Harding disease is a condition in which demyelinating neurological symptoms similar to multiple sclerosis (MS) coexist with Leber hereditary optic neuropathy (LHON) due to mitochondrial DNA (mtDNA) mutations. It was named after Harding et al., who reported in 1992 eight women with bilateral optic neuropathy and a family history of LHON, six of whom had neurological symptoms consistent with MS. 1)
A total of 88 cases have been reported in the literature. 1) Of these, 70.4% (62 cases) were women, with a male-to-female ratio of 2.38:1. 1) This contrasts with LHON, which predominantly affects men (93.1% male), suggesting that the female predominance in MS may play a role. The mean age of onset is 30.5 years. 1)
The prevalence of LHON is estimated at approximately 1 in 50,000 in Japan (with a total of about 4,000–5,000 patients nationwide), 1 in 31,000 in the UK, and 1 in 68,000 in Australia. 1) In 2015, LHON was designated as an intractable disease in Japan, and diagnostic criteria were established. The number of new cases is about 117 per year, with 47% occurring before the age of 30.
Acute Phase
Optic disc redness and hyperemia: Accompanied by peripapillary RNFL edema.
Retinal capillary dilation: Capillary dilation (telangiectasia) and increased vascular tortuosity are observed.
No leakage on fluorescein angiography: No leakage of fluorescent dye from the hyperemic optic disc. This is an important distinguishing feature from optic neuritis.
OCT findings: Optic disc swelling is reportedly present before the onset of symptoms.
Chronic phase
Optic disc atrophy: Disc atrophy remains after inflammation subsides.
RNFL thinning: OCT shows thinning of the inner retinal layers, mainly the papillomacular bundle.
Decreased best-corrected visual acuity: Often remains around 0.01. Light perception is preserved.
Preservation of the pupillary light reflex: Compared to other optic nerve diseases, the pupillary light reflex is preserved or only mildly impaired.
Differences from LHON include multiple episodes of visual loss and a longer interval before the second eye is affected (mean 1.66 years). 1) Differences from optic neuritis in MS include absence of eye pain and no leakage of dye from the optic disc on fluorescein angiography. These features provide important clues for differentiating Harding disease.
Harding disease is caused by pathogenic mutations in mitochondrial DNA (mtDNA). The three major mutations all encode complex I subunits, inducing apoptosis of retinal ganglion cells (RGCs) through impaired ATP synthesis and increased reactive oxygen species (ROS). 3)
The distribution of mutations in 88 cases of Harding disease is shown below.
| Mutation | Gene | Proportion in Harding disease |
|---|---|---|
| m.11778G>A | MT-ND4 | 69.3% (61/88 cases) |
| m.14484T>C | MT-ND6 | 12.5% |
| m.3460G>A | MT-ND1 | 10.2% |
Citation: Alorainy J et al. 20241)
The MT-ND4 mutation accounts for 90% of LHON cases in Asia and 70% in Europe.3) These three gene mutations also account for 95% of cases in Japan.
mtDNA is passed from mother to child (maternal inheritance). It is not passed on to the offspring of male patients.
The following three hypotheses have been proposed for the mechanism of coexistence of LHON and MS.1)
LHON is male-predominant (93.1% male), but 70.4% of patients with Harding disease are female. This is related to the fact that MS is more common in women. It has been hypothesized that genetic and environmental factors predisposing to MS trigger LHON in asymptomatic mtDNA mutation carriers. 1) Estrogen signaling may also be involved in the pathogenesis of LHON. 2)
Harding disease is defined as “a patient who meets the MS diagnostic criteria and has a major LHON mutation” (Pfeffer et al.).
In Japan, the diagnostic criteria for LHON established in 2015 are applied.
| Test | Main purpose/findings |
|---|---|
| mtDNA analysis / multi-gene panel | Identification of m.3460G>A, m.11778G>A, m.14484T>C |
| MRI | Confirmation of demyelinating T2 hyperintense white matter lesions |
| Fluorescein fundus angiography | Presence or absence of dye leakage from the optic disc (non-leakage is a distinguishing feature) |
| Optical coherence tomography (OCT) | Thinning of the papillomacular bundle |
| Electrocardiography (ECG) | Exclusion of pre-excitation syndrome |
Differentiation from toxic optic neuropathy, autosomal dominant optic atrophy (ADOA), occult macular dystrophy, cone dystrophy, compressive optic neuropathy, and neuromyelitis optica spectrum disorder (NMOSD) is necessary.
No established treatment exists. The current approach is described below.
While standard treatment has not been established, the following treatments have been attempted.
There is no established treatment, and symptomatic therapy is the mainstay. Some patients personally import idebenone, which is not approved domestically, for oral use, but evidence of its efficacy in Harding disease is lacking. 4)Currently, supplements such as CoQ10 and vitamin B complex, smoking cessation guidance, low vision care, and genetic counseling are provided.
The basic mechanism of LHON originates from impairment of complex I (mitochondrial respiratory chain) subunits due to mtDNA mutations. Impaired electron transport reduces ATP synthesis and simultaneously accumulates reactive oxygen species (ROS). This induces apoptosis of RGCs, leading to optic atrophy. 3)
There are differences in visual recovery rates depending on the mutation. The visual recovery rate for the MT-ND4 mutation is 4–25%, which is lower than that for MT-ND1 and MT-ND6 mutations. 3) In Japan, the most common mt11778 (MT-ND4) mutation has a visual improvement rate of only a few percent. The mt14484 (MT-ND6) mutation has the highest improvement rate.
The higher penetrance of LHON in males suggests the involvement of an X-linked nuclear gene. 3)
In the histopathology of LHON-MS, T cells and activated macrophages/microglia mediate tissue damage. The presence of inflammatory cells within LHON lesions is unusual and suggests an early immunological mechanism. 1) White matter changes involve not only MS-like demyelination but also vacuolation and diffuse myelin pallor. Mitochondrial dysfunction, autoimmune response, and molecular mimicry are hypothesized mechanisms of inflammatory demyelination. 1)
Neurological sequelae range from mild impairment to a course resembling relapsing-remitting MS.
Gene therapy (lenadogene nolparvovec; rAAV2/2-ND4) has been developed for LHON patients with MT-ND4 mutations. It introduces wild-type ND4 protein into the mitochondria of RGCs through allotopic gene expression. 2)
In the REVERSE study (2019), mean BCVA improvement of -0.308 LogMAR was reported in the injected eye and -0.259 LogMAR in the untreated eye. 2) This result suggested immunological transfer to the contralateral eye. Pooled analysis (RESCUE, REVERSE, RESTORE, REFLECT trials) showed approximately 21.5 ETDRS letter improvement compared to the control group. Bilateral administration was more effective than unilateral (approximately 12 vs 8 ETDRS letters), and meta-analysis showed rAAV2/2-ND4 was more effective than idebenone, both being more effective than natural history. 2)
Currently not approved by EMA or FDA, and efficacy for Harding disease has not been verified. While effects on the LHON component are expected, the impact on MS pathology is unclear. 2)
A method of intravitreal administration of iPSC-derived MSCs to directly transfer mitochondria to RGCs is being studied. F-actin-dependent tunneling nanotubes (TNTs) mediate the transfer. Prevention of RGC density reduction has been reported in Ndufs4 KO mice, but clinical trials have not yet been conducted. 3)
It has been suggested that 4-aminopyridine may improve visual evoked potentials (VEP) in MS, and its application to Harding disease patients with poor visual prognosis is being considered.
Development of biomarkers specific to Harding disease and prospective studies are needed. Treatment for this disease, which has the dual pathology of LHON and MS, should be individualized, and multidisciplinary collaboration is essential.
rAAV2/2-ND4 is a gene therapy for LHON patients with MT-ND4 mutations and is expected to be effective for the LHON component. 2) However, its efficacy for Harding disease has not been verified, and it is currently not approved by the EMA or FDA. Its impact on MS pathology is also unclear, and careful consideration is needed for its application to this disease.
