(SSC) Cerebrovascular Section Best Basic Scientific Paper Abstract Award Resident/fellow (2023 Award Winner): Prevalence of Somatic Activating KRAS Mutations in Pediatric and Adult Sporadic Brain Arteriovenous Malformations

Introduction: Sporadic brain arteriovenous malformations (bAVMs) are a potentially treatable cause of stroke disproportionately affecting young people. Non-inherited somatic activating mutations in KRAS have been reported in ~50% of bAVM specimens primarily from adults. We hypothesized that KRAS mutations would be associated with an earlier age at diagnosis, larger bAVM size, or earlier time to hemorrhage.

Methods: Sporadic bAVM tissue and clinical data were collected from patients seen at our institution. Genotyping was performed by digital droplet polymerase chain reaction to detect KRAS mutations (p.G12D, p.G12V or p.Q61H) in three batches and coded as presence/absence of any KRAS mutation (primary predictor). Age at diagnosis was dichotomized into adults (≥18 years) or children ( < 18 years). Regression analyses adjusting for genotyping batch were performed to test association of KRAS mutations between children and adults (logistic), with bAVM size (linear), or with time from diagnosis to hemorrhage (survival), censoring at treatment or last follow-up.

Results: We analyzed data from 221 patients: median age at bAVM diagnosis was 20 years; 44% were diagnosed as children; 53% were female; and 56% were ruptured on presentation. Median bAVM diameter was 2.1 cm (IQR: 1.4 – 3.0). KRAS mutations were detected in 53% of samples. Childhood bAVMs were significantly more likely to harbor KRAS mutations than those diagnosed in adulthood (OR=2.20, 95% CI: 1.24 – 3.91, p=0.007). KRAS-mutant bAVMs tended to be larger than KRAS-wildtype bAVMs, but this did not reach significance (+0.29 cm, 95% CI: -0.03 – 0.62, p=0.08). No association was observed for time to hemorrhage (HR=1.31, 95% CI: 0.91 – 1.88, p=0.15).

Conclusion : Somatic activating KRAS mutations are more prevalent in bAVMs diagnosed in childhood than adulthood. Further work is required to elucidate mechanisms of mutagenesis and bAVM progression, which may differ in children and adults.