Faecal mtDNA heteroplasmy as a non-invasive biomarker in m.3243 A > G mitochondrial disease

Background Quantifying mitochondrial DNA (mtDNA) heteroplasmy is critical for diagnosis, prognosis, and monitoring of mitochondrial disease, particularly that caused by the m.3243 A > G pathogenic variant, where tissue-specific heteroplasmy contributes to disease expression. However, current methods often rely on invasive or clinic-based sampling, limiting accessibility and repeat testing. We validate a non-invasive, decentralised approach using faecal-derived mtDNA to quantify m.3243 A > G heteroplasmy. Methods Fifty participants with genetically confirmed m.3243 A > G mitochondrial disease were recruited remotely and sent a postal stool collection kit. DNA was extracted from returned stool samples and heteroplasmy at nucleotide position m.3243 A > G within MT-TL1 was quantified using pyrosequencing. Where available, stool mtDNA heteroplasmy was compared with previously collected measurements from muscle, blood, urine, and buccal swabs. Associations were assessed using Pearson correlation, agreement between measurements was evaluated using Bland-Altman analysis, and diagnostic performance for identifying high muscle heteroplasmy was assessed using receiver operating characteristic (ROC) analysis. Results Here we show that 47 of 50 participants return a sample and 44 of 47 yield an analysable result. Faecal m.3243 A > G and heteroplasmy correlates with muscle heteroplasmy ( r = 0.638, R 2 = 0.407, p = 0.0141). For identifying individuals with a high muscle heteroplasmy ( ≥ 60%), a faecal cut-off of ≥56.4% provides 83.3% sensitivity, 50% specificity, and a positive predictive value (PPV) of 90.9%, with high overall accuracy (area under the curve (AUC) 0.92). Conclusions We demonstrate that faecal mtDNA heteroplasmy is a promising biomarker and validate the feasibility of remote, postal home sample collection with central laboratory analysis aligned with decentralised precision medicine. This approach may reduce reliance on clinic-based sampling, broaden access to laboratory testing, and enable longitudinal monitoring and remote data collection in decentralised clinical trials.