# Limited sensitivity of somatosensory evoked potentials as disease monitoring biomarkers in hereditary spastic paraplegias

## TL;DR

This study found that somatosensory evoked potentials are not sensitive enough to track disease progression in hereditary spastic paraplegias compared to clinical assessments.

## Contribution

The study demonstrates that SSEPs have limited sensitivity for monitoring HSP progression, which is novel for clinical trial endpoint selection.

## Key findings

- SSEP latencies showed minimal progression over 4 years compared to clinical outcomes.
- SPRS and mSPRS scores worsened significantly more than SSEP measures.
- SSEP-LL latency increased by 1.343 ms annually, but this was less sensitive to disease progression.

## Abstract

Hereditary Spastic Paraplegias (HSP) are a group of genetic disorders leading to the degeneration of long motor and sensory tracts in a progressive course. Clinician-reported outcomes (ClinROs) are the most commonly used endpoints for monitoring these diseases, but they have low sensitivity to detect progression. Therefore, identifying new monitoring biomarkers with higher sensitivity to change is crucial. Our objective was to compare the progression of Somatosensory Evoked Potential (SSEP) latencies over time with ClinROs in HSP.

A longitudinal study was conducted on 22 individuals with a genetic diagnosis (13 SPG4, 3 SPG5, 3 SPG7, 2 SPG10, and 1 cerebrotendinous xanthomatosis), with two evaluations over a 4-year interval of upper limb (UL) and lower limb (LL) SSEPs and the Spastic Paraplegia Rating Scale (SPRS) total score and motor items only (mSPRS).

In the follow-up time analysis, progression after 4 years was observed only for SPRS and mSPRS, with an annual progression of 1.12 points and 1.02 points, respectively. No statistically significant progression was observed for SSEPs. Disease progression modeled according to disease duration showed worsening in all outcomes. For each additional year of disease, the SPRS worsened by 0.834 points (95% CI 0.62 to 1.04, p < 0.001), mSPRS by 0.758 points (95% CI 0.55 to 0.96, p < 0.001), SSEP-UL latency by 0.164 ms (95% CI 0.03 to 0.3, p < 0.001), and SSEP-LL latency by 1.343 ms (95% CI 0.74 to 1.93, p < 0.001). Results for the SPG4 subgroup were similar to those for the overall HSP group.

The neurophysiological progression of sensory long tract dysfunction is even slower than the progression of motor findings measured by COAs in HSP. The low sensitivity to change of SSEPs identified suggests that they should not be used as primary endpoints in future clinical trials for disease-modifying drugs.