Clinical validation of a targeted methylation-based multi-cancer early detection test using an independent validation...

Area of Science:

• Oncology
• Genomics
• Biomarker Discovery

Background:

• Multi-cancer early detection (MCED) tests aim to improve cancer detection rates and patient outcomes when used with existing screening methods.
• The Circulating Cell-free Genome Atlas (CCGA) study previously demonstrated the potential of a blood-based MCED test using cell-free DNA (cfDNA) sequencing and machine learning for cancer signal detection and origin prediction.
• This substudy focused on validating a refined MCED test version for clinical screening applications.

Purpose of the Study:

• To validate a refined version of a blood-based multi-cancer early detection (MCED) test in an independent cohort.
• To assess the specificity, sensitivity, and cancer signal origin (CSO) prediction accuracy of the MCED test as a screening tool.
• To evaluate the MCED test's performance across different cancer stages and types.

Main Methods:

• A prospective, case-controlled, observational study design was employed.
• An independent validation set of 4077 participants (2823 with cancer, 1254 without) was included.
• The study measured specificity, sensitivity, and accuracy of cancer signal detection and CSO prediction using a refined blood-based MCED test.

Main Results:

• The MCED test achieved a specificity of 99.5% for cancer signal detection.
• Overall sensitivity was 51.5%, increasing with cancer stage (Stage I: 16.8% to Stage IV: 90.1%).
• Cancer signals were detected across over 50 cancer types, with 88.7% accuracy in predicting the cancer signal origin for true positives.

Conclusions:

• The validated MCED test demonstrates high specificity and accurate cancer signal origin prediction.
• The test successfully detected cancer signals across a wide range of cancer types.
• These findings support the feasibility of using this blood-based MCED test as a complementary tool to existing single-cancer screening methods.