Prevalence of SOD1 mutations in the Italian ALS population

Amyotrophic lateral sclerosis (ALS) is a degenerative disorder of adult life, characterized by a progressive impairment of lower and upper motor neuron function. The cause of this fatal neurodegenerative disease is unknown. Between 5% and 10% of patients have a family history of ALS (familial ALS [FALS]), but most cases of ALS occur sporadically throughout the community (sporadic ALS [SALS]).¹ Mutations in Cu/Zn superoxide dismutase (SOD1; OMIM 147450) are the most frequent genetic defects known to underlie ALS accounting for 20% of familial cases and 2% to 4% of apparently sporadic cases.¹ However, these estimates of SOD1 mutation prevalence have been exclusively based on case series and clinic referral cohorts. There are currently no published studies describing the frequency of SOD1 mutation in a population-based series of ALS cases.

In this article, we report the frequency and nature of SOD1 mutations in a large population-based cohort of Italian patients diagnosed with ALS over a 6-year period.

METHODS

RESULTS

During the 6-year study period from January 1, 2000, to December 31, 2005, 386 residents in the province of Turin (223 men and 163 women) were diagnosed with ALS, corresponding to a mean annual crude incidence rate of 2.91/100,000 population (95% CI, 2.63 to 3.21). Of these, 22 patients had a positive family history of ALS representing 5.7% of all cases. Patients with familial ALS were an average of 7 years younger than patients with sporadic ALS (FALS 57.4 [SD 12.2]; SALS 65.6 [SD 10.5]; p = 0.00001). Men and women are affected equally among familial cases, whereas there was a preponderance of men among the sporadic patients (male:female ratio, 1.3:1).

SOD1 analysis was performed in 325 patients (84.2% of the whole series), including all FALS cases. Forty-one patients did not give their consent for genetic analysis. Twenty patients were found only through secondary sources and were therefore not tested for DNA. Five patients carried a SOD1 coding mutation. Of these, three had a family history of ALS so that SOD1 mutations accounted for 13.6% of all familial cases diagnosed in the province of Turin during the study period (3 SOD1 cases, 22 FALS cases). These mutations consisted of the G41S mutation, which was seen in two cases from the same family, and the G93D mutation, which was seen in a single familial case. The remaining two SOD1 mutations were found in apparently sporadic cases representing 0.7% of all SALS cases (2 SOD1 cases, 303 SALS cases). These mutations were N19S and E133ΔE. The mean age at onset of the five SOD1 cases was 59.0 years (SD 15.3). The clinical characteristics of the patients with SOD1 mutations are reported in appendix e-1 on the Neurology® Web site at www.neurology.org.

Intronic SOD1 variants were found in 34 (11.2%) SALS cases (table 1). Intronic variances were not found in FALS or in patients with SOD1 mutations. The most frequent intronic variance was IVSI: 319 T>A, which was found in 25 cases (8.3%) and is in linkage disequilibrium with IVSIII: +34 A>C in 20 cases.

DISCUSSION

We have estimated the frequency of SOD1 mutations in a population-based series of patients with ALS. Nearly 85% of cases incident in the province of Turin were analyzed for SOD1 mutations. DNA samples were not available from the remaining 15% of cases because they did not give their consent or were not available for the analysis. However, clinical and demographic features of these subjects were not different from patients included in the study, and none of them had a positive family history of ALS.

Our study found that 6% of all cases of ALS diagnosed in Turin over the 5-year study period had a family history of ALS, a figure lower than the 10% rate of familial ALS that is typically quoted for ALS.⁵ However, this rate is based on clinical series from ALS referral centers, where patients are more likely to have a positive family history of ALS,⁶ leading to an overestimation the frequency of FALS. It is interesting to note that the frequency of SOD1 mutations among familial cases in our series (13.6% of all FALS) was similar to figures based on ALS referral cohorts.¹

A significant difference was found between the frequency of SOD1 mutations in sporadic ALS cases in our study (0.7%) and that previously reported from ALS referral centers (ranging from 0% to 6.3%) (table 2). This difference could be due to bias in the referral series due to the selection of younger patients, who have a greater risk of carrying a SOD1 gene mutation. Supporting this hypothesis is the observation that the highest frequencies of SOD1 mutations have been reported in series with the youngest mean age at onset,^7–9 whereas the lowest were found in series with the oldest age.^10,11

It is also possible that the WAVE screening method used in our study did not identify all of the SOD1 mutations within the Italian cohort. Though it is not as sensitive as direct sequencing, WAVE is known to be highly robust in terms of identifying point mutations.

The genesis of the mutations in the SALS cases is not clear. It is possible that these cases are FALS cases that have been diagnosed as SALS (so-called apparently sporadic cases). This may occur for a variety of reasons including decreased penetrance, poor diagnosis of ALS in the elderly in previous generations, or incomplete family history.¹⁴ Another possibility is that apparently sporadic ALS with SOD1 mutations are due to de novo mutations. To date, the Irish H80R is the only proven de novo mutation occurring in a SALS patient.¹² It was not possible to determine whether the SOD1 mutations identified in the Italian patients with SALS were due to de novo mutations because DNA samples were not available from the parents of the SALS cases. This is not surprising given that ALS is predominantly a disease of middle and older age groups.

Intronic variances were found in 34 patients diagnosed with SALS. There are no indications that these variances are pathogenetically related to ALS because none of them were in splicing sites and they have a similar frequency in normal controls as well.^4,13 The IVSI: +319 T>A variation was the most common in our series and was in linkage disequilibrium with the IVSIII: +34 A>C variation in 20 patients. This latter variation is the most frequent variation reported in other large series.^13,15

A possible limitation of our study is an underascertainment of the frequency of FALS cases due to their identification having mainly been based on obtaining an adequate family history from the patient. However, family history was collected using a standardized questionnaire. Despite this, it remains possible that patients are simply not aware of their family history. It is now increasingly recognized that frontotemporal dementia and ALS form a spectrum of disease, but collection of information concerning frontotemporal dementia among family members has not been prioritized until recently. Because SOD1 was sequenced in all familial and in most sporadic cases diagnosed during the time period, the estimation of the frequency of mutations of this gene in our series is highly accurate.

A second caveat is that the frequency of mutations of a gene can be different in different countries or even in different regions of the same country. For example, a cluster of G84D mutation of the SOD1 gene has been described in central Italy,¹⁶ but this mutation is quite rare in other Italian regions.^10,24 Another example is the A4V mutation, which accounts for more than 40% of cases in United States but rarely has been found in Europe.¹ Therefore, the estimation of SOD1 frequency performed in our study may not necessarily be generalizable to the figures from other countries.²⁵ Further population-based studies of SOD1 are warranted. The recently established collaboration between European ALS registries (EURALS) will prove to be an invaluable resource for replicating our study in other populations.¹⁷