|
Activity Number:
|
398
- Statistical and Computational Methods to Tackle Complex Diagnostic Challenges
|
|
Type:
|
Invited
|
|
Date/Time:
|
Thursday, August 12, 2021 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
WNAR
|
|
Abstract #319050
|
|
|
Title:
|
Stage Shift and Mortality Reduction by Screening with a Multi-Cancer Early-Detection Test: Microsimulation Modeling of a Randomized Controlled Trial
|
|
Author(s):
|
Jing Zhang* and Jerome Braun and Noah Simon and Earl Hubbell and Nan Zhang
|
|
Companies:
|
Grail Inc. and Independent Statistical Consultant and University of Washington and Grail Inc and Grail Inc
|
|
Keywords:
|
|
|
Abstract:
|
Microsimulation modeling can be an efficient and flexible alternative to randomized controlled trials (RCTs) to assess clinical utility. However, few existing models address multi-cancer detection. We built a novel microsimulation model assessing clinical utility of a methylation-based multi-cancer early detection (MCED) test and utilized Julia’s superior parallelism. We simulated the UK National Health Service (NHS) RCT piloting this MCED test. Individual cancer state transition was modeled under three different tumor growth rates. Our simulations found that detection of early stage cancers (I+II) increased from 62% to 75%+ with intervention under different tumor growth rates. Relative reduction of late stage cancers was 23%+. The power to detect such a reduction in late stage cancer (III+IV) count was 95%+. The resulting rate ratio for cancer death was ~0.80 with power to detect 90%+. The addition of an MCED test to usual care thus has the potential to shift cancer detection to earlier stages, help the NHS with the Long Term Plan of achieving 75% early stage detection, save lives and improve cancer care.
|
Authors who are presenting talks have a * after their name.
