Different patterns of spreading direction and motor neurons involvement in a cohort of limb‐onset amyotrophic lateral sclerosis patients from Southern Italy: Potential implication on disease course or progression?

Abstract Background Currently, there is a lack of knowledge concerning where the pathological process starts and how the neurodegeneration spreads during the course of amyotrophic lateral sclerosis (ALS). Aims This study aims to evaluate the spreading direction of the disease and the corresponding clinical characteristics in a cohort of patients with limb‐onset ALS. Patients and methods Consecutive incident ALS patients referring to an ALS tertiary center from Southern Italy, between 2015 and 2021, were recruited in the study. According to the initial directions of spread, patients were dichotomized into horizontal spreading pattern (HSP) or vertical spreading pattern (VSP) groups. Results Among 137 newly diagnosed ALS, 87 presented a spinal onset. Ten patients with pure LMN were not included in the study. All cases reported a clear spread direction. The frequency of HSP and VSP spreading was similar overall (47 vs. 30). The prevalence of HSP was higher (74% vs. 50%) in patients with upper limb‐onset (UL‐ALS), compared to patients with lower limb‐onset (LL‐ALS; p < .05). Conversely, the occurrence of VSP spread was threefold higher in patients with LL‐ALS, compared to UL‐ALS (p < .05). Patients with VSP showed a wider upper motor neuron impairment, whereas the involvement of LMN resulted greater in patients with HSP. Patients with HSP exhibited a greater drop of ALSFRS‐r sub‐score in the region of onset, while VSP showed a slighter but more diffuse reduction of ASLFRS‐r subscore in more body districts beyond the site of onset. Patients with VSP were also characterized by a higher median progression rate and an earlier median bulbar involvement, compared to HSP. Conclusions Our findings suggested investigating the spreading direction of ALS among patients with spinal onset, to better delineate the clinical profiles of patients with ALS, and predict an earlier impairment of bulbar muscle and a more rapid progression of the disease


BACKGROUND
Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disorder characterized by progressive degeneration of motor neurons in the cortex, the brainstem, and the spinal cord (Kiernan et al., 2011).
Heterogeneous motor manifestations are caused by simultaneous and progressive degeneration of upper motor neurons (UMNs) and lower motor neurons (LMNs) (Kiernan et al., 2011). Currently, through the available clinical and instrumental exams, it is still impossible to reveal if the pathological process starts in the motor neurons of the cortex, brainstem, or spinal cord. This latter limit is due to the lack of studies performed during the pre-symptomatic or very early phases of the disease. Indeed, patients with ALS commonly refer to a tertiary center after a median interval from symptom onset of 12 months (Zoccolella et al., 2006), during which simultaneous and progressive degeneration of UMNs and LMNs occur. Therefore, all the information regarding the primary pathologic focus and spreading pattern is based on the anamnestic history reported by patients during their first visit.
According to available population-based studies, the first symptoms occur in only one body region in 90%−98% of cases Turner et al., 2010). The subsequent progression is either confined largely to the same region (e.g., cervical or lumbosacral districts) before the involvement of other body districts or has a more diffuse spreading, with early and simultaneous involvement of several regions. Spread seems to be preferentially contiguous in either a horizontal, rostralcaudal, or caudal-rostral direction (Gargiulo-Monachelli et al., 2012;Körner et al., 2011;Ravits & La Spada, 2009).
In 2007,  reviewed the clinical history of 100 patients with ALS and found that both UMN and LMN signs spread outward to contiguous regions following the underlying, somatotopically arranged, neuronal anatomy. This is best seen when the onset of symptoms is spinal. For example, LMN clinical deficits may progress from one arm to the other, consistently with the neuronal anatomy of LMNs in the spinal cord. Instead, UMN clinical deficits progress from one arm to the ipsilateral leg, consistent with the somatotopic anatomy of the cerebral cortex .
Based on this postulation, we studied a cohort of consecutive patients with limb-onset ALS to evaluate the spreading direction of motor symptoms of the disease and the impact of the spreading direction on clinical characteristics.  (Chio et al., 2011). Patients were defined as definite, probable, possible, and suspected ALS according to the El Escorial criteria (Brooks, 1994) and their revised version of Airlie

Subjects and inclusion/exclusion criteria
House (Brooks et al., 2000). Considering the intrinsic divergent clinical profile of patients with suspected ALS (Schito et al., 2020) and primary lateral sclerosis (Gordon et al., 2006), patients with pure LMN and pure UMN were excluded from the analysis. Subsequent analyses focused only on patients with clearly identified, lateralized onset of weakness or wasting in an upper or lower limb ( Figure 1).

Spreading patterns
Progression of disease was defined by the extension of weakness or muscle wasting beyond the site of onset, established through a combination of anamnestic data and clinical evaluation. Patients were then divided into groups according to the different initial directions of spread, namely horizontal spreading pattern (HSP), vertical spreading pattern (VSP) or crossed spreading pattern. The direction of spread was considered horizontal when the first region of spreading consisted of cervical to cervical contralateral or lumbar to lumbar contralateral.
The vertical direction was defined as spread either from the cervical or lumbar region to the ipsilateral leg or arm. The crossed pattern consisted of those patients in whom symptoms spread in a diagonal pattern either from the cervical or lumbar region to the contralateral lumbar or cervical region.

Clinical evaluation
UMN burden was evaluated using the Penn Upper Motor Neuron scale (PUMNS) (Quinn et al., 2020). LMN burden was calculated using the score previously proposed by Devine et al. (2016). This latter score assigned a separate LMN score for each limb, ranging from 0 (no involvement) to 3 (significant, severe involvement), based on the Medical Research Council and muscle wasting.
The sum of LMN and UMN scores, calculated separately for upper and lower limbs, and the total LMN and UMN scores were retained for further analyses.
All patients were functionally evaluated using the ALS Functional  In order to avoid the confounding role of the disease severity at the clinical evaluation, ALSFRS-r and disease duration were included as covariates. Estimated marginal means of clinical measures were plotted with error bars to illustrate the differences between groups using a boxplot graph. Afterward, the above-mentioned analyses were performed separately in upper and lower limb-onset ALS patients to evaluate the differences between patients with HSP and VSP in the sum of the LMN and UMN scores of the arms and legs.

F I G U R E 1 Study population.
The association between number of preserved body districts and direction of spread, namely VSP and HSP, was tested using Glass rank biserial correlation (rg) (Glass, 1966).
A time-to-event analysis was performed based on anamnestic history reported by patients during the first visit: the development of bulbar symptoms (if occurred) was used as endpoints. Time to generalization was used for patients presenting with bulbar involvement at the first visit. Otherwise, the time elapsed between disease onset and clinical evaluation (namely disease duration) was used as a period of observation for the patients who did not experience the events.
Log-rank test was used to test for differences between horizontal and vertical spreading pattern groups.
Subjects were enrolled within the "SLAP-Dem Study," a populationbased study on rare neurodegenerative disease that was approved by

RESULTS
During the study period, 137 newly diagnosed ALS patients were identified, of which 87 (64%) presented a limb-onset. At the first visit, all spinal onset ALS patients had at least two limbs involved, and all of them reported a clearly defined spread direction. Ten patients (7% of the entire cohort and 11.5 % of spinal onset ALS) were pure LMN syndromes and were not included in the analysis, as stated above. No cases of pure UMN syndrome were identified. Among the remaining patients with ALS, 60 were identified as classic ALS (78%), five (6%) as pyramidal phenotype, six (8%) as flail arms, and six (8%) as flail legs.
No other spread directions were reported in our cohort. The clinical and demographic characteristics of the two groups are reported in Patients with upper limb-onset ALS (UL-ALS) more likely presented an HSP spreading direction (74% vs. 26% VSP; χ 2 = 4.7; p = .03), while the spreading was similar among patients with lower limb-onset (LL-ALS) ( Figure 2). Therefore, the risk of VSP was 2.9-fold higher in patients with LL-ALS compared to patients with UL-ALS (95% CI 1.01-7.61; p = .03). Conversely, no differences in the frequency of vertical/horizontal spreading patterns were found according to the side of symptoms onset (Figure 2).

Total population
The clinical profiles of the two subgroups are described in Table 1

Upper limb-onset ALS patients (UL-ALS)
Among patients with UL-ALS, those with HSP presented a higher LMN  Figure 3).

F I G U R E 2
Frequency of the site of the second segment spreading according to the site of onset.  Figure 3).

Assessment of disease severity expressed as ALSFRS-r score
No significant differences in total ALSFRS-r score were found between patients with VSP and HSP. However, subgroup analyses revealed that UL-ALS patients with HSP exhibited a statistically significant lower median value of ALSFRS-r-UL subscore, compared to VSP (

F I G U R E 3 Differences in clinical profiles of patients with amyotrophic lateral sclerosis (ALS) between vertical crossed spreading (VSP) and horizontal spreading pattern (HSP). (a) Upper limb-onset ALS patients and (b) lower limb-onset ALS patients.
were found in ALSFRS-r-B, ALSFRS-r-LL, or ALSFRS-r-R sub-score between HSP and VSP in patients with ALS with the onset in the lower limbs.

DISCUSSION
In the present study, we found different patterns of the disease spread- findings [Sekiguchi et al., 2014;Zhenfei et al., 2019]). Conversely, our study aimed to investigate the clinical characteristics of ALS patients according to two different spreading patterns, namely VSP and HSP.
Despite this latter methodological difference, our findings confirm the focal onset and clearly lateralized spreading direction of the disease in all patients of our cohort, as reported elsewhere (Brooks, 1991;Körner et al., 2011;. Similar to previous studies (Brooks, 1996;Fujimura-Kiyono et al., 2011;Körner et al., 2011), motor symptoms spread between adjacent regions either from rostral to caudal and vice versa, following the somatotopically arranged neuroanatomy of the cortex, or horizontally to the adjacent region, 14% [Kimura et al., 2006]). Previous studies have ascribed the "crossed pattern" to the lack of sensitive clinical and functional signs of the thoracic region, whose involvement may have gone unnoticed in patients progressing from lumbosacral to upper limb region and vice versa (Gromicho et al., 2020). Furthermore, Sekiguchi et al. (2014) used electromyography to test "contiguous" versus "multifocal hits" progression in 39 patients, and the authors proposed a "multifocal hits and local propagation" hypothesis. The absence of ALS patients with "crossed pattern" in our cohort may be explained by the relatively small size of our population, which did not allow us to incept this rare non-contiguous pattern of disease progression.
Another important observation of our study is the higher prevalence of HSP among upper limb-onset ALS patients. Since VSP seemed to reflect a preferential involvement of the somatotopically arranged neuroanatomy of the cortex, the lower VSP prevalence among patients with UL-ALS is probably due to the more susceptibility of cortical motor neurons innervating the lower limbs to neurodegeneration.
Specifically, the anterograde (or Wallerian) axonal degeneration of cortical motor neurons in ALS disease was previously demonstrated by numerous radiological studies (Agosta et al., 2010;Blain et al., 2011;Filippini et al., 2010), and it followed a neuronal insult, which could occur anywhere along corticospinal tract. Therefore, it could be speculated that longer motor neurons (e.g., those innervating lower limbs) are more susceptible to neuronal damage compared to shorter neurons (e.g., those innervating upper limbs).
Our study also showed that while the classic ALS phenotype could equally exhibit a VSP or HSP, different ALS phenotypes exhibited a preferential spreading pattern. Specifically, the pyramidal phenotype showed exclusively a VSP, while the flail arm and the flail leg exhibited only an HSP. These latter findings agree with previous studies which reported that higher UMN impairment (as found in pyramidal symptoms) was reflected by wider motor and extra motor cortical thinning and lower diffusion tensor imaging metrics of the corticospinal tract (Mezzapesa et al., 2013;Woo et al., 2014). Conversely, flail arm and flail leg phenotypes differ from healthy controls neither in cortical thick-ness (Schuster et al., 2013) nor in ALS peculiar neurophysiological index (e.g. split hand sign; Yang et al., 2015). Therefore, we agree with the previous authors who speculated that classical ALS and its variant could or could not originate from the motor cortex (Yang et al., 2015), and we suggested to delineate the spreading pattern in order to timely incept rare variants (e.g., pyramidal, flail arm, flail leg phenotypes) with undoubtedly different course of the disease (Schito et al., 2020).
To the best of our knowledge, our study for the first time evaluated the impairment of UMN and LMN according to different spreading directions, using well-known quantitative scores. A recent previous study adopted an exactly opposite methodological approach, evaluating the spreading directions after dichotomizing patients according to predominant UMN or LMN impairment (Gromicho et al., 2020). Differently from this latter study which reported that the next involved body part was the contralateral limb in both patients with predominant UMN or LMN signs, we observed that VSP has a wider impairment of UMN, while HSP exhibited a higher LMN burden. Our findings confirmed the abovementioned theory according to which VSP and HSP seemed to reflect a preferential somatotopically distribution of UMN and LMNs.
To date, only one study evaluated UMN involvement through quantitative scales in patients with different spreading directions (Hu et al., 2016), and the study reported similar results.
Furthermore, we demonstrated that the wider UMN impairment observed in VSP usually oversteps the site of onset even in the early phases of the disease, while the prominent LMN involvement in HSP is largely confined to the region of onset (e.g., cervical or lumbar), at least at the time of diagnosis.
The absence of statistically significant differences among ALSFRS-r total score between ALS patients with VSP and HSP has been previously reported (Hu et al., 2016). Nonetheless, we demonstrated that patients with HSP exhibited a greater drop of ALSFRS-r sub-score (e.g., 3-4 points) in the region of onset, while VSP showed a slighter (e.g. 1-2 points) but more diffuse reduction of ASLFRS-r subscore in more body districts beyond the site of onset. Since patients with VSP and HSP were characterized by prevalent UMN and LMN, respectively, it could be speculated that the UMN burden is more associated with the onset and spreading of the ALS disease rather than the LMN burden, as confirmed by epidemiological, clinical (Devine et al., 2014), and neurophysiological studies (Zoccolella et al., 2021).
These latter results were corroborated by the evidence that among spinal onset ALS patients, those with VSP (that more likely showed a UMN pattern) exhibited a shorter disease duration and a higher progression rate, and an earlier bulbar involvement, compared to HSP spread. Therefore, patients with VSP reached bulbar involvement and a higher degree of disability early compared to patients with HSP, suggesting a possible broader or even faster clinical course.
A possible explanation of these latter results laid in the neu- However, our study has potential limitations as it was a retrospective study conducted on a relatively small sample of consequent patients enrolled in a tertiary ALS center, which could have led to potential selection bias. Another potential limitation, consistent with the large part of available studies, is the chance of error in evaluating the progression pattern on a single point amnestic data in patients with a long progression time. However, we believe however that this chance of error in our study cohort is limited given the overall median disease duration of 22 months.
In conclusion, although preliminary, our findings suggest investigating and underlining the spreading direction of ALS among spinal onset cases, in order to predict an earlier involvement of bulbar symptoms and eventually a more rapid course of the disease.