Highly sensitive CRISPR-based method designed to detect low-frequency cancer mutations

Liquid biopsy is increasingly recognized as a promising tool for cancer detection and treatment monitoring, yet its effectiveness is often limited by the extremely low levels of tumor-derived DNA circulating in the blood.

To address this challenge, researchers led by Professor Junseok W Hur from Korea University College of Medicine, in collaboration with multiple partners, have developed MUTE-Seq, a highly sensitive CRISPR-based method designed to detect cancer mutations present at exceptionally low frequencies while simultaneously reducing sequencing costs and background error noise. The study was first made available online on 21 August 2025 and was later published in Volume 37, Issue 47 of Advanced Materials in 27 November 2025, where it was selected as the journal's front cover in recognition of its excellence.

At the core of this approach is FnCas9-AF2, an engineered high-fidelity CRISPR enzyme designed to recognize and discriminate even single-base mismatches with exceptional precision. The variant shows near-zero off-target activity and selectively cleaves perfectly matched wild-type DNA, thereby enriching the relative proportion of circulating tumor DNA before sequencing. This enrichment allows rare variants to rise above the intrinsic noise that commonly limits next-generation sequencing. As Prof. Hur explains, "Our findings suggest that the MUTE-Seq method has considerable potential for developing diagnosis panels aimed at detecting multiple low-frequency ctDNA for MCED, CDx, or MRD monitoring."

In performance evaluations using Sanger sequencing and next-generation sequencing, MUTE-Seq increased variant allele frequencies by up to tens of times, enabling detection of mutations present at approximately 0.005%, a level typically obscured by baseline sequencing error rates. In patients with acute myeloid leukemia, the method clearly identified minimal residual disease by amplifying weak NRAS mutation signals that are ordinarily undetectable. When applied in multiplex mode across multiple clinically relevant cancer hotspots as EGFR and KRAS, MUTE-Seq improved concordance between plasma and tumor tissue in patients with non-small cell lung cancer and pancreatic cancer, including early-stage cases in which ctDNA levels are extremely low.

Additional validation using cell-free DNA reference materials demonstrated substantial gains in sensitivity-ranging from twenty- to sixtyfold-while maintaining high specificity. Probit analysis determined a limit of detection of 0.034% variant allele frequency using 50 ng of input DNA, closely matching theoretical expectations based on Poisson distribution models. "Results suggested that even minute mutations can be detected using MUTE-Seq if they are present in blood samples," concluded Prof. Hur.

These results highlight the broad potential of MUTE-Seq to strengthen the accuracy of liquid biopsy testing. The method's ability to remove wild-type DNA before sequencing effectively acts as a noise-reduction step that can be integrated into standard laboratory workflows, with or without unique molecular identifiers. By elevating true mutation signals above background noise, MUTE-Seq offers a more reliable path for multi-cancer early detection (MCED), minimal residual disease (MRD) monitoring, and tracking of treatment-emergent resistance mutations. Together, these advantages position MUTE-Seq as a scalable and clinically adaptable tool with significant promise for improving precision oncology.