Identification of biomarkers in patients with mitochondrial disease using non-invasive imaging
Hossain C1,2, Hall J2,3, Ng Y1,3, Bright A1,3, Blamire A4,5, Turnbull D1,2,3, McFarland R1,2,3, Gorman G1,2,3
1 Wellcome Trust Centre for Mitochondrial Research, Newcastle University, Newcastle upon Tyne, UK
2 Institute of Health and Society, Newcastle University, Newcastle upon Tyne, UK
Heteroplasmy levels of the pathogenic m.3243A>G mutation differ between individuals and tissues and are not well correlated with clinical outcome. We compared heteroplasmy levels from three commonly sampled tissues (blood, urine and muscle) in 236 patients from the UK MRC Mitochondrial Disease Patient Cohort Study, evaluating which most reliably predicted disease burden and progression using linear and mixed models.
All heteroplasmy measures were highly correlated (R2=0.57-0.74). Urine levels showed a clear effect of sex, with males having 20% higher levels than females, leading us to develop a method to adjust levels for sex. Adjusted levels represent a surrogate for muscle levels, although do show higher intra-individual variation than blood and unadjusted urine levels.
In agreement with previous reports, we showed that blood levels decline exponentially by 1.85+/-1.15% year-1 (N=35), however, this model is a poor fit to our data (R2=0.06). Modelling the exponential decline using adjusted urine heteroplasmy levels (N=232) produced age-adjusted m.3243A>G blood levels linearly-correlated with muscle levels.
Blood heteroplasmy levels were better predictors of disease burden and progression than urine levels. However, variation explained by heteroplasmy, age and sex is low with R2 values for disease burden ranging from 0.20 (adjusted blood) to 0.12 (urine). Individuals with high levels tend to have a higher disease burden and progression, but there is high inter-individual variation, implying that unidentified factors heavily influence clinical outcome. Adjusted blood and urine levels were just as good predictors as muscle levels, suggesting overall that blood is a better tissue than urine for predicting disease burden and progression in m.3243A>G carriers when adjusted for age.