The Clinical Spectrum of Mitochondrial POLG Mutations

Abstract & Commentary

By Claire Henchcliffe, MD, DPhil, Assistant Professor, Department of Neurology, Weill Medical College, Cornell University. Dr. Henchcliffe is on the speaker's bureau for GlaxoSmithKline, Teva/Eisai, and Boehringer Ingelheim

Synopsis: POLG mutations result in an expanding array of phenotypes, from childhood encephalopathy to milder, later-onset of combinations of ataxia, epilepsy, myopathy, and progressive external ophthalmoplegia.

Sources: Horvath R, et al. Phenotypic Spectrum Associated with Mutations of the Mitochondrial Polymerase Gamma Gene. Brain. 2006;129:1674-1684; Tzoulis C, et al. The Spectrum of Clinical Disease Caused by the A467T and W748S POLG Mutations: A Study of 26 Cases. Brain. 2006;129:1685-1692.

The clinical heterogeneity of most mitochondrial disorders presents a diagnostic challenge to clinicians and geneticists. Two European studies have now identified the largest cohorts, to date, of individuals with mutations in the mitochondrial DNA polymerase gamma gene (POLG). Horvath and colleagues selected a large patient collection for genetic testing from several diagnostic centers based upon clinical findings of progressive external ophthalmoplegia (PEO), ataxia, or Alpers syndrome (children with developmental delay, refractory seizures, and liver failure), in the context of biochemical evidence of compromised mitochondrial function. They then carefully characterized those individuals who tested positive for POLG mutations. Age of symptom onset ranged from 6 months to 63 years. Despite overlap in presentations, 3 clinical groups were identified: a) childhood onset of fluctuating encephalopathy hepatopathy (n = 14, 90% male); b) PEO with limb myopathy, often in association with ataxia and sensorimotor neuropathy, usually with adult onset (n = 19, 53% male); and c) adult onset myopathy or ataxia without PEO (n = 5). Of those with childhood onset encephalopathy, 10/14 developed liver failure. Importantly, abnormal liver function followed sodium valproate treatment for seizures in 5 of these children, and mortality was high. No relevant family history could be elicited in 66% of those identified with POLG mutations.

Overall, a total of 89 different mutations were found in 38 patients. The A467T mutation was found in 12/19 cases presenting in childhood. In a complementary study focusing on specific POLG mutations (A467T and W748S), Tzoulis and colleagues describe clinical features in 26 patients belonging to 20 families, mostly from Norway. Age of symptom onset was more restricted (2-36 years). However, phenotypes overlapped with the prior study. Initial manifestations were epilepsy (n = 13), headaches (n = 7), and progressive gait unsteadiness (n = 6). Most developed a combination of epilepsy (n = 22) with status epilepticus, resulting in death in 9. They also developed headaches (n = 23), ataxia (n = 23), sensorimotor neuropathy (n = 25), and ptosis or progressive external ophthalmoplegia (n = 12). Liver failure was a cause of death in 2 and, as in the prior study, was associated in some with administration of sodium valproate. Brain MRI was abnormal in the majority, with high signal lesions on T2, or FLAIR sequences present variably in the occipital lobes, deep cerebellar nuclei, thalamus, and basal ganglia.


The POLG mutations described lead to a broad clinical picture ranging from a severe and rapidly progressive hepatocerebral syndrome with high early childhood mortality, to more slowly progressive and milder adult syndromes, such as isolated neuropathy. Although we have yet to understand the full spectrum of this disorder, both papers provide important new descriptive information and, impressively, discern patterns of symptoms that provide diagnostic clues. In addition, Horvath et al suggest a diagnostic algorithm to aid clinicians in navigating their way to a referral for molecular diagnosis. In this algorithm, it is suggested that children with classical Alpers syndrome be screened specifically for the A467T mutation, reserving more extensive sequencing for those who test negative in at least one POLG allele. In the absence of liver failure, or in more complex multi-system disorders, Horvath et al suggest muscle biopsy to quantify mitochondrial DNA prior to considering genetic testing. In adults, the algorithm is less helpful, with suggested muscle biopsy for those with "PEO/unexplained multi-system neurological disorder or isolated neurological disorder with a relevant family history," an unfortunately broad array of disorders. However, in adults, any combination of ataxia, PEO, myopathy, or sensorimotor neuropathy without other explanation should raise clinical suspicion of mitochondrial disease, such as an underlying POLG mutation. Brain MRI abnormalities may provide a tip off, and then following up, as Horvath et al suggest, with muscle biopsy, regardless of MRI results, may be indicated. Ultimately, diagnosis will depend upon an appropriate level of clinical suspicion, making detailed clinical descriptions, such as those presented in these papers, invaluable.