A Fast, Affordable, and Minimally Invasive Diagnostic Test for Cancer of Unknown Primary Using DNA Methylation...

Area of Science:

• Oncology
• Genomics
• Molecular Diagnostics

Background:

• Cancer of unknown primary (CUP) presents a diagnostic challenge, accounting for 3-5% of cancer cases.
• Tumor classification using DNA methylation profiles shows promise as a unique biological fingerprint.
• Current diagnostic methods for CUP are limited, necessitating innovative approaches.

Purpose of the Study:

• To evaluate the feasibility of cell-free reduced representation bisulfite sequencing (cfRRBS) for diagnosing CUP.
• To assess the efficacy of cfRRBS using formalin-fixed paraffin-embedded (FFPE) tissues and liquid biopsies.
• To develop a DNA methylation-based classification framework for CUP diagnosis.

Main Methods:

• Developed a cfRRBS technique for analyzing minimal amounts of fragmented DNA from FFPE and liquid samples.
• Created a custom cfRRBS reference dataset using 80 primary tumor FFPE samples and 15 healthy plasma samples.
• Built a non-negative least squares deconvolution-based classifier using entity-specific methylation regions.

Main Results:

• Accurate diagnosis was achieved for 27/30 FFPE samples (all CUPs) and 16/19 plasma samples (10/13 CUPs) with minimal DNA input (400 pg).
• Diagnosis was possible in 9/27 effusion samples with negative/inconclusive cytology, complementing routine analysis.
• Prediction accuracy was influenced by the cfDNA to high-molecular-weight DNA ratio and improved significantly with >7% predicted tumor percentage.

Conclusions:

• cfRRBS methylation profiling is a feasible, fast, and affordable method for diagnosing CUP from FFPE and liquid biopsies.
• The novel cfRRBS technique requires minimal DNA input and can be adapted for specific diagnostic needs.
• cfRRBS methylation profiling represents a valuable potential addition to the pathologist's diagnostic toolkit for CUP.