Award

Optimizing detection of circulating tumor DNA for lung cancer screening

Description: Cancers evolve over multiple decades from a single cell and form masses of billions of cells before they cause symptoms leading to clinical diagnosis. Cancer screening aims to find tumors before the progress to an incurable stage and thereby can decrease cancer mortality. Lung cancer is the leading cause of cancer-related death and is expected to be responsible for the deaths of 132,000 Americans in 2021 alone. The probability to develop lung cancer during a human’s life is 6.3% (1 in 16). If lung cancer is diagnosed at an early, localized stage, the 5-year survival rate is 59% compared to 6% if it is diagnosed a distant metastatic stage. Multiple recent studies have shown that lung cancer can be detected earlier through circulating tumor DNA (ctDNA) present in small blood samples. Because of the relatively low annual incidence of lung cancer in high-risk populations (1% per year), very large (>10,000 participants) and long-winded (~10 years) randomized clinical trials will be needed to assess the potential of ctDNA-based lung cancer screening. To optimize the design and interpretation of these clinical trials, we will need comprehensive mathematical and computational models of ctDNA-based screening. To accelerate the clinical evaluation and translation of ctDNA-based screening, we propose to develop computational and mathematical methods to estimate and optimize the expected benefits and costs of ctDNA-based lung cancer screening. We will use our recently developed mathematical framework of cancer evolution and ctDNA shedding as well as DNA sequencing samples of approximately 250 lung cancer patients and of 165 healthy subjects to compute the probability that a growing tumor will be detected with a given screening program. We will perform virtual ctDNA-based screening trials across subpopulations with various lung cancer risk profiles and estimate cost-effectiveness, cancer-specific mortality reduction, and life-years gained. We will compare these estimates to lung cancer screening with low-dose computed tomography (CT) which is the currently recommended screening modality but suffers from low uptake and exposes subjects to radiation leading to 1 radiation-induced lung cancer death for every 13 lung cancer deaths averted by CT-based screening. Moreover, our preliminary analyses suggest that dynamic screening schedules could eliminate more than 30% of screening examinations making screening potentially beneficial for a much larger fraction of the US population. Mathematical models have been pivotal to guide effective cancer screening recommendations. The successful completion of this young investigator proposal will lay the foundation for a new type of mathematical models to inform and improve ctDNA-based cancer screening and will help to accomplish the American Cancer Society’s mission to free the world of cancer.
Amount: $792,000
Funding type: award
Start date: 2022-01-01
End date: 2025-12-31
Investigators: Johannes Reiter (lead), Sylvia Holland, Garry Gold, Lisa Freitas, Sylvia Plevritis
Institutions: Stanford Medicine, Stanford University, The Board of Trustees of the Leland Stanford Junior University, CA - 166/166 - Stanford University Medical Center
Subfield: Cancer Research
Funded outputs count: 0
Funder award ID: RSG-21-152-01-CDP
OpenAlex ID: G2608088720