Abstract
Background: “Pure” autosomal dominant hereditary spastic paraparesis (AD-HSP) is clinically and genetically heterogeneous. There are at least seven genetic loci with varying ages at onset and disability. The SPAST gene at the SPG4 locus on chromosome 2p is the major disease gene for AD-HSP.
Objectives: To investigate whether there are distinct clinical features among families with AD-HSP due to SPAST mutations compared with families excluded from SPG4.
Methods: Nineteen families with “pure” AD-HSP were identified, and the clinical features of family members were compared using a standard protocol. With use of genetic studies, the families were divided into two groups for comparison: those with mutations in SPAST, the “mutation-positive” group, and those excluded from SPG4 on the basis of linkage studies, the “SPG4-excluded” group.
Results: Twenty-nine individuals from four families had mutations in SPAST, whereas 22 individuals from three families comprised the SPG4-excluded group; in 11 families, the pattern of linkage was unknown. In the one remaining family, no mutations were found despite strong linkage to SPG4. Different mutations were identified in the four SPAST pedigrees, but the clinical picture was similar in each. Comparison of the mutation-positive group with the SPG4-excluded group revealed an older age at onset (p = 0.03), more disability (p = 0.001), more rapidly progressive paraparesis (p = 0.044), and more cognitive impairment (p = 0.024) among affected individuals with SPAST mutations, not confounded by disease duration. Conclusion: Despite different mutations, SPAST families have a similar phenotype that can be distinguished from other genetic groups.
Hereditary spastic paraparesis (HSP) is characterized by slowly progressive spasticity and weakness of the lower limbs. There is clinical and genetic heterogeneity in HSP, and traditionally it is classified as “pure” or “complicated”1 depending on whether the paraparesis exists alone or in combination with other abnormalities. Autosomal dominant (AD), recessive, and X-linked inheritance occur in both forms.1-3⇓⇓
“Pure” AD-HSP is probably the most common form, and seven loci are identified on chromosomes 14q (SPG3), 2p (SPG4), 15q (SPG6), 8q (SPG8), 12q (SPG10), and most recently 2q and 19q.4-11⇓⇓⇓⇓⇓⇓⇓ The SPG4 locus on chromosome 2p was involved in 45% of cases in one survey,3 and the next most common are SPG3 and SPG8, whereas there are only single reports for the other loci. The SPG4 (also known as SPAST) gene encodes the protein “spastin,” which is a member of the AAA family of adenosine triphosphatases.12
Families with SPG4-linked HSP demonstrate variable expression and reduced or delayed penetrance, yet age at onset distinguishes it from other genetic subgroups.13 Mild, late-onset cognitive impairment occurs more often than expected in a population of “pure” AD-HSP kindreds,14 and there have been two reports of families with dementia and HSP linked to the SPG4 locus on chromosome 2p.15,16⇓ Cognitive impairment is a common feature of the phenotype at this locus, developing after age 40, and mild cognitive decline may be the first manifestation of the disease gene.17
We examine whether there are distinct clinical features of the HSP phenotype in families with mutations in the recently described SPAST gene. In particular, we examine the frequency of cognitive impairment and age at onset distributions in families with SPAST mutations and compare these with those in families where linkage to SPG4 has been excluded.
Methods.
Subjects.
In the course of a population-based study of HSP in Ireland, we identified 34 families with HSP; of these, 19 have “pure” AD-HSP. Two hundred and ten individuals of all ages from the 19 families were seen by two neurologists and underwent full general and neurologic examinations.
Clinical assessment.
Standard diagnostic criteria after Fink et al.3 were adopted for the study, and all clinical assessments were made blindly, that is, without knowledge of the genetic findings.
Age at onset, disease duration, disability, and age at which use of a cane or becoming wheelchair bound occurred were noted for all cases. Onset was defined for symptomatic individuals as the age at which symptoms were first noted by the patient or a close relative. Affected, asymptomatic individuals with signs of paraparesis were not included in age at onset calculations. Disease duration was calculated as age at examination minus age at onset, and disease severity was classified on a 6-point disability scale from 0 to 5: 0 = gene carrier, no signs; 1 = asymptomatic but affected with extensor plantar responses; 2 = symptomatic gait disturbance with minimal disability; 3 = unilateral support required to walk; 4 = bilateral support required to walk; 5 = wheelchair bound. With use of this scale, an estimate of the rate of progression was made for affected individuals. The severity index18 was calculated by dividing the disability score by disease duration and multiplying by 100.
Cognitive assessment.
Cognitive assessment was performed in subjects over age 40 using the Cambridge Cognitive Examination (CAMCOG), the cognitive portion of the Cambridge Mental Disorders of the Elderly.19 The CAMCOG is an extension of the Mini-Mental State Examination.20 It is a more detailed cognitive assessment, detects milder degrees of cognitive impairment, and avoids a ceiling effect, allowing distinction between individuals at the high end of the ability scale.21 The maximum total CAMCOG score is 107; a score of 60 to 80 indicates mild dementia, 35 to 59 moderate dementia, and <35 severe dementia, whereas a score of >80 is considered normal. CAMCOG were performed before clinical examination and without knowledge of genetic findings.
Assignment to linkage groups.
Following collection of all clinical data, the 19 families were divided into three groups based on linkage studies. Five pedigrees showed linkage to the SPG4 locus on chromosome 2p, and mutation analysis of the SPAST gene in these families revealed mutations in four—the “mutation-positive” group (peds 1002, 1006, 1010, and 1012). The fifth SPG4-linked family showed no evident mutations of the SPAST gene and was excluded from the analysis despite strong evidence of linkage (logarithm of the odds [lod] score +3.86 for D2S2374 at zero recombination). Three families comprised the “SPG4-excluded” group. Linkage to the SPG3, SPG4, SPG8, and SPG10 loci was excluded in ped 1008, whereas linkage to the SPG4 locus was excluded in ped 1004 and ped 1020. In the remaining 11 pedigrees, usually because of small family size, no linkage pattern could be established.
Genetic studies.
DNA extraction and PCR amplification.
DNA was extracted from peripheral blood using the QIamp system according to the manufacturer’s protocol (Qiagen, Crawley, UK). PCR were carried out in a 10-μL reaction volume using 100 ng of genomic DNA as template and 20 ng of each primer, as previously described.17 Microsatellite genotypes were determined for markers spanning SPG4 (D2S2255, D2S352, D2S2374, D2S367, D2S400, D2S2203, and D2S2347), SPG3 (D14S1068 and D14S978), SPG8 (D8S1799 and D8S1179), and SPG10 (D12S83 and D12S368). Primers to amplify microsatellite markers were fluorescently labeled and obtained from MWG Biotech (Ebersberg, Germany). In each case, the forward primer was labeled: D2S2203, D2S400, D2S2374, and D14S978 with TET (4,7,2,7–tetrachloro–6–carboxyfluoroscein); D2S352, D2S367 D2S2347, D14S1068, D8S1799, D12S83, and D12S368 with FAM (6–carboxyfluoroscein); and D2S2255 and D8S1179 with HEX (4,7,2,5,7–hexachloro–6–carboxyfluoroscein). The samples were run on an ABI 310 Prism (10 seconds, 15-kV injection; 24 minutes, 15-kV at 60 °C run (Perkin Elmer Biosystems, Foster City, CA). Analysis performed using Genescan software (Perkin Elmer Biosystems) was used to assign genotypes for all individuals. The markers used in this study have all been described elsewhere.22,23⇓
Linkage analysis.
Two-point linkage analysis was performed using the subprogram MLINK of the LINKAGE (version 5.0) program package24,25⇓ as implemented in FASTLINK,26 assuming equal recombination rates in men and women. HSP was analyzed as an AD disease with 90% penetrance and a gene frequency of 0.0001. An “affected-only” analysis was also performed to prevent skewing of linkage results in cases where the gene was nonpenetrant. In this instance, a phenotype was assigned only to affected individuals. Unaffected individuals were assumed to have an unknown status. For each marker, allele frequencies used were based on the CEPH (Centre d’Etudes du Polymorphisme Humain) database allele frequencies.
Mutation screening.
Mutation screening in peds 1001, 1004, 1002, 1006, 1010, and 1012 was performed by PCR amplification and sequencing of all 17 coding exons of SPG4, using genomic DNA as a template, as previously described.27 Exons were amplified using four sets of conditions and sequenced on an ABI377 (Perkin Elmer Biosystems). Total RNA was extracted from fresh blood drawn from affected patients in ped 1001 with the RNA isolation kit from Stratagene (La Jolla, CA) and used as a template for cDNA synthesis according to the Reverse Transcription System (Promega, Madison, WI) using random hexanucleotides to prime synthesis. The SPG4 coding sequences were amplified in four overlapping fragments using primers described.12
Statistical analysis.
Nonparametric tests of significance, the Mann–Whitney U and Wilcoxon signed rank tests, were used to compare variables in each of the genetic groupings. Comparison of proportions was made using the χ2 test, and all analyses were performed using SPSS version 8.0 (SPSS, Chicago, IL).
Results.
Subjects.
Assignment to linkage groups.
Seventy-five people from 18 families were found to be carriers of an abnormal AD-HSP gene. Sixty-seven had signs of paraparesis: 22 from the mutation-positive group, 21 from the SPG4-excluded group, and 24 from the remaining 11 families with unknown linkage. Seven individuals who carried mutations in SPAST had no abnormalities on examination, representing delayed or reduced penetrance of the gene. Of these, one individual, III-5 from ped 1002, has passed the maximal age at onset for her family, representing the only known case of nonpenetrance to date. Among the SPG4-excluded families, one individual was classed as an obligate carrier due to the presence of paraparesis in a sibling and offspring. The sex distribution for the whole population was 53% (40/75) women and 47% (35/75) men with similar proportions among the subgroups.
Genetic studies.
Linkage analysis.
DNA was extracted from all participating individuals. Evidence for linkage of HSP in peds 1002, 1006, 1010, and 1012 to SPG4 has previously been established with a maximum combined lod score of +5.99 for D2S2374 at zero recombination.17 We investigated peds 1004, 1008, and 1020 for evidence of linkage to SPG4. Polymorphic markers spanning ≈9cM surrounding the candidate SPG4 region were analyzed. The linkage analysis was performed as described. Linkage of HSP in ped 1008 to SPG4 was excluded using D2S367, D2S352, and D2S2374 (for further information, please visit the Neurology Web site at www.neurology.org). The maximum distance excluded by a single marker was for D2S367 (lod score = −2.55, θ = 0.15). The total distance excluded using these markers spanned the entire SPG4 region on chromosome 2p. Linkage to the SPG3 (chromosome 14), SPG8 (chromosome 8), and SPG10 (chromosome 12) loci has also been excluded by analysis of polymorphic markers spanning the candidate regions. Linkage of HSP in ped 1020 to SPG4 was excluded based on linkage and haplotype analysis. Due to the small size of this family and the limited usefulness of some markers, the entire SPG4 locus was excluded in small overlapping segments using D2S2203, D2S2347, and D2S2255. Haplotype analysis of genotype data in this family showed there was random segregation of alleles in the SPG4 candidate region among affected family members. Linkage of HSP in ped 1004 to SPG4 was also excluded by haplotype analysis (for further information, please visit the Neurology Web site at www.neurology.org). Mutation screening of the SPAST gene in ped 1004 did not identify a mutation segregating with the disease, thus supporting the argument that HSP in ped 1004 is not linked to the SPG4 locus on chromosome 2p.
Mutation screening.
Mutation analysis of all 17 coding exons revealed four different mutations segregating in peds 1002, 1006, 1010, and 1012.27 Despite the strong evidence for linkage of HSP in ped 1001 to SPG4 (lod score +3.86 for D2S2374 at zero recombination), no mutations were found. To date, only two SPG4-linked families including ped 1001 have been identified with no detectable mutation in the SPAST gene. Reverse transcription PCR of RNA extracted from affected patients in ped 1001 confirmed the absence of mutations in the coding sequence. It is possible that a promoter mutation causing reduced levels of spastin mRNA is responsible. Three of the four mutations identified appear to act by means of haploinsufficiency. Splice site mutations were identified in ped 1006 (intron 11) and ped 1012 (intron 13), disrupting normal splicing and resulting in aberrant transcripts, whereas a nonsense mutation identified in ped 1010 (exon 5) will cause premature termination of protein synthesis. The missense mutation identified in ped 1002 (N386K) occurs in the putative adenosine triphosphate-binding domain of spastin and might therefore cause a loss of activity.
Comparison of mutation-positive and SPG4-excluded families.
In the mutation-positive group, 7 of 29 family members (24.1%) had no signs of paraparesis and were thus nonpenetrant. Four of the 22 family members who did have paraparesis remained asymptomatic. In the SPG4-excluded group, 9 of 21 cases had signs of paraparesis but were asymptomatic at the time of examination. Thus, there was a trend toward fewer asymptomatic subjects in the mutation-positive (18.2%) than the SPG4-excluded group (42.9 %; p = 0.078).
Disability and progression.
The duration of illness was similar in both groups: for the mutation-positive group (n = 18), an average of 25.1 ± 11.1 years, and for the SPG4-excluded group (n = 12), 24 ± 13.7 years (p = 0.82). Disability score, however, was greater for the mutation-positive group (n = 22; median 4, mean 3.6 ± 1.6) than for the SPG4-excluded families (n = 21; median 2, mean 1.9 ± 1.2; p = 0.001). Disability scores in each of the four mutation-positive pedigrees were similar (p = 0.58). As a measure of the rate of progression of paraparesis, the severity index was calculated for all symptomatic individuals in both groups who had evidence of paraparesis for >1 year. Progression by this measure was faster among the mutation-positive group (median 16.75, mean 19.6 ± 9.5; n = 18) than the SPG4-excluded group (median 8.7, mean 11.9 ± 7.3; n = 10, p = 0.044) (see table 1). To investigate whether the age at which paraparesis became apparent has any influence on the subsequent course of the illness, we examined the relationship between age at onset and severity index in those with SPAST mutations. We found a modest positive correlation between age at onset and progression (r2 = 0.36, p = 0.0085), suggesting more rapid evolution in patients with later onset. This pattern was not seen in the SPG4-excluded patients. Severity index was similar among individual SPAST pedigrees (p = 0.76).
Clinical features in mutation-positive compared to SPG4-excluded individuals
Age at onset and anticipation.
Age at onset of paraparesis could be determined in 18 individuals with SPAST mutations and 12 from the SPG4-excluded group. Members of the mutation-positive group were older at onset with a mean age of 29.4 ± 16.9 years as compared with 15.1 ± 20.4 years for the SPG4-excluded group, a difference of 14.3 years (p = 0.03). Correspondingly, individuals in the mutation-positive group were older (53.4 ± 17.6 years) when examined than in the SPG4-excluded group (38.7 ± 21.2 years), by 14.7 years (p = 0.042). A frequency histogram showing the age at onset distributions for both groups is illustrated in figure 1. For the mutation-positive group, the distribution is approximately normal, with most cases becoming apparent in the fourth decade. The SPG4-excluded group is skewed to the left, showing a significantly younger onset, with most cases beginning in the first decade. Within individual SPAST pedigrees, ped 1002 tended to be younger at onset (12 ± 12 years; Kruskal–Wallis, p = 0.051) than the other three (ped 1010, 40.8 ± 11.7 years; ped 1012, 27 ± 9.7 years; ped 1006, 29.3 ± 24.2 years).
Figure 1. A histogram shows the age at onset frequency distribution for the mutation-positive ▪ and SPG4-excluded 
groups. The mutation-positive group shows a near normal distribution with maximal onset in the fourth decade, whereas the SPG4-excluded group is strongly skewed to the left.
To look for evidence of anticipation in our pedigrees, we examined all parent–child pairs for changes in age at onset across the generations. Of six parent–offspring pairs in those families with SPAST mutations, age at onset showed a tendency to drop from parent (34.7 ± 7.5 years) to child (22.3 ± 13.8 years; p = 0.078). However, the proportion of pairings with anticipation (5/6, 83.3%) rather than antianticipation (1/6, 16.7%) in this group did not differ from the null hypothesis of half and half (p = 0.22). In contrast, among six parent–offspring pairings in those with SPG4 excluded, there was a younger onset in children of affected individuals (parents 29.5 ± 21.9 years, offspring 2.5 ± 1.2 years; p = 0.025), and the proportion showing anticipation (6/6, 100%) was greater (p = 0.025).
Among affected individuals from both groups, the frequency of clinical findings was similar. Exceptions to this were wheelchair use (mutation positive 9/22, 40.9%; SPG4 excluded 1/21, 4.8%; p = 0.009) and abnormal vibration sensation in the lower limbs (mutation positive 15/22, 68.2%; SPG4 excluded 4/21, 19%; p = 0.002), which were both more common among the mutation-positive group (see table 2). In the mutation-positive group, several findings correlated with disease duration, including bladder dysfunction (p = 0.008), upper limb hyperreflexia (p = 0.011), bilateral lower limb clonus (p = 0.008), and abnormal vibration sensation (p = 0.001), as well as stick (p = 0.012) and wheelchair (p = 0.001) use. In the SPG4-excluded group, only bladder dysfunction (p = 0.04) correlated with advancing duration.
Frequency of clinical features in affected individuals according to linkage group
Cognitive assessment.
To examine whether cognitive decline in AD-HSP families is peculiar to the SPAST mutation-positive families; we compared CAMCOG scores for individuals over age 40 in both our linkage groups. CAMCOG scores from 18 subjects in the mutation-positive group (including one case of nonpenetrance, aged 66 years) and 9 in the SPG4-excluded group were compared. One mutation-positive individual also had chronic schizophrenia and dementia (CAMCOG = 37) and was therefore excluded from the analysis. Ages (mutation positive 59.1 ± 12.8 years, linkage excluded 55.7 ± 9.8 years; p = 0.35) and years of education (mutation positive 14.5 ± 1.75, linkage excluded 15.4 ± 1.9; p = 0.21) were similar for both groups, as were the CAMCOG scores among the individual SPAST families (p = 0.24). However, the median CAMCOG score for the mutation-positive group was lower (p = 0.011) at 86 (mean 83 ± 13.4, 95% CI 77.4 to 89.6) than for the linkage-excluded group with median 98 (mean 94.9 ± 6.6, 95% CI 89.8 to 100).
Dementia was more frequent among the mutation-positive group members. Seven of 18 individuals (38.9%) from the mutation-positive families had CAMCOG scores of ≤80, indicating cognitive impairment, whereas none of the linkage-excluded group had CAMCOG scores of ≤80 (p = 0.03). This difference remained when the analysis was confined to those with signs of a paraparesis (mutation positive 7/17, 41.2%, linkage excluded 0/9; p = 0.024). This mutation-positive dementia group comprised four men and three women, with average age of 70.6 ± 6.5 years and all seven confined to a wheelchair. The relationship between the CAMCOG scores for both linkage groups is illustrated as a scattergram in figure 2.
Figure 2. A scattergram shows the distribution of Cambridge Cognitive Examination (CAMCOG) scores for individuals over age 40 in both genetic groups. Error bars showing mean and 95% CI are presented. Lower scores are seen for the mutation-positive group.
Discussion.
In the course of our survey, we have identified 4 families with AD-HSP due to mutations in the SPAST gene on chromosome 2p, 3 families for whom involvement of this gene was excluded, and a further 11 families whose genetic status was unknown. One additional large pedigree with late-onset cognitive impairment15 had no detectable mutations in SPAST and was therefore excluded from the analysis; this was despite a maximal lod score of 3.86 for D2S237428 and may be explained by a mutation involving the promoter sequence or a deletion spanning several exons of the SPAST gene. Four different mutations were identified in each of the four families with involvement of SPAST: one missense (ped 1002), one nonsense (ped 1010), and two splicing mutations of introns 11 (ped 1006) and 13 (ped 1012). Despite this, there were no differences of phenotype between SPAST families in terms of the frequency of clinical signs including disease severity, rate of progression, and cognitive impairment. However, one family (ped 1002) with a missense mutation did tend toward a younger age at onset, in keeping with the findings of Fonknechten et al.27 Unlike the X-linked HSP genes, SPG1 and SPG2, where different mutations in the same gene cause clinically distinct conditions, allelic heterogeneity, therefore, does not appear to be a feature of the SPAST gene on chromosome 2p.
SPAST mutation-positive patients had a distinct evolution of their illness compared with SPG4-excluded families. Age at onset was later, and there was more rapid progression and correspondingly greater disability. Wheelchair use was more common, as was abnormal lower limb vibration sensation, and fewer family members were asymptomatic. In addition, cognitive impairment occurred more often, and CAMCOG scores were lower than in the SPG4-excluded families. Age-related dementia in gene carriers with paraparesis15-17⇓⇓ and cognitive decline in gene carriers without signs of paraparesis17 have been reported for the SPG4 locus. Cognitive impairment has not been found in reports of AD-HSP at other loci on chromosomes 14q,4,29,30⇓⇓ 15q,7 or 12q.9 In a family with severe uncomplicated AD-HSP linked to the SPG8 locus on chromosome 8q, one individual developed paraparesis at age 45 and senile dementia of the Alzheimer type at 79. All other family members had normal orientation, language, and memory.31 In another report of HSP at the SPG8 locus, 15 people had paraparesis but no dementia.32 Lower average CAMCOG scores and a greater frequency of dementia in those with SPAST mutations suggest that cognitive decline in AD-HSP is peculiar to the SPAST gene, perhaps indicating a unique pathologic process in subjects with this mutation.
Genetic penetrance in AD-HSP has been considered almost complete, if age dependent,3 though linkage studies suggest this may not be the case at the SPG4 locus.17,28⇓ Mutation analysis in our families revealed a high proportion of reduced expression and delayed penetrance. Seven of 29 (24.1%) family members carried mutations in SPAST without showing any abnormal symptoms or signs, despite having an average age of 34.6 ± 15.8 years (range 14 to 66 years)—5 years older than the average age at onset for those with paraparesis. Including the 4 individuals who remained asymptomatic, a total of 11 of 29 (37.9%) SPAST mutation carriers were therefore unaware of any abnormality. Only one individual, however, was older than the maximal age at onset for her family, representing true nonpenetrance.
Age at onset of paraparesis has been used to classify HSP in the past as type 1 or type 2.1 The postulate was that at least two genes were involved, with type 1 having onset before age 35 and a relatively static course and type 2 with onset after age 35 and more severe and rapidly progressive disability. However, rather than representing two distinct genes, this phenomenon of later onset correlating with more severe disease is seen within our families with SPAST mutations. In one report of SPG8-linked HSP, 6 of 1532 individuals required a wheelchair and the average age at onset was 29.6 years. In another SPG8-linked report with a mean onset age of 37.3 years, 8 of 1531 patients were confined to wheelchair. In comparison, of three families with AD-HSP linked to the SPG3 locus on chromosome 14q,4 only 1 of 32 patients needed a wheelchair and the ages at onset for these families were younger at 20, 7.5, and 6 years. Within our own study, the SPG4-excluded families had a younger onset and were less disabled overall, suggesting the phenomenon may indeed also be apparent between genetic loci.
Before cloning of the SPAST gene, anticipation was suggested to occur at the SPG4 locus5,16,33-36⇓⇓⇓⇓⇓ more often than not.13,18⇓ A CAG repeat expansion was posited as present in some pedigrees linked to this locus,37 though there was no obvious correlation between size of product and age at onset, and further analysis pointed to nonpathogenic expansions originating on chromosomes 18q and 17q.38 Although analysis of anticipation in neurodegenerative disorders is fraught with difficulty due to observation biases, gene cloning has formally ruled out expanded CAG repeats as the underlying mechanism in SPG4-linked AD-HSP. More recently, however, it has become apparent that additional mechanisms beyond unstable repeats may contribute to anticipation in other neurologic conditions.39 However, in keeping with current understanding, anticipation was not found in our pedigrees with SPAST mutations.
In the future, screening for SPAST mutations will allow us to refine further the phenotype of HSP at this locus and distinguish between subtypes of the condition on more than an arbitrary clinical basis. The relatively late age at onset in these families suggests that mutation screening will be a useful tool in genetic counseling, but the large proportion of reduced/delayed penetrance indicates caution in prognostic assessment. It will be of interest also in apparently sporadic cases of paraparesis to identify what proportion of these cases arise from spontaneous mutations in SPAST. Although this locus was previously thought to cause only “pure” HSP, our study suggests that cognitive impairment is a feature of SPAST mutations and that cognitive assessment should form part of the management of this disorder.
Acknowledgments
Acknowledgment
The authors thank the patients and their families for their participation, all adult and pediatric neurologists in Ireland, North and South, who have referred patients for the study, and the Health Research Board of Ireland for funding this research.
Footnotes
-
Additional material related to this article can be found on the Neurology Web site. Go to www.neurology.org and scroll down the Table of Contents for the December 26 issue to find the title link for this article.
- Received March 24, 2000.
- Accepted September 12, 2000.
References
Letters: Rapid online correspondence
REQUIREMENTS
You must ensure that your Disclosures have been updated within the previous six months. Please go to our Submission Site to add or update your Disclosure information.
Your co-authors must send a completed Publishing Agreement Form to Neurology Staff (not necessary for the lead/corresponding author as the form below will suffice) before you upload your comment.
If you are responding to a comment that was written about an article you originally authored:
You (and co-authors) do not need to fill out forms or check disclosures as author forms are still valid
and apply to letter.
Submission specifications:
- Submissions must be < 200 words with < 5 references. Reference 1 must be the article on which you are commenting.
- Submissions should not have more than 5 authors. (Exception: original author replies can include all original authors of the article)
- Submit only on articles published within 6 months of issue date.
- Do not be redundant. Read any comments already posted on the article prior to submission.
- Submitted comments are subject to editing and editor review prior to posting.
