The chromosomes stayed whole

Earlier attempts to edit human embryos with CRISPR-Cas9 kept running into the same wreckage: the enzyme cut both strands of the DNA, and the cell's repair sometimes deleted whole chunks of a chromosome. Dieter Egli's lab at Columbia used a different tool — a base editor that flips one DNA letter from A to G without ever making a double-strand break. The embryos grew to the blastocyst stage with no detectable large deletions, and the researchers pulled stem-cell lines from them that carried the intended edits.

"I would not call it a breakthrough, and it does not establish germline editing as safe." — Krishanu Saha

The buried detail is what the lab chose to fix. The edited sites — one in a cholesterol gene, one tied to fetal hemoglobin — were healthy, ordinary versions, not disease mutations. So this is not a demonstration of curing a sick embryo; it is a demonstration that the machinery works, run on genes that needed no fixing. One base-editing co-inventor called it a gateway to editing embryos for enhancement.

The limits are real: many embryos were mosaic, a patchwork of edited and unedited cells; off-target rates depended on the guide; and delivering the editor as RNA killed the embryos outright, so only a protein form survived. No embryo was implanted or carried toward a pregnancy. But the harder fact sits outside the biology. This crossed a US academic lab with almost no binding oversight — a 'gentleman's agreement' rather than a law — and it was privately funded, because federal money can't touch embryo-destroying research. A consumer-genomics company is reported to be lined up to pay for what comes next.