Skip to Main Content

An official website of the United States government

Evaluating risk factor-based strategies for oversampling of individuals at higher risk of cancer for bioscpecimen collection and prospective assessment of Multi-Cancer Detection assays

Principal Investigator
Name
Filip Pirsl
Degrees
Ph.D., Sc.M.
Institution
National Cancer Institute
Position Title
Postdoctoral Fellow
Email
About this CDAS Project
Study
PLCO (Learn more about this study)
Project ID
PLCO-1704
Initial CDAS Request Approval
Oct 11, 2024
Title
Evaluating risk factor-based strategies for oversampling of individuals at higher risk of cancer for bioscpecimen collection and prospective assessment of Multi-Cancer Detection assays
Summary
Multi-Cancer Detection (MCD) tests are a promising tool for the early detection of cancer, however, prospective studies of their performance are rare. MCD tests require large quantities of blood relative to amounts usually collected in cohort studies and such collections are expensive, making it infeasible to collect biospecimens from all participants in a cohort. Consequently, a prudent approach is to collect biospecimens for prospective evaluation of MCD on a subsample of participants. The purpose of this project is to develop and evaluate methods to efficiently sample a subset of cohort participants to be invited to provide additional biospecimen(s) for prospective assessment of MCD tests in the Connect for Cancer Prevention cohort.
In this project, we will apply various methods to sample individuals into a subcohort based on PLCO questionnaires and evaluate performance of approaches using outcome data. Some sampling strategies may be designed to target cancer types with common risk factors or specific cancer types and will be evaluated with respect to targeted cancers as well as all cancers. Findings obtained from PLCO data will be complemented by similar evaluations in other cohort data in the consideration of approaches for implementation in Connect.
Aims

The aim of this project is to develop and evaluate risk factor-based approaches to sample a subcohort of participants for prospective assessment of Multi-Cancer Detection tests to: a) minimize the size of the subcohort to reduce the number of participants on whom additional burdens are placed and to reduce study costs associated with biospecimen collection, storage, and assaying, among others; b) maximize the proportion of cancer cases among participants sampled into the subcohort to optimize statistical power, representativeness, and external validity (given constraints imposed by (a)); and c) maintain simplicity in the interest of communication to participants selected into the subcohort and invited to provide additional biospecimen(s), and ease of implementation in settings other than the Connect cohort.

Collaborators

Hormuzd Katki, Ph.D.; National Cancer Institute
Nicolas Wentzensen, M.D., Ph.D.; National Cancer Institute