Assessment of small fiber neuropathy in patients carrying the non-classical Fabry variant p.D313Y
Katharina von Cossel and Nicole Muschol contributed equally to this study.
Abstract
Introduction
The pathophysiological significance of the Fabry-related, non-classical variant p.D313Y still remains to be solved. This study assesses the involvement of the peripheral nervous system with respect to small fiber neuropathy and neuropathic pain in female patients carrying p.D313Y.
Methods
This study examined nine females carrying the Fabry-related p.D313Y variant by obtaining skin punch biopsies above the right lateral malleolus. Intraepidermal nerve fiber density was determined for each patient and compared to reference values matched for the patientʼs decade of life and sex. Moreover, each patient was characterized by a detailed neurological examination and by pain assessment via questionnaire.
Results
Compared to sex-matched lower fifth percentile reference values per decade, intraepidermal nerve fiber density was decreased in seven out of nine patients. Four patients reported acral paresthesias and neuropathic pain with an average visual analogue scale score of 7 out of 10 points. Two patients experienced acute pain crises. Six out of seven patients diagnosed with small fiber neuropathy had a their medical history of hypo- and/or hyperhidrosis.
Discussion
The diagnosis of small fiber neuropathy was made in seven out of nine females carrying the non-classical variant p.D313Y. Moreover, neuropathic pain and symptoms indicative of autonomic nervous system dysfunction seem to be common findings that may be of clinical significance and may warrant therapeutic intervention.
Abbreviations
-
- DRG
-
- dorsal root ganglia
-
- ERT
-
- enzyme replacement therapy
-
- FD
-
- Fabry disease
-
- Gb3
-
- globotriaosylceramide
-
- GLA
-
- alpha-galactosidase A
-
- IENFD
-
- intraepidermal nerve fiber density
-
- MSSI
-
- Mainz Severity Score Index
-
- MRN
-
- magnetic resonance neurography
-
- PNS
-
- peripheral nervous system
-
- QSART
-
- quantitative sudomotor axon reflex test
-
- QST
-
- quantitative sensory testing
-
- SFN
-
- small fiber neuropathy
-
- VAS
-
- visual analogue scale
1 INTRODUCTION
Fabry disease (FD) is a life-limiting, X-linked, lysosomal storage disorder caused by mutations in the gene encoding the enzyme alpha-galactosidase A (GLA).1 A reduction or a complete loss of enzyme activity leads to a progressive accumulation of glycolipids, resulting in a systemic multi-organ disease, including the peripheral nervous system (PNS) in terms of a painful small fiber neuropathy (SFN). While few men and many females develop cardiac, renal, and central nervous system manifestations, SFN seems to be a common feature of FD in both genders. This emphasizes the importance of the investigation of the PNS, making it an excellent diagnostic tool in the determination of early major organ involvement. More than 900 GLA-gene mutations have been described so far and can be divided into mutations causing classical FD and those causing disease modifying variants. While mutations causing classical FD are rare but well understood, recent screening studies indicate a high prevalence of disease modifying variants with an uncertain pathological significance.2-4 The p.D313Y variant is one of the most prevalent disease modifying variants and was initially regarded as a classical Fabry mutation.2-5 Subsequent studies, however, suggested p.D313Y to be a benign polymorphism with no need of therapeutic intervention.2, 6-8 Even more recent studies reported several p.D313Y carriers with clinical symptoms and laboratory markers that showed profound similarities with those of classical Fabry patients.9, 10 Moreover, previous imaging studies described alterations of the central and peripheral nervous systems in male and female p.D313Y carriers that are also commonly seen in classical FD.11, 12 These studies interpret the p.D313Y variant as a potentially pathological condition, resembling an early stage of classical FD. To further investigate the pathognomonic significance of p.D313Y, this study used skin biopsies as well as a detailed neurological examination and pain questionnaire as additional diagnostic markers for the investigation of PNS involvement.
2 METHODS
2.1 Clinical and demographic patient data, questionnaire assessment, and laboratory tests
The study was performed in accordance with the Declaration of Helsinki of 1964 and its later amendments, approved by the institutional ethics board (S398-2012) and written informed consent was obtained from all patients. In total, 15 females carrying the p.D313Y mutation were prospectively recruited as part of the imaging project “Investigation of PNS alterations in classical FD and disease modifying variants”, nine of whom agreed in having an additional skin biopsy taken. As patient data were anonymized before analysis, additional ethics committee approval was not required.
Apart from the p.D313Y mutation, further co-existing classical Fabry mutations in the GLA-gene were excluded in each patient by genetic testing. The Mainz Severity Score Index (MSSI)13 was determined in all patients. The total score is composed of four sections (general, neurological, cardiovascular and renal signs and symptoms) and its results are categorized by number of attained points as mild (≤18), moderate (19-38) or severe (>38). Overall, it has been shown that MSSI represents a useful tool for clinicians in evaluating the severity and progression of FD.14 Skin biopsies were carried out between August 2018 and April 2019 at the International Center for Lysosomal Disorders, University Medical Center Hamburg-Eppendorf. Patients with any other condition causing SFN, including chronic renal insufficiency (glomerular filtration rate <90 mL/min), diabetes mellitus, alcoholism, or acute or chronic immunological disorders were excluded from this study.
2.2 Pain assessment
All patients completed the pain questionnaire “pain-detect” of the German Research Network on Neuropathic Pain. It assesses pain intensity, pain duration, pain attacks, pain radiation and pattern as well as the quality of pain in general. The test´s scale ranges from 0 to 38 points, with a score below 12 points indicating absence and a score above 19 points indicating presence of neuropathic pain. Moreover, pain was assessed by visual analogue scale (VAS), an 11-point scale for pain intensity, ranging from no pain (0) to worst possible pain (10). The quality of neuropathic pain was surveyed and categorized into two different qualities: chronic burning pain (painful dysesthesia in hands and feet, low to moderate intensity) and episodes of painful crisis (with increased severity). Analgesic treatment and possible pain triggers were monitored.
2.3 Skin biopsies
To assess intraepidermal nerve fiber densities (IENFD), skin punch biopsies were performed and processed using a standardized procedure.15 8 mm punch biopsies were taken 10 cm above the right lateral malleolus. Skin samples were immediately transferred into Zamboniʼs solution and taken directly to the institute of neuropathology for further analysis. The biopsy specimen was fixed in a cryoprotective solution, cut in 50 μm sections and immunostained using a polyclonal panaxonal marker (PGP 9.5 antibody, Dako Z5116). IENFD was determined by counting the number of nerve fibers per mm under a light microscope at high magnification (×40) by an experienced neuropathologist (M.G.).
2.4 Normative reference values and statistical analysis
IENFD was assessed in each patient and compared to normative median and lower fifth percentile values for female sex and the individual corresponding decade of life.15, 16 Additionally, individual age-adapted normative reference values for the lower fifth percentile were calculated using a polynomial curve-fitting of order 2, described by the following equation: Age-adapted lower fifth percentile = (0.0007 × age2) - (0.2 × age) + 13. Subsequently, percentage difference between age-adapted lower fifth percentile and patients IENFDs was calculated individually as follows: IENFD difference = 100 × (Patient IENFD - Age-adapted lower fifth percentile)/(Age-adapted lower fifth percentile).
Statistical analyses were performed with GraphPad Prism 7.0 (GraphPad Software, La Jolla, California). Unpaired and two-tailed t-test was used to compare the IENFD specimen of each patient to normative reference values as described above. P-values <.05 were considered significant. All results are documented as mean values.
3 RESULTS
Overall, we included nine female patients (mean age 40 y, range 22-76 y) carrying the Fabry related GLA-gene variant p.D313Y. Patient 1, 5, 7 and 8 were blood relatives. The remaining patients were not related to each other. Patients 3 to 7 were neither on chaperone nor enzyme replacement therapy (ERT). Patient 1, 2, 8 and 9 received bi-weekly infusions of agalsidase-alfa [0.2 mg/kg body weight].
3.1 IENFD
Compared to normative reference values of the median, all p.D313Y carriers showed a reduced IENFD (P < .001) (Table 1). Compared to the lower fifth percentile reference values per decade, IENFD was decreased in seven out of nine examined patients carrying p.D313Y (P < .05). Compared to the lower fifth percentile reference values per age, eight out of nine p.D313Y carriers showed decreased IENFD (P < .05). In these eight patients, IENFD was decreased by 58.5% (range from 5.6% to 87.7%).
Patient | Age (at diagnosis) in years | Clinical features | IENFD (fibers/mm) | <5% norm/decade | IENFD change from norm/decade | <5% norm/age | IENFD change from norm/age | VAS/pain detect | MSSI | Lyso-Gb3 level (ng/mL) |
---|---|---|---|---|---|---|---|---|---|---|
1 | 52 (48) | Neurological: Acral paresthesias with neuropathic pain, stroke, TIA, cerebral vasculitis, depression, hypohydrosis. Cardiac: Hypertrophy of cardiac septum, enlarged left atrium, ECG abnormalities, arterial hypertension, NYHA II. Renal: Creatinine elevation. Others: Tortuositas vasorum retinae, cornea verticillata (stage 1), diarrhea, constipation, joint pain. | 0.9 | 4.3 | −80.2% | 4.5 | −81.1% | 6/26 | 20 | 3.9 |
2 | 46 (43) | Neurological: Stroke, depression, fatigue, tinnitus, vertigo, hypohydrosis. Others: Diarrhea, constipation, joint pain. | 2.7 | 5.7 | −52.6% | 5.3 | −48.9% | 0/1 | 12 | 0.7 |
3 | 76 (73) | Neurological: Vertigo. Cardiac: NYHA III, thickening of wall/septum, ECG abnormalities, arterial hypertension. | 1.0 | 2.2 | −54.5% | 1.8 | −45.7% | 0/2 | 12 | 0.7 |
4 | 20 (18) | Neurological: Acral paresthesias, depression, tinnitus, vertigo. Cardiac: NYHA III. Others: Diarrhea, constipation, joint pain. | 11.4 | 8.4 | +35.7% | 9.3 | +22.8% | n.a./5 | 15 | 1.4 |
5 | 22 (18) | Neurological: Vertigo, hypohydrosis. Cardiac: NYHA I. | 1.1 | 8.4 | −86.9% | 8.9 | −87.7% | 0/0 | 3 | 0.6 |
6 | 44 (44) | Neurological: Acral paresthesias with neuropathic pain, fatigue, reduced activity level, vertigo, hypohydrosis. Cardiac: NYHA I, valve insufficiency, ECG abnormalities. Others: Angiokeratoma, diarrhea, constipation, joint pain. | 0.8 | 5.7 | −86.8% | 5.6 | −86.5% | 8/26 | 14 | 0.9 |
7 | 40 (38) | Neurological: Acral paresthesias, tinnitus, vertigo, hypohydrosis. Cardiac: NYHA I, ECG abnormalities. Others: Muskuloskeletal pain. | 6.1 | 5.7 | +6.1% | 6.4 | −5.6% | n.a./2 | 10 | 0.7 |
8 | 32 (29) | Neurological: Acral paresthesias with neuropathic pain, depression, fatigue, vertigo, tinnitus, hypohydrosis. Cardiac: NYHA I. others: Musculoskeletal pain, angiokeratoma, joint pain. | 4.1 | 7.1 | −42.3% | 7.8 | −47.4% | 6/25 | 11 | 0.4 |
9 | 28 (24) | Neurological: Acral paresthesias with neuropathic pain, depression, tinnitus, vertigo, hypohydrosis. Cardiac: NYHA II, arterial hypertension. Others: Diarrhea, constipation, joint pain. | 3.0 | 8.4 | −64.3% | 8.6 | −65.1% | 6/24 | 11 | 0.6 |
- Abbreviations: ERT, enzyme replacement therapy; IENFD, intraepidermal nerve fiber density; MSSI, Mainz Severity Score Index; NYHA, New York Heart Association; SFN, small fiber neuropathy; VAS, visual analogue scale.
3.2 Pain
Six out of seven patients diagnosed with SFN reported hypo- and/or hyperhidrosis in their medical histories. Acral paresthesias accompanied by severe neuropathic pain was present in four patients diagnosed with SFN with a mean VAS score of 7 (Table 1). Certain trigger factors including physical activity, stress, cold weather, high humidity and fatigue led to pain exacerbations (Table 1). On average the patients experienced the first episode of neuropathic pain at an age of 22 y. Daily persistent neuropathic pain was reported by three patients, whereas acute episodes of burning pain were described by two patients. Joint pain was present in six patients. The reported duration of pain ranged from very short (a few minutes) to a daily consistent pain. Pain medication including pregabalin, gabapentin, acetaminophen, opioids and non-steroidal anti-inflammatory drugs like ibuprofen and metamizole was taken regularly by four patients.
4 DISCUSSION
As a principal finding of this study, skin biopsies of seven out of nine patients carrying p.D313Y showed a distinct reduction of IENFD so that a diagnosis of SFN was made according to recent guidelines.15, 16
SFN in FD has been shown to represent a length-dependent neuropathy with a relatively greater loss of innervation at calf height compared to the thigh level.17-19 While small nerve fiber dysfunction remains largely restricted to the calf level in females with classic FD, classically affected Fabry males additionally show an involvement of the thigh level at a later stage of the disease. These observations were explained by a length-dependent process, which starts at a distal site at a young age in both genders and ascends to a more proximal site only in classical affected males at a later stage.
In recent studies, however, we observed the potential key role of the dorsal root ganglion (DRG) in the development of SFN by using high resolution magnetic resonance neurography (MRN).11, 20, 21 Morphological alterations of DRG structure were explained as a consequence of a progressive lyso-globotriaosylceramide (Gb3) accumulation within the DRG due to its leaky blood-neuronal interface, while peripheral nerve morphology was not or only slightly altered. Gb3 deposits within DRG neurons have also been described ex-vivo.22-24 These findings suggest that the predominant pathophysiological process takes place at the level of the DRG, rather than a more distal nerve segment. This is in contrast to the theory that in small fiber sensory ganglionopathy the underlying pathological processes primarily take place within the sensory cell body and small nerve fiber degeneration occurs to the same extent at the proximal and distal site.
This discrepancy can also be observed in diabetic neuropathy, the most prevalent length-dependent neuropathy, which commonly only involves the acral regions of the lower extremities at an early stage. This observation led to the theory of a dying-back degeneration, that assumed a primary structural degeneration with an initial distal predominance and progression to more proximal levels over time.25, 26 However, by using high-resolution MRN of the lower extremity, we demonstrated a clear gradient of nerve lesions from proximal to distal with proximal predominance.27 This supports the contrary assumption that neuronal degeneration occurs at more proximal levels first and subsequently leads to length-dependent degeneration as a downstream consequence.
The question why the very proximal nerve segments seem to be more susceptible to the underlying pathophysiological processes remains to be solved in detail. In previous studies investigating classically affected Fabry males and females we also found a concomitantly decreased blood supply within the DRG. This observation led to the hypothesis that a hypoxic environment may cause neuronal apoptosis and/or a decreased nutritive microvascular supply of the distally located nerve segments, as they may represent a hemodynamic watershed region of nerve perfusion.20, 21 A definite proof of these assumptions could potentially be obtained through a sensory ganglion biopsy. Due to the invasive nature and potential side effects, this procedure, however, is not an option in the clinical evaluation of small fiber neuropathy.28
Although diagnostic criteria for SFN diagnosis have been developed, additional information is needed to determine whether SFN is related to FD or to other conditions causing damage to small nerve fibers. Thus, the presence of FD-specific SFN features should be obligatory to confirm diagnosis and start therapeutic intervention. Patients with classical FD usually report burning or shooting pain or painful pins and needles in hands and/or feet with pain attacks often initiated by typical trigger factors including heat or physical activity.29 In our cohort, four out of seven patients diagnosed with SFN experienced neuropathic pain with a moderate to severe intensity according to VAS score and reported certain trigger factors that led to pain exacerbation. As we excluded other conditions that are considered as risk factors for SFN, it can be assumed that SFN in patients with the Fabry related variant p.D313Y is a primary consequence of the underlying disease. According to recent guidelines, these four patients have to be diagnosed with “confirmed SFN due to FD,” while the 3 SFN patients without neuropathic pain, but autonomic nervous system dysfunction, should be classified as “probable SFN due to FD”.30 The SFN rate of 78% we found in our cohort is comparable to the rate of 50%-75% described by a previous study investigating IENFD in females with classical FD.30 This finding is of high interest since SFN could be present as the only manifestation of disease in otherwise asymptomatic males and females with classical FD.18, 31
Patients with a disease modifying variant, like p.D313Y, are biochemically different from classical Fabry patients, as they usually do not show elevated Gb3 levels that could be used as a sensitive and specific marker for the diagnosis of FD. Thus it has been questioned whether patients with the p.D313Y variant should be diagnosed with FD at all.6-8, 32 In a recent study, however, we report p.D313Y patients to have symptoms and organ manifestations comparable to classical FD.9 While cerebrovascular involvement and cardiac hypertrophy as classical complications in FD could also be observed in our p.D313Y cohort, it should be considered that those conditions might represent coincidental findings that are not related to a FD manifestation. Nonetheless, our cohort also included a p.D313Y patient with cornea verticillata and a lyso-Gb3 level above the reference. 9 Moreover, we found maltese cross particles in the urine of one patient, a finding that is considered 100% specific and sensitive for FD.10, 33 A further study described multifocal white matter lesions in young p.D313Y carriers, a finding that is linked to cerebral vasculopathy and normally only seen in classically affected Fabry patients, in older patients and/or in patients with classical cardiovascular risk factors.12 Hence, it is likely that mere coincidence cannot fully explain the high frequency of typical FD-complications in p.D313Y carriers.
ERT has been shown to not only prevent major organ complications, but also to reduce neuropathic pain. While IENFD did not increase after 12-18 mo of treatment,34 small nerve fiber function improved due to ERT.35, 36 Considering the fact that SFN and neuropathic pain may represent the sole manifestations of FD at initial diagnosis, the findings of this study support the approach that patients with the GLA-gene variant p.D313Y should be monitored closely for further Fabry-specific alterations and treatment should be considered in at least those p.D313Y carriers that show Fabry-related symptoms.
A limitation of this study is the relatively small sample size. FD is a rare disease and this study focused on a well-defined subpopulation of female patients with one mutation. Thus, the transfer of the results of this study to p.D313Y carriers in general should be made very carefully and individually with respect to the patientʼs medical history and symptoms. Second, quantitative sensory testing (QST) and quantitative sudomotor axon reflex test (QSART) as additional tests for sensory and autonomic small nerve fiber function were not performed in this study. In further studies, QST and QSART could potentially support the diagnosis of PNS involvement, especially in patients without pain. Third, conclusion about a length-depended process cannot be made in our study population since skin samples were only taken at the distal calf level.
In conclusion, SFN, neuropathic pain, and symptoms of autonomic nervous system dysfunction are common findings in patients carrying the GLA variant p.D313Y. The results of this study provide further evidence that p.D313Y represents a condition potentially requiring treatment, rather than a benign polymorphism.
ACKNOWLEDGMENTS
M.B. received grants from the German Research Council (SFB 1158). V.M. was supported by a grant from the Märta and Erik Karberg Foundation for Medical Research. T.G. is supported by a postdoctoral fellowship from the Medical Faculty of the University of Heidelberg, and an Else-Kröner Memorial grant. We thank Barbara Hofstadler for proofreading the manuscript. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Open Access funding enabled and organized by Projekt DEAL. [Correction added on 23 July 2021, after first online publication: Projekt Deal funding statement has been added.]
CONFLICT OF INTEREST
None of the authors has any conflict of interest to disclose.
ETHICAL PUBLICATION STATEMENT
We confirm that we have read the Journalʼs position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.
Open Research
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.