Autopsy Study Demonstrates Tofersen Distribution and SOD1 Reduction in SOD1-ALS

An autopsy case series of individuals with SOD1 amyotrophic lateral sclerosis (SOD1-ALS) who received tofersen demonstrated that the antisense oligonucleotide (ASO) reached somatic motor system tissues throughout the central nervous system and was associated with substantial reductions in SOD1 protein levels in the spinal cord. Overall, the findings provide the first emerging human postmortem evidence confirming the predicted distribution of tofersen and demonstrating robust SOD1 protein reduction in human motor system tissues.¹

"This case series presents the first emerging autopsy tissue data confirming the predicted distribution of tofersen and robust SOD1 protein reduction in human somatic motor systems tissues," study authors wrote.¹ "We anticipate that ongoing brain donation studies will address the aforementioned limitations of this study (small sample size, lack of robust SOD1 variant matching and dosing within the interval of 28 days before autopsy) and will refine our understanding of tofersen pharmacokinetics and pharmacodynamics, mechanism of action, and immune responses in patients."

Study Overview

Amanda J. Guise, PhD, lead study author and principal scientist at Biogen in Cambridge, Massachusetts, and colleagues conducted a cross-sectional autopsy tissue case series involving 8 individuals with SOD1-ALS who had received tofersen through clinical trials or an expanded access program. Autopsies were performed at 3 US academic medical institutions between 2018 and 2026, with analyses conducted between August 2020 and January 2026. The study included 5 male and 3 female donors ranging in age from 42 to 66 years.

Investigators measured tofersen concentrations in central nervous system tissues using hybridization enzyme-linked immunosorbent assay (ELISA) and evaluated SOD1 messenger RNA (mRNA) and protein levels using quantitative reverse transcription polymerase chain reaction and ELISA, respectively. Histological analyses were also performed to characterize tofersen distribution within motor neurons and glial cells, as well as the presence of misfolded SOD1 inclusions and potential immune responses.

Tissue samples from the 8 tofersen-treated donors were compared with samples from 15 tofersen-naive individuals with SOD1-ALS. Because of the small number of treated participants and substantial differences in SOD1 variants and dosing histories, investigators did not statistically compare the treated participants as a group.

Results

Tofersen concentrations in spinal cord and motor cortical tissues generally aligned with predictions from a physiologically based pharmacokinetic model, with 74% of observed tissue concentrations falling within 3-fold of predicted levels. Tofersen concentrations were highest in the lumbar spinal cord and generally decreased in more rostral regions, including the thoracic and cervical spinal cord and motor cortex.

Among 3 recently treated participants whose final tofersen dose was administered less than 3 months before autopsy, tofersen was detected in 93% to 99% of residual somatic motor neurons across lumbar, thoracic, and cervical spinal cord regions. Tofersen was also detected in residual motor neurons in the motor cortex, as well as in several cranial motor nuclei. Reactive astrocytes and microglia also demonstrated evidence of tofersen uptake.

The investigators observed substantial reductions in SOD1 mRNA and protein levels in spinal cord tissue. Among recently treated participants, lumbar spinal cord SOD1 protein levels were reduced by 45% to 84% compared with the mean levels observed in tofersen-naive SOD1-ALS controls. SOD1 mRNA levels were also reduced, with the largest reductions observed in lumbar spinal cord tissue.

Notably, the reductions in spinal cord SOD1 protein appeared to exceed the reductions previously observed in CSF SOD1, an indirect biomarker of target engagement used in tofersen clinical trials. One participant who received a final dose 6 weeks before autopsy demonstrated estimated SOD1 protein reductions of 66% in motor cortex, 68% in cervical spinal cord, and 84% in lumbar spinal cord, compared with a 35% nadir reduction in CSF SOD1.

The study also identified leptomeningeal or perivascular lymphocytic infiltrates in 5 of 7 tofersen-treated participants for whom histological tissue was available. These infiltrates were predominantly composed of CD20-positive B cells, with smaller populations of CD3-positive T cells and CD68-positive activated macrophages or microglia. The authors noted that these findings may represent an immune response to tofersen and could be related to the CSF pleocytosis commonly observed with treatment.

Clinical Context

SOD1-ALS accounts for approximately 2% of ALS cases and is associated with more than 200 pathogenic variants in the SOD1 gene. These variants can lead to the production, misfolding, and aggregation of SOD1 protein, which is believed to contribute to motor neuron toxicity and degeneration.

Tofersen is an antisense oligonucleotide designed to reduce SOD1 protein production by promoting RNase H–dependent degradation of SOD1 mRNA. It is the first approved disease-modifying therapy specifically indicated for individuals with SOD1-ALS. In clinical studies, tofersen has been shown to reduce CSF SOD1 and plasma neurofilament light chain levels, although the 28-week primary clinical end point of the phase 3 VALOR trial was not met.¹

As previously covered by NeurologyLive, longer-term analyses of the phase 3 VALOR trial and its open-label extension have suggested potential benefits associated with earlier initiation of tofersen, including slower functional decline and longer event-free survival in certain subgroups.² The current autopsy findings provide complementary evidence at the tissue level, demonstrating that tofersen reaches residual motor neurons throughout the somatic motor system and is associated with substantial reductions in SOD1 protein within the spinal cord.

The findings may also help clarify the relationship between tissue-level SOD1 reduction and previously reported changes in CSF SOD1, a biomarker of target engagement used in clinical studies of tofersen. Investigators noted that the observed caudal-to-rostral gradient in tofersen distribution and SOD1 reduction may have implications for understanding treatment effects across different regions of the motor system, although additional studies will be needed to determine whether these findings extend to other motor neuron diseases or patterns of disease involvement.

Limitations

The authors acknowledged several limitations, including the small sample size, heterogeneity in SOD1 variants and tofersen dosing histories, and limited matching between treated participants and tofersen-naive controls. Some participants had also missed scheduled tofersen doses before autopsy, restricting assessment of tissue SOD1 reduction during the standard 28-day dosing interval. In addition, substantial variability in tissue SOD1 concentrations among untreated donors and potential differences in protein metabolism across SOD1 variants complicated direct comparisons.

The investigators also noted that the presence of misfolded SOD1 inclusions in both treated and untreated donors precluded definitive conclusions about tofersen's effect on aggregate burden. They emphasized that larger autopsy cohorts with improved variant matching and more detailed analyses of soluble and insoluble SOD1 aggregates will be needed. Despite these limitations, the authors concluded that the findings provide the first human postmortem evidence of tofersen distribution and robust SOD1 protein reduction in somatic motor system tissues, with ongoing autopsy studies expected to further clarify the therapy's pharmacokinetics, pharmacodynamics, mechanism of action, and potential immune responses in SOD1-ALS.

REFERENCES
1. Guise AJ, Thanawala Sellon M, Roemer SF, et al. Antisense oligonucleotide tofersen distribution in the central nervous system of SOD1-ALS autopsy tissue donors. JAMA Neurol. Published online July 6, 2026. doi:10.1001/jamaneurol.2026.2195
2. Miller TM, Cudkowicz ME, Shaw PJ, et al. Long-term tofersen in SOD1 amyotrophic lateral sclerosis. JAMA Neurol. Published online December 22, 2025. doi:10.1001/jamaneurol.2025.4946