T2-relaxometry in a large cohort of hereditary transthyretin amyloidosis with polyneuropathy

Background: Previously, T2-relaxation time (T2app) and proton spin density (rho) detected nerve injury in a small group of ATTRv amyloidosis. Here, we aim to quantify peripheral nerve impairment in a large cohort of symptomatic and asymptomatic ATTRv amyloidosis and correlate T2-relaxometry markers with clinical parameters and nerve conduction studies (NCS). Methods: Eighty participants with pathologic variants of the transthyretin gene (TTRv) and 40 controls prospectively underwent magnetic resonance neurography. T2-relaxometry was performed, allowing to calculate tibial rho, T2app and cross-sectional-area (CSA). Detailed clinical examinations and NCS of tibial and peroneal nerves were performed. Results: Forty participants were classified as asymptomatic TTRv-carriers, 40 as symptomatic patients with polyneuropathy. rho, T2app and CSA were significantly higher in symptomatic ATTRv amyloidosis (484.2 +/- 14.8 a.u.; 70.6 +/- 1.8 ms; 25.7 +/- 0.9 mm2) versus TTRv-carriers (413.1 +/- 9.4 a.u., p < 0.0001; 62.3 +/- 1.3 ms, p = 0.0002; 19.0 +/- 0.8 mm2, p < 0.0001) and versus controls (362.6 +/- 7.5 a.u., p < 0.0001; 59.5 +/- 1.0 ms, p < 0.0001; 15.4 +/- 0.5 mm2, p < 0.0001). Only rho and CSA differentiated TTRv-carriers from controls. rho and CSA correlated with NCS in TTRv-carriers, while T2app correlated with NCS in symptomatic ATTRv amyloidosis. Both rho and T2app correlated with clinical score. Conclusion: rho and CSA can detect early nerve injury and correlate with electrophysiology in asymptomatic TTRv-carriers. T2app increases only in symptomatic ATTRv amyloidosis in whom it correlates with clinical scores and electrophysiology. Our results suggest that T2-relaxometry can provide biomarkers for disease- and therapy-monitoring in the future.