Muscle quantitative MRI as a novel biomarker in hereditary transthyretin amyloidosis with polyneuropathy: a cross-sectional study

Background The development of reproducible and sensitive outcome measures has been challenging in hereditary transthyretin (ATTRv) amyloidosis. Recently, quantification of intramuscular fat by magnetic resonance imaging (MRI) has proven as a sensitive marker in patients with other genetic neuropathies. The aim of this study was to investigate the role of muscle quantitative MRI (qMRI) as an outcome measure in ATTRv. Methods Calf- and thigh-centered multi-echo T2-weighted spin-echo and gradient-echo sequences were obtained in patients with ATTRv amyloidosis with polyneuropathy (n = 24) and healthy controls (n = 12). Water T2 (wT2) and fat fraction (FF) were calculated. Neurological assessment was performed in all ATTRv subjects. Quantitative MRI parameters were correlated with clinical and neurophysiological measures of disease severity. Results Quantitative imaging revealed significantly higher FF in lower limb muscles in patients with ATTRv amyloidosis compared to controls. In addition, wT2 was significantly higher in ATTRv patients. There was prominent involvement of the posterior compartment of the thighs. Noticeably, FF and wT2 did not exhibit a length-dependent pattern in ATTRv patients. MRI biomarkers correlated with previously validated clinical outcome measures, Polyneuropathy Disability scoring system, Neuropathy Impairment Score (NIS) and NIS-lower limb, and neurophysiological parameters of axonal damage regardless of age, sex, treatment and TTR mutation. Conclusions Muscle qMRI revealed significant difference between ATTRv and healthy controls. MRI biomarkers showed high correlation with clinical and neurophysiological measures of disease severity making qMRI as a promising tool to be further investigated in longitudinal studies to assess its role at monitoring onset, progression, and therapy efficacy for future clinical trials on this treatable condition. Supplementary Information The online version contains supplementary material available at 10.1007/s00415-022-11336-z.


Introduction
Hereditary transthyretin (ATTRv; "v" for variant) amyloidosis is a rare systemic disease caused by mutations in the transthyretin (TTR ) gene. Mutant TTR protein tends to misfold and accumulates as amyloid extracellular fibrils across different tissues and organs, especially peripheral nerves, and heart. Since its identification, more than 150 amyloidogenic mutations have been reported, with a broad phenotypic variability. The most frequent mutation is Val30Met (alternatively named p.Val50Met) which was the first one identified. ATTRv amyloidosis is a progressive, disabling and lifethreatening condition with a mean survival of 7-10 years after onset if left untreated [1][2][3].
Different therapies have been approved for ATTRv amyloidosis so far, including the TTR stabilizer tafamidis and, more recently, the RNAi agent patisiran and the antisense oligonucleotide inotersen [4,5]. Moreover, other treatment options are underway [6]. Therefore, there is a constant need for sensitive biomarkers to help establishing the disease onset, to track its progression and to monitor the drug efficacy.
Muscle quantitative MRI (qMRI) has been extensively used as an outcome measure in muscle diseases [12][13][14][15]. More recently, the quantification of intramuscular fat substitution (fat fraction, FF), which indirectly reflects axonal degeneration, showed high responsiveness to change over 12 months in patients with genetic neuropathies including Charcot-Marie-Tooth 1A (CMT1A) and hereditary sensory neuropathy type 1 due to SPTLC1 and SPTLC2 mutations [16,17]. In motoneuron diseases (amyotrophic lateral sclerosis, and spinal bulbar muscular atrophy) qMRI revealed significant fat substitution compared to controls. It also correlated with clinical measures, and identified distinct patterns of muscle involvement [18]. Moreover, STIR and relative T2-weighted signal turned out as objective surrogate markers of muscle denervation, and significantly increased over 12 months [18,19].
The aim of our cross-sectional study was to assess the role of qMRI of skeletal muscle as an outcome measure in ATTRv amyloidosis with polyneuropathy (ATTRv-PN) and compare it with previously validated and functionally relevant clinical outcomes.

Study design and patient recruitment
We performed a prospective cross-sectional study assessing muscle qMRI of the lower limbs in symptomatic patients diagnosed with ATTRv-PN (n = 24) who were enrolled among those who attended the Amyloidosis Research and Treatment Center (IRCCS Fondazione Policlinico S. Matteo) in Pavia (Italy) between September 2017 and August 2018.
Healthy controls (HCs), group-matched for age and sex, were also enrolled (n = 12). Exclusion criteria for all participants were pregnancy and safety-related MRI contraindications.

Data acquisition: clinical and functional testing and electrophysiological revision
All patients underwent detailed assessment by E.V. including demographic records, past medical history and full neurologic examination. Patients were rated using PND scoring system [20], NIS [7] and NIS-LL [7,8]. PND score was graded as follows: PND = 1 (sensory disturbances with preserved walking capability), PND = 2 (sensory-motor symptoms with unassisted gait), PND = 3 (sensory-motor symptoms with assisted gait), and PND = 4 (wheelchair-bound or bedridden). In a subset of n = 17 patients who underwent nerve conduction studies (NCSs) at the same time from MRI, compound muscle action potential (CMAP) of peroneal and tibial nerve, and sensory nerve action potential (SNAP) of sural nerve, measured form peak to peak, were reviewed in detail and considered for further analysis. For each patient, the most affected side was considered unless the asymmetry was due to other conditions (e.g., radiculopathy).

MRI data analysis: muscle quantitative MRI
A single observer (E.V.) with a 5-year training expertise and blinded to study groups outlined for each participant regions of interest (ROIs) on the 1st echo of the ME-SE sequence at mid-thigh (12 muscles) and mid-calf level (6 muscles) using ITK-SNAP software [21]. ROIs were then transferred to the 1st echo of the ME-GRE acquisition and manually adjusted to ensure proper alignment. All ROIs were verified by two expert neuroradiologists (A.P. and M.P.) with more than 5-year expertise in neuromuscular imaging.
The ME-SE sequence was processed using a bi-component extended phase graph algorithm, implemented in Python, for quantification of water T2 (wT2) [22,23], using an open-source toolbox [24]. The Fatty Riot algorithm was used offline for the calculation of fat/water images from the ME-GRE acquisition [25,26] and then FF maps were obtained (FF = F/F + W*100%) from each ROI. Average values of FF and wT2 were calculated for the global ROI and for each muscle at thigh and calf level. Figure 1 shows the muscles of the lower limbs which have been assessed. The tibialis posterior (TP) was not evaluated according to the poor quality of visualization at calf MRI.

Statistical analysis
Statistical analyses were performed with SPSS version 22 (SPSS, Armonk, NY) with a significance α level of 0.05. Quantitative measures are reported as mean ± standard deviation (SD) or median and interquartile range (IQR) as appropriate according to their distribution. For intergroup comparisons two-sample t test and Mann-Whitney U test were applied as appropriate. Correlations of MRI parameters with clinical and electrophysiological measures were investigated with Spearman (ρ) or Pearson coefficients as appropriate according to data distribution. Graphics were obtained using Prism-GraphPad version 9.2.0 (332).
Nine/12 (75%) HCs were males and median age at enrollment was 59 years (range 46-68). Demographic and clinical data of the participants are summarized in Table 1 and Supplementary Table 1.
No significant difference was seen between patients carrying Val30Met (p.Val50Met) and other mutations (data not shown).

Quantitative MRI parameters correlate with clinical outcomes in ATTRv amyloidosis
We next assessed the role of qMRI as a biomarker of disease severity by looking at the correlation between qMRI measures and previously validated scales of disability (PND score) and neurologic impairment (NIS and NIS-LL).

Muscle fat fraction and water T2 do not exhibit length-dependent changes in ATTRv amyloidosis
ATTRv amyloidosis typically presents with a lengthdependent pattern of weakness, namely lower limbs are more affected than upper limbs and distal limb segments are more affected than proximal ones [27].
In our cohort, calf muscles were significantly weaker compared to thigh muscles as measured by NIS-LL (NIS-LL score at calf vs thigh: median 7, IQR 28.5 vs median 0, IQR 6; p = 0.023).

Pattern of fat infiltration: posterior thigh involvement in ATTRv amyloidosis
Quantitative MRI identified a prominent involvement of the medio-posterior compartment of the thighs with a relative sparing of the quadriceps (median 12.8%, IQR 8.

Discussion
In this study, we assessed the role of muscle qMRI as a novel outcome measure in a cohort of ATTRv patients with polyneuropathy. We showed that qMRI reveals significant difference between ATTRv patients and healthy controls and strongly correlates with previously validated clinical measures.
ATTRv amyloidosis is a progressive and highly debilitating hereditary disease, which is fatal within a decade without treatment [1][2][3]. Early diagnosis is key to promptly start an appropriate anti-amyloidogenic treatment. Therefore, there is a need for reliable and objective measures to establish the disease onset, track its progression and monitor the response to treatments.
To date, most outcome measures used in ATTRv-PN are based on clinical examination and functional impairment. In particular, NIS, along with its subset NIS-LL, which was first designed to grade neurological impairment in diabetic neuropathy [28], has become the most used outcome measure in different clinical trials and observational studies in ATTRv amyloidosis [7,8,29]. However, it does not encompass the autonomic and cardiac involvement of the disease. Also, even if performed by expert and preliminarily-trained clinicians, NIS and NIS-LL are limited by intra-and inter-rater variability [4,30,31] along with patient's motivation. To better capture the multisystem involvement in ATTRv amyloidosis and reduce its variability [32], novel compound scales, NIS + 7 and mNIS + 7, have been developed [7,[9][10][11]. Although NIS + 7 and mNIS + 7 provide a thorough evaluation of ATTRv amyloidosis, they are time-consuming and require a bespoke setting and specific training of the examiners [33].
Quantitative MRI may represent an attractive option to overcome shortcomings in clinical examination as it is relatively rapid, with high inter-and intra-operator reproducibility of manual muscle segmentation [21] and analysis can also be automated in all ATTRv patients. In particular, previous studies have shown that muscle quantitative MRI has a very good reproducibility with an interclass correlation coefficient > 0.9 both for inter-and intra-rater agreement [34]. The future implementation of robust machine learning algorithms for automatic segmentation of muscles may be the key to overcome the limitations related to manual segmentation [35,36].
Recently, MRI neurography of sciatic and sural nerve was shown to be able to accurately distinguish patients with ATTRv amyloidosis from controls and, importantly, to detect subclinical and early nerve lesions in asymptomatic carriers [37,38]. Similarly, magnetization transfer ratio of the sciatic nerve showed promising results as it differentiated both symptomatic ATTRv patients and   asymptomatic carriers from healthy controls and correlated with electrophysiology [39]. Muscle qMRI has been extensively applied as disease biomarker for the study of muscle dystrophies and other myopathies, including Duchenne muscular dystrophy, where both fat fraction and water T2 content provided sensitive noninvasive measures of disease progression over time [12]. More recently, quantification of intramuscular fat showed high responsiveness to change over 12-month time in genetic neuropathies, and currently represents the most sensitive outcome measure for the assessment of the slow progression of CMT1A [16].
Prompted by these encouraging results, we decided to assess the role of muscle qMRI in ATTRv amyloidosis with polyneuropathy. We evaluated both acute (tissue water content as expressed by T2 signal) and chronic (fat replacement as expressed by FF) changes in the lower limb muscles. We found that water T2 and FF were significantly higher in ATTRv patients compared to healthy controls and were able to differentiate the two groups. Their increase might be explained by axonal damage due to the underlying neuropathy resulting in acute (as defined by water T2) and chronic (as defined by FF) denervation. More important, we observed a moderate to strong positive correlation between Correlation of muscle quantitative MRI measures with NCS parameters in ATTRv amyloidosis. Negative correlation between NCS parameters namely peroneal nerve compound muscle action potential (CMAP) (pink), tibial nerve CMAP (green), and sural nerve sensory nerve action potential (SNAP) (light-blue) and water T2 (wT2) and fat fraction (FF) at thigh (A) and calf (B) level muscle qMRI parameters (mean FF and wT2) at both thigh and calf level and patients' functionally relevant clinical measures (PND score, NIS, and NIS-LL scales). Therefore, muscle qMRI represents a reliable surrogate measure of disease severity in ATTRv amyloidosis, which is independent of participant's effort and with a high intra-rater agreement [21]. In addition, qMRI may help clinicians to monitor disease progression in ATTRv patients in more advanced stages of the disease, when widespread reduction or unexcitability of motor and sensory action potentials limits the role of neurophysiology in assessing progression and response to treatment.
Interestingly, in ATTRv, qMRI showed a similar degree of fat replacement and water content of thigh and calf muscles, which was unexpected considering the length-dependent involvement of lower limb muscles at examination. This finding differed from previous studies on slowly progressive neuropathies, CMT1A and hereditary neuropathy with liability to pressure palsies, where qMRI showed a higher degree of fat substitution at calves [40,41]. However, it is worth noting that previous pathological studies have shown in ATTRv amyloidosis conspicuous amyloid deposition in dorsal roots and sympathetic ganglia [42]. Also, previous MR neurography studies have detected in ATTRv carriers the presence of early and prominent changes of the proximal nerve tracts compared to distal ones [37], which is in agreement with our finding of significant fat replacement of proximal muscles of the lower limbs.
Finally, skeletal muscle MRI showed a characteristic pattern in ATTRv patients with a preferential involvement of posterior muscles and a relative sparing of quadriceps in the thighs, while in the calves, all muscles appeared to be similarly involved. This pattern has not been previously described in other acquired and genetic neuropathies. Indeed, patients affected by chronic inflammatory demyelinating polyneuropathy had fat infiltration both at biceps femoris and quadriceps [43] while in CMT1A, a predominant degeneration of antero-lateral compartments of calves was reported [40]. Therefore, although this observation warrants further confirmation in larger cohorts, it may be a clue to suspect ATTRv amyloidosis in patients with unexplained axonal neuropathy and, maybe in the future, to help differentiating ATTRv amyloidosis from other genetic and acquired neuropathies.
Our study has some limitations. First, we recruited mostly ATTRv patients with mild and moderate neuropathy, while the more advanced stages of the disease were underrepresented. Second, a muscle biopsy was not performed in any case. Therefore, we cannot rule out the presence of a coexisting myopathy. However, needle EMG was available for n = 17 patients and did not show myopathic changes in any of them. Also, amyloid myopathy is infrequent in previous case series [44]. Third, although the time required for muscle MRI was limited to 1 hour and was mostly well tolerated by all subjects, patients with more advanced neuropathy showed lower tolerance of a prolonged supine position, which partly explains their lower recruitment. However, in general, good collaboration from the patient was fundamental to obtain images suitable to quantitative analysis. Hopefully, in the near future compressed sensing and parallel imaging will allow to shorten the acquisition time making this evaluation easier to perform.
Longitudinal studies are warranted to assess the role of qMRI as noninvasive, objective, and sensitive biomarker for the diagnosis and monitoring over time of ATTRv-PN patients, especially in presymptomatic and early symptomatic stages of the disease.
Acknowledgements The authors thank the study participants and their families for their support.
Author contributions All authors contributed to the study conception and design. Material preparation, data collection, and analysis were performed by E.V., A.C., N.B., L.O. and A.P.. The first draft of the manuscript was written by E.V. and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Declarations
Conflicts of interest All authors certify that they have no affiliations with or involvement in any organization or entity with any financial interest or non-financial interest in the subject matter or materials discussed in this manuscript.
Ethics approval All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. The study was approved by the Ethic Committee of IRCCS Fondazione Policlinico S. Matteo (p-20170020469) and by local institutional review boards.

Consent to participate Informed consent was obtained from all individual participants included in the study.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.