This phase I pilot study aims to define the safety of a combined treatment of bevacizumab and whole brain radiation therapy for the treatment of patients with brain metastasis of solid tumors. If this therapeutic scheme confirms it's safety profile, the investigators can expect:

• first, to allow that all patients can receive bevacizumab for their advanced and/or metastatic cancer if necessary, even in case of brain metastasis.
• Secondly, if this trial confirm a synergic effect of the combination of angiogenesis inhibitors and brain radiotherapy for local control of brain metastasis, an improvement of the therapeutic results for these patients which have a poor hope of survival and for which none innovative approach is currently suggested.

Moreover, the investigators hope that the analysis of the different data of MRI evaluation - morphological and functional - will allow better definition of radiological evaluation of the therapeutic effect of angiogenesis inhibitors on brain metastasis.

Brain metastasis remain a common complication of many solid tumors, occurring from 10 to 20% of patients with cancer, and the incidence may arise up to 20-40% in autopsy series. Despite the treatment, the patient outcome is poor and brain metastasis become a major cause of death from cancer.

Some recent studies showed an increase incidence of brain metastasis; several explanations are evocated:

• The increased efficacy of brain imaging in screening of brain metastasis.
• The relative longer life expectancy of cancer patients in correlation with more efficient therapies (new chemotherapy drugs, targeted therapies) for visceral and bone metastasis, but without effect on the risk of brain metastasis (sanctuary effect),

Therapeutic management of brain metastasis: standards of care and improvement strategies

For patients with a poor survival hope, the treatment is limited to best supportive care with a palliative intent. At the opposite, a specific treatment of brain metastasis must be proposed when a real benefit may be expected.

Surgery or radiosurgery are indicated for limited brain involvement (1 to 3 metastasis).

In case of multiple localisations, whole brain radiation therapy (WBRT) is the most common approach; however this treatment is realized with a palliative intent: control of neurological symptoms and brain oedema.

In the literature, an objective response rate of 50-60 % is expected from WBRT, but the majority of these patients will relapse and died from brain metastasis, with a median survival time of 4 to 6 months.

Considering these poor results, it seems crucial to develop some innovative strategies, in order to improve the survival hope for these patients.

Different concomitant approaches were tested in association with WBRT (cytotoxics such as temozolomide, radio-sensitizers such as efaproxiral), and more recently combination of radiation therapy with therapy targeting different signalling pathway were developed, for example with lapatinib for HER2 pathway. Targeted therapies acting on neoangiogenesis seem also may be considered as an interesting axis of research.

Rationale utilisation of angiogenesis inhibitors for treatment of brain metastasis

Neo-angiogenesis plays an important role in the growth of brain metastasis, acting on different mechanisms such as tumor neo-angiogenesis (by development of new blood vessels), co-option with existing blood vessels and increased vascular permeability. Therefore, angiogenesis inhibitors, particularly agents targeting VEGF/VEGFR pathway, could be interesting in the treatment of brain metastasis.

In pre-clinical models, VEGF is the best characterized mediator of angiogenesis, and its role was pointed in the development of brain metastasis: increased transmigration of tumor cells through the blood-brain barrier by a VEGF-dependent disrupting effect, anti-apoptotic effect of VEGF by a paracrine action.

This data suggest a therapeutic effect of VEGF targeting agents on brain metastasis and a decrease of brain metastasis development has been observed in some anti-VEGF anti-sens therapy.

Moreover, VEGF involves vascular permeability and this mechanism is important in the development of peri-tumoral oedema.

Anti-angiogenic treatments are divided in two groups: on one hand, angiogenesis inhibitors (acting by normalization of the existing tumor vascularisation and through a decreased formation of new blood vessels) and on the other hand, vascular disrupting agents (which can destroy pathologic vessels).

For brain tumors, most of studies were focused on the role of angiogenesis inhibitors. Of special interest, anti-VEGF therapies, particularly the bevacizumab, were tested in clinical setting. Bevacizumab, a monoclonal recombinant antibody targeting VEGF, prevents the fixation of VEGF to VEGF receptors 1 and 2 on the endothelial cells, with a neutralizing effect on VEGF action.

In clinic, recent studies showed a significant benefit of bevacizumab (Avastin®) in the treatment of primary brain tumors, as shown by Vredenburgh et al. for patients with recurrent gliomas; these data confirm a synergic effect with chemotherapy. Similarly, a potential synergic action in a combined approach with radiation therapy seems to be an interesting therapeutic way.

Rationale for combining angiogenesis inhibitors with radiation therapy

In the tumor, the pathologic neo-angiogenesis reduces the diffusion of oxygen, and consequently promotes hypoxia. Hypoxia is one of the most causes of radio-resistance.

Ionizing radiations induce transient hypoxia in the central area of the tumor, increasing the production of VEGF and consequently the neo-angiogenesis. This "vicious circle" allows the development of optimal conditions for radio-resistance.

The normalization of pathologic vessels induced by angiogenesis inhibitors leads to a decrease of hypoxia and thus enhance the radio-sensitivity of tumor cells and endothelial cells.

In clinical setting, some studies were published about treatments combining radiation therapy and angiogenic inhibitors. The main study was conducted for rectal carcinoma (9), showing a safety profile of a combined scheme of bevacizumab and capecitabine in association with pre-operative pelvic irradiation. Efficacy data are encouraging (downstaging for 9 /11 patients, histological regression for 5 patients). Similar conclusions may be drawn from another study performed in patients with unresectable pancreatic carcinoma treated with a radio-chemotherapy including bevacizumab, leading to an objective response for 9 patients (20 %), 4 of them were operable at the end of radio-chemotherapy.

Interestingly, this combined approach seems interesting for brain tumors, as shown by a recent published trial conducted by Narayana in first-line treatment of gliomas. This phase 2 study combined temozolomide chemotherapy with bevacizumab administration for a dose of 10 m/kg at D14 and D28 during the course of radiotherapy (total dose of 59,40 Gy delivered to the tumor volume). An promising response rate was observed: 7% of complete response and 93% of partial response, the 1-year progression-free survival was 59,3%. The authors confirm the feasibility of this combined approach (13/15 patients completed their treatment). Three grade 3-4 haematological toxicities and 4 non-haematological toxicities (thrombo-embolism and HTA) are related. None intracerebral hemorrhage or death related to the treatment was reported.

These results indicate a possible way of innovative strategies in the therapeutic management of brain metastasis, by association of bevacizumab and whole brain radiation therapy. Furthermore, bevacizumab is now indicated for several advanced and/or metastatic cancer but its administration remains theorically contra-indicated for patients with brain metastasis, due to the fear of potential cerebral hemorrhage. However, this risk remains low, as shown by several studies.

Safety data were recorded with glioblastomas are reassuring, however, it is essential to determine the optimal dose level of bevacizumab when combined with irradiation of the entire brain at a dose of 30 Gy Thus for the phase 1 pilot study that we plan to carry, it is important to test several dose levels of bevacizumab in combination with irradiation of the brain: a precaution we have decided to use a conventional fractionation of 15 Gy in 2 fractions three weeks. Secondarily, given the good tolerability of this modality in combination with three levels of dose of bevacizumab used, we propose to continue the phase 1 according to the scheme commonly used in palliative situation (30 Gy in 10 fractions and 2 weeks) in combination with Avastin at a dose of 15 mg / kg, at which a maximum limiting toxicity has been observed.

Inclusion Criteria:

• 18 < age < 70 man or woman
• Patient with a cytologically or histologically proven primary solid tumor
• With measurable and inoperable brain metastasis,
• Without meningeal carcinomatosis ,
• ECOG performance status ≤ 2,
• No previous treatment with angiogenesis inhibitors less than 3 months before inclusion,
• No chemotherapy and/or immunotherapy less than 3 weeks before treatment,
• No contra-indication to bevacizumab,
• No proteinuria with urine dipstick for proteinuria > 2+
• Blood sample ≤ 7 days before inclusion:
• Hemoglobin ≥ 10 G/100 ml
• Neutrophils count ≥ 1500 /mm3
• Platelets ≥ 100 000 /mm3
• Normal coagulation test: INR ≤ 1,5 ET TCA ≤ 1,5 x LSN 7 days before inclusion
• A written informed consent must be obtained.

Exclusion Criteria:

• Haemorrhagic brain metastasis,
• Uncontrolled hypertension (systolic > 150 mm Hg and/or diastolic > 100 mm Hg). Patients with high initial blood pressure are eligible if entry criteria are met after initiation or adjustment of antihypertensive medication,
• Prior brain radiation therapy
• Concomitant anticoagulant treatment
• Significant surgical procedure less than 28 days before inclusion (1 day if minor surgical act)
• Significant cardio-vascular disease, eg:
• Cerebral vascular thrombosis/haemorrhage or myocardial infarction <6 months
• Congestive heart failure > 2 NYHA
• Uncontrolled coronary disease
• Prior venous and/or arterial thrombosis < 6 months before inclusion
• Severe concurrent uncontrolled medical disease,
• Any psychiatric disorder that might prevent the subject from completing the treatment or interfere with the interpretation of the study results,
• Pregnancy or breast feeding
• Individual deprived of liberty or placed under the authority of a tutor.