Preparing cfDNA Liquid Biopsy Samples Using LASSO Capture

Team Dynamic Matrices | 25 April 2026

Cell-free DNA is one of the least forgiving sample types in molecular biology. The target fragments are short, usually 130 to 180 base pairs for the mononucleosomal peak, and they sit in a background of genomic DNA released from lysed white blood cells, plasma proteins, and whatever inhibitors happened to travel along with the blood draw. Whatever signal is present, whether that is circulating tumor DNA from a solid tumor or fetal DNA in a maternal sample, is a small fraction of the total pool. Every step in the workflow, from the collection tube to the final elution, either protects that fraction or dilutes it further. Sample preparation is not a preamble to the real work of a liquid biopsy assay. It is the assay.

Why cfDNA punishes rough handling

Most extraction chemistries were designed for genomic DNA, where yield is high and fragment length barely matters. cfDNA breaks that assumption in two ways. First, the target is already fragmented, so any further mechanical shearing or nuclease activity during processing does not just reduce yield, it can shift the size distribution away from the biologically meaningful mononucleosomal peak. Second, the background is not inert. Lysed leukocytes release high-molecular-weight genomic DNA that swamps a plasma sample within hours of a blood draw if the tube is not stabilized properly. A protocol that looks fine on paper can quietly fail because the wrong tube was used, or because plasma sat too long before separation.

This is where most cfDNA workflows lose signal before the capture step ever begins. Traditional pulldown chemistries, whether biotin-streptavidin bead systems or antibody-based enrichment, add another layer of risk on top of that fragile starting material. Beads can adsorb short fragments non-specifically, wash steps can strip weakly bound targets, and the elution conditions needed to release captured DNA from a high-affinity interaction are often harsh enough to damage what little material survived the earlier steps.

Draw tube and plasma separation

The first decision that matters is the collection tube. Standard EDTA tubes are only reliable if plasma is separated within one to two hours of the draw, because white blood cells continue to lyse in the tube and release genomic DNA that contaminates the cfDNA pool. For any workflow where same-day processing is not guaranteed, a cell-stabilizing tube (Streck, PAXgene, or an equivalent) is worth the added cost. These tubes fix the leukocyte membrane and halt nuclease activity, keeping the background DNA contribution low for several days at room temperature.

Plasma separation itself should use a double centrifugation protocol: an initial low-speed spin to pellet cells, followed by a higher-speed spin on the resulting plasma to remove residual cellular debris and platelets. Skipping the second spin is a common shortcut that leaves platelet fragments in the plasma, which release additional DNA and inflate the apparent yield with material that is not the target circulating fraction.

Extraction chemistry and what to check before capture

Once plasma is separated, extraction should be optimized for short fragments specifically, not adapted from a genomic DNA kit. Silica column and magnetic bead kits validated for cfDNA typically differ from standard kits in their binding buffer chemistry, which is tuned to retain fragments below 200 base pairs that would otherwise wash through. Before moving to any selective capture step, it is worth confirming three things on the extracted material: total yield relative to input plasma volume, fragment size distribution by capillary electrophoresis, and the ratio of the mononucleosomal peak to any high-molecular-weight contamination. A sample with a large high-molecular-weight shoulder indicates genomic DNA contamination from cell lysis, and no downstream capture chemistry can fully separate that background from a target that overlaps it in size and sequence space.

Where selective capture changes the equation

Beyond extraction, many cfDNA workflows also need a targeted capture or enrichment step, whether that is hybridization-based panel capture for a gene set of interest or a broader effort to deplete unwanted background before sequencing. This is where affinity-based pulldown chemistries, whether biotin-streptavidin or antibody-based, tend to reintroduce the same problems that careful extraction just solved. Off-target genomic DNA fragments that share partial sequence homology with the target region are frequently co-captured, since broad affinity tags do not discriminate finely between similar sequences. And releasing bound material from a high-affinity interaction usually requires denaturing or high-salt elution conditions that can further fragment already-short cfDNA, undermining the size distribution that downstream library prep and variant calling depend on.

For applications like non-invasive prenatal testing or ctDNA monitoring in oncology, where the fraction of interest can be well under one percent of total cfDNA, this loss is not a marginal issue. It is often the difference between a variant call that clears the detection threshold and one that gets lost in background noise. This is why we built LASSO, a capture platform based on dynamic DNA crosslinking rather than fixed high-affinity binding. Because target capture and release both happen under mild, native conditions, off-target co-capture is reduced and elution does not require the harsh conditions that damage fragile cfDNA, which is a meaningful advantage for labs pairing careful extraction with a downstream selective capture step.

Practical checklist before capture

A few habits consistently separate reliable cfDNA capture runs from noisy ones. Process plasma within the stabilization window of the collection tube in use, and never refreeze and rethaw plasma more than once before extraction, since freeze-thaw cycles fragment DNA further and can lyse residual cells. Quantify input material with a fluorometric assay rather than absorbance alone, since cfDNA concentrations are often too low for A260 readings to be reliable, and always run a size distribution check before committing a sample to a full capture and library prep workflow. Finally, keep a genomic DNA contamination metric as a standing QC gate rather than an occasional spot check, since background creep is gradual and easy to miss until a batch of samples fails downstream.

The bigger picture

Liquid biopsy assays are judged on their limit of detection, and limit of detection is set as much by sample preparation as by the sensitivity of the downstream sequencing or PCR chemistry. Getting the collection tube, plasma separation, and extraction chemistry right protects the fragile signal cfDNA carries, and choosing a capture or enrichment method that is gentle enough to preserve fragment integrity gives every subsequent step in the pipeline a cleaner starting point to work from.