Measuring Life: From Single Cells to Liquid Biopsies

# Liquid biopsy: reading the body's molecular debris Watch the video first. Use this companion to revisit the reasoning and its evidence limits. It can. As cells die and are cleared, their DNA is chopped into short pieces, and some fragments circulate in the blood. This is cell-free DNA: short DNA fragments outside intact cells. The pieces tend to be nucleosome sized, because DNA wrapped around protein spools is protected from cutting. Nearly every organ contributes, so blood holds a molecular mixture from many cells. The key move is counting. A sequencer reads millions of fragments, so it can act as an enormous molecular counter. Instead of asking whether some DNA is present, ask how many fragments carry this chromosome, this sequence, or this mutation. Pregnancy is the classic example. In 1997, Dennis Lo's group found fetal DNA in the mother's plasma. Most of this so-called fetal cell-free DNA actually comes from the placenta, which usually shares the fetus's genome. It isn't intact fetal cells; it's fragments, mixed in with the mother's own DNA, which makes up the large majority. Only a little. With a tenth of the fragments placental, the extra copy raises chromosome twenty one's share by only about five percent. And chromosome twenty one is already a small slice of the genome. So the signal is a slight overrepresentation, hidden in a huge count. To see a five percent shift, you have to beat counting noise. If you count N independent molecules, the random sampling noise shrinks roughly as one over the square root of N. Count a hundred fragments from chromosome twenty one, and the noise is about ten percent: too big. Count ten thousand, and it's about one percent. More molecules counted means smaller noise, and smaller shifts you can detect. In 2008, Christina Fan, Quake and colleagues did this by shotgun sequencing cell-free DNA from maternal blood, about five million sequence tags per sample. In a small study of eighteen pregnancies, they identified every case of trisomy twenty one, eighteen and thirteen. That same year, Rossa Chiu, Dennis Lo and colleagues independently published a similar counting approach. Tests built on this principle are now offered around the world, and the lab estimates that in 2014 about a million women avoided invasive procedures such as amniocentesis. But clinically, cell-free DNA testing is screening, not diagnosis. A high-risk result is confirmed with diagnostic testing, such as amniocentesis or chorionic villus sampling. This is science, not medical advice. In 2011, Snyder, Quake and colleagues saw donor DNA rise during biopsy-proven rejection in heart transplant recipients. A 2014 prospective study of 65 patients found the signal performed comparably to the biopsy itself. So donor DNA is a molecular window on graft injury, but it doesn't replace every biopsy in every setting. ## Evidence guide ESTABLISHED MOLECULAR BIOLOGY: cfDNA is fragmented extracellular DNA, with placental contribution to maternal plasma. QUAKE-LAB PRIMARY RESULT · SMALL EARLY STUDY: Fan 2008 tested 18 pregnancies; independent Lo/Chiu contributions are credited. CLINICAL APPLICATION · SCOPE: prenatal screening is not diagnosis; transplant monitoring does not replace every biopsy. Educational science, not medical advice. Sources: [lo1997], [alberry2007], [fan2008], [devlaminck2014], [acog2020]. See the course bibliography and claim audit.