UMIN-ICDS Clinical Trial

Public title

Validation of the Clinical Utility of the CSGSA-AI Diagnostic Model for Early Cancer Screening

Acronym

Verification of Early Cancer Detection Tests

Scientific Title

Clinical Validation Study on the Diagnostic Accuracy and Clinical Utility of an Early Cancer Detection System Combining Tumor Markers, Comprehensive Serum Glycopeptide Analysis (CSGSA), and an AI Model

Scientific Title:Acronym

CSGSA-AI Clinical Validation Study

Region

Japan

Condition

Colorectal cancer, stomach cancer, lung cancer, pancreatic cancer, ovarian cancer, endometrial cancer

Classification by specialty

Gastroenterology	Hepato-biliary-pancreatic medicine	Hematology and clinical oncology
Gastrointestinal surgery	Hepato-biliary-pancreatic surgery	Chest surgery
Obstetrics and Gynecology

Classification by malignancy

Malignancy

Genomic information

NO

Narrative objectives1

The purpose of this study is to verify the effectiveness of the CSGSA a new early detection screening test using tumor markers and serum glycoproteins, which are obtained from blood and to further improve its accuracy.

Basic objectives2

Efficacy

Basic objectives -Others

Trial characteristics_1

Exploratory

Trial characteristics_2

Developmental phase

Primary outcomes

This study evaluates the diagnostic accuracy and clinical feasibility of an early cancer detection system that integrates tumor markers, comprehensive serum glycopeptide analysis (CSGSA), and an AI model.
Data obtained in this study will be analyzed using the AI model to assess its ability to distinguish between six types of cancer (colorectal, gastric, lung, pancreatic, ovarian, and endometrial) and a healthy control group. Specifically, the following metrics will be calculated to confirm that the results are comparable to those from previous retrospective studies:

Area Under the Receiver Operating Characteristic Curve (ROC-AUC)
Sensitivity
Specificity
Positive Predictive Value (PPV)
Negative Predictive Value (NPV)

Key secondary outcomes

1) Workflow of the research model
The proportion of cases where the entire workflow is successfully completed without critical errors or missing data.
This workflow encompasses: obtaining informed consent; participant registration of background information (e.g., height, weight, medical history, family history, and current symptoms); specimen collection and transportation; laboratory testing (tumor markers and glycopeptide measurements); automated data integration into the research database; and the generation of AI-predicted results.
2) Incidence of adverse events (Safety Assessment)
The number and incidence rate of adverse events associated with blood collection in this study, such as feeling unwell, nerve injury, or difficulty achieving hemostasis.

Study type

Observational

Basic design

Randomization

Randomization unit

Blinding

Control

Stratification

Dynamic allocation

Institution consideration

Blocking

Concealment

No. of arms

Purpose of intervention

Type of intervention

Interventions/Control_1

Interventions/Control_2

Interventions/Control_3

Interventions/Control_4

Interventions/Control_5

Interventions/Control_6

Interventions/Control_7

Interventions/Control_8

Interventions/Control_9

Interventions/Control_10

Age-lower limit

20

years-old

<=

Age-upper limit

75

years-old

>=

Gender

Male and Female

Key inclusion criteria

[Control Group (Healthy Individuals)]
Individuals undergoing routine health checkups or comprehensive medical screening at the Keio University Center for Preventive Medicine.
Individuals with no prior history of cancer and no current subjective symptoms, whom a physician has determined (e.g., via medical interview) to have an extremely low likelihood of having cancer.
Individuals aged 20 to 75 years at the time of informed consent (regardless of gender).
Individuals who have provided voluntary informed consent to participate in this study, documented via electronic records or similar means.
[Case Group (Cancer Patients)]
Individuals with a presumptive diagnosis of primary colorectal, gastric, lung, pancreatic, ovarian, or endometrial cancer, based on standard diagnostic methods (e.g., a combination of endoscopy/imaging and histopathology).
Note: If a definitive diagnosis has not been established prior to the test or surgery (at the time of blood collection), blood samples will be collected from all patients with suspected cancer.
Once a definitive diagnosis is confirmed (e.g., via postoperative histopathology), malignant cases will be assigned to the Case Group, while benign cases will be classified into the Control Group.
Patients who have not yet initiated any treatment (e.g., surgery, chemotherapy, radiation therapy, or hormone therapy) for the diagnosed cancer (i.e., treatment naive patients).
Individuals aged 20 to 75 years at the time of informed consent (restricted to females for ovarian cancer; regardless of gender for other cancer types).
Patients who have provided voluntary informed consent to participate in this study, documented via electronic records or similar means.

Key exclusion criteria

Individuals meeting any of the following criteria will be excluded from this study:
Individuals with severe renal, hepatic, respiratory, or cardiac dysfunction.
Individuals with an active infection.
Individuals who have already initiated treatment.
Any other individuals deemed ineligible for participation by the principal investigator or co-investigators.

Target sample size

760

Name of lead principal investigator

1st name	Wataru
Middle name
Last name	Yamagami

Organization

Keio University, School of Medicine

Division name

Department of Obstetrics and Gynecology

Zip code

1608582

Address

Shinanomachi 35, Shinjuku-ku, Tokyo

TEL

03-3353-1211

Email

gami.z8@keio.jp

Name of contact person

1st name	Yuya
Middle name
Last name	Nogami

Organization

Keio University, School of Medicine

Division name

Department of Obstetrics and Gynecology

Zip code

1608582

Address

Shinanomachi 35, Shinjuku-ku, Tokyo

TEL

03-3353-1211

Homepage URL

Email

y-nogami.a8@keio.jp

Institute

Keio University, School of Medicine

Institute

Department

Personal name

Organization

q

Organization

Division

Category of Funding Organization

Profit organization

Nationality of Funding Organization

Co-sponsor

Tokai University

Name of secondary funder(s)

Organization

Ethics Committee, Keio University School of Medicine

Address

Shinanomachi 35, Shinjuku-ku, Tokyo

Tel

03-3353-1211

Email

keio@esct.bvits.com

Secondary IDs

NO

Study ID_1

Org. issuing International ID_1

Study ID_2

Org. issuing International ID_2

IND to MHLW

Institutions

Date of disclosure of the study information

2026

Year

08

Month

01

Day

URL releasing protocol

Publication of results

Unpublished

URL related to results and publications

Number of participants that the trial has enrolled

Results

Results date posted

Results Delayed

Results Delay Reason

Date of the first journal publication of results

Baseline Characteristics

Participant flow

Adverse events

Outcome measures

Plan to share IPD

IPD sharing Plan description

Recruitment status

Preinitiation

Date of protocol fixation

2026

Year

08

Month

01

Day

Date of IRB

Anticipated trial start date

2026

Year

08

Month

01

Day

Last follow-up date

2028

Year

03

Month

31

Day

Date of closure to data entry

Date trial data considered complete

Date analysis concluded

Other related information

Serum samples (9 mL) will be collected from the participants to measure seven established tumor markers and 1,700 serum glycoproteins using mass spectrometry (LC-MS).
The acquired data will then be analyzed using an AI model (LightGBM algorithm) to assess cancer risk.

Registered date

2026

Year

05

Month

24

Day

Last modified on

2026

Year

05

Month

24

Day

Link to view the page

Value
https://center6.umin.ac.jp/cgi-open-bin/ctr_e/ctr_view.cgi?recptno=R000070566