Natural history of young-adult amyotrophic lateral sclerosis

Amyotrophic lateral sclerosis (ALS) is the most common and the best recognized form of the motor neuron diseases. It is characterized by relentless degeneration of both upper and lower motor neurons leading to progressive muscular paralysis, with death usually occurring 1 to 5 years after the onset.

In approximately 5% to 10% of patients, the disease is inherited, but genes have been identified only in a minority of cases.^1–3 Although in the past few years much progress has been made in the research, the cause of sporadic ALS remains unknown.

The clinical picture of ALS is a stereotypical one, resulting from a combination of signs secondary to dysfunction of both upper and lower motor neurons. However, clinical heterogeneity is a recognized feature of the disease, with respect to topographic distribution of the clinical signs, duration of the disease, and age at onset. Based on the clinical pattern in the initial phases of illness, different variants have been described, including spinal, bulbar, and pseudoneuritic forms,⁴ flail arm syndrome,⁵ and Mill's hemiparetic type.⁶ Most patients disclose the classic phenotype, in which lower motor neuron signs are prominent, whereas some patients present with a predominant upper motor neuron phenotype (p-UMN).⁴ A variable duration of the disease is observed: survival is 3 to 5 years in most patients, but a rapidly progressive course can also occur, whereas approximately 10% of patients survive for more than 10 years.^7,8

With regard to the age at onset, ALS affects people of all ages, but the incidence rate varies greatly in different age groups. The age-adjusted incidence rate is very low in the first four decades (<1.5 × 100,000/year), increasing abruptly around age 40 years, reaching its peak between ages 60 and 79 (10–15 × 100,000/year), and decreasing thereafter.^9–12 The significance of such a peculiar distribution by age is unclear. If ALS is a single disorder, the presence of a peak in the age-specific incidence curve suggests that the disease is the result of a time-dependent exposure to some risk factors whose nature is still unknown. According to this hypothesis, an early onset of the disease might reflect a major exposure to one or more of these risk factors. However, the existence of different mechanisms causing or triggering ALS in different age groups should also be considered.

In this report, we describe the clinical features and long-term follow-up of 57 patients affected by sporadic young-onset ALS, and compare them with 450 ALS patients with the common adult-onset form.

METHODS

Seven hundred five patients affected by motor neuron disease were observed at our Neurological Institute from 1987 to 2007. Our institute is the referral center for ALS in Lazio Region, and nearly 70% of patients who reside in this area are admitted to our center. The remaining patients address themselves to other neurological institutes in the same region. Approximately 70% of our patients are from Lazio Region; the rest come from other regions in central or southern Italy. Seventy-five percent of patients with onset before age 40 years came from Lazio Region. We excluded 152 patients because of insufficient clinical or follow-up data. Primary lateral sclerosis was diagnosed in 22 patients because at last follow-up (ranging from 36 to 187 months) they did not show any clinical or electromyographic evidence of lower motor neuron impairment. The remaining 531 patients were diagnosed as having ALS according to revised El Escorial–Arlie House Criteria¹³ for definite, probable, or laboratory-supported probable ALS. Every patient underwent serum immunoelectrophoresis, cranial and spinal cord MRI, electromyographic examination, evoked motor and sensory potential examinations, and functional respiratory tests. Familial ALS was defined in 23 patients who referred an affected relative; the remaining 508 patients were considered to be sporadic and were included in the present study. All patients were visited by us and were followed up with clinical and respiratory evaluation every 2 to 4 months. The follow-up period ranged from 2 to 226 months (mean 36.7, median 29).

Patients were defined as having juvenile ALS when the onset of the disease had occurred before age 20 years, patients with onset between 20 and 40 were classified as young-adult ALS, and those with onset after 40 were termed adult-onset type.

Based on clinical examination, ALS patients were divided into two phenotypes: p-UMN ALS and classic ALS. In the former group were those patients whose clinical manifestations were dominated by pyramidal signs, mainly severe spastic paraparesis, associated with one or more of the following signs: Babinski or Chaddock signs, hyperactive reflexes, dysarthric speech, and Hoffmann sign. Spastic paraparesis could be present at the beginning or in the fully developed stage of the disease. All these patients showed, by definition, clear signs of lower motor neuron impairment since the beginning of the disease.

The classic type was defined by the presence of prevailing lower motor neuron signs associated with slight to moderate pyramidal signs.

Univariate survival modeling was performed using Kaplan–Meier analysis and a log-rank sum test. The Cox proportional hazards regression model was used to assess the simultaneous effects of several variables on survival. Estimated survival was defined as the period of time from symptom onset to last follow-up or death/tracheotomy. Statistical analysis was performed using the program SPSS version 14.

RESULTS

Clinical features.

Classic phenotype.

The classic phenotype was observed in 395 of 507 patients (77.9%). Age at onset varied from 31 to 84 years (mean 60.8, median 62; figure 1A). This clinical pattern was found in 23 of 57 young-adult patients (40.4%) and in 372 of 450 patients (82.6%) with the adult-onset type (p < 0.0001). The distribution by age was unimodal in men and women, with a peak in the seventh decade of life in both sexes. The ratio of men to women was 1.4:1 in the whole population, and it did not vary substantially in different age groups. The ratio of men to women was 1.1:1 in the group of young-adult patients and 1.4:1 among patients with onset after 40 years. In the young-adult group, the first symptoms affected the legs in 6 patients (26%), the arms in 12 (52%), and the bulbar region in 5 (22%). Six patients received nocturnal noninvasive assisted ventilation.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1 Age distribution of patients separated by phenotype

(A) Three hundred ninety-five patients with classic phenotype. The distribution is similar in men and women, with a peak in the seventh decade of life. Male prevalence is present in all age groups. (B) One hundred twelve patients with p-UMN phenotype. Distribution is unimodal in women, with a peak in the sixth decade of life. Men show a first peak in the fourth decade of life and a second peak in the seventh decade of life. Men predominate markedly before age 40 years, but after this age, women are more numerous than men.

Sequencing analysis of the SOD1 gene was performed in 15 of 23 patients, with normal results in all cases, including the 12-year-old patient, who also underwent mutational analysis of the Alsin gene, with normal results.

p-UMN phenotype.

The p-UMN phenotype was observed in 112 of 507 patients (22.1%).

The age at onset ranged from 22 to 77 years (mean 50.2, median 51; figure 1B). Women showed a unimodal curve with a peak in the sixth decade of life, whereas men disclosed a bimodal distribution with a major peak in the fourth decade of life and a second peak in the seventh decade of life. This clinical phenotype was found in 59.6% of young-adult patients and in 17.4% of adult patients (p < 0.0001). Among young-adult patients, the ratio of men to women was 5.8:1, whereas in the adult-onset form, the ratio was 0.77:1 (table).

In the group of young-adult ALS, 10 patients (29.5%) first noted symptoms in the legs, 17 (50%) in the arms, and 7 (20.5%) in the bulbar region. On clinical examination, all patients with first symptoms in the lower limbs were found to have spastic paraparesis, which was frequently asymmetric.

Although the onset was variable, all patients disclosed a strikingly similar clinical pattern in the fully developed stage of the disease. This stage was characterized by marked spastic paraparesis, pseudobulbar speech, bilateral Babinski sign, clonic reflexes, and brisk Hoffmann sign. Muscular atrophy was obvious but confined to the upper limbs, bilaterally, with a slight degree of asymmetry, and it was more pronounced in distal than in proximal muscles. In this phase, muscular bulk was preserved in the legs, although the fasciculations were frequent and mild denervation activity was noted on EMG examination. Fasciculations and atrophy of the tongue were frequent. Respiratory muscle involvement and muscular atrophy in the lower limbs occurred at later stages of the disease. Five patients received nocturnal noninvasive assisted ventilation.

Mutational analysis of the SOD1 gene was performed in 20 of 34 patients, with normal results.

Survival.

Using Kaplan–Meier analysis, the median survival of 507 patients was 48 months (95% CI 42.19–53.80, range 6–226). Dividing patients into two groups according to age at onset, median survival was 71 months (95% CI 55.97–86.02) in young-adult patients compared with 44 months (95% CI 39.42–48.57) in adult patients (p = 0.001; figure 2).

• Download figure
• Open in new tab
• Download powerpoint

Figure 2 Kaplan–Meier plot of survival probabilities according to age at onset

The group of patients younger than 40 years (thick line) shows longer survival than does the group of patients older than 40 years (thin line).

When the 57 young-adult patients were stratified by clinical phenotype (figure 3), median survival was 56 month (range 6–106, 95% CI 48.65–63.34) months in the classic group compared with 74 months (range 10–226, 95% CI 60.61–87.38) in the p-UMN group (p = 0.03).

• Download figure
• Open in new tab
• Download powerpoint

Figure 3 Kaplan–Meier plot of 57 patients with onset before 40 years

The population is divided into two groups on the basis of clinical phenotype: classic phenotype (thin line) and predominant upper motor neuron phenotype (thick line).

Using the covariate Cox proportional hazards regression model in the 57 patients, after adjusting for survival factors such as sex, site of disease onset (bulbar vs spinal), and clinical phenotype (classic vs p-UMN), the p-UMN phenotype remained an independent predictor of survival (p = 0.007), whereas the other variables were not identified as independent prognostic factors (sex p = 0.06, site of onset p = 0.46).

The percentage of patients who survived at least 5 years was 50% in the p-UMN phenotype compared with 26% in the classic form (p = 0.12). The percentage of patients who survived more than 10 years was 17.6% in the p-UMN group, whereas no patient with classic phenotype had such long survival (p = 0.09).

DISCUSSION

One of the unsolved problems of ALS is whether the large spectrum of clinical manifestations merely reflects the phenotypic variability of one disease or indicates the existence of different conditions with heterogeneous mechanisms. In particular, whether distinct subgroups may be defined on the basis of age at onset remains to be elucidated.

Juvenile ALS is widely recognized as a separate entity because the disease usually discloses a slowly progressive course and it is nearly always familial.^14,15 The onset of juvenile ALS occurs before age 25 years, being in the first two decades of life in the majority of patients. The pattern of inheritance is autosomal recessive or, less frequently, autosomal dominant. Genes or loci have been identified only in a minority of these forms, and have been classified as ALS2,^16,17 ALS4,¹⁸ and ALS5.¹⁹

In most studies based on national registers or on large referral ALS populations, early-onset ALS commonly includes young-adult patients in which the disease starts in the third or fourth decade of life. Patients with childhood or juvenile onset are exceptionally registered or, in most instances, are not reported at all.^7,9–12 Young-adult ALS, which accounted for nearly 14% in a large series,⁷ is considered an extension of the common adult type, with which it shares similar clinical features and sporadic occurrence. However, some differences have been reported between young-adult ALS and the common adult form. Survival is significantly longer in young patients, so that early age at onset is widely recognized as the main positive prognostic factor in ALS.^20–22 Patients with onset before age 40 years show a survival of approximately 3.5 to 4 times longer than patients aged > 61 years.²¹ Greater neuronal reserve probably contributes to the better prognosis in younger patients, but different mechanisms can be involved. Furthermore, the site of initial clinical manifestations is different in relationship to the age at onset: the percentage of patients with bulbar onset is 15% in the young-adult type but increases gradually with age up to 43% after age 70 years.⁷

Few data are reported in the literature regarding clinical manifestations and natural history of young-adult ALS.

We analyzed 57 patients affected by young-adult ALS with age at onset between ages 20 and 40 years and compared them with 450 patients with the classic adult-onset type. Clinically, the 57 patients reported here usually showed a phenotype dominated by pyramidal signs. On the contrary, it is well recognized that the most common type of ALS (classic ALS) is characterized by signs of lower motor neuron disfunction, such as weakness, atrophy, cramping, and fasciculations, associated with few pyramidal signs, such as hyperactive deep reflexes, hypertonia, and pseudobulbar signs. Only in a minority of patients does severe spastic paraparesis dominate the clinical picture (p-UMN ALS).⁴ ALS with predominant pyramidal signs has also been referred to in the literature as “spastic form”²³ or “pyramidal type.”⁴ Accordingly with literature data, in our series of 507 patients, 22.1% had the p-UMN form and 77.9% exhibited the classic phenotype. However, 59.6% of the young-adult patients showed the p-UMN phenotype, which was observed only in 17.4% of adult patients (p < 0.0001). Similar data have been observed in a group of 51 young-adult patients in which the spastic form was more frequent (50%) than in adult patients (26%).²⁴ Interestingly, all our patients with the p-UMN form had a stereotypical topographic distribution of UMN and LMN dysfunction signs in the fully developed phase of the disease and showed the same pattern of spread of LMN signs. The hypothesis that motor neuron degeneration in ALS may be an orderly and sequential process^27,28 seems to be confirmed by our data in this peculiar phenotype.

Another remarkable finding in our series of patients with young-adult ALS was the different ratios of men to women in the two phenotypes. The ratio of men to women is 3–3.6:1 among young patients and decreases progressively in older ages, so that the proportion of males and females is equal after age 65 years.^7,21 Our findings confirmed these data, but when we stratified our patients by phenotype, we observed a striking male prevalence in the p-UMN group. Based on these findings, genetic or environmental factors linked to sex may play an important role in the development of young-adult ALS with p-UMN phenotype. It is worth noting that sex seems to be a strong risk factor also in other ALS phenotypes, such as ALS with bulbar onset, which is more prevalent in women,⁷ and flail arm syndrome, with male predominance.⁵

In all young-adult patients, the disease disclosed a progressive course, but survival was significantly longer compared with patients with onset after age 40 years, as shown by the Kaplan–Meier plot (figure 2). This finding is consistent with previous observations indicating the age at onset as the main prognostic factor in ALS.²² When survival was evaluated separately in young-adult patients with p-UMN and classic forms (figure 3), we found a longer duration of the disease in the former group. Younger onset and a predominance of pure upper motor neuron signs at presentation characterized a subset of ALS patients with long survival.²⁵ Using the Cox regression analysis, the p-UMN phenotype was identified as an independent risk factor determining survival of young-adult patients. The small number of patients does not allow us to draw any definite conclusions, and further studies are needed to establish whether p-UMN phenotype is a prognostic factor.

Sporadic ALS is considered a complex genetic disease resulting from interaction between environmental factors and specific susceptibility genes.²⁶ It is well recognized that patients with familial ALS, due to SOD1 mutation or to other unknown genes, have a mean age at onset of approximately 10 years earlier than patients with sporadic forms.² In young-adult sporadic ALS, genetic factors may play a major role, and the observation of a special phenotype in our series seems to strengthen this hypothesis. Our findings might be regional and genetically based, but it should be considered that similar data have been described in patients from Serbia.²⁴

Though clinical studies cannot solve the problem of whether the different phenotypes of ALS are separate entities, our findings support the idea that splitting ALS into clinical categories may be a helpful tool toward a proper approach in epidemiologic and biologic research.

Acknowledgment

Statistical analysis was performed by G. Marangi, MD, from Istituto di Genetica Medica, Università Cattolica del Sacro Cuore, Rome, Italy.