Skip to main content
. Author manuscript; available in PMC: 2026 Jul 8.
Published in final edited form as: Lancet Neurol. 2026 Jun;25(6):581–590. doi: 10.1016/S1474-4422(26)00123-7

Table 3:

Effect of sentinel variants on age at onset

rsID Linear model Linear model (APOE ε4 adjusted) Kaplan-Meier Cox regression
p β (95% CI) p β (95% CI) p (log-rank) p Hazard ratio (95% CI)
CNIH4 rs74546314 0·49 2·52 (−4·53 to 9·58) 0·68 1·52 (−5·79 to 8·48) 0·90 0·84 1·08 (0·51–2·24)
CCNG1 rs537168961 0·0068 −10·15 (−17·31 to −2·77) 0·013 −9·40 (−16·67 to −2·13) 0·0009 0·0065 5·72 (1·62–20·08)
RHOJ rs141931440 0·47 −2·09 (−7·74 to 3·55) 0·56 −1·66 (−7·35 to 4·02) 0·83 0·58 1·20 (0·63–2·27)

Results from the age at onset (linear model), adjusted APOE ε4 (linear model), Kaplan–Meier survival, and Cox proportional hazard risk ratio analyses of the top risk variants. rsID=Single Nucleotide Polymorphism Database reported variant identification number.