Sufficient funding could not be secured for the study

High-grade gliomas are the commonest primary malignant brain tumours in adults, affecting approximately 5000 people per year in the UK. Standard treatment comprises a combination of surgery, radiotherapy and chemotherapy; however this condition remains incurable and the average survival is approximately 18 months from diagnosis. There are a number of reasons for this. Firstly these tumours are highly invasive and involve important areas of brain making it impossible to remove them surgically or cure them with radiotherapy. In the majority of cases the tumour recurs within 2 to 3cm of the original site of tumour removal. Secondly, due to the presence of a barrier between the bloodstream and the brain, when drugs designed to kill tumour cells (chemotherapy) are given intravenously or orally, they frequently do not reach the tumour at a sufficient dose to have a beneficial effect. As the chemotherapy dose has to be very high for a sufficient dose to reach the tumour, drug-related side-effects are common.

Laboratory studies demonstrate that glioma tumour cells are sensitive to a number of different chemotherapies, including carboplatin. When given intravenously however, carboplatin does not reach a sufficient concentration in the tumour to have a beneficial effect. However, studies have shown that carboplatin can be infused directly into the brain at a concentration that is highly toxic to tumour cells, but not to normal brain tissue. Using very small tubes implanted around the tumour, the investigators are able to infuse carboplatin reliably and repeatedly into the area where tumours typical recur. In this study, the investigators intend to evaluate the safety of this approach and determine the optimal dose of carboplatin to administer. It is hoped that this study will also provide evidence of improved survival for patients with high-grade glioma.

Inclusion Criteria:

• Age 18 years old or over
• Male or female
• World Health Organisation performance status 0-2
• Life expectancy greater than 3 months
• Capacity to give informed consent
• Histologically confirmed glioblastoma multiforme. Patients with a previous history of a lower grade gliomas are eligible if histology at relapse confirms glioblastoma
• Progressive and/or recurrent disease confirmed by MRI
• Progressive disease, defined as 25% or greater increase in contrast-enhanced tumour volume on T1-weighted MRI
• Supratentorial disease
• Disease confined to a single quadrant of brain
• It must be feasible to achieve sufficient carboplatin distribution in the peritumoural tissue as defined by the principal investigator and/or trial coordinator. Feasibility may be determined through the use of appropriate software that uses diffusion imaging and fluid dynamics mathematical modelling to predict infusate distribution
• Recurrent disease following conventional treatment, including surgery (biopsy or debulking), radiotherapy and chemotherapy (temozolomide)
• More than 30 days since prior chemotherapy (42 days for nitrosureas or mitomycin)
• More than 90 days since radiotherapy or radiosurgery
• More than 7 days since tumour debulking or other neurosurgery
• More than 30 days since prior investigational agents or participation in another clinical research trial
• Platelet count > or = 100,000/mm3
• Absolute neutrophil count > or = 1000mm3
• Total bilirubin no greater than 1.5 x upper limit of normal (except patients with Gilbert's syndrome)
• AST and ALT < or = to 2 times upper limit of normal
• PT and APTT no greater than control
• Creatinine clearance > 50ml/min using Cockcroft Formula
• Fertile patients must agree to use effective contraception during and for 2 months after study treatment
• Negative pregnancy test if appropriate

Exclusion Criteria:

• Clinical evidence of raised intracranial pressure.
• Concurrent medical condition that would preclude general anaesthesia.
• Severe acute infection.
• Pregnancy or breast feeding.
• Documented allergy to carboplatin or cisplatin.
• Prior participation in a trial of biological therapy (e.g. monoclonal antibodies, gene therapy, oncolytic viral therapy, immunotoxin therapy).
• Prior local chemotherapy, including administration of biodegradable polymer wafers containing carmustine.
• Prior enrolment in this study.
• Concurrent anticancer drugs.
• Concurrent investigational therapies.
• Infratentorial or intraventricular tumour visible on MRI.