Skip to content

A Machine Learning–Based Risk Prediction Model for Immune-Related Cutaneous Adverse Events

A Machine Learning–Based Risk Prediction Model for Immune-Related Cutaneous Adverse Events

Status
Active, not recruiting
Phases
Unknown
Study type
Observational
Source
ChiCTR
Registry ID
ChiCTR2600125124
Enrollment
Unknown
Registered
2026-05-21
Start date
2026-06-01
Completion date
Unknown
Last updated
2026-05-25

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

Conditions

Cutaneous immune-related adverse event

Interventions

Training Set:None

Sponsors

Zhongshan Hospital, Fudan University
Lead Sponsor

Eligibility

Sex/Gender
All

Inclusion criteria

Inclusion criteria: 1. Adult patients with solid tumors who received immune checkpoint inhibitor treatment in the Department of Oncology at Zhongshan Hospital affiliated with Fudan University between January 1, 2020, and September 30, 2025; 2. Adult patients with solid tumors who received any ICI regimen (anti-PD-1/PD-L1/CTLA-4, monotherapy or combination therapy); 3. Must have laboratory test records within 24 weeks before ICI treatment (baseline) or within 2 to 12 weeks after ICI treatment (early post-treatment); 4. Must have at least 24 weeks of follow-up post-treatment, enabling the determination of skin adverse event outcomes; 5.Relevant clinical information, after data extraction and structuring, must meet the minimum key dataset requirements needed for predictive model construction, including demographic characteristics, basic tumor information, medication usage, and core laboratory indicators; 6.Previously signed informed consent for biobank sample donation, agreeing to donate their samples and data for all medical research.

Exclusion criteria

Exclusion criteria: 1. Key clinical information or laboratory test data are missing, and even after retrospective medical record review or follow-up cannot be completed, making it impossible to meet the minimum key dataset requirements necessary for predictive model construction; 2. Lack of clear follow-up records, making it impossible to determine whether immune-related skin adverse reactions occurred, or to accurately determine their time of occurrence and severity; 3.Having pre-existing active, systemically treated severe skin diseases before receiving immune checkpoint inhibitors, whose clinical manifestations are difficult to distinguish from subsequent skin adverse reactions, making it impossible to determine the occurrence or severity of immune-related skin adverse reactions.

Design outcomes

Primary

MeasureTime frame
Whether cutaneous immune-related adverse event occurred;

Countries

China

Contacts

Public ContactJi Yang

Zhongshan Hospital Affiliated to Fudan University

yang.ji@zs-hospital.sh.cn+86 180 0183 4888

Outcome results

None listed

Source: ChiCTR (via WHO ICTRP) · Data processed: May 30, 2026