CRISPR-enabled biosensing of cancer-associated DNA methylation in cell-free DNA: molecular architectures, molecular sampling constraints, and translational biosensor engineering

DNA methylation in circulating cell-free DNA (cfDNA) provides cancer-associated and tissue-of-origin information complementary to mutation-based liquid biopsy. Clustered regularly interspaced short palindromic repeats (CRISPR)-based sensing is challenging because 5-methylcytosine (5 mC) generally preserves Watson-Crick base pairing and is not inherently resolved by sequence-programmed CRISPR-associated (Cas) proteins. This targeted narrative review examines how methylated and unmethylated molecules are distinguished and converted into Cas-readable outputs under fragmented, low-input cfDNA conditions. We organize the literature into four discrimination routes comprising sequence conversion, enzymatic gating, affinity-assisted recognition, and native-state Cas modulation. Engineered molecular transduction is considered separately as a downstream enabling layer. We compare readouts in terms of quantitative fidelity and integration requirements. Sequence-conversion and enzymatic-gating systems have the broadest analytical evidence among the identified studies, largely from in vitro experiments. Direct affinity-assisted 5 mC-to-Cas sensing remains sparsely studied, and evidence for native-state sensing in patient-derived cfDNA is limited to exploratory observational comparisons. No native-state Cas methylation assay identified in this review has established clinically usable diagnostic performance. Reported analytical limits of detection and methylation-fraction discrimination cannot be extrapolated to clinical sensitivity without accounting for input, recovery, fragmentation, background, and decision thresholds. Translation requires validated biomarkers, fragment-matched calibration, orthogonal reference testing, reproducible device integration, and independent evaluation of locked assays in intended-use populations.