Introduction

Lambert–Eaton myasthenic syndrome (LEMS) is a pre-synaptic disorder of the neuromuscular transmission mediated by antibodies to P/Q-type voltage-gated calcium channels (VGCC) at the motor nerve terminal [1]. Nearly 50–60 % of patients with LEMS have an underlying tumour (T-LEMS) that in 90 % of cases is a small-cell lung cancer (SCLC). Non-tumour LEMS (NT-LEMS) is often associated with HLA-B8, DR3 haplotype and to other autoimmune diseases, such as rheumatoid arthritis, pernicious anaemia, thyroid disorders and Sjogren’s syndrome [2]. The antigenic stimulus for anti-VGCC antibody production in patients with SCLC-LEMS appears to be the tumour’s functional VGCC [3]. The trigger for the production of anti-VGCC antibodies in LEMS with no detectable lung cancer (non-SCLC-LEMS) is unknown.

LEMS treatment is based on a symptomatic approach and on therapies lowering circulating auto-antibody levels or interfering with their function [immunosuppressant drugs, plasmapheresis, intravenous immunoglobulin (IVIG)]. T-LEMS is a typical paraneoplastic disease as treatment of the underlying cancer significantly improves neurological symptoms, whose onset generally occurs before tumour detection. LEMS diagnosis is crucial not only to provide a proper treatment of the neurological disease but also to detect early in the course of the disease the possible underlying tumour [4]. However, early diagnosis is hampered by LEMS being a rare disease and by the fluctuating course of symptoms. Even when patient’s complaints have been properly related to a neuromuscular transmission disorder, LEMS is relatively often misdiagnosed as myasthenia gravis (MG) [5]. Moreover, LEMS symptoms in tumour patients may be ascribed to general conditions decline and to the consequences of the oncologic treatment, or may be masked by other paraneoplastic disorders associated with SCLC.

To address the issues related to LEMS management, the Italian Working Group on Myasthenic syndromes (GISMIA) developed the diagnostic and therapeutic recommendations presented in this paper. The aim of this review was to propose the GISMIA consensus on LEMS management, and to provide clinicians with a practical approach that could work within the specific framework of the Italian clinical practice setting.

Epidemiological overview on LEMS

LEMS is a rare disease, with scarce epidemiologic data mostly coming from national neuromuscular referral centres. The estimated annual incidence is nearly 10 times lower than that of MG and its frequency in SCLC patients is around 3 % [2]. In a 10-year study (1990–1999) run in a population of 1.7 million inhabitants in South Holland, the annual incidence rate of LEMS (0.48 × 10−6) and its prevalence (2.32 × 10−6) were 14 times and 46 times lower, respectively, than of MG in the same region [6]. Subsequent data on the same population showed that LEMS incidence in Netherlands rose to 0.75 × 10−6, with a prevalence of 3.42 per million, probably due to an improved recognition of the disorder [5]. Such data confirm that LEMS is quite rare and probably under-diagnosed.

On the basis of the Netherlands studies, in Italy the expected incidence and prevalence should be around 25–40 cases per year and 140–200 cases, respectively. Although LEMS registry data could clarify this aspect, the frequency of the disease in Italy seems to be quite lower than expected, most likely because of under-diagnosis.

LEMS diagnosis

Clinical findings

Diagnosis of LEMS is based on clinical data, electrophysiological studies (showing a pre-synaptic defect of the neuromuscular transmission) and immunological testing (anti-VGCC antibody assay). LEMS should be suspected in (mostly) adult patients presenting with proximal muscle weakness, particularly of legs, together with reduced/absent tendon reflexes. The clinical suspicion is strengthened by concomitant signs of autonomic dysfunction, with dry mouth, erectile dysfunction and constipation as the most frequently reported symptoms [7].

The distribution of muscle weakness during disease progression has been investigated in several studies [7–10]. Proximal leg muscle weakness is always present and is the most frequent complaint at onset; weakness of the arms is a common and early sign as well, reflecting a typical clinical pattern: weakness generally spreads proximally to distally and caudally to cranially, involving feet and hands and finally reaching cranial muscles [5]. The evidence of post-exercise facilitation (transient normalisation of tendon reflexes and improvement of muscle strength after a brief maximal contraction) is a disease hallmark, although it is evident in only 40 % of cases [11]. Bulbar symptoms are generally less severe than in MG. Ptosis and diplopia are less frequent and usually of moderate degree, even though patients with exclusively ocular signs have been described [12]. The disease is usually more rapidly progressive in T-LEMS patients.

Electrophysiological studies

A clinical suspicion of LEMS should be confirmed by electromyography (EMG). Although in these patients, weakness is more evident in proximal districts, electrophysiological findings are more easily detected on distal muscles [13]. LEMS diagnosis is based on the presence of the typical triad: (1) low amplitude of the compound muscle action potential (CMAP) at rest (0.1–6 mV); (2) further decrease of CMAP amplitude during low-rate (2–5 Hz) repetitive nerve stimulation (RNS); (3) CMAP amplitude increase during high-rate RNS (20–50 Hz) or immediately after a brief maximal voluntary contraction (15–20 s). The latter is the technique of choice as it is better tolerated [2]. Screening should be performed at least in two distal muscles. For technical convenience and relative comfort for the patient, the abductor digiti minimi is often studied.

A decrease of >10 % of the CMAP amplitude (decrement) during low-rate RNS does not discriminate between LEMS and MG; therefore, it is important to apply the whole protocol in every case where a neuromuscular transmission disorder is suspected. An increment in CMAP amplitude/area (post-tetanic/post-exercise facilitation) of 100 % or greater, in at least two tested muscles, is fairly typical of LEMS, with a sensitivity ranging from 84 to 96 %. A reduction in the normal limit for post-exercise facilitation from 100 to 60 % was reported to increase diagnostic sensitivity to 97 %, while retaining a specificity of 99 % [14]. Sensitivity is also increased by withdrawal of symptomatic medication 12 h before electrodiagnostic testing. However, low amplitude of CMAP as well as post-tetanic/post-exercise facilitation can be found in other disorders including overlap myasthenic syndrome, botulism, drug-induced paralysis and periodic paralysis, although, in these conditions, facilitation is less pronounced than in LEMS. Therefore, low amplitude of CMAP at rest or an incremental response per se does not establish a diagnosis of LEMS, but these findings must be evaluated within the clinical context. Furthermore, facilitation may not be evident in those rare cases presenting with severe weakness [8]. Low-frequency RNS during maximal contraction could increase EMG test sensitivity [15]. Single-fibre EMG might be more sensitive than RNS but it does not distinguish between MG and LEMS.

Antibody assay

Antibodies to P/Q VGCC are present in 75–90 % of LEMS patients; positivity rate can approach 100 % in T-LEMS cases, and is generally associated with high-antibody titres [16]. P/Q-type antibodies are highly specific to LEMS, having been only described in association with paraneoplastic SCLC-related cerebellar ataxia and in 2–3 % of SCLC patients without neurological dysfunction [16]. Seronegative LEMS (without detectable anti-VGCC antibodies) does not show distinctive clinical features, except for the lower association with SCLC [17].

Anti-VGCC antibody detection represents a highly specific diagnostic confirmation. The absence of these antibodies does not exclude a diagnosis of LEMS, though it reduces the probability of association with SCLC.

The diagnostic algorithm for LEMS we are proposing (Fig. 1) has been developed based on the Italian healthcare environment. While EMG studies can be performed in most neurological clinics, the availability of the anti-VGCC antibody assay is less widespread, even if commercial kits are available. As a matter of fact, both tests have a comparable diagnostic specificity, and antibody detection in a patient with consistent symptoms confirms the diagnosis of LEMS even without EMG confirmation.

Fig. 1
figure 1

LEMS diagnosis algorithm. EMG electromyography, anti-VGCC Ab antibodies to voltage-gated calcium channels. When LEMS is suspected on clinical grounds, EMG is usually the first diagnostic step. If EMG criteria are met, LEMS is highly likely. Anti-VGCC antibody (Ab) detection confirms the diagnosis, while a negative Ab assay does not exclude the diagnosis (anti-VGCC negative LEMS), rather it makes the chance of small-cell lung cancer association less likely. If EMG is negative, the diagnosis of LEMS relies on anti-VGCC antibody detection: a negative Ab assay in this context makes LEMS diagnosis very unlikely

Full size image

Antibodies against synaptotagmin [18] and M1 muscarinic acetylcholine receptors [19] have also been described occasionally in LEMS patients. Such antibodies are not routinely tested and currently have no diagnostic use.

T-LEMS: diagnosis of the associated cancer

Oncologic screening

Once LEMS is diagnosed, screening for a possible associated cancer is mandatory. More than 90 % of T-LEMS patients are affected by SCLC. Other types of tumours underlying LEMS (lung, prostate and breast carcinoma, thymoma and lymphoma) have been rarely reported [2, 5].

In LEMS patients, factors associated with a higher risk of SCLC are older age (≥50 years) at disease onset, rapid disease progression, high anti-VGCC antibody levels and being a smoker. All patients should undergo an oncologic screening as early as possible after LEMS diagnosis.

Oncologic risk assessment

SOX (sry-like high-mobility group box) proteins belong to different families of transcriptional factors that are expressed in the developing nervous system and in the adult cerebellum (Bergmann’s cells), as well as in some neuroendocrine cancers [20].

Anti-SOX antibodies, especially those against the main antigen SOX1, are important serological markers of SCLC. They are detected in SCLC-related paraneoplastic neurological syndromes (PNS) and are strongly associated with T-LEMS [20]. In particular, anti-SOX1 antibodies are found in 64–67 % of patients with SCLC-LEMS, in 22–32 % of SCLC patients without PNS, but in only 5 % of NT-LEMS cases [21, 22].

According to the guidelines of the European Federation of Neurological Societies (EFNS) [23], screening for SCLC should be performed with computerised tomography of the thorax (thorax-CT) followed, when negative, by positron emission tomography (PET) or integrated PET-CT. If the first screening is negative, oncologic surveillance should be continued by periodic screenings for at least 2 years after LEMS onset [23].

Assessing the oncologic risk in each patient is crucial to establish how long the search for SCLC should continue. As reported above, anti-SOX antibodies have a 67 % sensitivity as T-LEMS markers, but they are not extensively available. Neuron-specific enolase (NSE) has a sensitivity of 65 % as tumour marker in SCLC patients. NSE positivity is, however, dependent on the tumour-stage, and it is less frequent in patients with limited disease [23]. Recently, a clinical scoring system (Dutch-English LEMS Tumour Association Prediction [DELTA-P]) to predict SCLC in LEMS patients has been proposed [24]. DELTA-P consists of recording, within 3 months from the onset, the following parameters whose presence/absence are graded 1/0: (D) dysarthria, dysphagia, neck weakness (bulbar symptoms), (E) erectile dysfunction (in women this parameter is scored 0), (L) loss of weight ≥5 %, (T) tobacco use, (A) age of onset ≥50 years, and, (P) Karnofski performance status <70 [25]. DELTA-P score ranges from 0 to 6, and higher scores correlate with an increasing risk of SCLC association [5]. Thus, in LEMS patients, DELTA-P can be useful in deciding for how long oncologic screenings should be repeated when the first CT/PET-CT study is negative [5].

LEMS treatment

A treatment algorithm is shown in Fig. 2. Based on randomised clinical trials, symptomatic treatment of LEMS with 3,4-diaminopyridine (3,4-DAP) is effective and well tolerated [26], representing the first therapeutic approach in all patients. A novel calcium channel agonist (GV-58), which works slowing deactivation of the channels, is currently being evaluated for the treatment of neuromuscular weakness [27].

Fig. 2
figure 2

Treatment algorithm for LEMS patients. 3,4-DAP 3,4-diaminopyridine, SCLC small-cell lung carcinoma, IVIG intravenous immunoglobulin, T-LEMS tumour-LEMS, NT-LEMS non-tumour LEMS. Symptomatic treatment with 3,4-DAP represents the first therapeutic approach in all LEMS patients. In T-LEMS, oncologic treatment usually improves the neurological disease; for patients with disabling symptoms not satisfactorily controlled by 3,4-DAP, prednisone (associated with plasmapheresis or IVIG when needed) is indicated also at an early stage of the disease. In NT-LEMS, severely affected patients are treated with prednisone in the initial phase; in the subsequent course, some of these cases require immunosuppressants, such as azathioprine, cyclosporine A, or mycophenolate mofetil; rituximab has been used for refractory disease. Plasmapheresis and IVIG can be used to treat disease relapses in association with prednisone and/or immunosuppressants

Full size image

In T-LEMS, treatment of the associated tumour generally induces a significant improvement of neurologic symptoms. When symptoms are disabling, prednisone therapy is indicated and can be started during the oncologic screening, if necessary. In patients with severe disease, in whom a rapid therapeutic response is required, high-dose prednisone can be associated with plasmapheresis or intravenous immunoglobulin (IVIG). In T-LEMS patients, use of immunosuppressants other than steroids is controversial: the general trend is to avoid them as far as possible, even though there is no definite contraindication to azathioprine treatment [5].

In LEMS not associated with tumour (NT-LEMS), treatment is similar to that of MG. Patients with disabling symptoms, not adequately controlled with 3,4-DAP, are given prednisone starting at high dose with subsequent dosage tapering according to the clinical response. In patients with an inadequate response to prednisone, requiring high dosages or suffering from steroid-related adverse events, treatment with immunosuppressants is indicated (azathioprine as first choice, cyclosporine or mycophenolate mofetil as second-choice drugs). Patients with severe relapsing symptoms can be treated with plasmapheresis or IVIG in any phase of the disease [2, 28]. Some recent evidence has shown rituximab to be effective in treating LEMS, as in other immune-mediated diseases [29] (Fig. 2).

Conclusions

Many advances have been made in optimising tumour screening in LEMS as well as its treatment. The awareness of this rare disease is also improving, so it is crucial to have specific tools to guide diagnosis and treatment, based on good clinical practice. The algorithms presented in this review are the results of the Italian GISMIA Working Group consensus and are intended to provide a practical clinical approach to LEMS management.