[Practice guideline 'Brain metastases' (revision)]. / Richtlijn 'Hersenmetastasen' (revisie).

Improved survival of cancer patients results in an increase in the incidence of brain metastases. In addition, asymptomatic brain metastases are more often detected as a consequence of active screening. In patients with cancer and new neurological symptoms, MRI of the brain is indicated to assess the presence and number of brain metastases. Decisions concerning treatment of brain metastases should take place within a multidisciplinary team. Treatment is in the first instance focused on improvement or preservation of neurological functioning. The main treatment options for patients with brain metastases are whole brain radiotherapy, stereotactic radiosurgery/radiotherapy, and neurosurgical resection. The choice of treatment depends on the number and the location of the brain metastases, the general and neurological condition of the patient, the extent of extracranial tumour activity, and the expected results of treatment. The revised guideline supports the policy of whole brain radiotherapy not being the standard treatment following stereotactic radiosurgery or radiotherapy. In the case of complete resection, confirmed using early postoperative MRI, whole brain radiotherapy does not add to survival benefit, while patients may suffer from radiation-induced toxicity.

Diffusion tensor imaging and chemical shift imaging assessment of heterogeneity in low grade glioma under temozolomide chemotherapy.

Diffusion tensor imaging and multiple voxel magnetic resonance spectroscopy were performed in the MRI follow-up of a patient with a glioma treated with temozolomide chemotherapy. Tumor shrinkage was paralleled by reductions in choline level and by increases in apparent diffusion coefficient indicating decreased cellularity. Within the tumor, choline level and apparent diffusion coefficient showed a significant inverse correlation (P < 0.01). Fractional anisotropy distribution in the tumor correlated positively with N-acetyl aspartate level (P < 0.001), indicating that these parameters reflect (remaining) axonal structure. Tumor lactate level, also found to decrease under therapy, did not correlate with any other parameter.

Analysis of proliferation and apoptosis in brain gliomas: prognostic and clinical value.

With the introduction of new (immuno-)histochemical techniques it is now possible to assess rates of proliferation and apoptosis in brain gliomas using archival paraffin embedded material. As proliferation and apoptosis are related to tumour growth rate quantification of these processes has prognostic value and is related to tumour grading. In this study we assessed the proliferation rate by measuring the Ki-67 labelling index using the MIB-1 antibody (MIB-LI) and the apoptotic rate using the in situ labelling of DNA strand breaks with TUNEL (TUNEL-LI) in 315 supratentorial gliomas. MIB-LI and TUNEL-LI in astrocytomas (A) where significantly lower compared to anaplastic astrocytomas (AA), glioblastomas (GBM) and oligodendroglial tumours [oligodendrogliomas (O) and anaplastic oligodendrogliomas (AO)]. MIB-LI and TUNEL-LI were significantly lower in AA compared to GBM. In astrocytic tumours MIB-LI and TUNEL-LI appeared to be correlated. As the distinction between A and AA is of clinical value but can be difficult histomorphologically we analysed the prognostic value of MIB-LI and TUNEL-LI in gliomas with particular emphasis on A and AA. MIB-LI below 10% was of prognostic value in A and AA, O and AO but not in GBM on univariate survival analysis. TUNEL-LI was of no prognostic value. With multivariate survival analysis MIB-LI lost prognostic significance in O and AO. Astrocytomas with a gemistocytic component (AG) are similar to A with respect to survival and MIB-LI and TUNEL-LI. MIB-LI is of independent prognostic value in A and AA. Assessment of MIB-LI in A and AA can be used as an aid in distinguishing A and AA.

11C-tyrosine position emission tomography and 1H magnetic resonance spectroscopy of the response of brain gliomas to radiotherapy.

We monitored 10 patients with unresected (9) or partially resected (1) supratentorial gliomas with 11C-tyrosine position emission tomography (TYR-PET) before and after radiotherapy. TYR-PET tumour volumes were measured using a threshold technique. In seven patients the tumour volume decreased after radiotherapy, although all gliomas persisted on TYR-PET images. In eight patients the tumour protein synthesis rate (PSR) was calculated using a dynamic study protocol in combination with a PATLAK analysis. There were no changes in PSR after radiotherapy, but the PSR was calculated on the remaining tumour volume using the same threshold technique as before therapy, i.e. the decrease in tumour volume was not taken into account. In eight patients the PET data were compared with magnetic resonance spectroscopic imaging (1H-MRSI) performed simultaneously. Although there was no statistically significant correlation between TYR-PET volume changes and 1H-MRSI choline level we observed a simultaneous decrease in volume and choline in four patients.

The origin of lactate in peritumoral edema as measured by proton-magnetic resonance spectroscopic imaging.

Using in vivo proton-magnetic resonance spectroscopy (1H-MRS), which allows the measurement of metabolites of adequate tissue concentration, the origin of lactate in peritumoral edema has been assessed by comparison with lactate levels in the central and marginal areas of the tumor in 18 patients with cerebral gliomas. In the majority of cases lactate content in the area of peritumoral edema was lower than that in the tumor margin or tumor center, which is consistent with the assumption that the tumor is the source of lactate, which then reaches the surrounding area of edema by diffusion. In 3 of the 18 cases the amount of lactate in the peritumoral edematous tissue was higher than in the tumor, indicating that the lactate is locally produced on account of ischemia due to regional elevation of tissue pressure in the edematous area.

Concurrent chemo- and radiotherapy in patients with locally advanced carcinoma of the cervix.

BACKGROUND: The feasibility of concurrent chemotherapy and radiotherapy for advanced primary carcinoma of the cervix was evaluated and the results were compared to historical controls. PATIENTS AND METHODS: In a single institution study, patients (n = 74) with primary cervical carcinoma received 3 cycles carboplatin/5-fluorouracil concurrent with radiotherapy followed by salvage hysterectomy (group I). Treatment results were compared with those of a historical control group (n = 39) (group II), treated similarly but without chemotherapy. RESULTS: In group I median follow-up is 28 months (12-68+) and in group II 23 months (14-90+ months). The 5-year overall survival, progression-free survival and local recurrence free survival for group I and II are, respectively, 69% versus 38% (P < 0.003), 67% versus 38% (P < 0.005) and 84% versus 43% (P < 0.0001). Two patients in each group developed posttreatment enteritis. CONCLUSIONS: Radiotherapy with concurrent carboplatin and 5-fluorouracil resulted in a better overall survival, disease free survival and local disease free survival compared to historical controls. The toxicity of this schedule did not exceed that of radiation alone in historical controls.

Brain tumors: L-[1-C-11]tyrosine PET for visualization and quantification of protein synthesis rate.

PURPOSE: Positron emission tomography (PET) with the amino acid tracer L-[1-C-11]-tyrosine was evaluated in 27 patients with primary and recurrent brain tumors. MATERIALS AND METHODS: Patients underwent either static (n = 14) or dynamic PET (n = 13), with quantification of protein synthesis rate (PSR) and tumor-to-background ratio. Findings were compared with histologic findings. RESULTS: Primary brain tumor was proved in 22 patients histologically, as well as metastatic cancer of unknown origin, primary non-Hodgkin lymphoma, meningioma, atypical infarction, and vasculitis in one patient each. At PET, 20 of 22 primary tumors, the metastasis, and non-Hodgkin lymphoma were correctly depicted. A false-positive finding was obtained with the infarction, and the meningioma and vasculitis were not depicted. The calculated sensitivity was 92%; specificity, 67%; and accuracy, 89%. There were no statistically significant relationships between histologic findings, PSR, and tumor-to-background ratio. CONCLUSION: L-[1-C-11]-tyrosine is a valid tracer for diagnosis of brain tumors and allowed quantification of PSR.

Localised proton spectroscopy and spectroscopic imaging in cerebral gliomas, with comparison to positron emission tomography.

In 32 patients with gliomas, one- and two-dimensional proton magnetic resonance spectroscopy (1H-MRS) has been conducted, the latter allowing reconstruction of spectroscopic data into a spectroscopic image (MRSI), showing the distribution of the various metabolite concentrations over the cross-sectional plane. For lack of absolute concentrations, the measured concentrations of phosphocholine (CHOL), N-acetyl-L-aspartate (NAA), and lactate (LAC) were conventionally expressed in ratios relative to that of creatine (CREAT). Compared to normal brain tissue, an increased CHOL/CREAT ratio was found in all groups of tumours, in glioblastomas, high-, middle- and low-grade astrocytomas both at the margin and the core of the tumours, but in oligodendrogliomas only at the margin. This is consistent with an increased phosphocholine turnover in relation to membrane biosynthesis by the proliferating cells. The NAA/CREAT ratio was decreased in all groups of tumours, both in the centre and at the margin, reflecting replacement of functioning neurons by neoplastic cells. The LAC/CREAT ratio was elevated in the core of malignant gliomas, which may be the result of a prevailing glycolysis, characteristic of tumours, possibly in conjunction with hypoxia/ischaemia. In the perifocal oedema, there was neither elevation of the CHOL/CREAT ratio nor decrease of the NAA/CREAT ratio; an increased LAC/CREAT ratio therefore rather reflected ischaemia/hypoxia probably due to locally elevated pressure and compromised regional perfusion. In the normal brain, the metabolite ratios of grey matter did not differ from those of white matter. The frontal lobe and basal ganglia showed lower NAA/CREAT ratios than the other cerebral areas.(ABSTRACT TRUNCATED AT 250 WORDS)

On the principles underlying the diagnosis of brain tumours--a survey article.

A survey is given of the principles underlying the diagnosis of brain tumours. Traditionally diagnosis and localization of brain tumours have been based upon morphological criteria. Currently unsurpassed levels in imaging of anatomical details and topographical relations by the techniques of computed tomography (CT) and magnetic resonance imaging (MRI) have been achieved. The techniques of positron emission tomography (PET) and of magnetic resonance spectroscopy (MRS), which depict also metabolic and blood flow aspects, provide a refinement of our knowledge on the metabolism, structure and pathophysiological relations of a tumour to the surrounding parenchyma. Recent advances in the recording of function-related changes of the cerebral electro-magnetic field allow a better definition of critical functional areas.

Brain tumor delineation based on CT and MR imaging. Implications for radiotherapy treatment planning.

This paper deals with the impact MRI may have on radiotherapy treatment planning of brain tumors. The authors analyzed differences in size and position of treatment fields as indicated by three observers (two radiotherapists and one neuroradiologist) using CT or MR based radiotherapy planning procedures for cerebral-gliomas. Large differences in field size and position were found in non CT contrast enhancing tumors, all low grade gliomas. Small differences were found in contrast enhancing lesions including the high grade gliomas. We conclude that implementation of MRI in radiotherapy treatment planning leads to a greater precision in treatment fields for non CT contrast enhancing lesions.

Localized proton spectroscopy of inoperable brain gliomas. Response to radiation therapy.

With in vivo 1H-MRS resonances of several metabolites were simultaneously measured in cerebral gliomas and adjacent normal brain. 15 patients with inoperable brain gliomas all histologically verified were monitored with 1H-MRS and MRI before and after radiotherapy. 11 patients were evaluable. 1H-MRS technique evolved from single volume measurements to one dimensional and two dimensional 1H spectroscopic imaging. In all patients N-acetyl-aspartate signals were decreased in tumour areas compared to the normal brain hemisphere. No recovery was seen after radiotherapy. Choline signals were increased in tumour margins of high grade gliomas and more diffusely in low grade gliomas. In 5 patients the choline resonance decreased after radiotherapy, accompanied by a shrinkage of tumour diameter on MRI. Lactate signals were present in high grade and unspecified astrocytomas and absent in most low grade gliomas. In 3 patients the lactate signal disappeared after radiotherapy. These observations indicate the feasibility of 1H-MRS in monitoring metabolic responses on radiotherapy of brain gliomas.

Radiotherapy of primary brain tumours in the region of the third ventricle.

Patients (n = 18) with a primary brain tumour near the third ventricle and treated by radiotherapy were retrospectively analysed. Four different subgroups of patients, according to the histology (germ cell tumours, astrocytomas, other histologies, no histology) were separately discussed. Third ventricle tumours were more frequent in younger patients (less than 30 years). Five years actuarial disease free survival was 62%. Six out of 18 patients died of disease, the others are currently alive without evidence of disease.