cfDNA Mutation and Methylation Detection

cfDNA Mutation and Methylation Detection

1. Background

Liquid biopsy (Liquid Biopsy) is a revolutionary technology that is profoundly transforming the landscape of basic research and clinical diagnostics. By analyzing biomarkers in body fluids such as blood and urine, researchers can obtain key information about the physiological or pathological state of the organism without the need for invasive tissue biopsies. Among these, circulating cell-free DNA (cfDNA) has become one of the most promising biomarkers in the liquid biopsy field, as it carries genetic and epigenetic information from the tissue of origin.

However, cfDNA analysis faces two core challenges. The first is the "sensitivity" problem: target cfDNA molecules are present in extremely low amounts in body fluids, and the key mutation frequencies they carry are typically far below the detection limit of conventional next-generation sequencing (NGS) (error rate ~0.1–1%). The second is the "traceability" problem: cfDNA in body fluids is a mixture of signals from different tissues and organs throughout the body. Simply detecting an abnormal signal is insufficient; the more critical question is—where does this signal come from? Is it the primary tumor, a new metastatic site, or a transplanted organ?

These two challenges are particularly prominent in the emerging field of In Vivo gene therapy. In Vivo CAR-T therapies modify T cells in vivo by direct injection of viral vectors, but the random integration of integrated viral vectors may potentially activate proto-oncogenes or suppress tumor suppressor genes, introducing insertion mutation risks. Traditional peripheral blood cell testing has a "detection blind spot" and cannot comprehensively assess integration risk and off-target effects. Through ultra-sensitive cfDNA detection, integration events in various tissues throughout the body (lymph nodes, spleen, liver, etc.) can be non-invasively monitored, high-risk integration sites (such as those adjacent to key genes like MYC and TP53) can be identified, clone abundance changes can be dynamically tracked, and abnormal amplification signals can be detected in a timely manner.

We have introduced the MethylSaferSeqS ultra-high-sensitivity detection technology based on double-stranded molecular barcoding, which can elevate mutation detection sensitivity to an unprecedented level of 0.001%. We have further integrated tissue-of-origin analysis technology based on Differentially Methylated Regions (DMR) and Differentially Methylated Sites (DMS). DNA methylation patterns are highly specific across different tissues and cell types, serving as a unique "epigenetic zip code" for each tissue. By analyzing these tissue-specific DMRs and DMSs, we can deconvolve the complex cfDNA mixture in plasma, accurately identify and quantify cfDNA fractions from different tissue sources.

Figure 1. Liquid Biopsy Concept, Technical Challenges, and Solutions

2. Technical Principles

2.1 MethylSaferSeqS Technology

The core of MethylSaferSeqS technology lies in its unique double-stranded molecular barcoding (Unique Identifier, UID) and redundant sequencing strategy, aimed at fundamentally eliminating background noise errors introduced during PCR amplification and sequencing, thereby achieving ultra-high-sensitivity detection of extremely low-frequency mutations. The technology workflow first ligates Y-shaped adapters with a unique molecular barcode to both single strands of each original cfDNA molecule in the blood. Subsequently, single-primer amplification generates copies with biotinylated labels, and streptavidin beads are used to separate the template strand from the copy strand.

The separated original strand and copy strand are then independently subjected to library construction and PCR amplification, forming two independent "UID families." After sequencing, through bioinformatics analysis, only mutations detected in the UID families of both strands (Watson and Crick strands) from the same original DNA molecule are defined as true somatic mutations. This double-verification mechanism, also known as "duplex sequencing," can reduce the sequencing error rate to less than one in ten million, thereby dramatically improving the specificity and accuracy of detection, enabling it to accurately identify true mutation events with frequencies as low as 0.001% against complex background noise.

Figure 2. MethylSaferSeqS Technology Workflow

2.2 DMR/DMS Tissue-of-Origin Analysis

After detecting mutation signals, we still need to answer two key questions: "How much tumor signal is present?" and "Where does the signal come from?" Our integrated methylation analysis module provides precise answers through two complementary strategies: Differentially Methylated Sites (DMS) and Differentially Methylated Regions (DMR).

(1)   DMS Analysis: Precise Quantification of Tumor Burden

DMS (Differentially Methylated Sites) represents the methylation difference at individual CpG sites between tumor and normal tissue, serving as the "methylation fingerprint" of tumors. By evaluating the methylation status of tumor-specific DMS, the proportion of tumor-like sites (pTS) is calculated, thereby precisely quantifying tumor burden (TC), remaining reliable even at extremely low tumor content of 0.5–1%.

(2)   DMR Analysis: High-Precision Tissue-of-Origin Tracing

DMR (Differentially Methylated Regions) are continuous regions containing multiple CpG sites that exhibit stable methylation pattern differences between different tissues. We have constructed a DMR reference atlas covering multiple tissues, tumor types, and subtypes using public databases. Through signal deconvolution algorithms, this enables: distinguishing tumor from normal tissue (ctDNA+/− classification), inferring tumor molecular subtypes (such as ER status), and tracking tissue-of-origin (quantifying the contribution ratio of cfDNA from different tissues).

3. Technical Innovation and Advantages

MethylSaferSeqS + DMR/DMS technology provides unprecedented depth and breadth for liquid biopsy research. Its core advantages lie in breaking through the limitations of traditional methods and are applicable to multiple research scenarios requiring ultra-high-sensitivity detection and precise tissue tracing:

4. Application Scenarios

4.1 Gene and Cell Therapy

(1)   In Vivo Gene Therapy Safety Monitoring:

In Vivo gene therapy (such as In Vivo CAR-T) modifies T cells in vivo by direct injection of viral vectors, but the random integration of integrated viral vectors may potentially activate proto-oncogenes or suppress tumor suppressor genes, bringing insertion mutation risks. Through ultra-sensitive cfDNA detection, integration events in various tissues (lymph nodes, spleen, liver, etc.) can be non-invasively monitored, high-risk integration sites (such as those adjacent to key genes like MYC and TP53) can be identified, clone abundance changes can be dynamically tracked, and abnormal amplification signals can be detected in a timely manner. Combined with DMR tissue-of-origin analysis, the tissue location of off-target events can be accurately identified, meeting the FDA/NMPA regulatory requirements for 15-year long-term follow-up (LTFU) of gene therapy products, providing key data support for the long-term safety of therapies.

(2)   Ex Vivo CAR-T and Other Cell Therapy Persistence Tracking:

Quantitatively monitor the dynamic changes and long-term persistence status of therapeutic cells in the body.

4.2 Oncology Research

Minimal Residual Disease (MRD) Monitoring: After anti-tumor therapy, ultra-sensitively detect residual tumor DNA in the blood, providing precise basis for evaluating treatment depth and predicting recurrence risk.

Drug Resistance Mechanism Exploration: Dynamically track low-frequency drug-resistant mutations newly emerging during treatment, revealing new mechanisms of tumor progression and drug resistance.

Tumor Heterogeneity Research: In-depth analysis of genomic differences between primary and metastatic tumors, as well as the composition of different sub-clones within the tumor. DMR analysis can distinguish cfDNA from different metastatic sites.

4.3 Transplant Medicine Research

Donor-derived cfDNA Monitoring: Through recipient-specific DMR, ultra-sensitively detect donor-derived DNA fragments after organ or bone marrow transplantation, providing new research tools for early detection of rejection reactions, infections, or disease recurrence.

4.4 Complex Disease Research

Multi-organ Injury Assessment: Simultaneously monitor liver, kidney, heart, and other organ injury conditions through organ-specific DMR, providing comprehensive molecular markers for critical illness and drug toxicity research.

Somatic Mosaicism Research: Detect and quantitatively analyze low-frequency somatic mutations that accumulate with age in tissues or blood, exploring their relationship with aging and related diseases.

5. Example Results

(1)   Low-Frequency Mutation Detection

Based on MethylSaferSeqS double-stranded molecular barcode technology, somatic mutations with frequencies as low as 0.001% are detected in plasma cfDNA. Key attention is paid to single nucleotide variants (SNV) and small insertions/deletions (Indel) in tumor-related genes (TP53, PIK3CA, ESR1, BRCA2, EGFR, etc.), providing variant allele frequency (VAF), coverage depth, and clinical significance annotations.

Figure 3. Low-Frequency Mutation Detection Results

(2)   Tumor Burden Quantification (DMS Analysis)

Using the Differentially Methylated Sites (DMS) analysis method, evaluating the proportion of CpG sites (pTS) with tumor-specific methylation patterns in plasma cfDNA to precisely quantify tumor content (Tumor Content, TC). Analysis of 245 tumor-type-specific DMS sites, calculating tumor-like site proportion and 95% confidence intervals, achieving accurate estimation of 0.5–1% low tumor burden.

(3)   Tissue-of-Origin Analysis (DMR Analysis)

Using Differentially Methylated Regions (DMR) for tissue-of-origin deconvolution analysis of plasma cfDNA. Based on a tissue-specific DMR reference atlas (856 regions) constructed from public databases such as TCGA and ENCODE, evaluating the methylation status of reads mapped to different tissue DMRs, calculating tumor sample read proportion (pTR), determining ctDNA status (ctDNA+/−), and quantifying the contribution proportion of cfDNA from each tissue source (liver, hematopoietic system, lung, tumor, etc.).

(4)   Tumor Molecular Subtype Inference

Non-invasive inference of tumor molecular subtypes from plasma cfDNA based on subtype-specific DMR. For breast cancer samples, evaluating the methylation patterns of ER (estrogen receptor), HER2 (human epidermal growth factor receptor 2), and Ki67 (proliferation index) related DMRs, calculating subtype prediction scores through signal deconvolution algorithms, and determining molecular classification (such as Luminal A/B, HER2-positive, triple-negative, etc.), providing molecular evidence for precision therapy.

Figure 4. Tumor Molecular Subtype Prediction (Subtype-specific DMR)

(5)   Key DMR Methylation Pattern Validation

Select 5–6 key tumor-related gene DMR regions (ESR1, BRCA1, TP53, RASSF1, CDH1, etc.), demonstrating their methylation levels in normal tissue reference, tumor tissue reference, and patient cfDNA samples. Through heatmap visualization and comparative analysis, the consistency of patient sample methylation patterns with tumor reference is validated, evaluating the reliability of tissue-of-origin analysis and subtype inference results.

Figure 5. Key DMR Methylation Pattern Validation

6. Service Content and Sample Requirements

6.1 Service Content

*Service Turnaround Time: Standard process 30–40 business days

6.2 Sample Requirements

6.3 Sample Information and Transportation

(1)   Information Requirements: Each sample must be clearly labeled with the sample number and accompanied by a complete submission form indicating the sample type, origin, processing method, and research purpose.

(2)   Transportation Guidelines: Please use sufficient dry ice (for frozen samples) or blue ice (for refrigerated samples) for transportation, and choose reliable courier services to ensure the stability and safety of samples during transportation.

*Note: If samples do not meet the above requirements, it may affect data quality, extend the detection cycle, or cause experimental failure. For special sample types, please be sure to contact our technical support team for communication in advance.

7. References

[1] Crowley, E., et al. (2013). Liquid biopsy: monitoring cancer-genetics in the blood. Nature Reviews Clinical Oncology, 10(8), pp.472–484.

[2] Kaiser, J. (2019). CRISPR's unwanted anniversary. Science, 365(6459), pp.1224–1228.

[3] Newman, A. M., et al. (2016). Integrated digital error suppression for improved detection of circulating tumor DNA. Nature Biotechnology, 34(5), pp.547–555.