The inherited neuropathies are a large heterogeneous group which can be divided into those where the neuropathy is the sole or primary part of the disease, and those where it is part of a more widespread neurological or multisystem disorder (table 1). Here I will concentrate on the former group, especially Charcot-Marie-Tooth (CMT) disease and related disorders, with a particular emphasis on the diagnostic approach from a practising clinician’s perspective.

CHARCOT-MARIE-TOOTH DISEASE AND RELATED DISORDERS

Charcot-Marie-Tooth disease is not so much a single disease as a clinically and genetically heterogeneous group of inherited neuropathies, but for simplicity all types of CMT will just be referred to as CMT throughout this review. They are relatively common with an overall prevalence of 1 in 2500.¹ CMT is also referred to as hereditary motor and sensory neuropathy (HMSN) but CMT is the preferred term.

CMT is characterised clinically by distal muscle wasting and weakness, reduced tendon reflexes, impaired distal sensation and variable foot deformity, and neurophysiologically by a motor and sensory neuropathy. There is a wide variation in the age of onset and severity, depending to a large extent on the underlying genetic defect. Traditional clinical classifications differentiate between:

• CMT, with its clear motor and sensory involvement;

• hereditary sensory and autonomic neuropathy (HSAN), with much more sensory and autonomic but fewer motor features; and

• distal hereditary motor neuropathy (dHMN), which is only motor.

However, recent genetic findings have shown that certain forms of CMT and HSAN are very difficult to distinguish clinically despite being caused by mutations in different genes, and certain recently described genes for axonal CMT also cause forms of dHMN—that is, the same phenotype can be caused by different genes, and the same gene can cause different phenotypes. As there have now been more than 30 causative genes described for CMT, HSAN and dHMN, and as the technology is not yet advanced enough to allow rapid screening of all genes in an individual patient, it is crucial that clear diagnostic guidelines are available.

Is the neuropathy hereditary in an individual patient?

In the assessment of any patient with a potential hereditary neuropathy the first issue is whether the neuropathy really is hereditary. In certain cases it is easy—for example, a patient with an obviously affected parent making either autosomal dominant or X linked (if there is no definite male-to-male transmission) inheritance most likely, or affected siblings from a consanguineous marriage making autosomal recessive inheritance likely. Unfortunately with many patients it is not so straightforward because families are often small (especially in the UK, US and Northern Europe), extensive family histories are not available, relatives may not have lived long enough to develop the disease, and if mild the disease may not have been noticed by relatives. Non-paternity can also complicate the interpretation of a family history.

Several factors may help the clinician decide that the neuropathy in these “sporadic” patients is possibly hereditary:

• a long, slowly progressive history

• foot deformity such as pes cavus (fig 1) in an adult

• no definite sensory symptoms but clear sensory signs.

In the demyelinating forms of CMT (CMT1), neurophysiology can be very useful in distinguishing hereditary from acquired neuropathies because the motor conduction velocities are usually uniformly slow in the common hereditary neuropathies whereas there is often patchy slowing in the acquired neuropathies, such as chronic inflammatory demyelinating polyradiculoneuropathy (CIDP).

Is the hereditary neuropathy CMT or a related disorder?

Once the suspicion of a hereditary neuropathy has been raised, it is important to define as accurately as possible whether the neuropathy is most compatible with CMT, hereditary neuropathy with liability to pressure palsies (HNPP), HSAN or dHMN (table 1). Some tips are:

• CMT is the most common and therefore a priori the most likely diagnosis

• Although CMT is a sensory as well as a motor neuropathy, the patients may not have sensory symptoms and sometimes even no sensory signs (the most sensitive sensory sign is reduced distal vibration sensation).

• dHMN can be indistinguishable clinically from CMT; however, the sensory action potentials should always be reduced or absent in CMT, but always normal in dHMN.

• HNPP is usually easily distinguished from CMT because the patients normally have a clear history of recurrent pressure palsies; but in some patients who have accumulated neurological deficits, the phenotype can be very similar to CMT. The neurophysiological findings are usually more patchy in HNPP than in the common forms of CMT1.

• Most forms of HSAN have more sensory and autonomic, but less motor involvement than CMT. However, one form (HSAN1) is clinically very similar to a form of CMT2 (CMT2B) (neuropathic pain makes HSAN1 more likely).

CHARCOT-MARIE-TOOTH DISEASE

Charcot-Marie-Tooth disease is the most common of this group of disorders and most neurologists (adult and paediatric) and geneticists come across cases in their routine practice. Unfortunately, the classification of CMT is in a state of constant flux reflecting the rapid advances in the identification of the underlying causative genes; there are now 24 and the number is increasing rapidly (fig 2). Nonetheless, the most useful method of classification for the clinician is still a combination of neurophysiology and inheritance pattern.²

CMT is classified as either demyelinating (CMT1) if the median (or ulnar) nerve motor conduction velocity (MCV) is less than 38 m/s, or axonal (CMT2) if the median MCV is above 38 m/s (table 2). The concept of an intermediate form of CMT with median MCVs in the intermediate range (25–45 m/s) has been around since the 1970s and has recently become popular again in that it can be helpful in directing genetic diagnosis (see below). Although the inheritance pattern may be obvious (autosomal dominant, autosomal recessive or X linked), it is often difficult to ascertain because there are no other definitely affected family members, so the patient may be classified as “sporadic” until a genetic diagnosis has been established. Many of these sporadic patients have mutations in the common autosomal dominant genes (sometimes as de novo mutations) and others will be found to have mutations in the autosomal recessive genes.

In most UK, North European and US populations about 90% of cases of CMT are either autosomal dominant or X linked, whereas in countries with many consanguineous marriages, autosomal recessive CMT accounts for about 40%.³ It is therefore important to approach the diagnosis in a specific country with this in mind. CMT1 is consistently reported to be more common than CMT2 but as 75% of the genes have yet to be identified for CMT2, the true prevalence of CMT2 is unknown.

Autosomal dominant CMT1

This is the most common form of CMT in most populations. Patients usually present with the “classical CMT” phenotype in the first two decades of life with motor symptoms in the lower limbs—for example, difficulty walking or foot deformity. They usually have distal wasting and weakness and hyporeflexia affecting the lower more than the upper limbs (fig 1). Distal sensory loss and foot deformity are frequent findings. Neurophysiologically the median MCVs are below 38 m/s and the sensory action potentials (SAPs) are either reduced or absent. Nerve biopsy is no longer necessary to make the diagnosis because genetic testing is now widely available (if it is done it shows a demyelinating neuropathy with classical onion bulbs) (fig 3).

If there is a clear family history of autosomal dominant inheritance, if the patient is apparently “sporadic”, or if there is definite male-to-male transmission (ruling out X linked inheritance), then the most likely diagnosis is CMT1A secondary to the chromosome 17 duplication containing the peripheral myelin protein 22 gene (PMP22). In European populations this duplication accounts for 70% of all CMT1 cases.⁴ Mutations in the PMP22 gene can also cause CMT1A, and a wider spectrum of phenotypes including rare cases of HNPP and a more severe form of CMT1.

CMT1B is a less common form of CMT1 secondary to mutations in the human myelin protein zero gene (MPZ). Patients can present with the classical CMT1 phenotype but are more likely to have either a more severe early onset form of CMT or a much milder late onset form of CMT with median MCVs in the axonal range.⁵ Therefore because MPZ mutations can cause CMT1 and CMT2 (see below), MPZ-related CMT can be classified as intermediate CMT neurophysiologically.

The two other genes (EGR2 and LITAF) that can cause AD CMT1 are very rare, accounting for less than 1% of cases each, and they have no particular distinguishing clinical features. Mutations in EGR2 can also cause a more severe CMT1 phenotype. Mutations in NEFL were originally described as a cause of CMT2 but it is now recognised that some patients have median MCVs in the demyelinating range and would be classified as CMT1. Once again NEFL-related CMT could be classified as intermediate CMT neurophysiologically.

Severe CMT1 (HMSN III, Dejerine Sottas disease, congenital hypomyelinating neuropathy)

The severe cases of CMT1 used to be called HMSN III in older classifications and subdivided into Dejerine Sottas disease and congenital hypomyelinating neuropathy depending on the underlying pathology. They usually present in the first decade with very slow MCVs and a more severe neuropathy than classical CMT1. It was thought that most of these cases were recessive but we now know that they are usually secondary to de novo dominant mutations in the three genes that commonly cause CMT1 (PMP22, MPZ, EGR2). As can be seen in figure 2, the disease causing mutations in these three genes in Dejerine Sottas and congenital hypomyelinating neuropathy can occasionally be inherited in an autosomal recessive fashion.

Hereditary neuropathy with liability to pressure palsies

Hereditary neuropathy with liability to pressure palsies (HNPP) is an autosomal dominant condition usually caused by a deletion of the same 1.4 megabase portion of chromosome 17 that is duplicated in CMT1A. Point mutations in PMP22 rarely cause HNPP. Most patients present with episodic recurrent pressure palsies although atypical presentations, including recurrent focal transient sensory symptoms and a scapuloperoneal syndrome, have been described. Diagnostically an important point is that although patients may present with only one nerve clinically involved at a particular time, there is always a more generalised patchy demyelinating neuropathy neurophysiologically. Nonetheless, screening isolated pressure palsies for this mutation is not warranted unless there are more widespread nerve conduction abnormalities.

X linked CMT1

This is the second commonest form of CMT and is caused by mutations in the connexin 32 gene; over 300 mutations have been described (gap junction protein beta 1 gene). As expected for an X linked disorder, males are more severely affected than females. The MCVs in affected males are usually in the demyelinating range but can be in the axonal range, while the MCVs in the females are usually in the axonal range but can be in the demyelinating range. Connexin 32 related CMT is therefore another form of intermediate CMT neurophysiologically.

In recent years the phenotype associated with connexin 32 mutations has been clarified. Patients (especially males) have a rather patchy neuropathy, both clinically and neurophysiologically, with less uniform conduction slowing and more pronounced dispersion than in autosomal dominant CMT1.⁶ This can lead to difficulties in diagnosis in patients without a family history. I have seen two patients misdiagnosed as CIDP leading to unnecessary immunosuppressive therapy. The central nervous system is occasionally involved but this is usually mild (extensor plantars, mild deafness, abnormal brainstem evoked potentials), and there are some case reports of transient symptomatic white matter CNS involvement.⁷

Approach to the molecular diagnosis of autosomal dominant and X linked CMT1

Because in most UK, North European and US populations about 90% of cases of CMT are either autosomal dominant or X linked, a rational approach to molecular diagnosis is necessary (fig 4). In individual cases the clinician should seek expert advice if unsure which genetic test to request.

Autosomal recessive CMT1

In certain communities autosomal recessive CMT may account for 40% of all CMT cases, whereas in a typical northern European population it accounts for less than 10%. There are now 10 genes described that can cause autosomal recessive CMT1 (including the three genes—PMP22, MPZ and EGR2—that more commonly cause autosomal dominant or de novo dominant CMT1) (table 2). Unfortunately there is no one gene that is the most frequent cause of autosomal recessive CMT1, which makes the diagnosis difficult for a patient with presumed autosomal recessive CMT1. Further confusion arises as autosomal recessive CMT1 is called CMT4 in the genetic literature and hence the various forms of autosomal recessive CMT1 are classified as CMT4A, CMT4B, etc.

Although no one algorithm is suitable for the diagnosis of autosomal recessive CMT1, there are some simple clinical rules that can be used to aid diagnosis. Defining the phenotype accurately is a crucial first step. Although a history and physical examination are easily obtained, neurophysiology may be difficult because in severe cases it may not even be possible to distinguish CMT1 from CMT2, because the nerves can be unexcitable in both CMT1 and CMT2. Nerve biopsy can be particularly useful in these cases, as they are in autosomal recessive CMT1 in general, because specific features make a particular genetic diagnosis more likely.

The first question to answer is whether the inheritance is autosomal recessive. Obviously with consanguineous marriages, and especially with multiple siblings affected, this becomes very likely, but in other cases it may be difficult to prove. As a general rule autosomal recessive forms of CMT have an earlier onset and are more severe than autosomal dominant cases. They usually start as a length-dependant neuropathy but are more likely to progress to involve the proximal muscles and to result in loss of ambulation than in autosomal dominant patients. Certain clinical features, the ethnic background of patients, and specific neuropathological features, can aid the diagnosis (table 2):

• CMT4A secondary to mutations in GDAPI is usually an early onset progressive neuropathy and can be associated with diaphragmatic and vocal cord involvement (mutations in GDAP1 can also cause autosomal recessive CMT2 making GDAP1 related CMT another intermediate form of CMT neurophysiologically).

• Nerve biopsies showing focally folded myelin are characteristic of CMT4B1 (MTMR2 mutations) and CMT4B2 (MTMR13 mutations) although this has also been described with MPZ mutations and in CMT4F secondary to periaxin mutations.

• Severe and early scoliosis is seen with CMT4C due to mutations in the KIAA1985 gene. The patients also have characteristic nerve biopsy features including basal membrane onion bulbs and multiple cytoplasmic processes of the Schwann cells of unmyelinated axons.⁸

• Two forms of autosomal recessive CMT1 are largely confined to patients of Balkan gypsy origin. CMT4D secondary to NDRG1 mutations is characterised by a demyelinating neuropathy with a high prevalence of deafness. Tongue atrophy has also been described in this form of CMT. Another form seen in this gypsy population is CCFDN (congenital cataract, facial dysmorphism, and neuropathy syndrome) secondary to CTDP1 mutations.⁹

• A phenotype ranging from severe Dejerine Sottas to a milder neuropathy with mainly sensory involvement is seen in CMT4F secondary to periaxin mutations.

To summarise, although many different genes cause autosomal recessive CMT1, careful phenotyping together with consideration of the ethnic background can help guide the genetic diagnosis.

Autosomal dominant CMT2

It can be difficult to decide whether a patient has autosomal dominant CMT2 or an idiopathic axonal neuropathy because many adults present with a very longstanding, generally mild axonal neuropathy without an obvious family history. Because investigation so often fails to reveal an acquired cause, the possibility of autosomal dominant CMT2 is raised but is difficult to prove because so far only eight causative genes have been described (table 2) and they only account for about 25% of all the cases. Until all the causative genes are identified, we will not know the true prevalence of this condition. Furthermore, unlike the chromosome 17 duplication in autosomal dominant CMT1, there is no major causative gene for autosomal dominant CMT2. However, patients with autosomal dominant CMT2 can be divided into three different groups by phenotype which can then be used to direct genetic testing.

Most patients with autosomal dominant CMT2 present with the “classical CMT” phenotype, indistinguishable from autosomal dominant CMT1 until neurophysiology is done showing reduced motor action potentials (MAPS) with normal or near normal MCVs. The SAPs are reduced or absent. Nerve biopsies are rarely done, but if done show an axonal neuropathy without any specific diagnostic features. The major cause of the classical phenotype is in mitofusin 2 (MFN2) causing CMT2A¹⁰; mutations in this gene account for approximately 20% of all cases of autosomal dominant CMT2 in all populations tested to date. An important point about this gene is that many of the mutations (approximately 20%) are de novo, explaining why so many of these patients have normal parents. Patients with mutations in MFN2 may present early with a more severe phenotype and fairly rapid progression to proximal muscle involvement and loss of ambulation. Occasional patients also have brisk reflexes. Earlier last year mutations in MFN2 were also described as causing axonal CMT with optic atrophy (HMSN VI in previous classifications).¹¹ Prior to the identification of MFN2, one family had been described with a mutation in KIF1Bβ but this is an extremely rare cause of CMT2 with MFN2 now clearly established as the major cause. In patients with classical autosomal dominant CMT2 and no mutation in MFN2, MPZ and NEFL should be screened, although they rarely cause this phenotype. Two recently described heat shock protein genes (HSP27 and HSP22) also cause the classical CMT2 phenotype although this is rare (these two genes are of particular interest as they can also cause a purely motor phenotype, dHMN type II).

The second phenotype of autosomal dominant CMT2 has much more sensory involvement than is usually seen with CMT. These patients have been described as having an “autosomal dominant inherited neuropathy with prominent sensory loss and ulceromutilations”.¹² They were originally classified as CMT2B and usually presented in the second or third decade with typical motor CMT features but also severe sensory loss and sensory complications including ulcerations, osteomyelitis and amputations. The causative gene is RAB7.¹³ This presentation is very similar to that of patients with HSAN1 secondary to mutations in the SPTLC1 gene. Patients with SPTLC1 mutations usually have less motor involvement at presentation and they have neuropathic lancinating pain, an unusual feature in hereditary neuropathies. Therefore, if patients present with autosomal dominant CMT2 with prominent sensory features, both the SPTLC1 and the RAB7 genes should be initially screened; the presence of neuropathic pain suggests SPTLC1 should be screened first.

The third phenotype of autosomal dominant CMT2, classified as CMT2D, particularly affects the upper limbs. Patients present with wasting and weakness of the small muscles of the hand (this can be unilateral and misdiagnosed as thoracic outlet syndrome) with much later involvement of the distal lower limb muscles. Interestingly, as with mutations in HSP27 and HSP22, some of these patients have no sensory involvement and were in the past classified as dHMN type V. CMT2D and dHMN V have now been shown to be allelic conditions caused by the GARS gene.¹⁵ Some patients with the dHMN V phenotype (no sensory involvement) also have mutations in the BSCL2 gene which usually causes Silver syndrome (spastic legs and distal amyotrophy of the upper limbs) but can present (33% of cases) with just amyotrophy of the upper limbs. Therefore in patients presenting with CMT2/dHMN of upper limb onset, GARS should be screened if there is sensory involvement, and both GARS and BSCL2 if there is only motor involvement.

Approach to molecular diagnosis of autosomal dominant CMT2

Clinicians will want to screen the CMT2 genes in families with autosomal dominant CMT2, but also in individual patients in whom they suspect a hereditary neuropathy. This is particularly important in autosomal dominant CMT2 as mutations in the commonest causative gene, MFN2, often occur as de novo dominant mutations and will therefore be passed on to 50% of the children. The algorithm in figure 5 is a guide to the molecular diagnosis but in individual cases, as with CMT1, the clinician should seek expert advice if unsure what to request.

Autosomal recessive CMT2

There are only three known causative genes for AR CMT2 (table 2). In an individual case with CMT2 the possibility of autosomal recessive inheritance arises in the same circumstances as for autosomal recessive CMT1, especially in patients who are the product of a consanguineous marriage.

Mutations in lamin A/C (LMNA) are one cause.¹⁶ Most patients present in the second decade with a severe CMT phenotype including proximal muscle involvement, although some have a milder phenotype. Lamin A/C mutations have been associated with a wide spectrum of other phenotypes including Emery-Dreifuss muscular dystrophy, cardiomyopathy, and Dunniugan-type familial partial lipodystrophy.

The second form of autosomal recessive CMT2, ARCMT2B, has only been described in one Costa Rican family and is due to mutations in the MED25 gene. The phenotype is milder than with the other autosomal recessive CMT2 genes, more like classical CMT2.

Mutations in GDAP1 have been described above as a cause of autosomal recessive CMT1 (CMT4A) but they can also cause autosomal recessive CMT2 with a similar severe early onset phenotype which can include vocal cord paresis.¹⁷

X linked CMT2

One X linked form of CMT2 has been described linked to Xq24-Q26, but no gene has been identified.

Dominant intermediate CMT

As discussed above, it is increasingly recognised that many forms of CMT can present with MCVs in the intermediate range. These include patients with:

• X linked CMT due to CX32 mutations

• autosomal dominant CMT due to MPZ or NEFL mutations

• autosomal recessive CMT due to GDAP1 mutations.

In addition, two genes have recently been described (DNM2 and YARS) that cause classical CMT with intermediate MCVs and these have been classified as dominant intermediate CMT (table 2).

Hereditary neuralgic amyotrophy

Hereditary neuralgic amyotrophy (HNA) is an autosomal dominant condition characterised by recurrent episodes of typical brachial neuritis. The gene has recently been identified as the septin 9 gene (SEPT9).¹⁸

HEREDITARY SENSORY AND AUTONOMIC NEUROPATHIES

The hereditary sensory and autonomic neuropathies (HSANs) are much rarer than CMT but many of the genes have been identified (table 3). Generally they have more sensory and autonomic but less motor involvement than CMT. The sensory loss can lead to severe complications including recurrent injuries, ulcerations, osteomyelitis and amputations (fig 6).¹⁹ Education of the patients to try to prevent these complications is crucial, but even the most careful patient can develop them.

The commonest form in the UK is HSAN1 secondary to SPTLC1 mutations described above (more appropriately termed HSN as there is usually no autonomic involvement). In UK patients with the C133W SPTLC1 mutation, motor involvement can occur early in the disease course and the MCVs can be in the demyelinating range. Many of these patients were previously diagnosed as CMT. This disease is very difficult to differentiate from CMT2B secondary to RAB7 mutations although the lancinating pain in patients with SPTLC1 mutations is a useful guide to this diagnosis.

All other forms of HSAN are rare:

• HSAN II is an early onset autosomal recessive severe sensory neuropathy with prominent sensory complications, due to mutations in the HSN2 gene.

• The Riley-Day syndrome is a distinct autosomal recessive neuropathy seen in Ashkenazi Jews and characterised by mainly autonomic involvement, but it also involves the peripheral nervous system, particularly the sensory nerves. The causative gene is the IKBKAP gene.

• HSAN IV and V are both autosomal recessive neuropathies characterised by congenital insensitivity to pain. HSAN IV (also called congenital insensitivity to pain with anhidrosis) presents with a severe sensory neuropathy, anhidrosis and mental retardation and is due to mutations in the NTRK1 gene. HSAN V is similar but without the mental retardation or significant anhidrosis, described with both NTRK1 and also NGFβ mutations.²⁰

DISTAL HEREDITARY MOTOR NEUROPATHIES

The distal hereditary motor neuropathies (dHMNs) are a complex group of disorders (table 4); the forms that resemble CMT have already been discussed in the section on autosomal dominant CMT2. dHMN II is the classic form of autosomal dominant dHMN and is due to mutations in the HSP27 and HSP22 genes. The patients present with a syndrome typical for classical CMT but without any sensory involvement. dHMN I is a similar autosomal dominant disorder but with earlier onset for which no gene has been identified. There are many other forms but the genes are only known for a few of them:

• Mutations in GARS and BSCL2 cause dHMN V as described above.

• An unusual severe autosomal recessive form of dHMN, dHMN VI, presents in infancy with respiratory and distal limb involvement (called spinal muscle atrophy with respiratory distress type 1). This is due to mutations in the IGHMBP2 gene.

• Mutations in dynactin (DCTN1) cause one form of dHMN type VII, which is characterised by vocal cord paralysis and progressive weakness and atrophy of the face, hands and legs.

• Another very similar form, also called dHMN VII, with vocal cord paralysis has been mapped to a different locus.²¹

• Finally, missense mutations in senataxin (SETX) can cause a form of dHMN with pyramidal features, whereas nonsense mutations in the same gene cause autosomal recessive ataxia oculomotor apraxia type 2.

As can be seen in table 4, there are other forms of dHMN for which no causative genes have yet been identified, including an autosomal recessive form with pyramidal features seen in families in the Jerash region of Jordan.

CONCLUSIONS

The hereditary neuropathies are clinically and genetically heterogeneous. Although the area is complex and the number of causative genes is increasing rapidly, algorithms to guide the appropriate genetic testing for the more common autosomal dominant disorders, and careful phenotyping for the rare autosomal recessive disorders should aid diagnosis. One of the major issues in the UK and throughout the world is the lack of availability of testing for the rarer genes. Many countries (including the UK) offer routine diagnostic testing for the common CMT1 (chromosome 17 duplication, PMP22, MPZ, CX32) and CMT2 (MFN2) genes. Most of the other genes are only available through research laboratories but as these forms of CMT, HSAN and dHMN are rare, it is worth referring the patients to a neurologist with a special interest in these disorders before embarking on extensive genetic screening. It cannot be emphasised how important an accurate genetic diagnosis is for these patients. There are obvious benefits including diagnostic testing of other family members, accurate genetic counselling and prognosis, predictive and antenatal testing. Furthermore, an accurate diagnosis prevents patients from having invasive diagnostic tests—for example, nerve biopsies and lumbar punctures. In certain situations where inflammatory neuropathies are being considered, a genetic diagnosis can also prevent a trial of potentially dangerous immunosuppressive therapy. Finally, the era of specific treatment for genetic neuropathies is here. The first therapeutic trials have just begun (ascorbic acid for CMT1A) and because any treatments developed are likely to be gene specific, the case for an accurate genetic diagnosis will become even more convincing.