People with one broken LRRK2 copy are healthy

The research record

This is the project's file for this landmark, as written and checked. Sources are listed at the end; anything not yet confirmed is marked "Not yet verified".

Who

Nicola Whiffin (first author, then Broad Institute and Imperial College London), Eric Minikel, Daniel MacArthur (last author) and colleagues, with the gnomAD consortium, UK Biobank and the company 23andMe (Nature Medicine 26:869-877, published online 27 May 2020). The Europe PMC record lists NIH grants among the funders; I did not read the full funding statement.

What happened

Drugs that block LRRK2 (landmark 7) rest on a worry. Animals lacking LRRK2 showed changes in the lung and kidney, and monkeys given inhibitors showed lung cell changes (landmark 199). Nobody knew what lowering LRRK2 would do in people.

Nature does this experiment by chance. Some people carry a "predicted loss-of-function" (pLoF) variant: a typo in the gene that should stop it making working protein. The team searched three large datasets:

• gnomAD: 141,456 people with sequenced genomes or exomes
• UK Biobank: 49,960 people with sequenced exomes
• 23andMe: over 4 million people with genotype data

After strict checking of each variant they found 1,455 carriers with high-confidence pLoF variants. Lab tests on three variants, plus earlier work, showed reduced LRRK2 protein in 82.5% of the carriers. In the paper the variants were heterozygous, meaning one copy of the gene was affected. The authors found these variants were not strongly associated with any specific disease or trait.

Why it moved us toward a cure

It answered the question left open by the animal studies, with human data. Carrying less LRRK2 protein for a whole life did no obvious harm. This supported the safety case for LRRK2 inhibitors and antisense drugs, which were then moving into trials (landmark 110, landmark 145). It sits beside the lung changes seen in primates, which are covered in landmark 199, and helped justify giving these drugs to people.

Outcome

Advanced. The evidence is about carriers with one broken copy for a whole life. It does not show what happens with strong, near-complete drug blockade, and the biobank health records may miss mild or late effects. Later trials still had to test safety directly. The paper has an author correction (Nat Med 2021;27:355), which added a missing consortium author and a funding line; the findings were not changed. Two comments on it appeared (Nat Rev Genet 2020; Mov Disord 2020;35:1754).

Sources

• Whiffin et al., Nat Med 2020, DOI 10.1038/s41591-020-0893-5 (abstract read via Europe PMC, PMID 32461697) · Broad Institute record
• Author correction, Nat Med 2021;27:355, DOI 10.1038/s41591-020-01185-6 (text read at PMC8095267)
• Not yet verified: the phenotype tests and effect sizes in the full text; carrier ages; whether any carriers had both copies affected; how complete the biobank health records are; the full funding statement.