Skip to content

A prospective silent trial of using digital pathology-based artificial intelligence model to predict key genomic alterations in breast cancer to guide next-generation sequencing

A prospective silent trial of using digital pathology-based artificial intelligence model to predict key genomic alterations in breast cancer to guide next-generation sequencing

Status
Active, not recruiting
Phases
Unknown
Study type
Observational
Source
ChiCTR
Registry ID
ChiCTR2600119929
Enrollment
Unknown
Registered
2026-03-05
Start date
2026-03-05
Completion date
Unknown
Last updated
2026-03-09

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

Conditions

Breast Cancer

Interventions

NGS-recommended group:None
Non-NGS-recommended group:None

Sponsors

Fudan University Shanghai Cancer Center
Lead Sponsor

Eligibility

Sex/Gender
All

Inclusion criteria

Inclusion criteria: 1. Patients definitively diagnosed with breast cancer at our center (Fudan University Shanghai Cancer Center) who have completed or are scheduled to undergo next-generation sequencing (NGS) testing in accordance with routine clinical practice; 2. Availability of formalin-fixed paraffin-embedded (FFPE) tissue samples suitable for whole-slide imaging (WSI); 3. Availability of definitive NGS reports encompassing the mutational status of target genomic alterations of interest in this study, including germline BRCA1/2 and somatic PIK3CA, ESR1, AKT1, and PTEN mutations or copy number loss (CNL).

Exclusion criteria

Exclusion criteria: 1. Suboptimal sample quality: Tissue specimens exhibiting severe autolysis, cautery artifacts, crush artifacts, or insufficient tumor cellularity (50% of the region of interest (ROI); pathologist annotations, air bubbles, or staining artifacts obscuring >50% of the ROI; or severe blurring, out-of-focus areas, or stitching errors that compromise the evaluation of cellular morphology and tissue architecture. 3. Missing or invalid data: Cases where NGS testing fails quality control, resulting in the inability to generate a conclusive report on gene mutational status. 4. Treatment interference: Tissue samples collected following neoadjuvant therapy that induced significant morphological alterations (e.g., profound treatment-reactive stromal fibrosis or tumor cell degeneration) severely compromising feature extraction by the AI model.

Design outcomes

Primary

MeasureTime frame
Enrichment performance in the NGS-recommended population;

Secondary

MeasureTime frame
Enrichment performance in the overall model screen-positive population;Safety in the NGS-exempted population;

Countries

China

Contacts

Public ContactZhimin Shao

Fudan University Shanghai Cancer Center

zhimingshao@fudan.edu.cn+86 21 6417 5590

Outcome results

None listed

Source: ChiCTR (via WHO ICTRP) · Data processed: Mar 14, 2026