This pilot study assessed whether urine cell‑free DNA (cfDNA) methylation profiling can serve as a noninvasive diagnostic and surveillance tool for non‑muscle invasive bladder cancer (NMIBC) in a Thai patient cohort. The authors used cell‑free methylated DNA immunoprecipitation followed by sequencing (cfMeDIP‑seq) to characterize methylation differences between NMIBC patients and healthy controls.
The cohort comprised five patients with high‑grade NMIBC and three healthy volunteers. Midstream urine and blood samples were collected from all participants prior to transurethral resection of bladder tumor (TUR‑BT). The report specifies enrollment numbers and timing of sample acquisition relative to TUR‑BT but does not provide additional demographic breakdowns beyond case/control status in the abstract.
Urine cfDNA was extracted from midstream urine. The isolated cfDNA showed an atypical peak around 400 bp and was mechanically sheared to an optimal target fragment length of 150–200 bp prior to library preparation. Methylation enrichment and profiling were performed using cfMeDIP‑seq, a capture‑based approach designed for fragmented cell‑free DNA that avoids bisulfite conversion.
Sequencing data were analyzed to detect differentially methylated regions (DMRs) between NMIBC samples and healthy controls. Statistical thresholds were applied to control for multiple testing; the report provides adjusted P‑value cutoffs for called DMRs. Principal component analysis (PCA) was used to evaluate sample clustering based on methylation profiles.
In the initial comparison including all eight samples, the authors identified 82 significant DMRs at adjusted P < 0.001. Of these, 76 regions were hypermethylated in NMIBC versus controls and 6 were hypomethylated. Top hypermethylated genes reported in that analysis included IRX4, TBX5, and MIR124‑3.
Following removal of two outlier urine samples (labelled UC2 and UC3) to reduce cohort heterogeneity, a refined subset analysis returned a larger set of significant DMRs: 476 regions (449 hypermethylated). In this refined analysis, PCA showed clear segregation between NMIBC cases and controls, and MIR124‑3, IRX4, and LHX1‑DT emerged as the most significantly hypermethylated targets. The authors interpret these loci as promising, demographic‑specific biomarker candidates for detection and surveillance of NMIBC in their study population.
The study reports that two urine samples behaved as outliers (UC2 and UC3). Excluding these samples produced a stronger and more extensive methylation signature distinguishing cases from controls, suggesting that cohort heterogeneity and individual sample quality can materially affect cfDNA methylation analyses in small cohorts. The presence of outliers and the effect of their removal are explicitly documented in the abstract.
The authors highlight that cfMeDIP‑seq on mechanically sheared urine cfDNA bypasses the degradative limits associated with traditional bisulfite conversion approaches for methylation analysis of fragmented or degraded DNA. They propose that this workflow is technically validated within the context of this pilot and could form a foundation for noninvasive surveillance and diagnostic workflows for NMIBC, contingent on further validation in larger cohorts and independent studies.
The study protocol was approved by the Institutional Review Board of the Faculty of Medicine, Chulalongkorn University (IRB No. RA64/053). Participants provided informed written consent. The study was registered with the Thai Clinical Trials Registry under TCTR20210813003. The authors state that raw cfMeDIP‑seq datasets have been deposited in the NCBI Sequence Read Archive (SRA) and will be provided upon manuscript acceptance; additional minimal clinical and baseline data necessary to replicate findings are included in the manuscript and supporting information.
Key limitations apparent from the report include the small sample size (five NMIBC cases and three controls) and the impact of cohort heterogeneity as evidenced by two outlier urine samples. The manuscript is presented as a preprint and has not undergone peer review; therefore, findings should not be used to guide clinical practice without further validation. The authors suggest that the identified candidate loci—MIR124‑3, IRX4, LHX1‑DT, and TBX5—warrant additional investigation in larger, independent cohorts to confirm diagnostic performance and generalizability.
Overall, the study demonstrates that urine cfDNA methylation profiling by cfMeDIP‑seq can distinguish NMIBC from healthy controls in this pilot Thai cohort and identifies several hypermethylated targets as candidate biomarkers, while acknowledging the need for expanded validation.