Whole-exome sequencing of 42 adults with type 1b spontaneous spinal CSF leaks (Cedars-Sinai, 2006–2019) compared against 3 independent control cohorts (n=3,871 total controls) to identify causative genetic variants, with in vitro fibrillin-2 binding assays and CRISPR-Cas9 mouse models for functional validation.
Rare deleterious FBN2 variants were found in 9/42 (21%) patients vs. 5–8% of controls; enrichment was significant across all 3 control cohorts (OR 3.18–5.26, p=0.0003–0.041). Variants clustered in TGF-β binding protein-like (TB) domains and reduced dural fibroblast adhesion in vitro. All 3 Fbn2 mutant mouse lines showed significantly higher CSF leak rates (~73–79%) vs. wild-type (~57%, all p<0.001), while Marfan syndrome mice (Fbn1C1039G/+) showed increased meningeal compliance but no increased leak frequency.
- Small patient cohort (n=42), predominantly White women, limiting generalizability to other ethnicities and sexes. - Cases were admixed Americans; validation controls were Belgian, raising some population stratification concern despite focus on rare variants. - Mouse anatomy differs from humans (thinner dura, quadrupedal posture, no chronic pressure gradient), and the rapid leak-induction protocol cannot replicate decades-long dural compromise seen in patients.
Consider FBN2 genetic testing in patients with type 1b spontaneous spinal CSF leaks, especially those with subtle connective tissue features but no formal syndromic diagnosis. Identifying FBN2 variants may reframe these cases as fibrillinopathies and inform future targeted therapies as they are developed.
Explore related topics
