Incorporating Oncology Methods in CNS Drug Development

We previously covered the challenges that biopharmaceutical enterprises are facing when it comes to developing CNS medical products at ExL’s 2nd CNS Clinical Trial Forum. In this interview, Glenn Morrison, Executive Director of Clinical Development at Zogenix, will discuss how biopharmaceutical enterprises can apply techniques used in oncology clinical trials in CNS development.

Moe Alsumidaie: Why is there such a high failure rate in CNS studies?

Glenn Morrison

Glenn Morrison: If you asked ten experts, you would probably get ten different opinions. There are at least two factors that, in my opinion, contribute towards the high failure rates in CNS development. Firstly, in CNS studies, efficacy endpoints tend to be very subjective. Contrast this with oncology, where drugs are tested and approved for their effect on clinically meaningful endpoints; survival for example. Those endpoints are easy to identify, are reliable with repeated measures and are clinically meaningful. In CNS development, the primary endpoint for many disorders is subjective: HAM-D in depression, ADAS-Cog or CDR-SB in Alzheimer’s, EDSS in MS for example, and in many cases these scales are administered by the investigator which introduces variability from one study site to another, and although they are validated, they may not necessarily generate reproducible results. In many cases, the scales are administered to a caregiver rather than the patient themselves, and the caregiver gives an opinion of the patient’s progress or lack thereof, throwing additional variability into the mix, which impacts statistical analyses and study outcomes. Secondly, establishing clinical meaningfulness of endpoints is also challenging. For example, in Alzheimer’s disease approval is based on co-primary endpoints of cognition and function, and while you may show a significant change on a cognitive endpoint, without hitting on the functional endpoint, one is left with the question of what does that mean in the patient's future outcome? If CNS disorders treating a symptom may not be enough without demonstrating that it has an impact on a patient’s ability to function. Another related challenge to study endpoints is demonstrating proof of concept in Phase II trials; Phase II trials are becoming larger and longer, and so in many cases they are like mini Phase III studies. Accordingly, the cost of development increases and timelines lengthen and for many companies the risk-reward doesn’t justify the investment.

MA: Can you describe the biomarker process in oncology, and how targeted therapies are developed using this approach?GM: Today, one of the fascinating things in oncology is moving away from treating the whole body - a systemic approach to chemotherapy - to a more targeted molecular approach. Whether it means targeting specific genes or pathways or trying to directly get to the tumor. From a biomarker perspective, there have been significant advances in understanding the biology and researchers have come to realize that cancer is not one single disease and that patients respond in highly individualized ways based on their genetic profiles and disease stages. Anti-cancer agents are now being developed that are targeted - information at the level of individual genes, the genome, the proteome and signaling pathways must all be considered and consolidated in order to correlate changes within the cancer biology with clinical phenotypes and outcomes following treatments. This has driven the development of biomarkers that can be used to enrich the clinical trial population by identifying those most likely to benefit from the treatment. This reduces the risk to the non-responder population and, by allowing earlier assessment of therapeutic efficacy, substantially shrinks the costs of development – enabling smaller trial sizes to demonstrate efficacy.

MA: Why are biomarkers so important in CNS therapy development?

GM: In CNS, we lack a true understanding of the biology of the disease. Biomarkers, just like in oncology, can increase our understanding of the pathology that is at the root of CNS disorders. Having CNS biomarkers will help us develop various strategies that are more effective in identifying the markers and genes that contribute to the disease and the pathways that are particular to the disease. By specifically targeting those areas, we can speed up drug development by having smaller, more targeted studies. This could also lead to an increase in success rates, because we would be targeting a specific pathway that we know is relevant to the disease. For example, we have some understanding of the pathways that might contribute to Alzheimer’s disease, we know there is a beta amyloid and we know the role beta secretase and gamma secretase play in the amyloid pathway, however, there is some debate about how exactly they contribute to the disease. A number of compounds have been developed and tested that target the buildup of amyloid in the brain, but no study has yet been able to make the link between reducing amyloid deposition and actual clinical improvement - we can remove the amyloid from the brain, but we have yet to demonstrate that this causes any clinical improvement. Perhaps the approach should be to target specific pathways and genetic markers to first identify patient populations based on their genotype rather than their phenotype. Again, thinking about Alzheimer's disease, it may be that developing treatments that are specific for APOE E4+ homozygous patients is a more rational approach. Most companies are looking at the Alzheimer’s population as a homogenous group when there are llikely many different heterogeneous groups within the disease who might benefit differently to the experimental treatments. Having access to the right biomarkers will allow us to dissect problems like this. I have no doubt that future CNS development success will be based on targeted therapies that target specific pathways in specific patients at the right stage of the diesease.

MA: Can you elaborate on how biomarkers are currently being used in CNS development?GM: One of the major challenges for utilizing biomarkers in CNS is that unlike oncology where biopsies or tumor removal has routinely been used to study pathophysiology and create cellular models of disease for testing new therapeutics, brain tissue is rarely available during life. And so in CNS drug development we have to rely on functional and structural imaging, electrophysiology, and blood and cerebrospinal fluid-based measurements, which might yield glimpses into underlying pathological processes, especially when applied longitudinally during the years of risk and prodromal stages. There has been some attempt to try to use biomarkers in depression and schizophrenia, where we are trying to identify specific sub-groups within the general disease areas. If we crack that, it is going to be a huge advantage in CNS drug development because we will be able to develop specific target medications to treat these individuals. Because it is a challenge to get access to human brain tissue during life, we could use alternatives like induced pluripotent stem cells (iPSCs) - taking cells from patients with the disease, culturing them, and creating a disease in a dish model. iPSCs can be converted into neurons and glia in vitro and be used to study pathogenetic mechanisms of CNS diseases and these individualized cultures can serve as substrates for high throughput screening and testing novel therapeutics. So while this is still in the preclinical stage, there is some progress in understanding and identifying compounds that show impact against specific genotypes; a great example being motor neurons and cell death in ALS. Overall, with regards biomarkers in general, CNS development today is where oncology was 20 or 30 years ago. This is not necessarily a negative, but instead is a huge opportunity and in the coming years things are really going to explode as far as scientific advancement in CNS and drug development is concerned.

ExL’s 8th Proactive GCP Compliance conference is occurring from April 5-7 in Philadelphia.

Moe Alsumidaie, MBA, MSF is Chief Data Scientist at Annex Clinical, and Editorial Advisory Board member for and regular contributor to Applied Clinical Trials.