Superficial siderosis (SS) is a rare neurodegenerative disease primarily characterized by cerebellar ataxia, sensorineural hearing loss, and myelopathy (1–5). SS results from chronic or intermittent micro bleeding of small intradural vessels into the subarachnoid space, leading to the release of neurotoxic heme degradation products and their accumulation in the brain and spinal cord (1, 2, 5, 6). These processes trigger oxidative stress, ferroptosis, inflammation, and subsequent neurotoxicity, ultimately causing myelopathy (2, 6). Myelopathy, affecting approximately 53% of patients, typically presents with hyperreflexia, spastic gait, and urinary dysfunction (7).

Although prior studies have demonstrated the clinical efficacy of blood–brain barrier-permeable oral iron chelators (4, 8), their impact on myelopathy remains unexplored. To objectively assess spinal tract impairment, we aimed to evaluate the functional status of the corticospinal and sensory pathways in the nervous system of patients with SS.

This was a non-randomized controlled trial conducted in accordance with the principles of the Declaration of Helsinki (2013) (9). The study protocol was approved by the Institutional Ethics Committee of the Institute of Science Tokyo Hospital. Informed consent was obtained from each patient.

We recruited patients from October 6, 2010 to April 18, 2022. Of the 62 patients with SS who visited the Institute of Science Tokyo Hospital, 12 consented to participate. Participants underwent surgical closure of their dural defects and were divided into two groups based on the treatment approach (Supplementary Table 1). The chelator administration (CA) group comprised eight patients who received deferiprone (1,500 mg/day) for 36 months, whereas the non-administration control (NC) group included the remaining four patients. Nine participants with multiple sclerosis (MS) served as controls to represent demyelinating diseases (Supplementary Figure 1).

The median age at symptom onset was 51.0 (47.0–55.5) years, and baseline evaluations were performed at a median age of 66.7 (60.3–71.8) years. Male patients comprised 92% of the study cohort. No significant differences in age, height, or motor function were observed between the CA and NC groups. The demographic characteristics were similar between the SS and MS groups, whereas the median age of the MS group was significantly younger (Supplementary Table 1).

Abnormalities in MEP latencies and prolonged central motor conduction time (CMCT) were observed in 67 and 92% of patients with SS, respectively (Table 1).

Although CMCT values exceeded the normal limits in both the SS and MS groups, significantly longer CMCT and MEP latencies were observed in the MS group (Table 2). Moreover, MEP amplitude (MEP amp) and the amplitude ratio of MEP to CMAP (MEP/Mamp) were significantly better maintained in the SS group. The rMT was significantly lower in the SS group compared with that in the MS group. CECT findings were within normal limits in both groups.

In the SEP examinations, prolonged latencies or absence of P38 were observed in 58% of patients with SS. All patients with measurable N21–P38 CCT demonstrated significantly prolonged conduction times (Table 1). The N21 latencies were preserved.

Baseline MEP parameters did not differ between the CA and NC groups. Administration of the iron chelator led to a significant improvement in the CA group after 36 months of treatment (Table 3). PMCT did not change significantly in either group. Interestingly, one patient who had no measurable MEP at the baseline regained detectable responses following the intervention (Supplementary Figure 2). In contrast, no significant changes were observed in SEP parameters in either group (Table 4).

No adverse events were observed during or after the administration of the tests.

In this study, all patients with SS exhibited prolonged central motor and/or sensory tract conduction time. These conduction abnormalities were partially recovered with chelator administration. The relatively preserved MEP amplitude suggests that the pathophysiological mechanism underlying the conduction abnormalities in SS is likely different from that of MS, a typical demyelinating inflammatory disease.

Prolonged CMCT in SS suggests corticospinal tract damage caused by hemosiderin deposition, which leads to ferroptosis, oxidative stress, and inflammation (2, 7, 8, 14). The relatively preserved MEP amplitudes observed in patients with SS suggest that fewer central motor neurons and axons are affected.

Although prolonged N21–P38 CCT and abnormal P38 latency in SEP indicate sensory conduction impairments in the central nervous system (CNS), the preserved N21 latency suggests selective involvement of later sensory components. These findings align with the TMS results and emphasize the progressive impact of hemosiderin deposition on the central conduction pathways.

Proximal peripheral nerve function was preserved in the SS group, as evidenced by the normal CECT values. Pathological studies have confirmed limited hemosiderin deposition in peripheral nerves, which is consistent with the unimpaired functional results from electrophysiology (1, 15).

The iron chelator partially improved the CMCT in the treated patients, suggesting that neural injury can be partially reversed by removing iron and other heme-degrading products. Notably, one patient who initially had no measurable MEP displayed detectable responses after the intervention, further supporting the hypothesis that the chelator enhanced motor cortical excitability and restored conduction. The chelators appear to confer these benefits by reducing iron levels in the CNS, which may alleviate oxidative stress and ferroptosis, thereby enhancing neuronal excitability (2, 7, 8). Although the chelator partially improved central motor conduction, no significant recovery was observed in somatosensory evoked potentials. This may be because the dorsal column–medial lemniscus pathway is anatomically disposed closer to the spinal cord surface than the corticospinal tract, making it more susceptible to chronic inflammation and oxidative stress caused by subpial hemosiderin deposition (1, 2, 15).

Oligodendrocytes are the primary iron storage cells in the CNS (16). Their inability to process excess iron can lead to oxidative stress, ferroptosis, and other free iron-related toxicities, disrupting axon–oligodendrocyte interactions (15, 17). Axon–oligodendrocyte exchange of metabolic and structural support is essential for maintaining axonal integrity and proper myelination; disruption of this interaction may exacerbate axonal dysfunction. Iron chelators may help protect neurons by mitigating these effects.

This study has some limitations. The small sample size, owing to the rarity of SS, limits the generalizability of the findings. Further, the regulatory restrictions on iron chelator dosing might have influenced the therapeutic outcomes. Future studies with larger cohorts and optimized dosing regimens are required to validate these findings and further explore the potential of iron chelation therapy in SS.

In conclusion, hemosiderin deposition in the CNS causes significantly prolonged conduction time. The impairment was partially recoverable by removing deposited iron.