Size-Coded Hydrogel Microbeads for Extraction-Free Serum Multi-miRNAs Quantifications with Machine-Learning-Aided...

Area of Science:

• Biomedical Engineering
• Molecular Diagnostics
• Cancer Research

Background:

• Accurate lung cancer subtype classification is crucial for treatment and prognosis.
• Current liquid biopsy methods for microRNA (miRNA) detection require labor-intensive RNA extraction, impacting accuracy.
• There is a need for simplified, accurate methods for miRNA detection in liquid biopsies for lung cancer diagnostics.

Purpose of the Study:

• To develop an extraction-free method for quantifying specific microRNAs (miRNAs) associated with lung cancer directly from serum.
• To evaluate the utility of size-coded hydrogel microbeads combined with machine learning for lung cancer diagnosis and subtype classification.

Main Methods:

• Development of size-coded hydrogel microbeads immobilized with miRNA capture probes for direct serum miRNA quantification.
• Utilized porous microbead structure for efficient DNA cascade amplification and fluorescence signal generation.
• Employed flow cytometry for size-based microbead sorting and machine learning for data analysis of 108 serum samples.

Main Results:

• Achieved extraction-free quantification of specific miRNAs (miR-21, miR-205, miR-375) directly from serum.
• Demonstrated good accuracy in diagnosing lung cancer.
• Attained 80% accuracy in classifying lung cancer subtypes using machine-learning-assisted analysis of microbead fluorescence data.

Conclusions:

• Size-coded hydrogel microbeads offer a promising approach for simplified and accurate miRNA detection in liquid biopsies.
• This novel method facilitates improved lung cancer diagnosis and subtype classification.
• The extraction-free, machine-learning-assisted strategy holds potential for advancing non-invasive cancer diagnostics.