete remission, has been alive for 24 months. There was no neurotoxicity except for transient nausea and vomiting. Intra-arterial chemotherapy is effective against brain stem gliomas, producing remission with minimum side effects.

Surgical management of brain-stem tumors in children: results and statistical analysis of 75 cases

A study was made of 75 children treated between 1970 and 1990, with partial, subtotal, or total removal of three intrinsic and 72 exophytic or surface brain-stem tumors. In all cases, the goal of surgery was to remove as much tumor as possible. Extent of removal was defined according to data obtained from postoperative computerized tomography or magnetic resonance imaging, and was considered partial when only a small amount of tumor was removed, subtotal when a few cubic millimeters of tumor was left, and total when no residual tumor was seen on postoperative radiological investigations. An ultrasonic aspirator was used for the 43 most recent operations. Among tumor removals without the aspirator, 24 (75%) were partial, eight (25%) subtotal, and none total; with the use of the aspirator, the number of partial removals decreased to 44.5% while that of subtotal and total removals increased to 32% and 23.5%, respectively. There were 69 gliomas (92%) and 47 benign tumors (62.6%). Forty-nine patients were irradiated postoperatively, and 14 of the 23 patients whose benign tumors were removed totally or subtotally did not undergo irradiation. This study showed that: 1) the overall prognosis of patients with malignant tumors was poor and was not improved by surgery; 2) the survival rate of those with benign tumors was significantly (p < 0.01) lower after partial removal than after total or subtotal removal (52% and 94%, respectively, at 5 years); 3) comparison of means and proportions (Student's and chi-squared tests) between benign and malignant tumors showed a significant difference relating to patient age (p < 0.03), peritumoral hypodensity (p < 0.001), and preoperative duration of symptoms (p < 0.001); 4) stepwise logistic regression analysis confirmed that two of these three variables were related to malignancy: namely, patient age at surgery (p < 0.03) and presence of peritumoral hypodensity (p < 0.001); and 5) routine postoperative irradiation was contraindicated after total or subtotal removal of benign tumors.

Pre-irradiation chemotherapy and hyperfractionated radiation therapy 66 Gy for children with brain stem tumors. A phase II study of the Pediatric Oncology Group, Protocol 8833

BACKGROUND

Fewer than 20% of children with intrinsic brain stem tumors survive longer than 2 years. Although some improvement has been noted in recent trials using higher doses of hyperfractionated radiation therapy (HRT), the feasibility of pre-irradiation chemotherapy has not been explored in these patients with poor prognosis.

METHODS

Between February 1988 and March 1989, 37 patients were entered onto a Phase II Pediatric Oncology Group study for evaluating the feasibility, response, and toxicity of treating children with high-risk brain stem tumors with chemotherapy followed by HRT (66 Gy). Chemotherapy consisted of four cycles of cisplatin (100 mg/m2) plus cyclophosphamide (3 g/m2).

RESULTS

Of 32 eligible patients, 65% improved clinically during the first 2-3 cycles of chemotherapy; 75% of those improving were weaned from steroids. On neuroradiology review of scans before and after chemotherapy, 3 patients had partial responses (PR, > 50% shrinkage), 23 had stable disease (SD), and 6 had progressive disease (PD). The median survival was 9 months. The three patients who attained a PR on chemotherapy were among the longest survivors at 38 plus, 44 plus, and 40 months. Toxicities included profound but brief marrow suppression, transient electrolyte-renal dysfunction, and ototoxicity. Brain stem swelling from intravenous fluids caused transient deterioration in two patients. Six patients developed an unusual syndrome of transient marrow suppression after HRT.

CONCLUSIONS

This study suggests that pre-irradiation chemotherapy can be successfully added to the treatment of patients with brain stem tumors with both clinical and objective responses noted, but that other agents must be identified to overcome the apparent development of drug resistance and to improve survival.

Brain-stem glioma growth patterns

During the last decade, several authors have reported that certain brain-stem gliomas may be associated with a better prognosis than others. In this paper, retrospective correlations between the pathological findings and the magnetic resonance (MR) imaging appearance of 88 brain-stem gliomas are established. The authors propose an anatomical hypothesis that helps identify glioma growth patterns in general and that clarifies why cervicomedullary, dorsally exophytic, and focal tumors have a more favorable prognosis. According to this hypothesis, growth of benign gliomas of the brain stem is guided by secondary structures such as the pia, fiber tracts, and the ependyma, which in turn leads to stereotypical growth patterns that are clearly identified on MR images. The authors believe that this hypothesis, in conjunction with clinical data, may help establish selection criteria for the surgical treatment of specific brain-stem lesions.

Recurrent brainstem gliomas treated with oral VP-16

12 patients: (7 males and 5 females) with recurrent brainstem gliomas were treated with the oral topoisomerase inhibitor VP-16 (Etoposide). Patients ranged in age from 3 to 49 years with a median age of 7 years. All patients had been previously treated with radiation therapy (conventional fractionation: 4; hyperfractionation: 8) and 5 had received prior nitrosourea-based chemotherapy at time of tumor recurrence. Tumor recurrence was documented by radiographic tumor enlargement utilizing brain MRI with gadolinium enhancement (12) and clinical neurologic deterioration (9). Two patients underwent biopsy pathologically documenting tumor recurrence. Each cycle of therapy consisted of 21 days of VP-16 (50 mg/m2/day) followed by a 14 day rest followed by an additional 21 days of VP-16 (50 mg/m2 day). Complete blood counts were followed bi-weekly and a neurologic examination and brain MRI scan with contrast were performed prior to initiation of each cycle of therapy. Treatment related complications included: partial alopecia (5); diarrhea (5); weight loss (4); neutropenia (2); and thrombocytopenia (4). No patient required transfusion or antibiotic treatment of neutropenic fever. There were no treatment related deaths. 12 patients were evaluable of whom 6 demonstrated a radiographic response (1 complete; 3 partial; 2 stable disease) with a median duration of response of 8 months. In summary; oral VP-16 is a well tolerated and relatively non-toxic chemotherapeutic agent with apparent activity in this small cohort of patients with recurrent brainstem gliomas.

Contemporary management of adult and pediatric brain stem gliomas

The aim of this study was to analyze the survival of a series of patients diagnosed as having brain stem tumors by computerized tomography scan, magnetic resonance imaging and/or biopsy and treated with megavoltage irradiation. Fifty-three patients presenting to the Peter MacCallum Cancer Institute with a diagnosis of brain stem tumor from January 1980 to July 1989 were reviewed. There were 32 pediatric (age < or = 16) and 21 adult patients. The median age at presentation was 12 years (range 2-73 years). Eighteen patients had biopsy proof of glioma, the rest were diagnosed on the basis of CT and/or MRI appearance. Eighty-seven percent of patients received 44-55 Gy in 1.67-2.25 Gy fractions to the brain stem. Seventy-seven percent of patients showed improvement following treatment, 6% were stable, 11% progressed, and 6% were not evaluable. The estimated median survival from presentation for all patients was 34 months with 49% surviving at 3 years. The estimated median time to progression was 19 months with 34% not progressed at 3 years. There was no significant survival difference between patients with biopsy-proven glioma and those patients where the diagnosis was made radiologically. Survival was not significantly affected by age-group or sex. Patients with symptoms for more than 9 months prior to presentation had significantly longer survival than those with shorter duration of symptoms (p = 0.002). This paper presents the survival of patients with brain stem tumors diagnosed and treated by contemporary radiological and radiotherapeutic techniques.

The role of image-directed biopsy in the diagnosis and management of brainstem lesions

The appreciation that brainstem tumours do not comprise a homogeneous pathological group and that up to 20% of the preoperative radiological diagnoses of brainstem lesions prove at operation to be wrong, has established the need for a firm histological diagnosis prior to treatment. Current neuro-imaging modalities may have increased the diagnostic accuracy and the detection rate of intrinsic brainstem lesions, but open exploration in cases without an exophytic component is still associated with a low diagnostic yield and considerable morbidity. A series of 72 brainstem lesions approached stereotactically with CT or MRI guidance is presented. A transcortical frontal precoronal trajectory was used in 58 and a suboccipital transcerebellar route in 14. Haematoma was diagnosed preoperatively in 16 cases and therapeutic aspiration was planned. In 56 cases the diagnosis was uncertain, although intrinsic tumour was suspected. A histological diagnosis was established in 52 cases, although in the remaining four cases a tumour was excluded. Unexpected findings occurred in over 15% of the cases. There were no operative deaths and the morbidity was low. In no case was there a permanent neurological deterioration directly related to the procedure, although there was a transient deterioration in two patients and another patient required early re-aspiration of a haematoma.

Image directed stereotactic surgery for brain stem lesions

Advances in neurological imaging may have increased the diagnostic accuracy and the detection rate of intrinsic brain stem lesions, but a histological diagnosis is still an essential requirement for rational and appropriate management. Open exploration allows biopsy and resection in cases where an exophytic component is present. The surgical inaccessibility and the resultant morbidity of these approaches, however, associated with a low diagnostic yield in cases with no visible surface abnormality, are important limiting factors. A series of 45 brain stem lesions stereotactically approached with CT or MRI guidance is presented. A transcortical frontal pre-coronal trajectory was used in all of them. Haematoma was preoperatively diagnosed in 10 cases and the procedure was for therapeutic aspiration. Of 35 cases where the diagnosis was uncertain, although intrinsic tumour was suspected, positive results were obtained in 33, while unexpected findings of granuloma, lymphoma, angioma, leucoencephalopathy, vasculitis and radiation necrosis were found in over 10% of the cases. There were no operative deaths and the morbidity was low. In no case was there a permanent neurological deterioration directly related to the procedure, although there was a transient deterioration in two patients and one patient required early reaspiration of a haematoma. Image directed stereotactic approaches to brain stem lesions can combine a high degree of accuracy (offering positive histological diagnoses) with a low operative morbidity. MRI directed biopsies can complement CT guided ones thus increasing the number of suitable cases and improving the success rate. The frontal precoronal transcortical trajectory provides safe access to the majority of the brain stem targets.

Prolonged response to carboplatin in an infant with brain stem glioma

Adults and children with brain stem gliomas have a mean survival time of 15 months after radiation therapy (XRT). Infants with this tumor present additional complexities for treatment because of possible neurotoxicity of the radiation to the developing brain. We report a 15-month-old child with biopsy-proven brain stem glioma with clinical and radiographic evidence of disease progression. She was treated with 24 monthly courses of carboplatin without radiation therapy and has had a 39+ month response. The clinical response started after 3 months and the radiographic evidence was documented at 10 months by magnetic resonance imaging. The toxicity was minimal. Longitudinal neuropsychological assessment demonstrated continued improvement at 36 months post diagnosis but with some motor functioning below expected age levels. Cervico-medullary astrocytoma in a young patient may be the appropriate clinical setting for future trials of chemotherapy without XRT.

Stereotaxic biopsy of brain-stem tumours in children

Stereotaxic biopsies were performed in a series of 14 patients with clinical and computed tomographic evidence of a brain-stem lesion. Seven of these cases were in the paediatric age group (under 15 years of age). Biopsies were obtained by means of the Leksell stereotaxic system, through the transcerebellar (three cases) or transfrontal route (four cases). A definite tumour diagnosis could be obtained in all cases. No patients were subjected to open biopsy, as the further management of their tumours did not warrant surgery. One patient developed transient oculomotor nerve palsy. There was no mortality related to the biopsy procedure in this series. The biopsy technique and its utility and safety are discussed.

Radiation therapy for brain stem tumor with special reference to CT feature and prognosis correlations

Seventy-nine patients with tumors of the medulla oblongata, pons, and midbrain treated by radiation therapy between 1962 and 1987 were analyzed. According to histology obtained in 28 cases, 61% were high-grade astrocytomas and 39% were low-grade astrocytomas. The 5- and 10-year survival rates of 71 patients who completed radiotherapy were 17 and 15%, respectively. Various potential prognostic factors were analyzed, and patients with a midbrain tumor, with a low-grade tumor, and without cranial nerve paresis were found to have a better prognosis. Computed tomography (CT) was performed both before and after radiotherapy on 42 patients. The volume of the tumor estimated from the CT image was various, but the tumors could be classified into three types according to the pattern of contrast enhancement (CE). Twenty-three tumors showed no CE, 15 showed a diffuse CE, and 4 showed a ring CE. No-CE tumors were generally hypodense on plain scan and were mostly large, whereas diffuse- or ring-CE tumors were relatively small. A complete response (CR) or partial response was obtained with 77% of the no-CE tumors, 64% of the diffuse-CE tumors, and 50% of the ring-CE tumors. The CR rate was higher in the tumors with a smaller volume. The 5-year survival rates for diffuse-CE, no-CE, and ring-CE tumors were 33, 15, and 0%, respectively. The 5-year survival rate was 56% for tumors smaller than 6 cm3, 16% for tumors between 6 and 20 cm3, and 0% for tumors larger than 20 cm3. In conclusion, (a) ring-CE tumors exhibited a poor prognosis; and (b) a diffuse-CE tumor, or a small or middle-sized no-CE tumor is most likely to be controlled by conventional radiation therapy.

Prognostic factors and results of therapy for adult thalamic and brainstem tumors

This report is a retrospective analysis of 83 adults (greater than 16 years of age) with histologically proven or presumed primary neoplasms of the thalamus, hypothalamus, midbrain, pons, and medulla. Patients were treated with combined surgery and postoperative irradiation or with irradiation alone at the Washington University Medical Center (St. Louis, MO) from January 1950 through December 1984. Histologic analysis confirmed the diagnosis of tumor in 21, including nine with well-differentiated astrocytoma, four with astrocytoma with anaplasia, and eight with glioblastoma multiforme. Overall and disease-free survivals at 5 years were 28.7 and 23.2%, respectively. A statistical analysis was performed to ascertain the prognostic importance of the following variables: age, race, gender, duration of symptoms, cranial nerve paresis, primary site, extent of surgery, histology, and irradiation dose. The only factor identified by univariate analysis to be critical for survival was primary location of disease. Patients with supratentorial (thalamus/hypothalamus, midbrain) tumors had a 10-year disease-free survival of 15.4% compared to 29.6% for those with infratentorial (pons, medulla) tumors (P = 0.07). Patients with lesions of the pons had a 5-year disease-free survival of 35.8% compared to 13.8% for those with tumors of the thalamus (P = 0.05). Increasing irradiation dose was not correlated with superior survival. Factors evaluated but established to be insignificant were age (P = 0.27), race (P = 0.63), gender (P = 0.27), duration of symptoms (P = 0.19), cranial nerve paresis (P = 0.71), histologic type (P = 0.16), and extent of surgery (P = 0.94). Follow-up for 13 surviving patients ranged from 2.6 to 28.7 (mean, 12.0) years. Neurologic deficits in surviving patients were absent in 15% (two of 13), mild in 62% (eight of 23), and moderate in 23% (three of 13). One case of brain radionecrosis was identified (6000 cGy, 200 cGy daily).

Posterior fossa stereotaxic biopsy using the Brown-Roberts-Wells stereotaxic system. Technical note

Stereotaxic biopsy has been shown to be a reliable means of diagnosing posterior fossa lesions. The authors describe a technique for infratentorial transcerebellar stereotaxic access to posterior fossa parenchymal lesions using the Brown-Roberts-Wells apparatus in its standard commercial configuration. The necessity for tissue diagnosis of these lesions is briefly discussed.

Multivariate analysis of prognostic factors in pediatric and adult thalamic and brainstem tumors

A multivariate analysis of prognostic variables was performed on a retrospective review of 136 patients with presumed or histologically proven primary lesions of the thalamus and brainstem treated by combined surgery and post-operative irradiation or by irradiation alone from January 1950 through December 1983. Overall survival for all patients at 5 and 10 years was 34.4 and 27.8%, respectively. Follow-up of 33 living patients ranged from 3 to 22 years. Prognostic variables analyzed by univariate analysis and found to be of significance (p less than 0.05) were race, duration of symptoms, extent of surgery (i.e. subtotal excision), and dose of irradiation. Further evaluation by Cox regression analysis revealed these same factors to be of prognostic significance (p less than 0.05). It is of importance to note that age and tumor site were not significant prognostic variables in the multivariate analysis but were significant by univariate analysis. The 5-year overall survival for patients with thalamic tumors was 59.5 and 20.9% for children and adults, respectively (p = 0.006). The 5-year overall survival for patients with pontine lesions was 46.6 and 16.0% for adults and children, respectively (p = 0.01). Only one patient was known to have expired due to a complication of therapy. Neurologic deficits and functional ability was normal or mild in 57.6% of the surviving patients.

Brain stem gliomas: a clinicopathological analysis of 23 histologically proven cases

The clinical and anatomic pathological findings in 23 patients with histologically proven brain stem glioma were reviewed. Eleven patients had malignant tumors (astrocytoma, grade III, and glioblastoma multiforme) and the remaining had low grade ones. The primary site of the tumor was the pons in 15 of 21 patients examined, followed by the medulla oblongata and midbrain. Continuous cephalad and caudad involvement of pontine tumors was evident in 14 patients, and was more extensive in the high grade gliomas. At autopsy, the tumor spread via the CSF was observed in all patients with high grade but not low grade tumors. Open surgical posterior fossa exploration was performed on 12 patients. Volume reduction, by partial removal of the tumor and/or by cyst evacuation, was performed on seven, with good results. Nineteen patients were given radiation therapy, with temporary improvement of the clinical conditions. While none of the patients with malignant tumors survived more than 15 months after diagnosis, those with low grade tumors had a five-year actuarial survival rate of 50.0%, and three survived more than ten years after the diagnosis. Our data indicate that brain stem gliomas are not a homogeneous group of tumors as far as the clinical and pathological features are concerned. This heterogeneity shows the need to design specific treatments for these tumors.

Prognostic criteria for experimental protocols in pediatric brainstem gliomas

Prognostic criteria were developed to select children with high-risk brainstem gliomas for experimental treatment with hyperfractionated radiation. Patients were required to have (a) tumors confined primarily to the brainstem; (b) positive findings in two of three groups of neurologic signs (cranial nerve deficits and long-tract and cerebellar signs), and (c) symptom duration of less than 6 months. The 18-month survival of only 1 of 33 evaluable patients suggests that these criteria did in fact select tumors of poor prognosis. Such noninvasive predictors provide a valuable means of selecting for experimental treatment those children who would not be expected to benefit from standard therapies for malignant brainstem tumors. These criteria do not require routine biopsy for prognostic purposes, and avoid exposing the children to unnecessary risk.

Brain stem expansive lesions: stereotactic biopsy for a better therapeutic approach

CT diagnosed brain stem malignant lesions were in the past almost always treated with radiation therapy (RT). Eventually this turned out to be a grave mistake. With stereotactic serial biopsies of all brain stem expanding lesions we have been able to verify the histological nature in all cases but two and to prevent a blind therapeutic approach. In 24 patients bearing CT diagnosed expansive lesions into the brain stem 68 samples were taken during 24 stereotactic procedures. In 8 patients surgical removal of the expanding lesion was attempted after stereotactic biopsy.

Irradiation of primary thalamic and brainstem tumors in a pediatric population. A 33-year experience

A retrospective analysis of 70 pediatric patients (less than 16 years of age) with histologically proven or presumed primary neoplasms of the thalamus, hypothalamus, and brainstem, treated with combined surgery and postoperative radiotherapy or radiotherapy alone at the Washington University Medical Center from January 1950 through December 1983, is reported. Overall survival for all patients at 5 and 10 years was 34.9% and 32.7%, respectively. Follow-up of the 22 surviving patients ranged from 3.0 to 20.0 years (median, 10.6 years). Statistical analysis of multiple prognostic factors was performed. Prognostic factors found by single variate analysis to significantly influence survival were primary site of disease, extent of surgery, race, cranial nerve paresis at diagnosis, and dose of radiation. Factors evaluated but found to be insignificant were age at diagnosis, duration of symptoms before diagnosis, sex, and volume irradiated. Multivariate analysis revealed that only total radiation dose and race were of prognostic significance.

Cystic schwannoma of the trochlear nerve mimicking a brain-stem tumor. Case report

A case is reported of a rare cystic schwannoma of the fourth cranial nerve which was interpreted as a probable intrinsic brain-stem lesion. The clinical approach to brain-stem tumors in terms of empirical treatment or surgical biopsy is discussed.

Brain-stem tumors in childhood: a prospective randomized trial of irradiation with and without adjuvant CCNU, VCR, and prednisone. A report of the Childrens Cancer Study Group

Seventy-four children with a brain-stem tumor diagnosed between 1977 and 1980 were entered into a prospective study in which exploration and assessment for resection were optional, radiation treatment using standard methods was required, and randomization occurred with regard to the use of adjuvant chemotherapy (1-(2-chloroethyl)-1-nitrosourea, vincristine, and prednisone) or no further treatment. The overall 5-year survival rate was 20% and was not improved by the adjuvant chemotherapy program. An increased risk of infection was associated with the adjuvant therapy.

Diagnosis and management of pediatric brain-stem gliomas

The authors reviewed the cases of 49 children, ranging in age from 9 months to 15 years, who were diagnosed by computerized tomography (CT) as having brain-stem glioma. Four distinct groups of brain-stem gliomas were identified based on CT scan characteristics: Group I included isodense contrast-enhancing tumors that were dorsally exophytic into the fourth ventricle; Group II(a) included hypodense nonenhancing intrinsic tumors of the brain stem; Group II(b) included intrinsic tumors of the brain stem with hyperdense exophytic components extending ventrally and laterally into the cerebellopontine and prepontine cisterns; Group III included intrinsic cystic tumors with contrast-enhancing capsules; and Group IV included focally intrinsic tumors of the brain stem that were isodense and enhanced brightly on administration of contrast medium. The clinical presentation, efficacy of surgical intervention, pathology, and prognosis of these tumors were correlated within these groupings. Eleven patients had Group I tumors, all of which were surgically resected; 10 of the 11 lesions were proven to be low-grade gliomas. These patients had an excellent prognosis; 10 of the 11 survived, with a mean follow-up period of 4.5 years. There were 18 patients with Group II(a) tumors; although tumor biopsy was attempted on eight of these, pathological diagnosis at the time of surgery was made in only one case. These patients did poorly; the mean survival time was 6.2 months. The seven Group II(b) tumor patients demonstrated a similarly poor prognosis: all of them died within 23 months of diagnosis, with a mean survival time of 12 months. Only two of six patients undergoing biopsy had sufficient tissue for histological verification. Three of the four patients with Group III tumors died; their mean survival time was 11.5 months. Successful histological examination was carried out in all four cases. The nine Group IV tumor patients did reasonably well; seven of these patients remain alive, with a mean follow-up period of 2.3 years. Histological diagnosis was obtained in three of the seven patients who were explored in this group. This classification system has proven to be of value in determining prognosis and efficacy of surgical intervention.

Brain stem tumors in children: results of a survey of 62 patients treated with radiotherapy

A survey was conducted by the Radiation Oncology Discipline Committee of the Pediatric Oncology Group to provide an estimate of the outcome of treatment with modern radiotherapy of brain stem tumors in children. Thirteen of thirty-eight institutions submitted data for 62 patients who had been diagnosed between 1972 and 1981, who had completed planned treatment with radiotherapy, and for whom follow-up information was available. The results were poor. Fourteen patients were alive with no evidence of recurrence between 11 and 154 (median 50.5) months post-treatment, two were alive with disease at 22 and 24 months, and forty-six died of disease at intervals ranging from 2 to 30 (median 8.7) months from initiation of treatment. Actuarial survival was 50% at 1 year, 29% at 2 years, and 23% at 5 years. The age of the patient, the surgical approach, and the volume and dose of radiotherapy all appeared to correlate with outcome on univariate analysis; but on multivariate analysis a statistically significant survival advantage was found only for patients greater than 5 years of age who had undergone surgery as part of their treatment. Possible explanations for these findings are discussed, and the roles of surgery, radiotherapy, and chemotherapy are reviewed.

Stereotaxic biopsy of intrinsic lesions of the brain stem

Despite improved brain-stem imaging by magnetic resonance and high-resolution x-ray computerized tomography, definitive diagnosis and therapy of intrinsic lesions of the brain stem require histological verification. A stereotaxic approach to brain-stem lesions provides a high yield of positive histological diagnosis with a low incidence of morbidity. A series of 14 stereotaxic procedures performed on 12 patients with intrinsic lesions of the mesencephalon, pons, and medulla is reviewed. A detailed description of the transfrontal approach used by the authors is presented. Definitive pathological diagnosis was obtained in all patients. There was no operative mortality and only one case of permanent neurological deficit. The significance of accurate histological diagnosis in the therapy of brain-stem lesions is discussed.

Brainstem glioma: an analysis of 85 cases

The study analyses 85 cases of brainstem glioma in the past 35 years, 69 of which include patients under 16 years of age. The incidence of brainstem glioma was 2.4% of all intracranial tumours, and 9.4% of intracranial tumours in children. There were two peaks in age distribution, in the first and in the fourth decades. In children, the tumours were located mainly in the pons, so VIth and VIIth cranial nerve palsies, and pyramidal and cerebellar signs were frequently seen. In adult cases, the tumours ranged in location from the midbrain to the medulla, so neurological symptoms caused by lesions of the whole brainstem axis were seen. The left side was dominant in both age groups. The choice of treatment was steroid administration and radiation. Chemotherapy was not effective. Even after these treatments, the median survival period from onset was no longer than 10.5 months. We conclude that the treatment of brainstem gliomas in children should be distinguished from adult cases, which in the latter may be considered to be merely one of the gliomas which may occur at any other sites. Since brainstem gliomas in children may be congenital, we must redirect our treatment of these lesions to treatment of congenital tumours.

Pediatric brain stem tumors: patterns of treatment failure and their implications for radiotherapy

There have been conflicting opinions regarding the correct volume to be used in radiotherapy fields for brain stem tumors of childhood. Whereas many clinicians recommend limited fields designed to cover the tumor volume with a margin, some have advocated whole brain radiotherapy. Using our clinical experience at Duke University Medical Center, we have made an attempt to determine the proper irradiation volume in this group of tumors. We have evaluated 38 patients with brain stem tumors in children less than 18 years of age. The most common presenting symptoms were headache, ataxia, and hemiparesis. Thirteen patients had a histologic diagnosis made prior to treatment or post-mortem. All had either an anaplastic astrocytoma or a glioblastoma multiforme. Tumors were located in the thalamus, hypothalamus, or midbrain in 9 patients and in the pons or medulla oblongata in the remaining 29 patients. All patients received a course of radiotherapy. The mean minimum tumor dose was 52.6 +/- 5 Gy given at 1.7 to 2.0 Gy/fraction. Twenty-three patients received radiation to a limited field and 14 received whole brain irradiation. In one patient, the field size could not be ascertained. The five year survival of the total group was 39%. The survival of patients with thalamic, hypothalamic, or midbrain tumors was 73% compared with 28% for those with tumors of the pons or medulla oblongata (p = 0.0159). Eighty-eight percent of the tumor recurrences in evaluable patients (22/25) occurred within the radiotherapy fields. Patients were stratified for tumor location and no difference was observed in survival or relapse-free survival among those individuals treated with limited irradiation fields or whole brain irradiation fields. When our results are examined in conjunction with previously published data, the bulk of existing evidence supports the use of limited fields for irradiation of brain stem tumors of childhood.

Long-term results of surgically treated brainstem gliomas

Long-term results of 16 surgically treated brainstem gliomas are reported. The results have been reviewed with reference to the clinical and pathological criteria defined by Hoffman and co-workers (1980) in the attempt to differentiate a distinct group of benign tumours. A good correlation has been obtained. The usefulness of surgical treatment of brainstem gliomas has been stressed in view of the survival time as well quality of survival.

New therapeutic approaches for brain tumors

In the last decade, advances in the treatment of primary neuroglial tumors of the central nervous system in adults have been modest. Theories regarding their resistance to treatment have changed little, although now the heterogeneity of anaplastic astrocytomas is recognized. The most effective chemotherapeutic agents--the nitrosoureas and procarbazine--have been used for more than a decade, with no comparably active drugs identified in the meantime. The authors have initiated clinical trials using inhibitors of polyamine synthesis and augmentation of the known effectiveness of irradiation. These programs will be described with preliminary observations.

Supratentorial malignant gliomas of childhood. Results of treatment with radiation therapy and chemotherapy

Twenty-seven patients aged 1 to 18 years harboring supratentorial (20 in the cerebrum and seven in the thalamus) malignant gliomas were treated between 1975 and 1982. There were four glioblastomas multiforme, 14 anaplastic astrocytomas, and nine malignant gliomas. All patients had a subtotal resection or biopsy as the initial procedure and received postoperative radiation therapy (RT). Fifteen of 27 patients were treated by RT alone; 14 had tumor progression with a median time to tumor progression (MTP) of 65 weeks. Twelve patients were treated with chemotherapy as an adjuvant to RT; only seven had tumor recurrence, with an MTP of 130 weeks. Of the 21 patients with recurrent tumors in both groups, 18 were treated with chemotherapy alone, or chemotherapy with a second surgical procedure or second course of RT. For all histological grades of tumor, the MTP for first recurrence was 75 weeks and the median survival time was 180 weeks. Age at initial diagnosis was found to be a statistically significant prognostic factor, with patients younger than 10 years of age surviving longer than patients aged over 10 years (p = 0.02).

Brain stem tumours-diagnosis and surgical treatment

The diagnosis of brain stem tumours is mainly based on clinical findings and CT examinations: The CT examination including intravenous or intrathecal administration of contrast media together with electronic image-reconstruction greatly influences the indications for direct operation. According to our interpretation, direct surgical treatment is apparently successful under the following circumstances: - in tumours with external growth, which allows a decompression of the tumour without harming the surrounding structures in the brain stem. - in tumours with a marked cystic component, especially if the compressive effect is causing the features and progress of the symptoms.