Mary Black

Ph.D., M.S.

Foresite Labs

VP, Head of Human Genomics

PLCO (Learn more about this study)

PLCO-1356

Oct 16, 2023

Genomic Discovery: Therapeutic Targets for Non-Cancer Conditions

Naturally occurring human genetic variants that impact the expression or function of a given therapeutic target can be used to predict the efficacy and/or safety of that therapeutic. Nelson et al. 2015 and King et al. 2019 have shown that drug targets with human genetic evidence are two to three times as likely to result in an FDA approved therapeutic than targets without this support. Ochoa et al. 2022 examined therapies approved by FDA in 2021 and found 65% were supported by genetic evidence of drug target effects. This evidence can, in part, be generated from well-designed human genetic studies in large cohorts or case-control sets.

This application will utilize all available Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial datasets to support ongoing research in genomic discovery and validation of therapeutic targets and biomarkers across a wide range of adult diseases. Analyses include genetic association (common- and rare-variant, variant-set, gene-based, etc.), fine-mapping, co-localization, Mendelian randomization, and other approaches in combination with data from internal and publicly available databases, as well as development and validation of polygenic risk score (PRS) approaches applicable to multiple ancestries. Phenotypes of interest include non-cancer phenotypes (eg. obesity: BMI, weight, waist-hip comparison; asthma; hypertension; heart attack; stroke; osteoporosis; arthritis) reported by questionnaire. We have applied for genetic data available under dbGaP Study Accession: phs001286.v3.p2, comprised of array data on ~110K participants of the PLCO study. In the present application, we are requesting demographic, medication exposure, and behavioral information collected in the baseline, medication use, supplemental, and brief (follow-up) questionnaires in order to account for confounders and assess both genetic and non-genetic stratifiers of drug target gene effects.

All analyses derived from the requested datasets will only be used in aggregated form and will abide by dataset-specific limitations. Access to the data will be limited to the collaborators named in our application (i.e. our staff) and will abide by all applicable data use certifications. The proposed research does not involve patient identification and is consistent with the data use agreement and consent requirements.

The specific Aims of the project are as follows:

Aim 1: Test association between disease-specific PRS (eg. BMI PRS) and disease/trait assessed at baseline and follow-up time points (eg. BMI or obesity).
Aim 2: Identify pharmacomimetic variants (eg. functional variants in drug target genes) and test their associations with diseases/traits assessed at baseline and follow-up time points.
Aim 3: Assess whether individuals with 'high background genetic risk' defined by the distributions of each PRS are likely to benefit more than all-comers from specific therapies based on genetic evidence for drug target modulation.
Aim 4: Examine whether the results in Aim 3 additionally vary by demographic and comorbid features of PLCO participants.