Isolation and ddPCR analysis of cell-free DNA from plasma for lung cancer research

Abstract:

This study presents a fast and straightforward method for isolating circulating cell-free DNA (cfDNA) from plasma samples using the Ultra pure MagPrep^® Cell-Free DNA Isolation Kit. Following isolation, digital droplet polymerase chain reaction (ddPCR) analysis was employed to detect lung cancer mutation targets, utilizing a thermal cycler for precise PCR reactions.

Section Overview

• Introduction
• Principle
• Materials & Methods
• Results
  • cfDNA Binding Capacity
  • Synthetic Matrix
  • Benchmark against competitors – cfDNA purification from sample matrix
  • Analysis in Complex Matrices
  • Benchmark against competitors – cfDNA purification from sample matrix (plasma)
  • Clinical Validation
• Discussion
• Related Product Category
• References

The isolation and analysis of cfDNA were successfully performed from synthetic matrices as well as control plasma spiked with cfDNA reference standards. Sensitivity and specificity were demonstrated by detecting lung cancer mutations, including PIK3CA-E545K, EGFR-L858R, and EGFR-T790M, with a limit of detection (LOD) of 0.1% allele frequency (AF) from 0.5 mL spiked plasma.

The fraction of positive droplets was analyzed using a droplet reader, confirming the presence of target DNA. Clinical validation of this rapid method, based on magnetized silica particles, was performed on an automated system (Kingfisher) using 0.5 mL pooled plasma from stage 4 breast cancer and stage 4 prostate cancer patients, successfully confirming the EGFR mutation at 1.37% AF. By employing TaqMan probes and Poisson statistics for absolute quantification, we achieved accurate gene expression analysis of target molecules.

These results demonstrate that the Ultra pure MagPrep^® Cell-Free DNA Isolation Kit efficiently facilitates high-throughput cfDNA purification for downstream applications, significantly reducing processing time compared to traditional column-based methods. This innovative approach underscores the potential of liquid biopsy techniques for cancer biomarker detection and enhances methodologies in cancer research.

Introduction

Cell-Free DNA (cfDNA) Isolation and ddPCR Analysis for Lung Cancer Mutations: Utilizing PCR Reactions, TaqMan Probes, and High-Throughput Techniques

Circulating cell-free DNA (cfDNA) is released into body fluids from cancer-associated necrotic tissues or apoptotic cells that shed their genomic DNA or release DNA-containing exosomes. DNA is usually degraded into mono- or di-nucleosome size fragments usually with a characteristic size of 170-180 bp protected by histones. Liquid biopsies readily employ cfDNA isolation from these fluids and characterization of mutations present within the purified cfDNA. By screening for characteristic biomarkers and epigenetic modifications, such as DNA methylation, researchers can enhance early cancer detection, optimize treatment choices, and monitor drug resistance.

Lung cancer remains a leading cause of death globally, with the American Cancer Society estimating that in 2023, it will account for approximately 238,340 new cases and around 127,070 deaths in the U.S. Consequently, monitoring lung cancer mutations in cfDNA purified from plasma is increasingly adopted in liquid biopsy techniques for non-invasive early cancer screening. Mutations in the epidermal growth factor receptor gene (EGFR) are common in non-small cell lung carcinomas, with EGFR-L858R recognized as a primary driver mutation in lung adenocarcinomas. This mutation stabilizes EGFR autophosphorylation, activating signal transduction cascades that lead to cell proliferation, invasion, metastasis, and angiogenesis. Tyrosine kinase inhibitors (TKIs) are the first-line treatment for patients diagnosed with these mutations, although resistance often develops, notably in the form of the EGFR-T790M mutation. Therefore, it is crucial to monitor several actionable cancer mutations through robust screening of purified cfDNA using advanced downstream applications such as digital droplet PCR (ddPCR), multiplex qPCR panels, next-generation sequencing (NGS), and others.

Given the challenges of conducting invasive biopsies post-treatment, developing less-invasive genetic tests that utilize peripheral blood containing tissue-derived cfDNA is imperative. As circulating cfDNA levels can be limited in peripheral blood, efficient isolation procedures must be coupled with sensitive detection systems, like ddPCR, to enable accurate diagnosis of cancer-specific mutations.
In this study, we employed the DNA-free Ultra pure MagPrep^® cfDNA Isolation Kit for both manual and high-throughput cfDNA isolation from plasma samples. This kit is also applicable for other biological fluids, such as urine, cerebrospinal fluid (CSF), and saliva. The buffers in the kit are designed to maximize the performance of magnetized silica particles, resulting in a rapid and straightforward protocol for cfDNA isolation. These buffers are compatible with automation systems, such as Kingfisher, and the user-friendly protocol does not require Proteinase K or heating steps, with all components shipped and stored at ambient temperature.

All kit components, including MagPrep^® particles and buffers, are rigorously tested to ensure ultra-purity, being free from any Nickase, DNase, RNase, and, most importantly, DNA-free, as verified by qPCR for 16S and 18S RNA genes and plasmid origins. The DNA-free feature of the Ultra pure MagPrep^® Kit guarantees reliable results in downstream applications without the risk of false positives(DNA-free) or negatives (nuclease-free). The purified cfDNA is devoid of any PCR inhibitors, making it suitable for various downstream applications, including qPCR, ddPCR, NGS, and bisulfite sequencing. We will demonstrate the successful isolation and detection of three lung cancer-associated mutations (PIK3CA-E545K, EGFR-L858R, and EGFR-T790M) using ddPCR with cfDNA purified from spiked plasma and patient-derived samples.

Principle

Enhancing Lung Cancer Mutation Detection Using Magnetic Particles and Advanced PCR Techniques

The MagPrep^® magnetic particles provide a straightforward, cost-effective, and rapid method for isolating nucleic acids, specifically circulating cell-free DNA (cfDNA), from plasma samples. During the isolation process, the magnetized silica particles are incubated with a matrix containing cfDNA and MagPrep^® Binding Buffer. This binding process allows the cfDNA to attach to the particles, while unwanted materials are efficiently washed away using a magnet. The isolated cfDNA is then eluted with MagPrep^® Elution Buffer, ensuring high purity and quality for downstream applications
This method can be automated or performed manually in under 30 minutes, enabling high-throughput cfDNA purification without the presence of PCR inhibitors. The MagPrep^® silica particles, with a particle size of 100-200 µm, exhibit exceptional binding efficiency for both RNA and DNA, facilitating reliable isolation of target DNA sequences. The magnetic clearance of the suspension takes less than 15 seconds. Thus, these properties make MagPrep^® particles particularly useful for high-throughput applications in automation systems.

Materials and Methods:

Isolation of cfDNA

Cell-free DNA (cfDNA) isolation was performed using 0.5 mL of either synthetic matrix I (Horizon Discovery, HD817), synthetic matrix II (Horizon Discovery, HD917), or plasma (BioIVT) spiked with a Multiplex cfDNA reference standard (Horizon Discovery, HD 870). The isolation procedure followed the protocols outlined in the Ultrapure MagPrep^® Cell-Free DNA Isolation Kit (Millipore Sigma, 10113) user guide, as well as the QIAamp MinElute ccfDNA kit (Qiagen, 55204) and the MagMAX Cell-Free DNA Isolation Kit (Thermo/ABI, A29319) user guides. Plasma samples from stage 4 breast cancer and stage 4 prostate cancer patients were pooled (BioIVT) to ensure a representative analysis of cancer-associated cfDNA.
Human genomic DNA (Promega G304A) was sonicated in a polystyrene clear tube (TPX plastic) using a water-based sonicator (Diagenode Bioruptor). The sonication process consisted of four cycles at high intensity, with 30 seconds ON and 30 seconds OFF, for a total of 15 minutes per cycle, while fresh ice was added to the water bath after each cycle. This method effectively reduced the genomic DNA to cfDNA size (approximately 173 bp), which was subsequently verified using a Bioanalyzer.
The cfDNA purification was executed using the Kingfisher Flex automation system, ensuring high-throughput capabilities for efficient processing.

ddPCR Analysis

Purified cfDNA was subjected to digital droplet polymerase chain reaction (ddPCR) analysis using the QX200 system, coupled with an automated droplet generator from Bio-Rad. Each ddPCR reaction was set up with 8 µL of purified cfDNA, 11 µL of 2X ddPCR Supermix for Probes (No dUTP) (Bio-Rad, 186-3024), and 1.1 µL of a 20X mutation-specific assay mix, which included proprietary formulations from MilliporeSigma for EGFR-L858R and PIK3CA-E545K mutations, and the EGFR-T790M assay from Bio-Rad (assay ID: dHsa MDV2010019).

The PCR cycling parameters were optimized as follows: Initial denaturation at 95°C for 10 minutes.
40 cycles of denaturation at 95°C for 30 seconds and annealing/elongation at 60°C for 1 minute for EGFR-L858R and EGFR-T790M, while PIK3CA-E545K was set at 55°C for 1 minute.
A final extension step at 98°C for 10 minutes.
Post-amplification, samples were analyzed using the Bio-Rad droplet reader, with the data processed using the QX Manager 1.2 software. MLH1 assay was used as published.

Results

cfDNA Binding Capacity

The cfDNA binding capacity of MagPrep^® particles was determined to be 2 µg/mg, demonstrating their efficiency in isolating target DNA sequences. Increasing amounts of sonicated human DNA were spiked into 0.5 mL of plasma and incubated with 0.5 mg of MagPrep^® particles. The isolated cfDNA was then analyzed using ddPCR with MLH1 and PIK3CA gene assays. The positive droplet counts for both assays indicated that the cfDNA binding capacity maxed out at 2 µg/mg of particles. Further increases in DNA input to 3.5 µg/mg did not yield higher event counts, and at 4 µg/mg, a reduction in event counts was observed, suggesting that excess DNA inhibits the binding capacity.

Synthetic Matrix

Initial proof-of-concept (POC) and cfDNA isolation protocol optimization experiments were conducted using synthetic matrix I, which was spiked with known lung cancer mutation reference standards (Horizon Discovery HD817).

Sensitivity:
The sensitivity of the method was demonstrated by successfully detecting the PIK3CA-E545K mutation at a 0.1% allele frequency (AF) using 8 µL (10 ng) of purified cfDNA extracted from a 0.5 mL synthetic matrix I spike-in sample (Fig. 2).

Specificity:
Specificity was confirmed as no positive events were detected from the control (healthy) matrix sample (wildtype in Fig. 2).

Benchmark against competitors – cfDNA purification from simple matrix:

The performance of the MagPrep^® cfDNA Isolation Kit was benchmarked against competitor-Q, a column and magnetic beads-based system, demonstrating comparable efficacy in cfDNA purification from a simple matrix. However, competitor-T’s magnetic beads-based kit did not perform well in this synthetic matrix I, as mentioned on the Horizon Discovery website (Table 1). All three kits exhibited a limit of detection (LOD) of 0.1% allele frequency (AF) when tested in Horizon Discovery’s synthetic matrix II (HD917), with results not shown here.

Analysis in Complex Matrices

In addition to synthetic matrices, we performed cfDNA isolation from human healthy pooled plasma (BioIVT) spiked with a Multiplex cfDNA reference standard.

Sensitivity: The sensitivity of the MagPrep^® method demonstrated by ddPCR analysis was confirmed by detecting three different lung cancer mutations—PIK3CA-E545K (Fig. 3), EGFR-L858R (Fig. 4), and EGFR-T790M (Fig. 5)—at a 0.1% AF using 8 µL (10 ng) of purified cfDNA extracted from 0.5 mL plasma spiked with the Multiplex cfDNA reference standard set (Horizon Discovery HD780, 160 ng).

Specificity: The specificity of the assay was demonstrated as no ddPCR positive events were detected from the control (healthy) plasma samples (WT in Figs. 3-5).

Benchmark against competitors – cfDNA purification from complex matrix (plasma):

The performance of the Ultra pure MagPrep^® cfDNA Isolation Kit was rigorously evaluated in comparison to competitor-Q, a column and magnetic beads-based system, and competitor-T, which is also magnetic beads-based. The ddPCR analysis of purified cfDNA demonstrated that the MagPrep^® kit is equally effective as competitor-Q while exhibiting greater sensitivity than competitor-T. This comparison was substantiated through the analysis of three different lung cancer mutation targets, with results summarized in Tables 2-4.

Table 2 presents the sensitivity (LOD) comparison of cfDNA isolation kits for the PIK3CA-E545K ddPCR assay. cfDNA was purified from 0.5 mL of plasma (BioIVT) spiked with cancer mutations (HD780, 160 ng) at various allele frequencies (AF): 0% (wildtype), 6.3%, 1.3%, and 0.13% AF. Both the MagPrep^® and competitor-Q kits successfully detected the PIK3CA-E545K mutation at 0.13% AF, while competitor-T’s kit failed to do so (as indicated in red). All three kits effectively detected mutations spiked at 6.3% and 1.3% AF, with data demonstrating high reproducibility across three experiments (N=3).

Table 3 outlines the cfDNA isolation kits' sensitivity (LOD) comparison for the EGFR-L858R ddPCR assay. The same procedure was followed, utilizing 0.5 mL of plasma spiked with cancer mutations (HD780, 160 ng) at various allele frequencies (AF): 0 % - WT, 5 %, 1 % and 0.1 % AF. The MagPrep^® and competitor-Q kits detected the EGFR-L858R mutation at 0.1% AF, while competitor-T’s kit showed limited detection capability (circled in red). Purified cfDNA was screened for the lung cancer mutation EGFR-L858R by ddPCR. All kits successfully identified mutations spiked at 5% and 1% AF, reinforcing the robustness of the results (N=3).

Table 4 provides a sensitivity (LOD) comparison for the EGFR-T790M ddPCR assay. cfDNA was purified using either the MagPrep^® cfDNA isolation kit, competitor-Q, or competitor-T kit from 0.5 mL of plasma (BioIVT) spiked with cancer mutations (HD780, 160 ng) at various allele frequencies (AF): 0 % - WT, 5 %, 1 % and 0.1 % AF. Purified cfDNA was screened for the lung cancer mutation EGFR-T790M by ddPCR. All three kits successfully detected the EGFR-T790M mutation at a LOD of 0.1% AF, with results showing high reproducibility (N=3).

Clinical Validation

Clinical validation was performed using pooled plasma samples from stage 4 breast cancer and stage 4 prostate cancer patients. cfDNA isolation was conducted using the Kingfisher automated system, demonstrating the efficiency of the Ultra pure MagPrep^® cfDNA Isolation Kit (Fig. 6).

Discussion

In this study, we successfully isolated cell-free DNA (cfDNA) from spiked plasma and pooled plasma samples from breast and prostate cancer patients using the Ultra pure MagPrep^® cfDNA Isolation Kit, employing both manual and automated protocols. The ddPCR analysis results of the purified cfDNA demonstrated exceptional sensitivity, achieving a limit of detection (LOD) of 0.1% allele frequency (AF). This level of sensitivity is comparable to traditional methods, which are often more time-consuming and labor-intensive.

The MagPrep^® kit notably outperformed another magnetic beads-based kit, yielding high-quality cfDNA that is compatible with sensitive DNA amplification techniques such as digital droplet PCR (ddPCR). The use of a droplet generator in the ddPCR setup allowed for precise control of the PCR reaction, enabling absolute quantification of target DNA sequences Furthermore, the high-quality cfDNA obtained from the Ultra pure MagPrep^® kit was suitable for various downstream applications, including next-generation sequencing (NGS), microsatellite instability (MSI) assessment, tumor mutational burden (TMB) evaluation, minimal residual disease (MRD) monitoring, DNA methylation analysis, and miRNA profiling. These applications underscore the versatility of cfDNA isolation in cancer biomarker detection and highlight the potential of liquid biopsy techniques in advancing cancer research methodologies. The automated system (Kingfisher) streamlined the cfDNA purification process, significantly enhancing throughput and efficiency. By minimizing the time required for sample processing, we have established a robust workflow that can be integrated into clinical settings for routine plasma sample analysis.

In conclusion, our results affirm the effectiveness of the Ultra pure MagPrep^® cfDNA Isolation Kit in isolating high-quality cfDNA suitable for ddPCR analysis, paving the way for improved detection of lung cancer mutations and other critical biomarkers in plasma samples.