Radiosurgery: Gamma Knife versus adapted linear accelerator
Project lead: Stefan Mathis
Project team: Stefan Mathis
Duration: May 2010 - September 2010
Suggested by: KAGes
Publication: HTA Project reprot No. 47 - http://eprints.hta.lbg.ac.at/901
In a number of diseases of the brain and associated structures the methods of microsurgery are limited. For such situations radiosurgery is an option. Radiosurgery is defined by the biological inactivation or destruction of intracranial 3D targets by high-precision ionising radiation during on single session. One important indication is the treatment of brain metastases. Different forms of applying radiosurgery exist including gamma knife, adapted linear accelerators and proton beam systems. For this assessment we compared two of these – gamma knife and adapted linear accelerator – with a focus on clinical effectiveness and cost.
The gamma knife is a radiation device that bundles approximately 200 rays from radioactive sources (cobalt 60) by collimation. The resulting focus with a diameter of a few millimetres is called isocenter. A patient is treated by positioning his head in such a way, that the isocenter and the patient’s lesion overlap. In this position the patient remains fixed (by a stereotactic frame), until the therapeutic dose is reached according to the dose distribution plan. Because of the high dose gradient (high dose in the centre, low doses in the neighbourhood) the surrounding tissue can be protected from doses above their radiation tolerance level. The gamma knife is used for pathologies of the head with a diameter of less than 3 centimetres. Radiosurgery is applicable to patients with primary brain/head tumours (menigioma, schwannoma, pituitary adenomas, glioma, melanoma), with brain metastases, vascular malformations and a number of functional diseases (for instance: trigeminal neuralgia). The comparator in this assessment is an adapted linear accelerator (LINAC). LINAC is a radiation device which emits X-rays. For radiosurgical use the linear accelerator must be adapted by specific equipment that bundles the rays, sometimes referred to as 'stereotactical extension'. In use are tertiary collimators (multileaf or ring collimators) which allow the radiographer to direct the radiation focus as precise as is necessary for radiosurgery. In contrast to the gamma knife, the adapted linear accelerator has only one source of rays. This emitter head must therefore be moved around the patient, otherwise (with a single entrance point) the tolerance of the tissue in the pathway of the ray would be exceeded quickly and –as a consequence- the therapeutic target dose in the target lesion could not be reached. Moving parts (a complex feature of the LINAC system) require a vast amount of maintenance and quality control.
A systematic search in medical databases was performed and completed by hand searching. The systematic search-strategy included all studies written in English or German. The included population was defined as patients with primary brain tumours, brain metastases, arteriovenous malformations or trigeminal neuralgia. To be considered for inclusion, studies had to evaluate gamma knife or LINAC based radiosurgery. Only studies with a high level of quality in terms of their study design (RCTs, prospective cohort studies) were included.
For the cost comparison the findings from economic literature were analysed and experts interviewed about the market situation and the cost components during the radiosurgery procedure.
Out of a total of 742 records only few clinical studies made statements on the comparison of both methods. All of them were of low evidence quality (consequently not meeting the inclusion criteria), therefore no definite conclusions can be drawn. Statements from those studies indicate that gamma knife and adapted LINACs have comparable clinical effects. From studies on the dose distribution a slight advantage in precision for gamma knife (especially with very small target volumes) and some advantages in homogeneity for the adapted LINAC (in bigger target volumes) can be reported. The clinical relevance however remains unclear in terms of prospective controlled studies.
Three assessments (all from Germany) with a similar pool of study questions, to this assessment were identified. One HTA from DAHTA evaluated options for meningioma patients, another HTA (also from DAHTA) addressed options for patients with brain metastases. A third german HTA report about brain metastases was completed by the medical service agency of a major German insurance (Medizinischer Dienst des Spitzenverbandes Bund der Krankenkassen e. V./MDS). All these reports revealed a lack of high level evidence from head-to-head comparison studies. From the best available evidence equality can only be assumed. A general lack of evidence was observed regarding most of the indications of radiosurgery.
A cost comparison was performed regarding annual costs. The higher acquisition costs of the gamma knife (~4,000.000 €) are compensated by its longer life span (~20), yet gamma sources need to be replaced every 5-7 years costing approximately 700.000 €. An adapted linear accelerator costs ~3,000.000 € with a life span of ~10 years. Due to the complexity of a LINAC system, more maintenance, quality control and therefore personnel resources should be expected. On the other hand, a LINAC system can also be used for non-radiosurgical indications, such as fractionated (more than one session) radiation or extracranial indications. Other cost factors depend on variables that are given by the local context (e.g.: intended indications, case numbers, existing equipment and experience)
There are statements from studies indicating a similar effect of both technologies – gamma knife and adapted linear accelerator. However, the strength of this evidence is low. In addition most of the indications for radiosurgery seem to be questionable and should be evaluated by randomised controlled studies which compare radiosurgery with the latest available treatment options. In the interest of a growing population of affected people, studies that evaluate the benefits and harms of radiosurgery should predominantly be funded. If both modalities are available the dose distribution characteristics indicate the use of gamma knife for small lesions (better conformity, efficient workflow), while the adapted LINAC has advantages with bigger volumes (faster, better coverage, homogeneity). From a cost perspective the LINAC system seem to be more versatile in its use and therefore recommended, when radiosurgery case numbers are small. Local variables (consented indications, case numbers, existing equipment, experience of personnel) should be carefully evaluated to allow a decision that meets criteria of an efficient patient care.