Article Text

PDF

New cases of adult-onset Sandhoff disease with a cerebellar or lower motor neuron phenotype
  1. C C S Delnooz1,
  2. D J Lefeber2,
  3. S M C Langemeijer1,
  4. S Hoffjan3,
  5. G Dekomien3,
  6. M J Zwarts1,
  7. B G M Van Engelen1,
  8. R A Wevers2,
  9. H J Schelhaas1,
  10. B P C van de Warrenburg1
  1. 1Department of Neurology, Donders Institute for Brain, Cognition and Behaviour, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands
  2. 2Laboratory of Pediatrics & Neurology, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands
  3. 3Department of Human Genetics, Ruhr-University Bochum, Bochum, Germany
  1. Correspondence to Dr Bart P C van de Warrenburg, Radboud University Nijmegen Medical Centre, Department of Neurology (935), PO Box 9101, Nijmegen 6500 HB, The Netherlands; b.vandewarrenburg{at}neuro.umcn.nl

Abstract

Sandhoff disease is a lipid-storage disorder caused by a defect in ganglioside metabolism. It is caused by a lack of functional N-acetyl-β-d-glucosaminidase A and B due to mutations in the HEXB gene. Typical, early-onset Sandhoff disease presents before 9 months of age with progressive psychomotor retardation and early death. A late-onset form of Sandhoff disease is rare, and its symptoms are heterogeneous. As drug trials that aim to intervene in the disease mechanism are emerging, the recognition and identification of Sandhoff disease patients—particularly those with atypical phenotypes—are becoming more important. The authors describe six new late-onset Sandhoff cases demonstrating cerebellar ataxia or lower motor neuron (LMN) involvement combined with, mostly subclinical, neuropathy. Two different mutations were found: IVS 12–26 G/A and c.1514G→A. In patients with either progressive cerebellar ataxia or LMN disease in the setting of a possibly recessive disorder, Sandhoff disease should be suspected, even when the onset age is over 45 years.

  • Motor neuron disease (MND)
  • spinal muscular atrophy (SMA)
  • sandhoff disease
  • cerebellar ataxia
  • ganglioside

Statistics from Altmetric.com

Introduction

Typical Sandhoff disease presents before 9 months of age with progressive psychomotor retardation and early death, usually before age 4 years. The cause is a lack of functional N-acetyl-β-d-glucosaminidase (HEX) A and B due to mutations in the HEXB gene. In a few cases, disease onset occurs in adolescence or adulthood. This late-onset form of Sandhoff disease is rare, and its symptoms are heterogeneous, thus posing a challenge to the clinician to prevent a significant diagnostic delay.

Different treatment strategies for Sandhoff disease are emerging, and trials with substrate deprivation therapy are currently ongoing in typical Sandhoff disease.1 This development further emphasises the importance of an early diagnosis, also in those with an onset in adulthood or otherwise atypical presentations.

In order to highlight the clinical spectrum, age at onset and diagnostic approach of late-onset Sandhoff disease, we describe here six new cases.

Patients

Case descriptions

Patients 1, 2 and 3 are siblings. Their parents were consanguineous. Patient 1 is a 44-year-old Moroccan woman who first presented at the age of 30 years with a 2-year history of weakness in her arms and a progressive difficulty climbing stairs and getting up from a supine position. Neurological examination showed generalised muscle atrophy and weakness, particularly of her forearms and proximal leg muscles; no signs of cerebellar, sensory or autonomic dysfunction; and normal tendon reflexes with downgoing plantar responses. At age 33, symptoms had progressed, and she had begun to experience muscle cramps. Muscle atrophy and weakness were more severe (MRC proximal 3/5, distal 4/5), and tendon reflexes were now low.

Patient 2, the brother of patients 1 and 3, presented at the age of 26 years following 4 years of progressive leg weakness. During the last 2 years, he had been experiencing cramps in his calves and had noticed ‘twitching’ of the muscles of his limbs and trunk. Examination showed generalised muscle atrophy and weakness, mostly of the proximal muscles (MRC proximal 4–/5, distal 5/5); fasciculations in the arms, upper legs and trunk; normal cerebellar and sensory functions; generalised areflexia; and normal plantar responses.

Patient 3, a sister of patients 1 and 2, presented with progressive leg weakness noticeable as difficulty in walking and climbing stairs that began at the age of 28. Neurological examination showed atrophy of the quadriceps and calve muscles; generalised weakness mainly in the lower limbs and predominantly proximal (MRC proximal 4/5, distal 5/5); no signs of sensory, cerebellar or autonomic dysfunction; normal tendon reflexes in the upper limbs, but low to absent lower-limb reflexes.

Patient 4 is a 37-year-old Dutch man. He had one affected brother (described below) and three unaffected siblings. There was no consanguinity and no family history of note. He first experienced gait problems at age 18 years. Difficulties with speech and swallowing, a balance disorder and urinary incontinence gradually developed. At times, the patient experienced painful muscle cramps. Neurological examination showed a cerebellar dysarthria; a spastic-ataxic gait; mild appendicular ataxia; mild length-dependent, deep sensory disturbances; brisk tendon reflexes; and normal plantar reflexes.

Patient 5 is the 35-year-old brother of patient 4. Complaints of slurred speech, difficulties carrying cups and muscle cramps started at the age of 30. Walking difficulties, clumsiness of the forearms and hands, and swallowing problems manifested gradually. Neurological examination showed jerky ocular pursuit movements and square-wave jerks; prominent cerebellar dysarthria; gait ataxia and milder ataxia of the extremities; mildly increased tone in the lower limbs; normal sensory examination; and brisk tendon reflexes without pathological plantar responses.

Patient 6 is a 64-year-old woman of Turkish descent. Her parents were consanguineous. She presented at the age of 47 years with walking difficulties in the form of reduced walking speed and a staggering gait. Subsequently, sensory changes occurred with paroxysmal tingling mainly at the left side and sensory loss at both feet. She also mentioned balance difficulties and urinary incontinence. Neurological examination at age of 62 showed a broad-based and spastic gait; limb ataxia, more pronounced at the legs; sensory disturbances consisting of a reduced vibration sense distally in the legs and diminished pain perception in a sock-and-glove distribution; brisk tendon reflexes of the arms, but absent reflexes of the legs; and normal plantar responses.

Neurophysiological and imaging studies

Nerve conduction (NCS) and electromyography (EMG) studies in patients 1 and 2 showed evidence of a generalised LMN involvement with an additional sensory neuronopathy. EMG in patient 3 showed no abnormalities at the time of diagnosis. Re-examination 3 years later indicated a sensorimotor axonal neuropathy. A needle examination could not be performed due to anxiety. NCS and EMG in patients 4, 5 and 6 were all indicative of a predominantly sensory axonal neuropathy. The full NCS/EMG data are provided as supplemental data (see appendices).

Brain MRI in patients 1–3 at diagnosis did not reveal any abnormalities. Four years later, a repeat scan showed mild cerebellar atrophy in all (figure 1A). MRI in patients 4 and 5 showed more marked atrophy of the cerebellum, most pronounced in the vermis (figure 1B). In patient 6, the brain MRI showed some aspecific white-matter lesions, a small hyperintense lesion in the right thalamus (probably explaining her left hemisensory symptoms) and cerebellar atrophy.

Figure 1

(A) Sagittal T1-weighted brain MRI of patient 2 showing mild cerebellar midline atrophy. (B) Sagittal T1-weighted brain MRI of patient 4 showing cerebellar midline atrophy.

Biochemical and genetic studies

The biochemical methods are described in the supplement. We found a profound decrease in N-acetyl-β-d-glucosaminidase activities (total and A isoform) in plasma samples of all patients, compatible with a diagnosis of Sandhoff disease. In leucocytes, total HEX activities were clearly reduced with mildly reduced levels of HEXA (table 1), showing a significant overlap with heterozygous carriers.

Table 1

Patient characteristics

In urine, the characteristic oligosaccharides for infantile Sandhoff could not be identified. Mutation analysis showed two different mutations in our six patients. Direct sequencing of exons 2–14 of the HEXB gene and flanking intronic regions showed a homozygous IVS12–26 G/A mutation in the first family (patients 1–3) and in patient 6. In the second family (patients 4 and 5), a homozygous missense mutation (c.1514G→A; p.Arg505Gln) was found.

Discussion

Sandhoff disease is a lipid-storage disorder caused by a defect in ganglioside metabolism due to mutations in the HEXB gene that disrupt the function of the N-acetyl-β-d-glucosaminidase isoforms HEXA and HEXB.2 As research aimed at treatment strategies is emerging, diagnosing these patients correctly and timely is crucial, and thorough knowledge of the phenotypes is therefore essential. Several cases of adult-onset Sandhoff disease have been reported, mostly with an onset in the third or fourth decade.3–12 Overall, the main clinical features of adult-onset Sandhoff disease consist of a cerebellar syndrome (50% of cases) or LMN disease (42%); the remainder mainly show autonomic dysfunction. Some case reports mention the presence of either neuronopathy or axonopathy, which was mostly clinically mild or only detectable by EMG/NCS.13 14 Adult-onset Tay–Sachs disease can present with similar clinical features, but biochemically there is a deficiency of only HEXA.15 16

Lower motor neuron involvement was the prominent feature in three of our six late-onset Sandhoff disease patients, in whom symptoms started in the third decade. In patients 4–6, cerebellar ataxia was the main and presenting feature. Two of the patients displayed autonomic disturbances in the form of urinary incontinence. Although neurophysiological studies showed a sensory neuronopathy or axonal neuropathy in all patients, only patients 4 and 6 had corresponding symptoms.

In GM2 gangliosidosis, brain MRI typically reveals signs of cerebellar or cerebral atrophy, thalamic hyperintensity and white-matter signal changes. There appears to be little correlation between clinical signs, disease severity and the imaging.16 All our patients showed various degrees of cerebellar atrophy. The cerebellar atrophy was most pronounced in the two siblings with the predominant cerebellar syndrome. Interestingly, however, the siblings with a LMN syndrome also had some cerebellar atrophy, yet there was no cerebellar involvement clinically. The question is whether their weakness masked a possible coexisting ataxia. Conversely, muscle ultrasound indicated the presence of fasciculations in the right quadriceps muscle in the two brothers with the cerebellar phenotype, which might suggest some motor neuron involvement in the ataxia patients as well (data not shown).

Until now, a few HEXB mutations have been identified in late-onset Sandhoff disease. In six out of nine previously reported late-onset cases, the c.1514G→A (p.Arg505Gln) missense mutation was found; spinocerebellar ataxia was the main clinical correlate in all.3–6 Other HEXB mutations have been described, but all in just one or two individuals. Our patients were homozygous for the IVS12–26 G/A or c.1514G→A mutation. As the IVS12–26 G/A mutation has been found in only two cases of juvenile Sandhoff disease, our patients therefore seem to be the first with adult-onset disease associated with the IVS12–26 G/A mutation.17 18 Still, there is significant phenotypic variability, as three of our patients with this mutation had prominent LMN disease, while one had a predominant cerebellar syndrome. The c.1514G→A mutation is more common and has been described in six additional adult-onset cases, presenting with a spinocerebellar syndrome similar to our patients 4 and 5, perhaps suggesting a correlation between this phenotype and genotype.

In conclusion, Sandhoff disease should be considered in the differential diagnosis of unexplained hereditary spinocerebellar ataxia and LMN, in a setting that suggests a recessive disease, even if the age at onset is above 45 years. The diagnostic work-up should include an assessment of total N-acetyl-β-d-glucosaminidase and N-acetyl-β-d-glucosaminidase A activities in leucocytes and plasma, followed by HEXB mutation analysis.

Appendix

Neurophysiological studies

Patient 1

NerveStimulation siteRecording siteLat (ms)Lat (ms)Amp (μV)Dist (cm)CV (m/s)
Sural leftMidline posterior lower legLateral malleolusNRNRNR12NR
Superficial peroneus leftLateral to tibialis anterior tendonAnkleNRNRNR10NR
Median leftHand palmThird finger1.502.057.707.5050.00
Wrist2.803.704.907.5057.70
Radial leftForearmEPL1.902.453.9010.0052.60

Sensory-conduction studies

NerveStimulation siteRecording siteLat (ms)Dur (ms)Amp (mV)Dist (cm)CV (m/s)
Median leftWristAPB2.955.4012.806.00
ElbowADM6.605.9012.4023.5064.40
Ulnar leftWristADM2.555.107.906.00
Under elbow6.005.008.4022.0063.80
Above elbow7.255.307.908.0064.00
Arm8.805.006.608.0051.60
Peroneal commune leftAnkle EDBEDB + TA4.255.400.707.00
Inferior fibular head EDB11.956.200.4032.0041.60
Inferior fibular head TA3.5012.753.0010.00
Superior fibular head EDB13.6014.750.207.0042.40
Superior fibular head TA5.2513.153.107.0040.00
Posterior Tibula leftMedial malleolusAH3.954.509.307.00
Popliteal fossa14.5553.158.5041.0038.70

Motor-conduction studies

NerveInsertionFibrPositive spikesFascDischFormDur (ms)PatternAmp (μV)
Anterior tibial leftN++++++PF+10–15Poor1000–2000
Rectus femoris leftNNDNDNDND
Interosseus dorsi leftN+PF+10–20Poor4000–7000
Biceps brachii leftNPF++NDPoor4000–7000

Needle examination

Patient 2

NerveStimulation siteRecording siteLat (ms)Lat (ms)Amp (μV)Dist (cm)CV (m/s)
Sural leftMidline posterior lower legLateral malleolus012.00
Superficial peroneus leftLateral to tibialis anterior tendonAnkleNRNRNR10.00NR
Median leftHand palmThird finger1.602.105.007.5046.90
Wrist3.254.057.207.5045.50
Radial leftForearmEPL2.352.857.3010.0042.60

Sensory-conduction studies

NerveStimulation siteRecording siteLat (ms)Dur (ms)Amp (mV)Dist (cm)CV (m/s)
Median leftWristAPB
ElbowADM
Ulnar leftWristADM2.206.2513.006.00
Under elbow6.606.1012.7023.0052.30
Above elbow8.806.4511.7010.0045.50
Arm10.806.3510.509.0045.00
Peroneal commune leftAnkle EDBEDB + TA5.005.102.907.007.00
Inferior fibular head EDB13.755.102.4035.5040.60
Inferior fibular head TA15.655.002.807.0036.80
Superior fibular head EDB3.804.5011.7010.00
Superior fibular head TA14.605.259.9046.5043.10
Posterior Tibula leftMedial malleolusAH2.206.2513.006.00
Popliteal fossa6.606.1012.7023.052.30

Motor-conduction studies

NerveInsertionFibrPositive spikesFascDischFormDur (ms)PatternAmp (μV)
Anterior tibial leftN+PF++>20Poor7000–10000
Rectus femoris leftN++PF++10–20Single unit7000–10000
Interosseus dorsi leftNPF+Normal unitsModerate2000–4000
Biceps brachii leftN+PF++Normal + polyfasic unitsInterference2000–4000

Needle examination

Patient 3

NerveStimulation siteRecording siteLat (ms)Lat (ms)Amp (μV)Dist (cm)CV (m/s)
Sural leftMidline posterior lower legLateral malleolus012.00
Sural rightMidline posterior lower legLateral malleolus012.00
Superficial peroneus leftLateral to tibialis anterior tendonAnkle010.00
Superficial peroneus rightLateral to tibialis anterior tendonAnkle010.00
Median rightHand palmThird finger1.452.0039.507.0048.30
Wrist3.003.6527.807.0045.20
Radial rightForearmEPL2.052.5021.2010.0048.80

Sensory-conduction studies

NerveStimulation siteRecording siteLat (ms)Dur (ms)Amp (mV)Dist (cm)CV (m/s)
Median rightWristAPB2.605.4021.406.00
ElbowADM7.005.8020.0027.0061.40
Ulnar rightWristADM2.405.9516.606.00
Under elbow6.406.0515.7022.5056.30
Above elbow7.706.1515.807.0053.80
Arm8.456.2015.804.0053.30
Peroneal communal rightAnkle EDBEDB + TA3.807.156.507.00
Inferior fibular head EDB11.007.805.6032.0044.40
Inferior fibular head TA12.257.255.606.0048.00
Posterior tibula rightMedial malleolusAH2.605.4021.406.00
Popliteal fossa7.005.8020.0027.0061.40
Posterior tibula leftMedial malleolusAH4.604.1023.9024.20
Popliteal fossa13.705.0517.0020.4044.50

Motor-conduction studies

Patient 4

NerveStimulation siteRecording siteLat (ms)Lat (ms)Amp (μV)Dist (cm)CV (m/s)
Sural leftMidline posterior lower legLateral malleolus0.0012.00
Superficial peroneus leftLateral to tibialis anterior tendonAnkle0.0012.00
Median leftHand palmThird finger1.651.953.508.0048.50
Wrist3.253.852.708.0050.00
Radial leftForearmEPL1.652.153.5010.0060.60

Sensory-conduction studies

NerveStimulation siteRecording siteLat (ms)Dur (ms)Amp (mV)Dist (cm)CV (m/s)
Ulnar leftWristADM2.755.4513.806.00
Under elbow6.006.0013.2024.0073.80
Above elbow7.555.7012.708.554.80
Peroneal commune leftAnkle EDBEDB + TA3.254.1513.707.00
Inferior fibular head EDB10.705.1011.5038.0051.00
Superior fibular head EDB12.805.1011.509.5045.20
Posterior tibula leftMedial malleolusAH2.755.2014.608.00
Popliteal fossa11.956.759.3046.0050.00
Posterior tibula rightMedial malleolusAH3.354.8019.208.00
Popliteal fossa12.905.8512.8047.0049.20

Motor-conduction studies

NerveInsertionFibrPositive spikesFascDischFormDur (ms)PatternAmp (μV)
Extensor hallucis longus leftNPF+NInterferenceN

Needle examination

Patient 5

NerveStimulation siteRecording siteLat (ms)Lat (ms)Amp (μV)Dist (cm)CV (m/s)
Sural leftMidline posterior lower legLateral malleolus0.0012.00
Sural rightMidline posterior lower legLateral malleolus0.0012.00
Superficial peroneus leftLateral to tibialis anterior tendonAnkle0.0010.00
Median leftHand palmThird finger1.802.201.308.0044.40
Wrist3.554.203.208.0045.70
Radial leftForearmEPL0.0010.00
Ulnar leftWristFifth finger3.204.552.6014.0043.80

Sensory-conduction studies

NerveStimulation siteRecording siteLat (ms)Dur (ms)Amp (mV)Dist (cm)CV (m/s)
Ulnar leftWristADM2.155.7015.806.00
Under elbow6.205.7515.4024.0059.30
Above elbow7.605.9014.608.0057.10
Peroneal commune leftAnkle EDBEDB + TA4.754.5515.607.00
Inferior fibular head EDB12.105.2016.9034.5046.90
Superior fibular head EDB14.155.2014.109.0043.90
Posterior tibula leftMedial malleolusAH2.954.9018.708.00
Popliteal fossa12.005.6515.1046.0050.80
Posterior tibula rightMedial malleolusAH2.854.8016.908.00
Popliteal fossa12.555.3513.0045.0046.40

Motor-conduction studies

NerveInsertionFibrPositive spikesFascDischFormDur (ms)PatternAmp (μV)
Extensor hallucis longus leftNPF+NInterferenceN

Needle examination

Patient 6

NerveStimulation siteRecording siteLat (ms)Amp (μV)Dist (cm)CV (m/s)
Sural rightMidline posterior lower legLateral malleolusNRNR12.00
Superficial peroneus rightLateral to tibialis anterior tendonAnkleNRNR10.00

Sensory-conduction studies

NerveStimulation siteRecording siteLat (ms)Amp (mV)Dist (cm)CV (m/s)
Peroneal commune leftAnkle EDBEDB + TA3.702.90
Inferior fibular head EDB12.202.5034.0040.00
Posterior Tibula rightMedial malleolusAH3.202.10
Popliteal fossa12.801.5037.0039.00

Motor-conduction studies

NerveInsertionFibrPositive spikes
Anterior tibial LeftN
EDB leftN
Short flexor hallucisN

Needle examination

–, absent; AH, abductor hallucis; ADM, adductor digiti minimi; Amp, amplitude; APB, abductor pollicis brevis; CV, conduction velocity; Disch, discharge; Dist, distance; Dur, duration; EDB, extensor digitorum brevis; Fasc, fasciculations; Fibr, fibrillations; Lat, latency; N, normal; NR, not responsive; PF, polyfasic units; TA, tibialis anterior; +/++/+++, gradation of presence.

Appendix

Laboratory studies

Total N-acetyl-β-d-glucosaminidase and N-acetyl-β-d-glucosaminidase A activities were determined essentially as described previously,19 using fluorescent 4-methylumbelliferyl N-acetyl-β-d-glucosaminide and its 6-O-sulfated analogue, respectively. A liquid-handling system (MulitPROBE II, Perkin-Elmer) was used for automated pipetting and incubation in 96-well plates, after which the reaction was stopped by addition of glycine buffer (pH 10.6) and fluorescence was read at 460 nm using a Victor2 (Perkin elmer).

There was a profound decrease in N-acetyl-β-d-glucosaminidase activities (total and A isoform) in plasma samples of all patients, compatible with a diagnosis of M. Sandhoff. In leucocytes, total N-acetyl-β-d-glucosaminidase activities were clearly reduced with mildly reduced levels of N-acetyl-β-d-glucosaminidase A. Considerable overlap was found with the ranges of heterozygous carriers. In fibroblasts of patient 5, a normal activity was found. After preheating of the fibroblast lysates at 52°C for 30 min, a strong reduction was observed in activity to 10% of untreated fibroblasts, while values for controls were reduced to 80% upon preheating of the samples. Thus, results obtained in fibroblasts should be interpreted with caution when adult Sandhoff disease is considered. The most discriminating element in the diagnosis of adult Sandhoff disease is the N-acetyl-β-d-glucosaminidase activity (total and A-isoform) in plasma.

References

View Abstract

Footnotes

  • Competing interests None.

  • Patient consent Obtained.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.