Abstract
Introduction
MND incidence and clinical phenotype vary by region and accurate statistics about geographical location and clinical features of patients is necessary for care planning and provision. There is no single source of data about people with MND in England, Wales and Northern Ireland to facilitate this and understand MND aetiology in the UK overall. The MND Register for England, Wales and Northern Ireland was set up in 2015 for the purpose of collecting and standardising data from clinic and population registers to understand MND across the UK.
Methods
Data are collected from services that see people over 16, resident in England, Wales or Northern Ireland with a diagnosis of MND that has been confirmed by a neurologist. A minimum dataset consisting of demographic and clinical phenotype data are collected, with some sites also contributing a more comprehensive dataset that includes interventional and functional outcome measures. Data are standardised and uploaded to a central database for analysis. Since 2021 the MND Register has used an opt-out method of inclusion in order to improve case ascertainment and data quality. Each service that contributes to the project defines an area made up of postcode districts where they are confident that all people with ALS in that area will be included in their caseload and therefore contributed dataset. Crude age- and sex-specific incidence rates will be estimated using population census records for the relevant postcodes from Office of National Statistics census data. These rates will be standardised to the UK population structure using direct standardisation. Lifetime risk will be calculated using the cumulative risk and current probability methods. Case ascertainment will be estimated using capture-recapture methodology.
Results
32 centres sent data in the most recent transfer for the period up to 31 April 2023. The catchment areas of the services included cover approximately 70% of the area of England, Wales and Northern Ireland.
Discussion
The MND Register for England, Wales and Northern Ireland continues in its aim to become the single source of standardised data in this geographical catchment area.
Abstract
Introduction
The incidence and phenotype of amyotrophic lateral sclerosis (ALS) varies by country and genetic aetiology, however there are few studies that cover the full disease course in large, multinational populations. We describe the clinical characteristics of people with ALS in eight European countries by genetic subtype and country of origin.
Methods
This study includes people with a diagnosis of ALS, including classic ALS and progressive muscular atrophy subtypes, from nine registers in eight European countries (The Republic of Ireland, Spain, Netherlands, Italy, Belgium, France, Sweden and the UK). Clinical, genetic results and functional outcome data were collected locally, and variables were standardised between datasets.
The dataset includes demographics, clinical phenotype descriptors and intervention information. Genetic testing information on four genes: C9ORF72, FUS, SOD1 and TARDBP were collected. Functional outcome was measured using the ALS functional rating scale (revised). Age and site of onset of motor symptoms, diagnostic subtype, age of diagnosis and sex were compared between data source, genetic subtypes and disease severity categories such as use of gastrostomy, non-invasive ventilation (NIV) and ambulatory support. Patients with information on full disease course were included in the comparison of gastrostomy, NIV and ambulation. Use of ambulatory support was defined by a score of less than 3 on question 8 (walking function) of the ALSFRS-R, at more than 80% of a person’s disease course. Fast and slow progression was quantified using averaged monthly delta ALSFRS-R change; “fast” progression was defined as those people with a decrease in ALSFRS-R in the top quartile, “slow” those in the bottom quartile and “intermediate” progression defined as the middle two quartiles.
Results
The overall median age of onset is 64 years (IQR 55, 71), the youngest age of onset in genetic subtypes is FUS at 51 years (IQR 39, 63) and the oldest is C9ORF72 at 60 years (IQR 29, 91). The genetic subtype with the slowest progression in this dataset is SOD1, and C9ORF72 had the fastest progression. Overall, 30% of people were recorded as bulbar onset and 67% as limb onset. 77% of people that utilise ambulatory support have limb onset and 55% of people who use gastrostomy have bulbar onset. The proportions of people with bulbar and limb onset who utilise NIV is similar to the overall population – 30% of people with bulbar onset and 65% of people with limb onset. However, there is a higher proportion of respiratory onset cases in this category, 4.4% compared with 2.2% in the overall population.
Discussion
The dataset collected as part of this phase of the Precision ALS project can provide detailed insights into the ALS phenotype and patient journey, the database is an invaluable resource for ALS researchers and clinicians.
Abstract
Background
ALS is a rare disease, and to fully elucidate its complexity and generate data with sufficient statistical power, extensive international cooperation is required. Furthermore, as artificial intelligence (AI) and related advanced analytical technologies are increasingly applied to healthcare with the potential to transform many aspects of patient care, as well as administrative processes, data must be collated, stored and accessed in a format that enables these evolving technologies. To achieve this potential, sustainable harmonised ICT systems are required with the capacity to support appropriate long-term collection, collation, and interrogation of multi-sourced clinical data for multi-modal analytics.
Aim
To generate a bespoke pan-European platform for ALS that enables population-based data collection and collation, and to provide novel technologies that enhance agnostic data driven analysis that provide new insights into disease heterogeneity.
Methods
A consortium comprising clinical, data science and industry expertise partnered as the PRECISION ALS initiative (www.precisionals.ie). A series of use-cases has been generated through interdisciplinary engagement with relevant consortium partners. A core set of clinical and socio-economic data points have been digitised in a PRECISION ALS data collection application. The PRECISION ALS digital application captures harmonised data across participating European sites in a GDPR compliant manner.
Results
The PRECISION ALS data collection application and bespoke data platform respectively provide a digital harmonised process for capturing clinical data at the point of clinical contact and storing and collating data suitable for multimodal analytics. The process is governed by a robust clinical, data science and legal framework that addresses the complex challenges inherent to successful long-term collaborative European multisite health-related research.
Conclusions
PRECISION ALS is a GDPR compliant system for sustainable pan European multimodal data collection and advanced analytics that will provide large scale sustainable real-world datasets. The platform will contain multimodal data of sufficient scale to elucidate the complexity of ALS and support a precision medicine-based approach towards new therapeutics. Clinical, data science and procedural components meet the requirements for sustainability of multimodal data collection, collation, retention, and analysis, while also protecting and respecting data privacy rights of subjects.
Abstract
Background
The ALS etiology involves genetic and non-genetic factors, while the established risk factors include older age, family history, and male sex (Citation1). This study explores the role of BMI, smoking, and head injury and their interaction with C9orf72 carrier status on ALS risk and functional decline.
Methods
Newly diagnosed ALS cases (n = 280) at Karolinska University Hospital between October 2015 and June 2022 were compared to their healthy siblings (n = 79) and partners (n = 128) using a case-control design. Information on BMI, smoking, and history of head trauma was obtained through the Euro-MOTOR questionnaire completed at the date of diagnosis or recruitment. Conditional logistic regression models adjusted for age and sex were used to compare cases and controls. We further examined the interaction between the above factors, and C9ORF72 repeat expansion status. The disease progression of ALS cases was assessed every six months using the ALSFRS-R scale. Linear mixed effects models adjusted for age, sex, site of onset, diagnostic delay, and progression rate were used to calculate the average change in the ALSFRS-R score over time.
Results
The average increase in BMI by 1 Kg/m2 over the 25 years before diagnosis or recruitment was associated with a 5% decrease in the risk for ALS (0.95 95%CI: 0.91, 0.99). This association was significant among males and individuals without the C9 mutation (C9−). Smoking and history of head injury and their interaction with C9 mutation (C9+) status were not significantly associated with the risk for ALS. Higher BMI was associated with a slower functional decline among C9+ cases, whereas the effect among C9− was smaller and not significant (p of interaction = 0.01). Among C9− cases, the history of head injury was associated with a decrease in the average ALSFRS-R score over time by 2.1 (95% CI: −5.2, 0.6). Among C9+ cases, this association was null. A history of multiple head injuries, head injuries occurring at the workplace, and head injuries leaving permanent damage were associated with faster functional decline over time −8.79 (95% CI: −17.87, 0.27), −9.1 (95% CI: −15.9, −2.3), and −10.6 (95% CI: −17.1, −3.9), respectively. The association between smoking status and the average ALSFRS-R score over time was not significant, and no interaction with the C9 carrier status was recorded.
Discussion
Low BMI was identified both as a risk factor for ALS risk and for a faster functional decline. An interaction was also identified between BMI and C9 mutation status. History of head trauma was associated with ALS functional decline over time, especially if the trauma left permanent damage.
Acknowledgements
This study was supported by the European Research Council (ERC) Starting Grant [802091], and the US CDC [R01TS000324-01–00].
Reference
- Longinetti E, Fang F. Curr Opin Neurol. 2019;32(5):771–76.
Abstract
Background
ALS is a fatal neuromuscular disease. 90% of ALS cases are sporadic. Environmental contaminants, such as lead (Pb), have been linked to ALS, although contaminant sources remain inconclusive.
Objectives
This study developed a geospatial analysis procedure to comprehensively detect the impact of airborne Pb, including pure Pb, Pb compounds, and their combination on ALS using data on 695 ALS mortality cases (diagnosed during 2013–2015) and 3,519 controls in Ohio.
Methods
Each subject has age, sex, and migration history information. The airborne Pb data is the US EPA AERMOD data covering 2000–2020. We first estimated yearly Pb exposures for each subject based on location in that year and calculated the cumulative, median, maximum, and minimum exposure of each subject for periods 1, 2, up to 12 years before the index year. We then performed logistic regression to detect any differences between cases and controls considering age, sex, and population density as confounding factors.
Results
Our findings show (1) The impacts of pure Pb and Pb compounds are considerably different. Pure Pb tends to be significantly and positively associated with ALS in areas with lower population density (OR 1.6 ∼ 3.5), and the association has an at least 5-year lag effect. While the positive association of Pb compounds with ALS is generally stronger (OR 1.7 ∼ 4.2), and it is much less related to population density and migration history. (2) The sum of pure Pb and Pb compounds (pure Pb + Pb compounds) results in a more complex association pattern with ALS. It is mostly positive, with the strongest in most rural or urban areas (OR 1.7 ∼ 3.6), and tends to lag at least 8 years in urban areas. (3) Cumulative, maximum, and median exposures to airborne Pb for a given period before the index year generate relatively consistent and positive results. However, the calculated minimum amount of pure Pb appears unrelated to ALS. Supporting prior work linking Pb to ALS, our spatiotemporal approach also has the potential to identify sources of neurodegenerative contaminants for potential disease mitigation.
Discussion
Further study and advanced analytical methods are required to understand the effect of environmental risk factors on ALS pathogenesis.
Abstract
Background
Various risk factors may influence the likelihood of developing neurodegenerative disorders, including factors such as gender and aging. Hormonal differences between genders and age groups could potentially explain these variations, as indicated by evidence from prior epidemiological and clinical research. These studies have highlighted the intricate relationship between sex hormones and neurodegenerative disorders. However, it is important to acknowledge that observational studies may be influenced by unavoidable confounding variables and cannot establish causation. Therefore, the existence of a causal association and the direction of the effect between sex hormones and neurodegenerative disorders remain subjects of controversy.
Objectives
We aimed to investigate the potential causal relationship between major sex hormones and the risk of neurodegenerative disorders.
Methods
We conducted a two-sample Mendelian randomization analysis using summary statistics from genome-wide association studies of sex hormones including bioavailable testosterone, total testosterone, estradiol, and progesterone, together with neurodegenerative disorders including Alzheimer’s disease, Parkinson’s disease, and amyotrophic lateral sclerosis (ALS).
Results
Individuals genetically predisposed to higher bioavailable testosterone exhibited a decreased risk of ALS (OR: 0.80, 95% CI: 0.67–0.94, p = 8.45E-03). In the sex-stratified analysis, such reverse causal association was observed specifically in females (OR: 0.86, 95% CI: 0.77–0.97, p = 0.015), whereas no significant association was found in males (OR: 0.90, 95% CI: 0.80–1.01, p = 0.08). Conversely, we did not identify any association between estradiol, progesterone, and other neurodegenerative disorders. These results were robust across all the sensitivity analyses.
Discussion
These findings enhance our understanding of the involvement of sex hormones in neurodegenerative disorders, and suggest the potential therapeutic implications of targeting testosterone as a treatment option for ALS. The results will also facilitate therapeutic management and drug discovery in future clinical trials.
Abstract
Background
There is great interest in developing genetically targeted therapies for ALS and FTD, especially since a gene-targeted therapy was recently approved in the United States. Treating neurodegeneration early in the disease process with the goal of retaining or prolonging a patient’s full function is an important one to the people impacted by genetic ALS and FTD. Estimating the population of people who hold these genetic mutations without having symptoms (or asymptomatically) is important to prepare for the future where early initiation of therapies will be possible.
Method
Looking at prior publications on: (1) what percent of the disease harbors any of the causative mutations, (2) what the incidence is of the disease in question, and (3) the average age of diagnosis of the causative genetic mutations can provide an estimate of the population of carriers. Accounting for reduced penetrance impacts the size - if penetrance is complete, the population of asymptomatic carriers will be lower. The further penetrance is reduced, the larger the size of the asymptomatic carrier population. Given the remaining uncertainty over the penetrance of most ALS and FTD causative mutations it is prudent to acknowledge any penetrance assumption at this time is tentative.
Results
Using the lowest estimation of disease incidence (1.7/100k ALS and 2.9/100k FTD) and the low range of what amount of each disease is familial (5% ALS and 20% FTD), and assuming a 90% penetrance for each gene provides for a total asymptomatic ALS & FTD genetic carrier population of 147,431 in the United States. Using the highest estimation of disease incidence (2.4/100k ALS and 4.1/100k FTD) and the high range of what amount of each disease is familial (10% ALS and 50% FTD), and maintaining a 90% penetrance for each gene provides for a total asymptomatic genetic ALS & FTD carrier population of 574,387 in the United States. For each major gene, the range using the above low and high estimates is c9orf72 56,211–154,476; GRN 13,375–118,180; MAPT 11,088–90,132; SOD1 6,131–11,786; TARDBP 3,576–5,934; FUS 1,141–2,060. We can also account for the share of genetic carriers in inherited ALS and FTD families that do not yet have an identified gene and those with less commonly known ALS or FTD genetic mutations. For this group in ALS it is 8,069–22,737 and in FTD it is 47,840–16,908.
Discussion
Given the significant population of people who are known to be at great risk for genetic ALS and FTD, a strategy for enhanced neurological care for this population is imperative to plan for the introduction of disease-modifying therapies at the earliest point the disease process can be confirmed to have begun.
Abstract
Background
Facial Onset Sensory and Motor Neuronopathy (FOSMN) is a rare syndrome with only about 100 cases reported worldwide (Citation1). Key symptoms are initial asymmetrical facial paresthesia and/or sensory deficits followed by bulbar symptoms and spreading of sensory and motor deficits from the face to the scalp, neck, upper trunk, and upper extremities (Citation2). Several autopsy cases showed neurodegenerative changes and, in some cases, intraneuronal TAR DNA-binding Protein of 43 kDa (TDP-43) inclusions were detected. Since some genetic mutations related to ALS were also reported in FOSMN cases, a similar pathomechanism to ALS is assumed (Citation1).
Objectives
To investigate the prevalence of FOSMN in Japan and clarify the characteristics of this disease.
Methods
A questionnaire on FOSMN was mailed to 1,214 randomly selected neurology facilities in Japan. The primary survey asked whether each facility had experienced FOSMN cases in 2019. Facilities with affirmed cases were sent a second questionnaire, asking for details on disease characteristics, neurological findings, clinical courses, laboratory findings, and treatments.
Results
In the primary survey, we received answers from 604 facilities (49.8%), leading to an estimated number of 35.8 FOSMN cases in Japan. The secondary survey collected detailed clinical and laboratory data on 21 cases. Decreased or absent corneal and pharyngeal reflexes were present in over 85% of the examined cases, and electrophysiological findings revealed blink reflex test abnormalities in 94.1%. The pattern of spread of motor and sensory symptoms varied between cases, with motor-dominant cases and sensory-dominant cases having distinctive characteristics. Cases with motor-dominant progression mimicked ALS. Immunotherapy was performed in 81%. All received intravenous immunoglobulin (IVIg) and temporary efficacy was observed in 35.3%, more during the initial stages of disease progression.
Discussion
This is the first nationwide epidemiological survey of FOSMN in Japan. The clinical course of FOSMN has a wide spectrum and motor-dominant cases exhibited severe conditions resembling ALS. Decreased or absent corneal reflexes and abnormalities in blink reflex test were recognized in the early stage of disease and could be a diagnostic hallmark of FOSMN, distinguishing it from ALS. Clinical interventions tended to be effective in earlier disease phases, thus swift diagnosis is desirable.
Funding
Acknowledgments
We would like to thank all those who participated in this study for taking the time to respond to the survey thoughtfully.
References
- de Boer EMJ, Barritt AW, Elamin M et al. Neurol Clin Pract. 2021;11:147–57.
- Vucic S, Tian D, Chong PS et al. Brain. 2006;129:3384–90.
Abstract
Background
Treatment of ALS is likely to be most effective early in the disease process. This suggests prevention policies are essential, including acting on actionable risk factors and secondary prevention to slow disease progression with very early therapeutic interventions. Key to the implementation of such prevention measures is the identification of at-risk individuals, preferably long before disease onset. Administrative healthcare data such as electronic health records are increasingly being used in the study of risk factors for different health outcomes, and can potentially identify individuals at risk of developing ALS.
Objectives
To identify predictors or associates of future ALS diagnosis through electronic health data.
Methods
Data from electronic health records in Australia (Australian Institute of Health and Welfare), France (French National Health Data System) and Sweden (Swedish Patient Register) were used to compare historic healthcare usage between ALS patients and matched control individuals. Associations between prescribed medications and ALS in the periods of 2–5, 5–10, and >10 years before the disease index date were tested at the third level of the Anatomical Therapeutic Chemical (ATC) Classification. Differential diagnosis of health conditions as classified by ICD-10 codes were additionally tested in the French and Swedish cohorts. Meta-analysis was performed across countries to identify robust associations.
Results
In the cross-country meta-analysis, prescription of drugs used in diabetes [A10; OR (95% CI) 0.7 (0.56–0.87)], anti-hypertensives [C02; 0.82 (0.69–0.97)] and vasoprotectives [C05; 0.75 (0.64–0.9)] during the 5-10 years before diagnosis were associated with a decrease subsequent ALS diagnosis. Five drugs were associated with an increased odds of diagnosis of ALS in the subsequent 5–10 years: stomatological preparations [A01; 1.17 (1.01–1.37)], drugs for acid related disorders [A02; 1.08 (1.01–1.16)], corticosteriods fo systemic use [H02; 1.1 (1.03–1.17)], psychoanaleptics [N06; 1.12 (1.04–1.2)] and nasal preparations [R01; 1.15 (1.03–1.29)]. Similarly, we found that polyneuropathies (G60) and neuromuscular diseases (G70) diagnosed during 5–10 years before the diagnosis were associated an increased odds of subsequent ALS diagnosis, while diabetes (E10) was associated with a decreased odds of ALS diagnosis. Comparison with similar analyses for Alzheimer's and Parkinson's disease demonstrate both shared and unique risk factors for future diagnosis across neurodegenerative diseases.
Discussion
The observation of differential health care usage substantially before diagnosis of ALS indicates the potential for the identification of at-risk individuals in the decade before disease onset. The use of electronic health records enables the calculations of risk scores for many individuals (if not the entire population) at a relatively low cost and at the point of care.
Funding
Abstract
Background
Amyotrophic Lateral Sclerosis (ALS) is still an incurable disease. Death usually occurs a median of 2–4 years after symptoms onset, usually because of respiratory failure (Citation1). There are controversial evidences on whether patients’ survival has increased over time. Here we analyzed the survival trend of a large cohort of ALS patients over two subsequent 9-year epochs.
Methods
All patients from the Piemonte and Valle d’Aosta ALS Register (PARALS) (Citation2) and who were diagnosed in the 2002–2019 period were considered. Patients were then classified as belonging to the first (2002–2011) or second epoch (2012–2019) based on their diagnosis date. Survival was considered as time from diagnosis to death or tracheostomy. Censoring date was set at 10 years after disease onset. The effect of the diagnosis epoch on survival was assessed using Kaplan-Meier curves with log-rank test and a Cox analysis adjusted by sex, onset age, phenotype at diagnosis, ALSFRS preslope (defined as 48-ALSFRS at diagnosis/diagnostic delay expressed in months), King’s staging system.
Results
A total of 2533 patients were considered for the analysis. 1198 (551 males) and 1335 (597 males) were diagnosed during the first and second epoch, respectively. Being diagnosed during the second epoch resulted to be a positive independent prognostic factor (median survival 1st epoch =2.51 years, 95%CI 2.40–2.61; median survival 2nd epoch =2.67, 95%CI 2.53–2.80) with an HR =0.84, 95%CI 0.76–0.92. Interestingly, the beneficial effect of diagnosis epochs was preserved among patients with milder disease severity at the time of diagnosis (King’s stage 1, 2 and 3) but was not evident among patients more severely affected (King’s stage 4).
Conclusion
Our study showed that ALS survival increased during an 18-year period. Such increase is likely attributable to an improvement in the care of ALS patients. Accordingly, the increase was not observed among patients who were more severely impaired at the time of their diagnosis.
References
- Feldman EL, Goutman SA, Petri S, et al. Amyotrophic lateral sclerosis. Lancet. 2022;400:1363–1380.
- Chiò A, Mora G, Moglia C, et al. Secular trends of amyotrophic lateral sclerosis: The Piemonte and Valle d’Aosta Register. JAMA Neurol. 2017;74(9):1097–1104.
Abstract
Background and objectives
To investigate the underlying reasons for variability in incidence rate of Amyotrophic Lateral Sclerosis (ALS) within the Irish population between the years 1996 and 2021.
Methods
The Irish ALS register was used to calculate incidence and to subsequently extract age at diagnosis (age), year of diagnosis (period), and date of birth (cohort) for all incident patients within the study period (n = 2771). An age-period-cohort (APC) model using partial least squares regression (PLSR) was constructed to examine each component separately and their respective contribution to the incidence while minimising the well-known identifiability problem of APC effects.
A dummy regression model consisting of 5 periods, 19 cohorts and 16 age groups was used to examine non-linear relationships within the data over time. The confidence intervals for each of these were estimated using the jackknife method.
Results
The non-linear model achieved R² of 99.43% with two-component extraction. Age-variation was evident with those from the ages 65–79 contributing significantly to the incidence as expected. However, those aged 25–60 contributed significantly less than expected. Each successive period has shown an increase in the regression model coefficient suggesting an increasing incidence over time, independent of the other factors examined. A Cohort effect was demonstrated showing that those born between 1927 and 1951 contributed significantly to the incidence, more than the other birth cohorts examined.
Discussion
This is the first study to identify age, period, and cohort effects within the Irish ALS population. Ireland suffered from two waves of mass migration in recent history; the age effect demonstrated is likely due to this significant demographic shift. Changes in disease categorisation, changing competing risks of death and improved surveillance likely underpin the period effects, whereas the challenging economic situation during 1927–1951 likely explains the cohort effect. Together, these effects demonstrate both demographic and environmental factors that underpin ALS pathobiology and how they can be epidemiologically modified at a whole-population level.
Abstract
Objective
Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disease affecting motor neurons. ALS occurs in people of all ages, and the incidence rises significantly in people among 65–80 years. The proportion of old-onset ALS in China were low and may have distinct clinical and genetic manifestations. We confirmed that “old-onset” group defined as age at onset of 60 years and over was more appropriate for analyzing old-onset ALS features in the Chinese population. In this study, we aimed to investigate the natural history and remarkable psychiatric state of ALS with age at onset over 60 years in China.
Materials and methods
We collected all ALS cases from 2017 to 2020 in our center and focused on old-onset ALS patients particularly, by analyzing the clinical data, including the ALS onset and disease progression. Anxiety, depression, cognitive function, and sleep quality were assessed to reflect the psychiatric state. In order to analyze the influencing factors of disease progression, we defined slow-progress group as delta FS <0.47 and moderate or rapid-progress group as delta FS >0.47.
Results
The proportion of old-onset ALS increased gradually. A total of 193 old-onset ALS patients were included in this study. The median age at onset of old-onset ALS was 65 years with the quartile from 62 to 68 years. When compared with 446 non-old-onset ALS, old-onset ALS showed distinct clinical presentation, with lower ALS Functional Rating Scale-Revised at diagnosis and faster rate of progression. Remarkably, old-onset ALS were suffering from worse psychiatric state, including serious anxiety and depression, as well as worse cognitive function with sleep quality. The abnormal psychiatric state was more pronounced in female patients of old-onset.
Conclusions
In the current study, ALS patients with old onset showed unique clinical features. Severe psychiatric conditions and faster progression in the early stage of the disease of old-onset ALS indicated the need for more social and psychiatric support in this crowd.
Abstract
Background
Post-mortem studies on primary lateral sclerosis (PLS) are rare and limited to less than 30 cases in the literature. Autopsy studies analysing the presence of transactive DNA binding protein 43 (TDP-43) inclusions are even more sparse. Neuropathological features seen in PLS include degeneration of the upper motor neuron (UMN) with TDP-43 pathology and variable lower motor neuron (LMN) involvement (Citation1,Citation2). To date, there is no international consensus on or gold standard for the neuropathological substrate in PLS.
Method
This study presents clinical and post-mortem finding of patients clinically diagnosed with PLS, and evaluates the interpretations of clinical and post-mortem data by an expert panel using an e-module.
Results
Five PLS cases with post-mortem neuropathological examination were included. Age of onset ranged from 36 to 73 years and disease duration from 8 to 19 years. All patients had UMN symptoms in three regions without LMN symptoms or EMG abnormalities at disease durations of ≥4 years. One case had neuropathological findings consistent with amyotrophic lateral sclerosis (ALS) (degeneration and TDP-43 pathology of UMN and LMN); two cases exhibited similar findings but with less prominent LMN findings, indicating ALS or PLS; one case met the neuropathological criteria for progressive supranuclear palsy (PSP); and the final case displayed extensive old myelin loss in various brain regions and the spinal cord, without signs of a proteinopathy. The expert panel, consisting 11 neurologists, differed in which ancillary investigations were deemed required but missing, in their differential diagnosis, and in their most likely diagnosis prior to knowing the autopsy results. Experts changed their final diagnosis in several cases (42%) based on the autopsy results. After autopsy results were revealed, opinions on the final diagnosis still varied, but were mostly within the MND spectrum (ALS, PLS, UMN-dominant ALS).
Discussion
This study highlights the discrepancies among neurologists in terms of diagnostic tests, differential diagnosis, and even final diagnoses in cases presenting with a UMN-dominant syndrome (fulfilling the criteria for PLS). Three of five cases clinically diagnosed with PLS received a different neuropathological diagnosis (ALS, PSP, and undetermined). This study emphasises the need for neuropathological criteria to diagnose PLS and for consensus on ancillary investigations required to reach high levels of certainly for antemortem PLS diagnosis.
Acknowledgements
We would like to thank Dr. Corcia, Dr. van Damme, Dr. Hoglinger, Dr. Ingre, Dr. Litvan, Dr. Morris, Dr. Pijnenburg, Dr. Raaphorst, Dr. Seelaar, Dr. Ticozzi and Dr. Vandenberghe for participating in this study.
References
- MackenzieIRA. Neuropathology of primary lateral sclerosis. Amyotroph Lateral Scler Frontotemporal Degener. 2020;21(sup1):47–51.
- MackenzieIRA, BriembergH. TDP-43 pathology in primary lateral sclerosis. Amyotroph Lateral Scler Frontotemporal Degener. 2020;21(sup1):52–8.
Abstract
Background
Times to clinically relevant events can be a valuable primary or secondary outcome in observational and interventional studies, complementing linear outcomes such as functional rating scales and progression biomarkers. Time to death or permanent assisted ventilation (PAV) has been a useful secondary outcome in ALS clinical trials but has limited utility because for most participants the outcome, fortunately, is not reached before the end of the study. There are several clinically relevant events in ALS that occur prior to death or PAV for most people living with this disease. However, there are a variety of possible pathways for a patient to progress from being healthy to death. We show how one can use semi-competing risk models to model time-to-event through different pathways of events among those with ALS, using as an example gastrostomy as an intermediate event before death.
Methods
We obtained a data set from the ALS Natural History Consortium, which collects baseline and longitudinal information on ALS patients from 9 participating multidisciplinary ALS clinics. Data from 1600 participants were used. We implemented semi-competing risk models that incorporate as predictors age, sex, ALS phenotype, diagnostic delay, ALSFRS-R at baseline, and ALSFRS-R rates of change. The pathways we considered were (i) healthy to gastrostomy, (ii) healthy to death, and (iii) gastrostomy to death. We used the outputs from the semi-competing risk models to create intuitive visualizations that show how different patient profiles impact time-to-event across each of these pathways.
Results
We calculated the hazard ratios (HRs) to quantify the association between the predictors with the event in the considered pathways. For instance, the baseline fine motor subscore and the rate of change of this subscore was a strong predictor along the pathway from healthy to death (p < 0.05), but both were not predictive of time-to-gastrostomy or time-to-death from gastrostomy (p > 0.05). Using these data and their associated HRs, we calculated risk prediction intervals that provide information on how a set of characteristics is predictive of the time-to-event within each of the considered pathways.
Discussion
We used semi-competing risk models to create intuitive visualizations that show how certain characteristics impact time-to-event through different possible pathways among those with ALS. These models are flexible and can be expanded towards the use of different events or combinations of events to create the different pathways. For our next steps, we plan to include biomarker data to assess their influence on time-to-event across the different pathways among those with ALS. Time-to-event prediction intervals using semi-competing risk models can be used in clinical trial design, pharmacoeconomic studies, and personal planning for people living with ALS.
Abstract
Background
Times to clinically relevant events are a valuable primary or secondary outcome in observational and interventional studies, complementing linear outcomes such as functional rating scales and progression biomarkers. Time to death or permanent assisted ventilation has been a useful secondary outcome in ALS clinical trials but has limited utility because for most participants the outcome, fortunately, is not reached before the end of the study. There are several clinically relevant events that occur prior to death or permanent assisted ventilation for most people living with this disease. This study utilizes the Clinic-based Multicenter ALS Natural History Data Collection, which provides critical information about medical history, ALS subtypes, and ALS progression. We developed prediction models for several of these times to events that can be used for clinical trial modeling and personal planning.
Methods
Landmark time-to-event analysis was implemented to determine the effect of patient characteristics on disease progression. Longitudinal data from 1600 participants in the ALS Natural History Consortium dataset were used. Five outcomes in the ALS disease progression were considered: loss of ambulation, loss of speech, gastrostomy, non-invasive ventilation usage, and forced vital capacity 50% or below the predicted value. In each of these landmark models, time-varying covariate values, such as ALSFRS-R, at landmark time “s” are treated as fixed covariates in a traditional time-to-event Cox regression model from s onward. We implemented 6 different landmark times for each outcome, set every six months from date of diagnosis. Covariates in our model include age at diagnosis, sex, ALS phenotype, diagnostic delay, ALSFRS-R scores at the landmark time, and ALSFRS-R rates of change from baseline.
Results
Our models identified several meaningful relationships between covariates and times to events. Forest plots illustrated how the effect of specific predictors varied across landmark times. In particular, ALSFRS-R subscores and diagnostic delay were consistently among the strongest predictors. For each of our models, we presented risk prediction intervals for a random set of patient characteristics to illustrate how covariates affect risk of experiencing an outcome. We calculated cross-validated c-index values for each of these models, yielding values that range from 0.62 to 0.75, indicating good predictive capability for the models.
Discussion
Landmarking provides a method to compare the changing effects of covariates, biomarkers, and other prognostic factors as the condition changes over time. It is an efficient, continually-updating individualized risk prediction model that is intuitive for both clinicians and patients alike. Importantly, landmarking allows patients to understand how their risk of progression changes in real-time with their unique profile of characteristics. The final product of this analysis will be consolidated into an Rshiny web application for ease of visualization and communication of results that will be accessible to clinicians and patients.
Abstract
Background
Previously we reported on predictors of communication impairment and outcomes of patients with amyotrophic lateral sclerosis (ALS) using tracheostomy and invasive ventilation (TIV). Differences in outcomes of long-term mechanical ventilation (LTMV) users in Japan and USA have not been previously investigated.
Objective
This study aimed to compare the use of TIV between patients with ALS in Japan and those in the USA.
Methods
We enrolled 213 patients with ALS (n = 123, Ohio; n = 90, Tokyo) using TIV at home from 1997 to 2022. Data were collected based on sex, age of onset, age at death, use of noninvasive ventilation (NIV) prior to TIV, duration of months on TIV, emergency TIV, and the causes of deaths. Data analysis was performed using SPSS, version 27. The Pearson's chi-square, Kaplan-Meier and log-rank tests were used for the comparison. P-values <0.05 were considered statistically significant.
Results
The following demographics were observed for Ohio vs Tokyo: male,70(56.9%) vs. 56(62.2%) and median onset age, 56.0(95%CI 52.65–59.34) years vs, 58.0(95%CI 55.22–60.78) years; no significance differences were noted for either between the two countries. Pre-NIV use was reported to be 22(17.9%) and 34(37.8%) (p = 0.002) for Ohio and Tokyo, respectively, with median months from onset to NIV being 17.0 (95%CI 14.6–19.4) and 25.0 (95%CI 18.4–31.6) months (p = 0.003). Median months from onset to TIV were 16.8 (95%CI 14.2–19.4) and 36.0 (95%CI 30.5–41.5) months (p < 0.001), rates of emergency TIV 118(95.9%) and 36(42.4%) (unknown 17(20%)), median TIV duration were 34.0(95%CI 22.8–45.2) months and 81.0 months (95%CI 63.4–98.5) months (p < 0.001), and deaths at the last follow-up were 123 (100%) and 40 (47.1%) for Ohio and Tokyo, respectively. The causes of deaths of patients in Ohio and Tokyo were as follows: 27.5 and 22.5% from cardiac arrest, 25.0 and 22.5% from pneumonia, 36.7 and 0% from planned withdrawal, 5.8 and 10.0% from accidental withdrawal, and 5.0 and 45.0% from another cause, respectively.
Discussion and conclusion
There were significant differences in respiratory therapy in the two countries although onset age and sex were same. Even though a general comparison cannot be made due to disparities in healthcare systems, the decision -making support and delivery of care may have contributed significantly to the difference. The cause of death due to planned withdrawal was high in the United States but non-existent in Japan. Further research on use of LTMV including ethical, emotional, and social aspects, is required.
Funding
References
- Nakayama Y, Shimizu T, Mochizuki Y, et al. Predictors of impaired communication in amyotrophic lateral sclerosis patients with tracheostomy-invasive ventilation. ALSFTLD 2015 amyotrophic lateral sclerosis and frontotemporal degeneration. 2016;17(1–2):38–46.
- Nakayama Y, Shimizu T, et al. Non-motor manifestations in ALS patients with tracheostomy and invasive ventilation. Muscle Nerve. 2018:57:735–41.
Abstract
Background
People with ALS die in hospital, at home with specialised palliative home care or in a palliative unit. They are less likely to have specialized palliative care (SPC) than patients with cancer. For equal care, it is necessary to study how the care is provided last time in life. The aim was to study the possible associations between demographic and clinical factors, including age, sex, and frailty, with acute healthcare utilization in the last month of life, measured by emergency room (ER) visits, admissions to acute hospitals and, acute hospitals as place of death, among patients with ALS. A second aim was to study whether receipt of SPC affects above-mentioned healthcare utilization.
Method
This was an observational, retrospective study from Sweden, based on Region Stockholm´s administrative data warehouse (VAL). Data was retrieved for 2015–2021. All deceased patients (n = 448) ≥18 years with a diagnosis of ALS (motor neuron disease G12.2 according to ICD-10 classification) were included. Emergency room (ER) visits, and hospital deaths were outcome measures. As explanatory variables age, sex, frailty measured by the Hospital Frailty Risk Score (HFRS), receipt of SPC or being a nursing home (NH) resident were used. T-tests were used for age, Wilcoxon Rank Sum test (Mann Whitney U test) were used for comparisons with skew distributions and Chi-square tests for comparison of proportions. Initially, univariable logistic regression analyses were performed for relevant variables, which then were entered into fully adjusted logistic regression models.
Results
Of 448 persons who died in ALS, 99 persons were NH residents. Mean age was 70.5 years, 54% were men, 58% were classified as frail according to the Hospital Frailty Risk Score (HFRS), and 49% received SPC. The receipt of SPC was equal in relation to gender, socio-economic standing, frailty, and age <75 years. Patients >75 years, with dementia and/or living in NH residents were less likely to receive SPC (p = 0.01, p = 0.03 and p = 0.002, respectively). Receipt of SPC reduced ER visits (29 vs. 48%, p < 0.0001) and deaths at hospital (12 vs. 48%, p < 0.0001). Patients who were frail, had a higher risk of ER visits and were more likely to die in acute hospital (p = 0.0002 and p = 0.004). NH residents were less likely to have ER visits and to die at hospital (p = 0.002 and p = 0.005).
Discussion
Our results show that the access to SPC is equal in relation to gender, socio-economic standing, frailty, and age ≤75 years. However, older patients and those living in NH residents were less likely to receive SPC. ER visits and deaths at hospital reduce if the patient has access to SPC. The results indicate partly unequal receipt of palliative care. It is of importance to individualize the care regardless of age or form of residence.