Natural History of Leigh Syndrome: A Study of Disease Burden and Progression

Abstract

Objective: This observational cohort study aims to quantify disease burden over time, establish disease progression rates, and identify factors that may determine the disease course of Leigh syndrome.

Methods: Seventy-two Leigh syndrome children who completed the Newcastle Paediatric Mitochondrial Disease Scale (NPMDS) at baseline at 3.7 years (interquartile range [IQR] = 2.0-7.6) and follow-up assessments at 7.5 years (IQR = 3.7-11.0) in clinics were enrolled. Eighty-two percent of this cohort had a confirmed genetic diagnosis, with pathogenic variants in the MT-ATP6 and SURF1 genes being the most common cause. The total NPMDS scores denoted mild (0-14), moderate (15-25), and severe (>25) disease burden. Detailed clinical, neuroradiological, and molecular genetic findings were also analyzed.

Results: The median total NPMDS scores rose significantly (Z = -6.9, p < 0.001), and the percentage of children with severe disease burden doubled (22% → 42%) over 2.6 years of follow-up. Poor function (especially mobility, self-care, communication, feeding, and education) and extrapyramidal features contributed significantly to the disease burden (τ_b ≈ 0.45-0.68, p < 0.001). These children also deteriorated to wheelchair dependence (31% → 57%), exclusive enteral feeding (22% → 46%), and one-to-one assistance for self-care (25% → 43%) during the study period. Twelve children (17%) died after their last NPMDS scores were recorded. These children had higher follow-up NPMDS scores (disease burden; p < 0.001) and steeper increase in NPMDS score per annum (disease progression; p < 0.001). Other predictors of poor outcomes include SURF1 gene variants (p < 0.001) and bilateral caudate changes on neuroimaging (p < 0.01).

Interpretation: This study has objectively defined the disease burden and progression of Leigh syndrome. Our analysis has also uncovered potential influences on the trajectory of this neurodegenerative condition. ANN NEUROL 2022;91:117-130.

FIGURE 1

The Newcastle Paediatric Mitochondrial Disease Scale (NPMDS) ratings for each item in this cohort of children with Leigh syndrome. The percentages of children with Leigh syndrome were rated as normal (green), mild (yellow), moderate (amber), and severe (red) in selected items of the NPMDS. Baseline and follow‐up NPMDS scores were arranged alongside each other for comparisons in respective items: (A) current function, (B) system‐specific involvement, and (C) clinical assessment. *p < 0.05, **p < 0.01, ***p < 0.001. [Color figure can be viewed at www.annalsofneurology.org]

FIGURE 2

The interitem relationships of the Newcastle Paediatric Mitochondrial Disease Scale (NPMDS) items in this cohort of children with Leigh syndrome. Any positive Kendall tau‐b (τ _b) correlation coefficients are shown in darker shades. The correlation matrix at baseline (left) and at follow‐up assessments (right) were similar. CPEO = chronic progressive external ophthalmoplegia. [Color figure can be viewed at www.annalsofneurology.org]

FIGURE 3

The change of total Newcastle Paediatric Mitochondrial Disease Scale (NPMDS) scores for all participants. (A) A bee swarm plot of the total NPMDS scores for each individual patient in this study at baseline (left swarm) and at follow‐up (right swarm). Horizontal bars indicate the median scores. The median score rose from 18 at baseline to 24 at follow‐up NPMDS assessments. The difference between these two assessments was significant at p < 0.001. (B) A vector graph showing these changes according to their ages. Each individual line shows the change in NPMDS scores from baseline to follow‐up assessments for every patient in this study. [Color figure can be viewed at www.annalsofneurology.org]

FIGURE 4

The vector plots for Newcastle Paediatric Mitochondrial Disease Scale (NPMDS) scores at baseline and at follow‐up in this study based on their genotypes. The grey vector lines indicate patients who had died since the end of study. aδ represents the change of NPMDS scores per annum. Vector plots are categorized into several groups according to genotypes: (A) complex I assembly factor and structural protein variants (NDUFV1, NDUFS1, NDUFA9, NDUFA13, NDUFAF6, NDUFAF8), (B) mtDNA‐encoded complex I subunits variants (MT‐ND1, MT‐ND4, MT‐ND5, MT‐ND6), (C) pathogenic SURF1 gene variants affecting complex IV assembly, (D) MT‐ATP6 gene variants, (E) pathogenic variants affecting pyruvate dehydrogenase complex (PDHA1, PDHX), and (F) other pathogenic nuclear gene variants. Children without a known genetic diagnosis are not shown here. **p < 0.01. [Color figure can be viewed at www.annalsofneurology.org]

FIGURE 5

The Kaplan–Meier curves for survival probability from mortality and from severe disease burden (Newcastle Paediatric Mitochondrial Disease Scale [NPMDS] score > 25). (A) Survival probability for those children with disease onset at 6 months and younger. (B) Probability of NPMDS > 25 for disease onset at 6 months and younger. (C, D) Probability of SURF1 cases compared to other nuclear and mitochondrial pathogenic variants from mortality (C) and severe disease burden (NPMDS > 25; D). There were significant differences between the SURF1 cases and the other two genotypes. (E) The survival probability of different disease burden based on NPMDS scores (mild, 0–14; moderate, 15–25; severe, >25). Those children with severe disease burden (NPMDS > 25) had significantly worse survival than those who had mild or moderate disease burden. (F) The survival probability of disease progression shows that those who had a change of >3 points on the NPMDS scores per annum also had significantly poorer survival outcomes. [Color figure can be viewed at www.annalsofneurology.org]

FIGURE 6

Neuroimaging of this cohort of children with Leigh syndrome. (A) A representative neuroimaging changes in these children with Leigh syndrome: (i) medulla, (ii) cerebellar nuclei, (iii) pons, (iv) midbrain, (v) thalami, (vi) globus pallidus, (vii) putamen, (viii) caudate, (ix) white matter involvement, (x) thin corpus callosum, (xi) generalized atrophy. (B) The correlation matrix of Φ values of positive findings for each region of the brain based on the neuroimaging images of 63 patients. Green represents positive nominal‐by‐nominal association. (C) The point biserial Pearson correlation factor for each abnormal neuroimaging change against the disease burden (follow‐up Newcastle Paediatric Mitochondrial Disease Scale [NPMDS] scores), disease progression (NPMDS score change per annum), and death. Blue indicates positive correlation. *p < 0.05, **p < 0.01. [Color figure can be viewed at www.annalsofneurology.org]