Effect of Metformin on Behaviour and the Brain in Children Treated for a Brain Tumour

A critical barrier to improving the quality of life of children/adolescents living with cancer is that our curative therapies, which include a combination of surgery, chemotherapy and radiation, have toxic effects on healthy tissue, resulting in long-term problems. This is evident for children and adolescents who survive medulloblastoma - a brain tumour requiring aggressive therapy: they experience brain injury and cognitive impairment. There are few therapies for restoring cognitive function and promoting brain growth in survivors; however new work in regenerative medicine offers a possible alternative. The drug metformin promotes brain growth in animal models by activating neural stem cells. In a pilot trial with 24 participants, we found that metformin was safe and tolerable for use in children/adolescents treated with cranial radiation for a brain tumour and may improve cognition and promote white matter growth. In this multi-site clinical trial, we will test the efficacy of treatment with metformin for brain repair and cognitive recovery in medulloblastoma survivors. If we find that metformin promotes cognitive improvement and brain growth in paediatric survivors of medulloblastoma, this may offer a viable therapeutic approach that may improve quality of life of these cancer patients and provide a model for treatment of late effects in other paediatric cancers.

This study is designed to test the efficacy of metformin in a 16-week multi-centre, phase III, double-blind, randomized placebo-controlled superiority trial with two parallel conditions (metformin versus placebo). Participants will be randomly assigned to one of the two treatments where they will either complete a 16-week cycle of metformin or a 16-week cycle of placebo. Participants will be randomized using Research Electronic Data Capture (REDCap) to ensure allocation concealment. The randomization code will not be released until the participant has been recruited, consented and passed screening. Outcome assessments will be conducted at Baseline (intelligence quotient (IQ) testing will also be conducted at Screening), immediately following the completion of week 16 treatment (Post-Intervention, Week 17), and 24 weeks following completion of the intervention (6 Month Follow-Up, Week 41).

The primary endpoint is cognitive function in children/adolescent survivors of medulloblastoma at Post-Intervention (Week 17) compared to Baseline (Week 1). We hypothesize that 16 weeks of treatment with metformin will be associated with better cognitive outcomes than 16 weeks of treatment with placebo. Cognitive outcomes will be measured using tests of working memory, declarative memory, and processing speed.

The key secondary outcome will be diffusion MRI within the corpus callosum at Post-Intervention (Week 17) compared to Baseline (Week 1). We hypothesize that 16 weeks of treatment with metformin will be associated with increased white matter growth in the corpus callosum compared to 16 weeks of treatment with placebo. Increased white matter growth will be measured using diffusion MRI metrics.

Exploratory outcomes have been selected to investigate broader metformin-induced changes in the brain and cognition.

1. We hypothesize that 16 weeks of treatment with metformin will promote global white matter growth in the brain more so than 16 weeks of treatment with placebo at Post-Intervention (Week 17) compared to Baseline (Week 1). White matter growth will be assessed using diffusion MRI metrics of myelin and fiber structure.

2. We hypothesize that 16 weeks of treatment with metformin will result in greater increases in hippocampal volume compared to that 16 weeks of treatment with placebo at Post-Intervention (Week 17) compared to Baseline (Week 1). Structural MRI measures of hippocampal volume will be explored.

3. We hypothesize that 16 weeks of treatment with metformin will result in superior performance on measures of attention, executive functioning, and intelligence compared to 16 weeks of treatment with placebo at Post-Intervention (Week 17) compared to Baseline (Week 1). Tests of attention, executive functioning, and intelligence will be used.

4. We hypothesize that all outcome measures will continue in the predicted direction at 24 weeks (6 Month Follow-Up, Week 41) compared to Baseline (Week 1)/ screening (Day

   0. following completion of 16 weeks of metformin compared to 16 weeks of placebo.

5. We also hypothesize that 16 weeks of treatment with metformin will yield better outcomes in females compared to males for all measures and that these findings will persist at 24 weeks (6 Month Follow-Up, Week 41) following the intervention compared to Baseline (Week 1).

6. We hypothesize that 16 weeks of treatment with metformin will result in improved ratings of global health as reported by the parent/guardian at Post-Intervention (Week 17) compared to Baseline (Week 1).

Metformin is a well-studied medication with a broad clinical experience in children including polycystic ovarian syndrome, diabetes, and obesity. The youngest age of use is 2 years old. The proposed dose and the schedule of administration of metformin is based on safety and toxicity data obtained from our pilot trial and previous use in paediatric populations. One hundred and twenty (120) English speaking and twenty (20) French speaking participants - aged 7 years to 17 years and 11 months - will be recruited from up to 19 sites across Canada and Australia.

Analysis of covariance (ANCOVA) will be used to examine the effects metformin versus placebo for each outcome in English speaking participants, controlling for Baseline outcome measurements. For French speaking participants, IQ testing will be completed, but not Cognitive testing as French-Canadian translations are not available. Results from IQ testing will be examined to explore the effects metformin versus placebo controlling for Screening visit outcome measurements. By focusing on a disease that requires some of the most aggressive therapy used in modern protocols, and by targeting the patients most vulnerable to the harmful effects of treatment we hope to provide a model of intervention that can then be applied to other cancers and actively promote brain health and cognitive recovery.