Genome-wide and phenome-wide association studies to inform drug development

The development of a single drug takes considerable time and resources, yet the failure rate in clinical trials ranges from 86% at phase one to 40% at phase 3. Drugs with strong genetic evidence, generally from genome wide association studies (GWAS), have double the success rate of those that do not. Genetic mimics of drugs investigated through phenome wide association studies (PheWAS) have successfully predicted drug side effects and identified possible repurposing opportunities. However, many phenotypes have been understudied in GWAS and are absent in PheWAS. To address this, I developed new PheWAS software then used that software alongside GWAS methods to investigate chronic sputum and neuropathic pain phenotypes to discover genetic insights to aid drug development.

In the largest known GWAS of chronic sputum I discovered six genome-wide significant findings (P<5x10-8). My findings provide strong evidence for the role of FUT2 in mucus production and as a potential target for drug development.

I identified eight areas of development in PheWAS methods that I addressed with the development of the R package DeepPheWAS. Through analyses I showed the utility of DeepPheWAS, including demonstrating DeepPheWAS being used in published analyses of lung function and thyroid stimulating hormone for novel pathway specific risk scores. I also described an example of how DeepPheWAS had been used by industry collaborators Orion to help inform funding decisions for drug development.

Finally, I performed a GWAS of 32 neuropathic pain phenotypes developed using the DeepPheWAS package. I found seven unique genome-wide significant loci. Sentinel variant rs12277922 for all-cause neuropathic pain showed consistent effect size across the other neuropathic pain phenotypes and nearest gene (intronic) NCAM1 has been implicated in neural sensitisation and degeneration of intervertebral discs. This provides useful evidence for further follow up as a potential target for drug development.