AI-Assisted Non-Contrast CT for Multi-Cancer Screening

Official title

A Prospective Cohort Study Evaluating the Utility of Artificial Intelligence-Assisted Non-Contrast Computed Tomography for Multi-Cancer Screening in Asymptomatic Individuals Undergoing Routine Health Examinations

Overview

Cancer poses a major public health challenge in China. Early detection can improve treatment outcomes and survival rates. In this study, we will conduct a large-scale, prospective, multi-center cohort study to evaluate the utility of AI-assisted non-contrast CT for multi-cancer screening. The study aims to enroll 1 million asymptomatic participants undergoing routine health examinations, using an AI imaging model based on non-contrast CT to detect seven cancers such as lung, liver, gastric, colorectal, esophageal, pancreatic, and breast cancers. Positive cases will be required to be referred to Shanghai Changhai Hospital for further imaging and care based on National Comprehensive Cancer Network (NCCN) and American College of Radiology (ACR) guidelines. The goal is to assess the AI model's diagnostic performance for seven cancer types, especially for early-stage, resectable tumors.

Detailed description

Cancer has become a major public health issue in China, seriously affecting population health, the economy, and social development. In 2022, there were an estimated 4.82 million new cancer cases and 2.57 million cancer-related deaths. Lung cancer, liver cancer, gastric cancer, colorectal cancer, esophageal cancer, pancreatic cancer, and breast cancer are the seven leading causes of cancer-related mortality. A successful earlier detection strategy would allow patients to receive timely interventions, improve treatment outcomes, enhance overall survival, and reduce the complexity and cost of treatment.

In this study, we will conduct a large-scale, prospective, multi-center cohort study, aiming to evaluate the utility of AI-assisted non-contrast CT for multi-cancer screening. The population consists of individuals who have undergone non-contrast abdominal or chest CT scans at Meinian Onehealth Health Examination Center or Shanghai Changhai Health Examination Center, with an expected enrollment of 1 million participants. A multi-cancer screening model via non-contrast CT, developed by Alibaba DAMO Academy, will be integrated into the PACS system of health examination centers. The imaging AI model will be used to automatically detect various cancerous lesions, including lung cancer, liver cancer, gastric cancer, colorectal cancer, esophageal cancer, pancreatic cancer, and breast cancer. Subjects identified with positive lesions by the AI model will be required to be referred to Shanghai Changhai Hospital for further imaging examinations (such as contrast-enhanced CT, MRI, Endoscopy, etc.) to confirm the final disease status and formulate a treatment plan. Additionally, the medical team should follow care pathways developed based on guidelines from NCCN and ACR, and if necessary, patients will be directed to the multidisciplinary team (MDT) clinic for specific cancer types to determine the diagnostic procedures. The ultimate goal of this study is to comprehensively assess the diagnostic performance metrics of the AI model for each of the seven cancer types individually. These metrics include, but are not limited to, sensitivity, specificity, and positive/negative predictive value. Particular emphasis will be placed on evaluating the model's efficacy in detecting early-stage, resectable tumors. The overarching aim is to determine whether the implementation of this AI-assisted screening approach could potentially lead to improved overall survival rates through earlier detection and intervention.

Interventions

• Diagnostic test AI-Assisted Non-Contrast CT for Multi-Cancer Screening
  Participants identified by the AI model as having potential cancerous lesions, including those suspected of lung, liver, gastric, colorectal, esophageal, pancreatic, and breast cancer, will be required to undergo blood tests (for tumor markers) and additional imaging studies (such as contrast-enhanced CT, MRI, Endoscopy, etc.) to confirm the diagnosis of cancerous lesions.

Primary outcome measures

• Diagnostic yield [Time frame: 3 years]
• Incidence [Time frame: 3 years]
• Resectable rate [Time frame: 3 years]

Secondary outcome measures (1)

• Survival time [Time frame: 3 years]

Eligibility criteria

Inclusion criteria

• Subject is able and willing to provide informed consent and sign an informed consent form.
• Subject has undergone an abdominal or chest non-contrast CT scan.

Exclusion criteria

• Subject has been diagnosed with one of the following cancers within the last five years: lung, liver, stomach, colon, esophageal, pancreatic, or breast cancer;
• Subject has any medical condition that contraindicates high-resolution MRI/CT/Endoscopy;
• Subject cannot be followed up or is participating in other clinical trials.

Criteria are shown verbatim from the registry (in English). Final eligibility is always assessed by the study center.

Healthy volunteers: Yes

Study design

Allocation

N/A

Model

Single group

Masking

Open label

Primary purpose

Diagnostic

Study locations

China · 1 center

• Changhai Hospital — Shanghai

Publications

• Han B, Zheng R, Zeng H, Wang S, Sun K, Chen R, Li L, Wei W, He J. Cancer incidence and mortality in China, 2022. J Natl Cancer Cent. 2024 Feb 2;4(1):47-53. doi: 10.1016/j.jncc.2024.01.006. eCollection 2024 Mar. PMID 39036382
• Siegel RL, Giaquinto AN, Jemal A. Cancer statistics, 2024. CA Cancer J Clin. 2024 Jan-Feb;74(1):12-49. doi: 10.3322/caac.21820. Epub 2024 Jan 17. PMID 38230766
• Cao K, Xia Y, Yao J, Han X, Lambert L, Zhang T, Tang W, Jin G, Jiang H, Fang X, Nogues I, Li X, Guo W, Wang Y, Fang W, Qiu M, Hou Y, Kovarnik T, Vocka M, Lu Y, Chen Y, Chen X, Liu Z, Zhou J, Xie C, Zhang R, Lu H, Hager GD, Yuille AL, Lu L, Shao C, Shi Y, Zhang Q, Liang T, Zhang L, Lu J. Large-scale pancreatic cancer detection via non-contrast CT and deep learning. Nat Med. 2023 Dec;29(12):3033-304 PMID 37985692
• Schrag D, Beer TM, McDonnell CH 3rd, Nadauld L, Dilaveri CA, Reid R, Marinac CR, Chung KC, Lopatin M, Fung ET, Klein EA. Blood-based tests for multicancer early detection (PATHFINDER): a prospective cohort study. Lancet. 2023 Oct 7;402(10409):1251-1260. doi: 10.1016/S0140-6736(23)01700-2. PMID 37805216
• Gao Q, Lin YP, Li BS, Wang GQ, Dong LQ, Shen BY, Lou WH, Wu WC, Ge D, Zhu QL, Xu Y, Xu JM, Chang WJ, Lan P, Zhou PH, He MJ, Qiao GB, Chuai SK, Zang RY, Shi TY, Tan LJ, Yin J, Zeng Q, Su XF, Wang ZD, Zhao XQ, Nian WQ, Zhang S, Zhou J, Cai SL, Zhang ZH, Fan J. Unintrusive multi-cancer detection by circulating cell-free DNA methylation sequencing (THUNDER): development and independent validation stud PMID 36849097
• Klein EA, Richards D, Cohn A, Tummala M, Lapham R, Cosgrove D, Chung G, Clement J, Gao J, Hunkapiller N, Jamshidi A, Kurtzman KN, Seiden MV, Swanton C, Liu MC. Clinical validation of a targeted methylation-based multi-cancer early detection test using an independent validation set. Ann Oncol. 2021 Sep;32(9):1167-1177. doi: 10.1016/j.annonc.2021.05.806. Epub 2021 Jun 24. PMID 34176681
• Hackshaw A, Clarke CA, Hartman AR. New genomic technologies for multi-cancer early detection: Rethinking the scope of cancer screening. Cancer Cell. 2022 Feb 14;40(2):109-113. doi: 10.1016/j.ccell.2022.01.012. Epub 2022 Feb 3. PMID 35120599

Identifiers

NCT: NCT06632886 · AI-MCScreen

Primary sources (government registries)

• ClinicalTrials.gov (US NIH)

View this study on ClinicalTrials.gov ↗