SEP 30, 2020 9:30 AM PDT

Common risk variants and complex disorders

Speaker

Abstract

Successful sequencing of the human genomic brought great expectations for precision medicine. However, early efforts to understand common diseases under the “one gene; one disease” paradigm resulted in widespread failures in finding simple answers to complex questions, such as “the missing heritability of complex diseases.” This dilemma is part of a larger problem, that of Western medicine built on the germ theory of disease. In this paradigm, cases and controls are defined by the presence or absence of advanced clinicopathologic criteria – a crude “top-down” approach based on clinical features rather than etiology. The solution to improving the management of human disease must begin within a new paradigm of medicine (e.g. a bottom-up mechanistic definition of Precision Medicine), to address the descriptions and uses of genetic data based on mechanistic principles.

The ACMG/AMP and others have defined “pathogenic” genetic variants based on levels of evidence that a variant is associated with a Mendelian disorder (PMID:25741868). However, these high-impact variants tend to be rare, and are only responsible for a subset of common disorders. The challenge is in understanding how to use more common genetic variants that are clearly associated with disease (e.g. in a GWAS), but are neither sufficient nor necessary to cause that disease! We argue that continuing with a population-based, top-down approach alone cannot resolve this question of the “missing heritability” because there are too many variables for too few homogenous patient subsets.

Understanding how common variants contribute to disease demands that genomics, cell biology, physiology, pathophysiology, and medicine be integrated into epidemiology-based disease models. Using mechanistic modeling, we can view risk variants as “pathogenic” with variable impacts and effect sizes under specific conditions. This approach facilitates the determination of complex disease mechanisms in an individual patient.