FIGURE 1.

Cortical regions with a significant interaction between EYO and mutation carrier status in relationship to image intensity metrics in the baseline cohort. Panels (a) through (d) demonstrate the negative log10 of P‐value for the interaction term between EYO and mutation carrier status in the linear model predicting image intensity metrics. This value is multiplied by the sign of the beta coefficient of the interaction term, so that areas in which mutation carriers demonstrated increasingly lower image intensity metrics with advancing EYO are coded as Cyan and areas with increasingly higher image intensity metrics with advancing EYO are coded as Orange. The colors intensities are inversely proportional to the P‐value of the interaction. Panels (e) through (h) demonstrate the negative log10 of P‐value for the ANOVA test comparing the models with (Model 3) and without (Model 1) the quadratic term of EYO^2. As a result, areas in which the addition of the quadratic EYO² term improved the prediction of image intensity metrics are shown with different intensities of Green. Panel (i) is example of a region where there was no interaction between mutation carrier status and EYO in predicting T1‐μ. Panel (j) is example of a region with a significant linear interaction between EYO and mutation carrier status and no improvement in the model with the addition of the EYO² term in predicting T1‐σ. Panel (k) is example of a region with a significant improvement in model prediction with the addition of the EYO² term over the linear term in predicting FLAIR‐μ. EYO, estimated years to onset of symptoms; image intensity metrics, image intensity metric based on mean or standard deviation of the intensity in the cortical regions in the T1 or FLAIR images (T1‐μ, T1‐σ, FLAIR‐μ, and FLAIR‐σ).