Polygenic risk score: use in migraine research

Abstract

Background: The latest Genome-Wide Association Study identified 38 genetic variants associated with migraine. In this type of studies the significance level is very difficult to achieve (5 × 10^{- 8}) due to multiple testing. Thus, the identified variants only explain a small fraction of the genetic risk. It is expected that hundreds of thousands of variants also confer an increased risk but do not reach significance levels. One way to capture this information is by constructing a Polygenic Risk Score. Polygenic Risk Score has been widely used with success in genetics studies within neuropsychiatric disorders. The use of polygenic scores is highly relevant as data from a large migraine Genome-Wide Association Study are now available, which will form an excellent basis for Polygenic Risk Score in migraine studies.

Results: Polygenic Risk Score has been used in studies of neuropsychiatric disorders to assess prediction of disease status in case-control studies, shared genetic correlation between co-morbid diseases, and shared genetic correlation between a disease and specific endophenotypes.

Conclusion: Polygenic Risk Score provides an opportunity to investigate the shared genetic risk between known and previously unestablished co-morbidities in migraine research, and may lead to better and personalized treatment of migraine if used as a clinical assistant when identifying responders to specific drugs. Polygenic Risk Score can be used to analyze the genetic relationship between different headache types and migraine endophenotypes. Finally, Polygenic Risk Score can be used to assess pharmacogenetic effects, and perhaps help to predict efficacy of the Calcitonin Gene-Related Peptide monoclonal antibodies that soon become available as migraine treatment.

Keywords: Migraine genetics; Genome-Wide Association Studies; Polygenic Risk Score; pleiotropy; endophenotype..

Fig. 1

AVENGEME [7] was used to calculate the study power (y-axis) given different target sample sizes (x-axis) for migraine. For the calculation we used assumptions derived from Gormley et al.: Discovery cohort = 375.000; variance explained in the discovery sample = 0.1463; and a prevalence of 0.158. Further, we assumed that the fraction of NULL SNPs is 0.95 and that the outcome is binary. The effects from Gormley et al. are used to weigh the SNPs for calculating the PRS. We present six curves representing two different prevalences of migraine in the target sample (circle = 0.2 and triangle = 0.5) given three different P_T in the discovery sample (black = 5 × 10^− 8, red = 1 × 10^− 4, and blue = 0.05