Atypical processing of tones and phonemes in Rett Syndrome as biomarkers of disease progression

Rett Syndrome (RTT), a neurodevelopmental disorder caused by *MECP2* gene mutations, presents significant challenges in assessing auditory processing due to severe motor impairments and limited expressive language. Existing research using auditory brainstem responses (ABR) suggests largely unaffected subcortical processing, while studies on auditory evoked potentials (AEP) assessing cortical stages indicate significant impairments. Previous AEP studies have reported atypicalities, delayed processing, and abnormalities in later components (e.g., mismatch negativity, MMN). However, there's a lack of systematic investigation into the initial sequence of AEP components (P1, N1, P2, N2) in response to speech-like sounds. This study aimed to address this gap by using high-density EEG to record AEPs in response to both simple tones and complex phonemes in females with RTT, re-analyzing data from a previous oddball study focusing on standard stimuli to achieve high signal-to-noise ratios. The goal was to map cortical auditory processing abilities and explore the potential of AEP components as biomarkers for RTT.

Early electrophysiological studies in RTT primarily focused on the auditory brainstem response (ABR), revealing mostly unaffected subcortical auditory processing. However, studies focusing on later cortical stages of auditory processing using AEPs have shown significant impairments. Studies have reported atypical AEPs, delayed auditory processing (prolonged peak latencies of Pa, N1, and P2), atypical developmental changes in N2 latency, and substantial delays in the mismatch negativity (MMN) response. Impairments in auditory sensory memory for duration, as measured by the MMN, were also found. While some research used frequency-specific tone-pip stimuli, studies examining AEPs in response to more complex sounds like speech were lacking. Previous research on speech processing using AEPs in RTT reported abnormalities in the latency range of 200–500 ms, but these studies did not assess earlier cortical processing stages due to limitations in trial repetitions. This highlights the need for a systematic study of the initial AEP components in response to speech sounds to assess the integrity of successive processing stages.

This study involved 13 female RTT patients (mean age 12.9, range 3.9–20.6 years) and 21 age-matched typically developing (TD) female controls. RTT diagnosis was confirmed clinically and genetically. Symptom severity was assessed using the Rett Syndrome Severity Scale (RSSS). Participants listened to auditory stimuli (tones and phonemes) in a sound-attenuated booth while watching a silent movie. An oddball paradigm was used with standard and deviant stimuli. In the Tone condition, 1000 Hz and 500 Hz tones served as standards and deviants; in the Phoneme condition, /ba/ and /da/ served as standards and deviants. High-density EEG data were recorded using a Biosemi ActiveTwo system. Data processing involved epoching, bandpass filtering (1–20 Hz), artifact rejection, and channel interpolation. AEPs were calculated by averaging trials, baselining, and re-referencing to the average of TP7 and TP8 sites. Amplitude analyses focused on FCz, FC3, and FC4 electrodes. Statistical analyses included ANOVAs (group and stimulus type as factors) and t-tests to compare AEP component amplitudes and latencies. ROC analysis and split-half reliability were used to evaluate the potential of AEP components to segregate RTT from TD participants and assess the internal consistency of the measure. Pearson correlations assessed relationships between AEP components, age, and RSSS.

Grand-averaged AEPs showed expected P1, N1, P2, and N2 components in TD participants, while RTT participants showed clear deviations. ANOVA revealed a significant Stimulus-type by Group interaction for P1, due to the greater P1 amplitude to phonemes compared to tones in TD participants, which was absent in RTT. N1 amplitude was reduced for phonemes compared to tones in both groups. Crucially, P2 amplitude was substantially smaller in RTT than TD, regardless of stimulus type, with near-perfect separation between groups despite the wide age range. Individual subject data showed only three RTT participants had P2 amplitudes exceeding the minimum value observed in TD participants. A significant negative correlation between P2 amplitude and RSSS in RTT participants (r(12) = -0.62, p = 0.032) was found, suggesting the relevance of P2 to clinical manifestation.

The findings highlight atypical early auditory cortical processing in RTT. The absence of the typical P1 modulation by stimulus complexity in RTT suggests altered early sensory processing. The significantly reduced P2 amplitude in RTT, independent of stimulus type, is a robust finding with potential as a biomarker for RTT severity. The correlation between P2 amplitude and RSSS supports this potential. The similarity of reduced P2 amplitude in RTT patients to findings in animal models strengthens the translational relevance of this finding. Future research should explore longitudinal studies to track P2 changes with disease progression and investigate whether P2 is responsive to therapeutic interventions.

This study provides evidence of atypical auditory processing in RTT, particularly a significant reduction in the P2 component of the AEP. The robust nature of this finding and its correlation with symptom severity strongly suggest its potential as a valuable biomarker for monitoring disease progression and evaluating treatment response. The concordance between human and animal model findings further supports the translational potential of this biomarker. Further research is needed to validate these findings longitudinally and explore the use of P2 as a surrogate endpoint in clinical trials.

The study primarily included female participants, limiting generalizability to males with RTT. The relatively small sample size, while allowing for high signal-to-noise ratio analyses of AEP components, may limit the generalizability of the findings. The cross-sectional design restricts conclusions regarding the developmental trajectory of auditory processing abnormalities. The study focused on the initial AEP components and did not fully investigate later components beyond the N2.