Using AI to Select Women for Supplemental MRI in Breast Cancer Screening (ScreenTrustMRI)

Image Analysis With Artificial Intelligence to Increase Precision in Breast Cancer Screening - the ScreenTrust MRI Substudy: a Prospective Trial of AI to Select Women for Supplemental Screening MRI

This is a prospective clinical trial aiming to determine the ability of an AI pipeline to identify women who would benefit from supplemental MRI in terms of decreasing the number of cancers having a significantly delayed detection

Study Overview

• Status: Active, not recruiting
• Conditions: Breast Cancer
• Intervention / Treatment: Other: AI selection for supplemental breast MRI

Detailed Description

All women attending mammography screening at Karolinska University Hospital will have their mammograms analyzed by AI (Figure 1). The specific AI-implementation (AI tool) in this study is a result of AI predictions from three equally weighted component AI models analyzing mammograms: (i) masking predictor, (ii) risk predictor and (iii) cancer signs predictor (by one commercial CAD model and one in-house academic CAD model); the age of the woman is also taken into account by multiplying the score with (110-age)/70. The purpose of the age factor is to attain a relatively similar proportion of MRI exams in the lower and higher age groups. The aim of the AI tool is to identify women with the highest probability of having a delay in cancer detection, i.e., having had a false negative screening mammogram.

An AI-based framework has been developed by researchers at Karolinska Institute (led by Dr. Fredrik Strand) and Royal Institute of Technology (led by Dr: Kevin Smith). The specific AI-implementation (AI tool) in this study is a result of AI predictions from three equally weighted component AI models analyzing mammograms: (i) masking predictor, (ii) risk predictor and (iii) cancer signs predictor (by one commercial CAD model and one in-house academic CAD model); the age of the woman is also taken into account by multiplying the score with (110-age)/70. The purpose of the age factor is to attain a relatively similar proportion of MRI exams in the lower and higher age groups. The aim of the AI tool is to identify women with the highest probability of having a delay in cancer detection, i.e., having had a false negative screening mammogram. The specific AI tool and its settings will remain the same during the study. For each examination, the AI tool will produce an AI Joint Score and an AI Masking Score. The AI Masking Score cut-off point was defined by the median of examinations collected during the initial period of March 1 to March 24, 2021. The cut-off point of the AI Joint Score was defined by the 92nd percentile of the initial population. Women meeting these criteria will be invited to the study, and randomized to MRI or no-MRI (standard-of-care).

A Signa Premier 3T MRI scanner from GE Healthcare will be used. The MRI protocol will contain a T2-weighted Dixon sequence and a T1-weighted dynamic contrast enhanced series, and will remain the same through the course of the study. All MRI exams will be assessed by two radiologists, where the second reader will have access to the assessment of the first reader. In case of disagreement, a consensus discussion between two radiologists will be held. The MRI exams will be assessed according to BI-RADS, and follow-up will depend on the BI-RADS category (Figure 2). Women with BI-RADS 1-2 will have no further diagnostics and will be sent a 'healthy letter'. Women with BI-RADS 3 to 5 will be recalled for 2nd look ultrasound. Women with BI-RADS 4-5 will be included in the regular process for established cancer suspicion and be discussed in a multidisciplinary team conference. For women with BI-RADS 3, the follow-up will be handled within the breast radiology unit.

• Study Type: Interventional
• Enrollment (Estimated): 2500
• Phase: Not Applicable

Contacts and Locations

Study Locations

• Sweden
  • Stockholm, Sweden, 17164
    • Karolinska University Hospital

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: 40 years to 74 years (Adult, Older Adult)
• Accepts Healthy Volunteers: No

Description

Inclusion Criteria:

• Complete four-view screening mammography examination

Exclusion Criteria:

• Women in surveillance program referred from the hereditary cancer unit
• Breast implants
• Prior breast cancer
• Breast feeding
• MRI contraindication requiring radiologist assessment
• AI Tool unable to process mammograms due to technical reason

Study Plan

How is the study designed?

Design Details

• Primary Purpose: Screening
• Allocation: Randomized
• Interventional Model: Parallel Assignment
• Masking: None (Open Label)

Number of Arms

2

Arms and Interventions

Participant Group / Arm	Intervention / Treatment
Experimental: Supplemental MRI; Women randomized to MRI will be examined using a shortened MRI protocol on a Signa Premier 3T MRI scanner. The MRI examination will be reviewed by two radiologists and assigned BI-RADS score. Appropriate clinical work-up will follow according to the BI-RADS score. BI-RADS 3 or higher at initial MRI will be recalled for a second look ultrasound.	Other: AI selection for supplemental breast MRI; An AI tool will generate scores used to determine eligibility. Women randomized to MRI will be examined in an MRI scanner.
No Intervention: No MRI (standard-of-care); Standard-of-care. Both arms will have had a regular screening mammography examination prior to randomization. The "No MRI" arm will have no further intervention.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Significantly Delayed Breast Cancer Detection per 1000 women	Composite end-point defined by either: 1. Interval Cancer, 2. Cancer with lymph node metastasis, 3. Cancer with invasive component larger than 15 mm	Until 27 months from study inclusion (includes cancer detected at subsequent screening within this time frame). Cancer detected at the initial screening mammography or MRI shall not be included.

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
MRI-detected breast cancer	Breast cancer detected at the initial screening MRI for women in the Intervention arm of the study	Diagnosis during work-up within 2 months of the initial screening MRI
Invasiveness	In situ and/or Invasive cancer	All diagnosed breast cancer within 27 months of study inclusion
Histology	Ductal, Lobular, Mucinous, Tubular, Other	All diagnosed breast cancer within 27 months of study inclusion
Lymph node metastasis	0 nodes, 1-3 nodes, 4 or more nodes	All diagnosed breast cancer within 27 months of study inclusion
Tumor size	Size (in millimetre) for the invasive and the in situ component	All diagnosed breast cancer within 27 months of study inclusion
Receptor status	ER positive/negative, PR positive/negative, HER2 positive/negative	All diagnosed breast cancer within 27 months of study inclusion
Age	Age of the woman	At study inclusion
Distribution of AI scores	Histogram, mean, median and dispersion measures for the AI Scores	At study inclusion
BI-RADS codes	For each MRI examination, the BI-RADS code for fibroglandular volume, background enhancement and breast lesions	At study inclusion and until end of 27 month follow-up
Biopsy result	Pathology assessment of biopsy: normal tissue, benign lesion, cancer in situ, invasive cancer	Diagnosis during work-up within 2 months of the MRI examination
Participant questionnaire	Participant questionnaire replies for MRI contraindications and for Breast cancer-related history	At study inclusion

Collaborators and Investigators

• Sponsor: Karolinska University Hospital
• Collaborators: Region Stockholm; KTH Royal Institute of Technology; Bröstcancerförbundet, Sweden
• Investigators: Principal Investigator: Fredrik Strand, MDPhD, Karolinska University Hospital

Study record dates

Study Major Dates

• Study Start (Actual): April 1, 2021
• Primary Completion (Actual): April 1, 2023
• Study Completion (Estimated): July 1, 2025

Study Registration Dates

• First Submitted: March 31, 2021
• First Submitted That Met QC Criteria: April 5, 2021
• First Posted (Actual): April 6, 2021

Study Record Updates

• Last Update Posted (Actual): October 5, 2023
• Last Update Submitted That Met QC Criteria: October 4, 2023
• Last Verified: October 1, 2023

More Information

Terms related to this study

• Keywords: Artificial Intelligence
• Additional Relevant MeSH Terms: Skin Diseases; Neoplasms; Neoplasms by Site; Breast Diseases; Breast Neoplasms
• Other Study ID Numbers: KSRAD001

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: YES
• IPD Plan Description: The final trial dataset will be available for the research team of the principal investigator. Pseudonymized data can be made available for external research audit. Anonymous data may be shared with academic researchers.
• IPD Sharing Time Frame: Available during study time and until 2 years after publication of any manuscript resulting from the study
• IPD Sharing Access Criteria: Data Transfer Agreement must be signed. Request must be made by an academic researcher at an internationally recognized university.
• IPD Sharing Supporting Information Type: STUDY_PROTOCOL; SAP; ICF

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No