Skip to content

Exploratory Study on Diagnostic Performance of Multi-Cancer using Machine Learning Models with Urinary Biomarkers

Exploratory Study on Diagnostic Performance of Multi-Cancer using Machine Learning Models with Urinary Biomarkers - Exploratory Study on Diagnostic Performance of Multi-Cancer using Machine Learning Models with Urinary Biomarkers

Status
Recruiting
Phases
Unknown
Study type
Observational
Source
JPRN
Registry ID
JPRN-UMIN000053883
Enrollment
900
Registered
2024-03-18
Start date
2024-04-30
Completion date
Unknown
Last updated
2026-06-29

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Lung cancer, breast cancer, colon cancer, stomach cancer, esophagus cancer, ovary cancer, kidney cancer, urothelial cancer, prostate cancer, cervix cancer, brain tumor

Interventions

None listed

Sponsors

Craif Inc.
Lead Sponsor

Eligibility

Sex/Gender
All

Inclusion criteria

Inclusion criteria: For Cancer Patients and Brain Tumor Patients To participate in this study, individuals must meet all of the following criteria (1) to (4): 1. At the time of signing the consent form, the individual must be a Japanese national aged 18 years or older. 2. Using a consent document approved by the ethics review committee of each research institution, written consent must be obtained based on the free will of the individual or their proxy for participation in this study. 3. The individual must have been pathologically diagnosed with one of the following cancers or brain tumors with a clear primary site. The histological type and stage of cancer, as well as the classification and malignancy of brain tumors, are not relevant. Lung cancer, breast cancer, colorectal cancer, stomach cancer, esophageal cancer, ovarian cancer, kidney cancer, urinary tract epithelial cancer, prostate cancer, cervical cancer 4. At the time of urine specimen collection, the individual must not have a history of treatment for cancer or brain tumors. For Healthy Adults To participate in this study, individuals must meet all of the following criteria (1) to (3): 1. At the time of signing the consent form, the individual must be a Japanese national aged 50 years or older. 2. Using a consent document approved by the Institutional Review Board (IRB) of each research institution, written consent must be obtained based on the free will of the individual. 3. Based on medical evaluation including medical history, physical examination, vital signs (blood pressure, pulse rate), and clinical laboratory tests, individuals must be determined by a physician to be in good health.

Exclusion criteria

Exclusion criteria: Individuals who meet any of the following criteria (1) to (5) at the time of specimen collection are not eligible to participate in this study: (1) Those who are pregnant or may be pregnant. (2) Women who are menstruating. (3) Those with a history of malignant tumors or brain tumors. (4) Those currently participating in intervention trials (including clinical trials) other than this study. (5) Any other individuals deemed inappropriate by the physician of the research institution.

Design outcomes

Primary

MeasureTime frame
The specimens collected in this study are allocated to training and test datasets in approximately a 2:1 ratio. Using either miRNA expression levels or DNA methylation rates as features, machine learning is employed to construct predictive models optimized for the diagnostic performance of each cancer type and brain tumor. The diagnostic performance for each cancer type and brain tumor is evaluated using statistical measures such as AUC values, sensitivity, and specificity. In the analysis based on miRNA expression levels, for cancer types currently targeted by miSignal, a comprehensive evaluation with existing predictive models (e.g., through cross-validation) is conducted. As for other urinary biomarkers, since it is in an exploratory stage, specific evaluation criteria will be established using these specimens.

Secondary

MeasureTime frame
This study is a non-interventional clinical research, and there are no health risks to the study participants. Therefore, no information regarding safety is collected.

Countries

Japan

Contacts

Public ContactMotoki Mikami

Craif inc. Clinical Development

clinicaltrial@craif.com03-6801-8334

Outcome results

None listed

Source: JPRN (via WHO ICTRP) · Data processed: Jul 3, 2026