Kaplan-Meier plot showing the survival rates of patients with frontotemporal lobar degeneration (FTLD) without amyotrophic lateral sclerosis (ALS) (solid line; n = 11), those in whom the onset of FTLD symptoms/signs preceded those of ALS (FTLD-ALS) (dashed line; n = 9), and those in whom the onset of ALS symptoms/signs preceded those of FTLD (ALS-FTLD) (dotted line; n = 23). Survival times were significantly shorter in patients with FTLD without ALS than in those with FTLD-ALS or ALS-FTLD (P < .001).
Findings shown include the severity of neuronal loss, gliosis, phosphorylated TAR DNA-binding protein of 43 kDa (pTDP-43) pathological changes, and aggregations of macrophages and the presence of Bunina bodies in the lower motor neuron systems. The severity of each pathological change was graded as 0 (none [−, not colored]), 1 (mild [+, green]), 2 (moderate [++, yellow]), or 3 (severe [+++, red]). Neuropathological changes became increasingly severe in those in whom amyotrophic lateral sclerosis (ALS) symptoms/signs preceded those of frontotemporal lobar degeneration (FTLD; ALS-FTLD), as well as the FTLD-ALS (FTLD symptoms/signs preceding those of ALS) and FTLD without ALS groups (Spearman rank order). Cx indicates cervical cord; Lx, lumbar cord; NA, not assessed; Sx, sacral cord; TDP-43, TAR DNA-binding protein of 43 kDa; and Tx, thoracic cord.
Findings in patients with type A (A-D) , B (E-H), and C (I-L) pathological changes. In a patient with type A pathological change, cerebral coronal sections showed cortical atrophy of the parasylvian region (A). Transverse section of the cervical cord showed marked myelin pallor in the corticospinal tract (B). Microscopically, the frontal cortices showed marked neuronal loss (C) and phosphorylated TAR DNA-binding protein of 43 kDa (pTDP-43)–positive neuronal inclusions and short dystrophic neurites (D). In a patient with type B pathological change, the cerebral cortex showed severe temporal atrophy (E), neuronal loss (G), and pTDP-43–positive neuronal inclusions (H). The corticospinal tract showed mild myelin pallor (F). In a patient with type C patholgoical change, the frontal and temporal cortices showed severe atrophy (I), marked neuronal loss (K), and pTDP-43–positive long dystrophic neurites (L). The corticospinal tract showed marked myelin pallor (J). Klüver-Barrera staining (A, B, E, F, I, and J), hematoxylin-eosin staining (C, G, and K), and pTDP-43 immunohistochemistry (D, H, and L) were performed. Scale bars represent 1 cm (A, E, and I), 3 mm (B, F, and J), 100 μm (C, G, and K), and 20 μm (D, H, and L). Original magnifications are ×1 (A, B, E, F, I, and J), ×200 (C, G, and K), and ×400 (D, H, and L).
Patients with type A (A-H), type B (I-L), and type C (M-P) pathological changes. A patient with type A pathological change showed mild neuronal loss (A), phosphorylated TDP-43 (pTDP-43)–positive skeinlike cytoplasmic inclusions (B), nuclear inclusions (C), and glial inclusions (D), Bunina bodies (E [arrow] and F) in the spinal anterior horn, and dystrophic neurites (G). In a patient with type B pathological change, neuronal loss (I), pTDP-43–positive skeinlike cytoplasmic inclusions (J), and Bunina bodies (K) were markedly observed. In a patient with type C pathological change, the spinal anterior horn showed mild neuronal loss (M), pTDP-43–positive skeinlike cytoplasmic inclusions (N), and dystrophic neurites (O). Double immunohistochemistry for choline acetyltransferase (ChAT) and pTDP-43 revealed cytoplasmic inclusions (violet [arrows]) present within the cytoplasm of a ChAT-positive spinal motor neuron (brown [asterisks]) of patients with type A (H), B (L), or C (P) pathological change. Hematoxylin-eosin staining (A, E, I, K, and M), pTDP-43 immunohistochemistry (B, C, D, G, J, N, and O), cystatin-C (F), and double immunohistochemical analysis for pTDP-43 and ChAT (H, L, and P) were performed. Scale bars represent 100 (A, I, and M), 20 (G, L, and O), and 10 (B-F, H, J, K, N, and P) μm. Original magnifications are ×100 (A, I, and M), ×400 (G, L, and O), and ×1000 (B-F, H, J, K, N, and P).
eFigure 1. Grading of Neuropathological Changes in the Spinal Anterior Horns.
eFigure 2. A TAR DNA-binding protein of 43 kDa (TDP-43)–positive inclusion in a control material.
eFigure 3. Ubiquilin 2–positive structures in patients enrolled in the present study.
Riku Y, Watanabe H, Yoshida M, Tatsumi S, Mimuro M, Iwasaki Y, Katsuno M, Iguchi Y, Masuda M, Senda J, Ishigaki S, Udagawa T, Sobue G. Lower Motor Neuron Involvement in TAR DNA-Binding Protein of 43 kDa–Related Frontotemporal Lobar Degeneration and Amyotrophic Lateral Sclerosis. JAMA Neurol. 2014;71(2):172-179. doi:10.1001/jamaneurol.2013.5489
TAR DNA-binding protein of 43 kDa (TDP-43) plays a major role in the pathogenesis of frontotemporal lobar degeneration (FTLD) and amyotrophic lateral sclerosis (ALS). Although a pathological continuity between FTLD and ALS has been suggested, the neuropathological changes of the lower motor neuron (LMN) systems have not been assessed in TDP-43–associated FTLD (FTLD-TDP), to our knowledge.
To investigate a pathological continuity between FTLD-TDP and ALS by comparing their respective neuropathological changes in the motor neuron system.
Design and Setting
A retrospective clinical medical record review and a semiquantitative neuropathological evaluation of the cranial motor nerve nuclei and spinal cord were conducted at autopsy. We included 43 patients with sporadic FTLD-TDP, type A, B, or C, from 269 consecutively autopsied patients with TDP-43 proteinopathy. Patients were categorized as having FTLD without ALS, FTLD-ALS (onset of FTLD symptoms/signs preceded those of ALS), or ALS-FTLD (onset of ALS symptoms/signs preceded those of FTLD).
Main Outcomes and Measures
Neuronal TDP-43 pathological changes and neuronal loss.
Forty-three patients were included in the clinical analysis, and 29 from whom spinal cords were obtained were included in the neuropathological analysis. Survival time was significantly shorter in the FTLD-ALS and ALS-FTLD groups than in the FTLD without ALS group (P < .001). At neuropathological examination, 89% of patients in the FTLD without ALS group showed aggregations of TDP-43 in the spinal motor neurons. The LMN loss was most severe in ALS-FTLD, followed by FTLD-ALS and FTLD without ALS. All the patients with type A or C FTLD-TDP were included in the FTLD without ALS group, and all those with type B pathological changes were in the FTLD-ALS or the ALS-FTLD group. Lower motor neuron loss and TDP-43–positive skeinlike inclusions were observed in all pathological subtypes.
Conclusions and Relevance
The LMN systems of FTLD-TDP frequently exhibit neuropathological changes corresponding to ALS. Thus, a pathological continuity between FTLD-TDP and ALS is supported at the level of the LMN system.
Frontotemporal lobar degeneration (FTLD) is a sporadic or familial neurodegenerative disease that clinically encompasses frontotemporal dementia, language disorder, and motor symptoms.1 Immunohistochemical profiles show that approximately half of patients with FTLD present with tau-positive disease, but the other half primarily exhibit an accumulation of TAR DNA-binding protein of 43 kDa (TDP-43), referred to as FTLD-TDP.2- 5 Currently, the cortical TDP-43 pathological changes in sporadic FTLD-TDP are classified into 3 subtypes: A, B, and C.6- 8
TAR DNA-binding protein of 43 kDa is also a major disease protein in amyotrophic lateral sclerosis (ALS), which is characterized by upper motor neuron and lower motor neuron (LMN) involvement.9 The pathological features of LMN involvement in ALS include neuronal loss, gliosis, TDP-43–positive neuronal inclusions with skeinlike or round shapes and glial inclusions, and Bunina bodies.10
Some patients exhibit symptoms of both ALS and FTLD, and the cerebral cortices of patients with FTLD and ALS almost always show type B TDP-43 changes.7,11- 13 Thus, a pathological continuity between FTLD and ALS has been proposed based on brain TDP-43 pathological findings. Studies of the cerebral cortex, including the motor cortex, and subcortical gray matter have shown common TDP-43 pathological findings in FTLD, FTLD with ALS, and ALS.12,14- 17 However, the neuropathological features of LMN systems in FTLD-TDP have not been investigated comprehensively, particularly in the spinal cord, although characterization of these features is necessary to confirm the pathological relationship between FTLD and ALS.
In this study, we investigated LMN pathological findings in patients with sporadic FTLD-TDP who clinically demonstrated FTLD, FTLD with ALS, or ALS. We also investigated the correlation between TDP-43 pathological subtypes (type A, B, and C) and LMN involvement to further elucidate the continuity of FTLD and ALS.
We enrolled 269 consecutively autopsied patients with sporadic and adult-onset FTLD, FTLD with ALS, or ALS in which pathological aggregation of TDP-43 was confirmed at the Department of Neuropathology, Institute for Medical Science of Aging, Aichi Medical University, from 1988 to 2012. All patients had been clinically evaluated by neurological experts in the affiliated hospitals of Nagoya University School of Medicine or Aichi Medical University. Permission to perform an autopsy and archive the nervous system tissues for research purposes was obtained from family members after death. The clinical data on the included patients were obtained from case notes made at diagnosis and at an advanced stage of illness. We initially excluded 216 of the 269 patients because they did not present clinical FTLD symptoms. In 53 patients, FTLD or FTLD with ALS was diagnosed according to the diagnostic criteria of FTLD and ALS.1,18 Moreover, we subclassified FTLD with ALS into FTLD-ALS (onset of FTLD symptoms/signs preceded those of ALS) and ALS-FTLD (onset of ALS symptoms/signs preceded those of FTLD) groups.
The enrolled patients were categorized into 3 groups: FTLD without ALS, FTLD-ALS, and ALS-FTLD. The FTLD symptoms were categorized into 2 groups: behavior-variant frontotemporal dementia and language impairments.14 The LMN symptoms/signs were defined by progressive muscular weakness, muscular atrophy, fasciculation, or electromyographic findings. We excluded patients with Alzheimer disease–associated neurofibrillary pathological abnormalities that were more advanced than Braak stage IV,19 those with argyrophilic grain disease, and those with invalid clinical data. Finally, 43 patients were included in the clinical analysis (11 with FTLD without ALS, 9 with FTLD-ALS, and 23 with ALS-FTLD). For comparison, we prepared 13 age-matched controls (mean [SD] age at death, 68.2 [6.9] years) who had no diagnosis of any neurodegenerative disease, dementia, or cerebrovascular disease.
The information regarding sex, age at onset, disease duration, and duration between onset of FTLD and ALS was collected from clinical notes. Causes of death were classified as respiratory failure due to respiratory muscle weakness, pneumonia, or other. The last category comprised systemic diseases other than respiratory failure or pneumonia, including cancer, ileus, infections, and renal failure. Information on the subtypes of dementia, clinical data on motor symptoms, and electromyographic results were also collected.
For the neuropathological analysis, we excluded patients who had received a respirator/tracheotomy (n = 11) or whose spinal cord was not available (n = 3). In total, 29 patients were included in the neuropathological analysis and divided into 3 groups: FTLD without ALS (n = 9), FTLD-ALS (n = 8), and ALS-FTLD (n = 12). The tissues were fixed in 20% neutral-buffered formalin. The paraffin-embedded tissue blocks were cut at a thickness of 4.5 μm. We evaluated sections from the spinal cord and whole brain. The whole spinal cord was examined at each segmental level, but only the cervical cord was available in 3 patients and the sacral cord was not available in another 3.
For routine neuropathological examinations, the sections were stained with hematoxylin-eosin and Klüver-Barrera. Immunohistochemical studies were performed using a standard polymer-based method with the EnVision Kit or anti-goat immunoglobulin (Dako). As primary antibodies, we used antibodies to the following: anti-ubiquitin (ubiquitin, monoclonal mouse, 1:250; Millipore), anti–TDP-43 (TARDBP, polyclonal rabbit, 1:2500; ProteinTech), anti–phosphorylated TDP-43 (pTDP-43 ser409/410, polyclonal rabbit, 1:2500; CosmoBio), anti–phosphorylated tau (AT-8, monoclonal mouse, 1:4000; Innogenetics), anti–β-amyloid (β-amyloid 6F/3D, monoclonal mouse, 1:100; Dako), anti-CD68 (CD68, monoclonal mouse, 1:200; Dako), anti–cystatin C (cystatin C, polyclonal rabbit, 1:200; Dako), p62 N-terminal (p62N, polyclonal guinea pig, 1:100; Progen), anti–ubiquilin 2 (UBQLN-2 5F5, monoclonal mouse, 1:5000; Abnova), and anti–choline acetyltransferase (ChAT, polyclonal goat, 1:100; Millipore). Diaminobenzidine (Wako) was used as the chromogen.
Antigens were retrieved with trypsin for anti-CD68 immunohistochemistry and with 95°C 3 mmol/L citrate buffer at 95°C for 20 minutes, followed by 5-minute incubation in 98% formic acid for anti-p62N, anti–TDP-43, anti–pTDP-43, and anti-ChAT immunohistochemistry. To confirm the presence of TDP-43–positive inclusions within the cholinergic motor neurons, we performed double immunohistochemistry using anti–pTDP-43 and anti-ChAT antibodies. Spinal cord specimens were prepared from 3 patients with type A, 3 with type B, and 2 with type C. Initially, the specimens were immunostained with the anti-ChAT and anti-goat immunoglobulin antibodies and diaminobenzidine. The anti-ChAT antibody was inactivated in distilled water at 100°C for 20 minutes, followed by immunohistochemistry with pTDP-43 and violet pigmentation using a VIP Peroxidase Substrate Kit (SK-4600; Vector).
For the semiquantitative neuropathological analysis, 2 investigators (Y.R. and M.Y.) observed the specimens containing the facial and hypoglossal nuclei and the anterior horn of the spinal cord. They evaluated the severity of LMN neuropathological changes that are indicative of ALS (neuronal loss, gliosis, aggregation of macrophages, TDP-43–immunopositive neuronal inclusions, and Bunina bodies) and graded neuronal loss and gliosis using Klüver-Barrera and hematoxylin-eosin staining. The investigators also evaluated the aggregations of macrophages rather than rod-shaped microglia using anti-CD68 immunohistochemistry and identified Bunina bodies using hematoxylin-eosin staining and anti–cystatin C immunohistochemistry. They scored the severity of neuronal loss and gliosis as grade 0 (none), 1 (mild), 2 (moderate), or 3 (severe) (eFigure 1 in Supplement. The appearance of TDP-43–positive inclusions was scored as grade 0 (none), grade 1 (1-5 neuronal inclusions per 5 fields; ×20 objective), grade 2 (6-10 inclusions), or grade 3 (≥11 inclusions) using anti–pTDP-43 immunohistochemistry.
Pathological cortical TDP-43 subtypes were identified according to current neuropathological criteria, using specimens from the frontal lobes, temporal lobes, and hippocampus.5 For FTLD-TDP, type A was defined as the presence of neuronal cytoplasmic inclusions predominantly in the neocortex layer 2 and short dystrophic neurites; type B, as a predominance of neuronal cytoplasmic inclusions in all cortical layers; and type C, as a predominance of long dystrophic neurites in layer 2 and cytoplasmic inclusions in the dentate granular cells of the hippocampus. Our patient series did not include type D, which is characterized by numerous short dystrophic neurites and neuronal intranuclear inclusions in association with valosin-containing protein gene mutations. We also evaluated pathological changes in the upper motor neuron systems that include the primary motor cortex and corticospinal tract (CST). We evaluated the presence or absence of neuronal loss and gliosis in the primary motor cortex and myelin pallor, as well as the aggregation of macrophages in the CST.
Our study focused on sporadic FTLD-TDP, and patients with familial histories of FTLD or ALS, dementia, or other neurodegenerative diseases were excluded. However, FTLD or ALS associated with chromosome 9 open reading frame 72 (C9ORF72) hexanucleotide expansion exhibits pathological aggregation of TDP-43 and, in some cases, low penetration,20,21 although these mutations are extremely rare in Japan.22 It was recently reported that the pattern of ubiquilin abnormalities in ALS and FTLD corresponds well with the presence of C9ORF72 hexanucleotide expansion.23 The UBQLN-2–positive, p62-positive, but TDP-43–negative thick dystrophic neurites are abundantly present in patients with C9ORF72 hexanucleotide expansion, predominantly in the hippocampus and cerebellum. Because the materials for a genetic study were not available for a large proportion of our patients, we histologically screened C9ORF72 hexanucleotide expansion with the absence of UBQLN-2 and p62N-positive thick dystrophic neurites in the temporal lobes and cerebella of all patients.
The Mann-Whitney test was applied to continuous variables between 2 groups, and the Kruskal-Wallis test was applied to the analysis of continuous variables among 3 groups. The χ2 test was used for categorized variables among 3 groups. Spearman rank correlation coefficient analyses were applied to univariate correlations between the clinical groups and severity of pathological changes. Survival curves were constructed using the Kaplan-Meier method. The end point of clinical course was defined as death or the introduction of a respirator or tracheotomy. The significance level for all comparisons was set atP < .05. All statistical tests were 2 sided and were conducted using the PASW 18.0 program (IBM SPSS).
Patient characteristics are summarized in the Table. The mean (SD) time from symptom onset to death or respirator or tracheotomy administration was 50.5 (58.4) months across all patients. The survival time from symptom onset did not differ significantly between the FTLD-ALS and ALS-FTLD groups but was significantly shorter for the FTLD without ALS group than for the FTLD-ALS or ALS-FTLD group (Figure 1 and Table; P < .001). The most common cause of death for the ALS-FTLD and FTLD-ALS groups was respiratory failure, but patients with FTLD without ALS commonly died of other systemic diseases (P < .001). Frequencies of dementia subtypes did not significantly differ between the clinical groups.
With regard to motor symptoms/signs, 3 patients in the FTLD without ALS group had hyperreflexia, 1 had the Babinski sign, and 1 had spasticity, but none had a clinical diagnosis of progressive lateral sclerosis (PLS) according to the published diagnostic criteria of PLS.24 Patients with FTLD-ALS or ALS-FTLD generally exhibited both upper motor neuron and LMN symptoms/signs except for 3 who exhibited only LMN symptoms/signs. Based on the electromyographic data, active denervation potentials (positive sharp waves and fibrillation potentials18) were identified in 3 patients with FTLD-ALS and 12 with ALS-FTLD but not in any of those with FTLD without ALS.
The results of semiquantitative pathological evaluations of the 3 clinical groups are summarized in Figure 2. In the FTLD without ALS group, 8 of 9 patients (89%) showed pTDP-43–positive neuronal inclusions. In addition, neuronal loss and gliosis in the spinal anterior horns were observed in 5 of 11 patients (45%) and Bunina bodies were present in 4 (36%). The pathological changes in LMN systems were most severe in the ALS-FTLD group, followed by the FTLD-ALS group, and were rather mild in the FTLD without ALS group. Among control patients, 1 had a pTDP-43–positive glial inclusion in the lumbar anterior horn, but this patient did not show neuronal loss, gliosis, or Bunina bodies (eFigure 2 in Supplement).
According to cortical TDP-43 pathological findings,5 29 patients were classified into 3 subtypes: A (n = 6), B (n = 20), or C (n = 3). Patients with FTLD without ALS showed type A or C disease, whereas those with FTLD-ALS or ALS-FTLD all showed type B disease (Figure 2 and Figure 3). For all the subtypes, the LMN system showed neuropathological changes that were indicative of ALS, including pTDP-43–positive neuronal and glial inclusions, neuronal loss, and gliosis. In patients with type A disease (Figure 4A-H), the severity of neuronal loss and gliosis in LMN systems ranged from none to moderate. Five patients (83%) in this group had pTDP-43–positive, skeinlike cytoplasmic and/or nuclear inclusions (Figure 4B and C), and 4 (67%) had Bunina bodies (Figure 4E and F) in the LMNs.
All 20 patients in the type B group (Figure 4I-L) showed neuronal loss, gliosis, and pTDP-43–positive skeinlike cytoplasmic inclusions in the LMN systems, and 18 (90%) had Bunina bodies. Among the 3 patients with type C disease (Figure 4M-P), 1 (33%) had mild loss of the LMNs (Figure 4M), and all 3 (100%) had pTDP-43–positive skeinlike cytoplasmic inclusions in the LMNs (Figure 4N). Unlike patients with the other subtypes, those with type C disease lacked Bunina bodies. Moreover, thick dystrophic neurites were prominent in the spinal anterior horn in patients with type A or C disease but rarely present in those with type B disease (Figure 4G and O). These dystrophic neurites were larger in diameter (8-12 μm) than those found in the cortices. In a double immunohistochemical analysis, pTDP-43–positive inclusions were found within the cytoplasm of ChAT-positive neurons in patients with type A, B, and C disease (Figure 4H, L, and P).
In the primary motor cortex, neuronal loss and gliosis were evident in 5 patients with FTLD without ALS (56%), 2 with FTLD-ALS (25%), and 3 with ALS-FTLD (25%). Myelin pallor in the CST was evident in 6 patients with FTLD without ALS (67%), 1 with FTLD-ALS (12%), and 2 with ALS-FTLD (17%). Aggregations of macrophages in the CST were evident in 4 patients with FTLD without ALS (44%), 5 with FTLD-ALS (62%), and 6 with ALS-FTLD (50%).
No patients showed any cerebellar UBQLN-2–positive or p62N-positive structures. In the temporal lobes, UBQLN-2–positive structures were occasionally observed in 8 patients, but abundant, thick, and aggregatelike structures, which are found in patients with C9ORF72 expansions, were not observed (eFigure 3 in Supplement). We presumed that our patients did not have C9ORF72 expansions.
Our study demonstrated that pTDP-43–associated pathological changes were common in the spinal anterior horns of the FTLD without ALS, FTLD-ALS, and ALS-FTLD groups. Neuronal loss and gliosis were most severe among the ALS-FTLD group, followed by the FTLD-ALS and then the FTLD without ALS groups. Our results clearly demonstrated the pathological continuum among TDP-43–associated FTLD and ALS, even at the LMN level.
Although the FTLD without ALS group that lacked LMN symptoms showed a loss of LMNs, the degree of neuronal loss and TDP-43 disease were generally mild in this group. Experiment data using ALS mouse models revealed that symptoms developed when approximately 29% of spinal motor neurons were lost.25 Further investigation will be needed to clarify whether LMN involvement occurs in a later stage of illness or progresses very slowly compared with cerebral involvement in FTLD without ALS.
Our results revealed that the FTLD-TDP types A, B, and C were associated with neuropathological changes corresponding to ALS in the spinal motor neurons. The severity of neuronal loss and pTDP-43 disease in the spinal motor neurons may differ quantitatively among these neuropathological subtypes. Based on cortical TDP-43 pathological findings, patients in the type B group had severe neuronal loss and diffuse pTDP-43–positive neuronal inclusions, which were entirely identical to ALS, whereas these changes were mild in the type C group. In type A, LMN pathological findings were diverse regardless of clinical duration; their severity and extension may be heterogeneous among patients with type A disease, unlike those with type B or C disease. Indeed, type A disease has also been identified in the FTLD with ALS phenotype in sporadic or familial (C9ORF72 expansion or progranulin gene mutations) form.2,5,21,26 Dystrophic neurites were prominent in the spinal anterior horn of patients with type A or C disease. In our patient series, Bunina bodies were observed in most patients with type A or B disease but were absent in those with type C disease, findings consistent with those of previous studies.3,17
Several studies have demonstrated that some patients with FTLD-TDP, particularly type C, showed marked CST degeneration.3,11,17,27 We also observed a marked myelin pallor in the CST in 67% of patients with FTLD without ALS, 12% with FTLD-ALS, and 16% with ALS-FTLD (50% for type A, 15% for type B, and 100% for type C). Some patients showed neuronal loss or gliosis in the primary motor cortex to varying extents. Furthermore, patients with FTLD without ALS often exhibited severe degenerative changes in broad areas of the frontal cortices. The broad involvement of the frontal lobes might also contribute to the CST degeneration because CST fibers arise not only from the primary motor cortex but also from the premotor cortex and supplementary motor areas.28
Two limitations of our study is that the evaluation of slight or very mild muscle weakness was not completed and that there were few patients with electromyographic data in the FTLD without ALS group. However, our clinical data demonstrated that patients with FTLD without ALS had significantly longer survival times than those with FTLD-ALS or ALS-FTLD. These prognostic data correspond well to previous results.29,30 In addition, the causes of death differed considerably between the FTLD without ALS group and the FTLD-ALS and ALS-FTLD groups. Respiratory failure was observed in patients with FTLD-ALS or ALS-FTLD but not in those with FTLD without ALS, and respiratory failure was strongly associated with severity of LMN loss. These results support the view that classification of FTLD based on the presence of LMN involvement was applicable in this study.
In conclusion, the LMN systems of FTLD-TDP generally show neuropathological changes that are indicative of ALS, although the severity of pathological changes differs among clinical phenotypes or subtypes of cortical TDP-43 disease. A pathological continuity between FTLD-TDP and ALS is supported by evidence of LMN involvement.
Accepted for Publication: October 16, 2013.
Corresponding Author: Gen Sobue, MD, Department of Neurology, Nagoya University Graduate School of Medicine, Tsurumai 65, Showa-ku, Nagoya, Japan (firstname.lastname@example.org).
Published Online: December 30, 2013. doi:10.1001/jamaneurol.2013.5489.
Author Contributions: Drs Sobue and Yoshida had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Riku, Watanabe, Yoshida, Sobue.
Acquisition of data: Riku, Watanabe, Yoshida, Masuda, Senda, Sobue.
Analysis and interpretation of data: Riku, Watanabe, Yoshida, Tatsumi, Mimuro, Iwasaki, Katsuno, Iguchi, Ishigaki, Udagawa, Sobue.
Drafting of the manuscript: Riku, Watanabe, Yoshida, Sobue.
Critical revision of the manuscript for important intellectual content: Tatsumi, Mimuro, Iwasaki, Katsuno, Iguchi, Masuda, Senda, Ishigaki, Udagawa, Sobue.
Statistical analysis: Riku, Watanabe, Masuda, Senda.
Administrative, technical, or material support: Riku, Yoshida, Tatsumi, Mimuro, Iwasaki, Iguchi, Udagawa.
Study supervision: Watanabe, Yoshida, Katsuno, Senda, Sobue.
Conflict of Interest Disclosures: None reported.
Funding/Support: This work was supported by grants-in-aid from the Research Committee of Central Nervous System Degenerative Diseases by Ministry of Health, Labour, and Welfare and from Integrated Research on Neuropsychiatric Disorders, carried out under the Strategic Research for Brain Sciences by Ministry of Education, Culture, Sports, Science, and Technology of Japan.
Role of the Sponsor: The funding agencies had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Additional Contributions: We thank all the patients, their families, and the staff in the affiliated hospitals for providing autopsy materials and clinical data.