Montreal Cognitive Assessment (MoCA) performance in Huntington’s disease patients correlates with cortical and caudate atrophy

Introduction

Huntington’s Disease (HD) is an autosomal dominant inherited neurodegenerative disease (Huntington’s Disease Collaborative Research Group, 1993). The hallmark of the HD neuropathology is caudate neurodegeneration; although several cortical regions are also affected (Rosas et al., 2008; Tabrizi et al., 2009; Aylward, 2014). HD clinical manifestations include motor, cognitive, and psychiatric deficits. Even though HD is classically considered as a motor disorder, the cognitive decline may occur even before the clinical manifestations of the motor symptoms start (Paulsen, 2011). Although it has been suggested that cognitive deficits are mainly linked to striatal degeneration (Kassubek et al., 2004; Peinemann et al., 2005; Aylward et al., 2013), the extensive atrophy of the cortical mantle also contributes considerably to the progression of these alterations (Johnson et al., 2015).

Cognitive assessment in HD patients has been performed using several tests, including the Cognitive Functioning evaluation of the Unified Huntington’s Disease Rating Scale (UHDRS) (Mestre et al., 2016). However, the analysis of the possible brain structures contributing to the cognitive impairment in HD has only been based on correlation analyses with the UHDRS cognitive component, namely, no other conventional cognitive assessment instrument has been associated with the structural brain atrophy in HD. One of the most used cognitive screening tests in research settings and clinical practice is the Montreal Cognitive Assessment (MoCA). This screening test shows a high sensitivity to detect mild cognitive impairment (MCI) in the evaluated participant (Nasreddine et al., 2005). Besides, the MoCA global performance can be used to detect, follow up, and predict the progression of cognitive impairment in a number of clinical populations (Julayanont et al., 2014; Sivakumar et al., 2014; Nijsse et al., 2017; Hendershott et al., 2017). Likewise, the MoCA total score has also shown a relationship with MRI functional markers associated with cognitive impairment (Cheewakriengkrai et al., 2014; Porges et al., 2017) and with atrophy of cortical and subcortical nuclei (Ritter et al., 2017; Ogawa et al., 2019). The aforementioned evidence suggests that the MoCA test is useful not only to evaluate cognitive performance but also to explore its relationship with functional and structural brain deterioration.

In HD patients the MoCA test has been evaluated to determine its usefulness and sensitivity to detect the presence of potential cognitive impairment. Some reports have demonstrated that the MoCA test presents high sensitivity and specificity to detect cognitive impairment, even suggesting that it presents higher sensitivity than other brief screening instruments (Videnovic et al., 2010; Mickes et al., 2010; Ringkøbing et al., 2020). Consequently, the MoCA test is a suitable monitoring and screening tool for assessing cognitive dysfunction in HD patients (Videnovic et al., 2010; Mickes et al., 2010; Bezdicek et al., 2013; Gluhm et al., 2013; Ringkøbing et al., 2020) across a wide range of disease severity stages (Mild, Moderate, and Severe) (Gluhm et al., 2013). However, there is no current evidence of the association of MoCA global performance with the pattern of brain structural atrophy exhibited by HD patients. Therefore, the present study aimed to determine the relationship between the MoCA global performance and the cortical and subcortical gray matter deterioration in prodromal/early HD patients.

Methods

Results

MoCA cut-off determination and MoCA global performance comparison

The AUC ± SD of MoCA total score was 0.794 ± 0.07 with 95% CI of [0.647–0.941], and sensitivity and specificity of 72.7% and 86.4%, respectively. MoCA test showed an AUC significantly greater than 0.5 (p = 0.000), which demonstrates its discriminative ability between HD patients and healthy controls. The maximum likelihood of Youden’s index J was 0.59 and showed the optimum cut-off point was < 26 pts coinciding with its suggested original cut-off score (Fig. 1). Comparison for MoCA total score between healthy control (Mean ± SD = 27.90 ± 1.41; range = 6) and HD patients (Mean ± SD = 24.18 ± 4.17; range = 16) showed significant differences (W = 384.5; p = 0.000) with a large effect size (Cohen’s r = 0.50; Fig. 2A), indicating a large difference between group medians. Thirteen HD patients (59%) presented an impaired cognitive performance with scores ranging between 14 to 25 pts, and nine patients (41%) had no impaired cognitive performance with MoCA scores ≥ 26 pts.

Discussion

In this study, we tested the hypothesis that MoCA global performance correlates with a specific brain atrophy pattern in prodromal/early HD. Our analyses identified that the MoCA total score correlates with bilateral caudate and left amygdala volume differences and with extensive cortical thinning of areas that belong to the temporo-occipital, and parietal and superior frontal cortices.

MoCA has demonstrated adequate sensitivity to detect cognitive impairment in HD patients (Mickes et al., 2010; Bezdicek et al., 2013; Gluhm et al., 2013). Here, we corroborate the ability of MoCA to distinguish between healthy controls and prodromal/early HD patients according to their cognitive performance. Besides, we determine the appropriate MoCA cut-off score (< 26 pts) for this HD cohort, which coincided with the original cut-off point suggested previously to HD (Nasreddine et al., 2005; Bezdicek et al., 2013; Ringkøbing et al., 2020).

The analysis of the prodromal/early HD patients’ MoCA total score showed a significant difference with respect to the control group. This result support previous reports showing that symptomatic as well as prodromal HD patients exhibit cognitive decline assessed using brief screenings or comprehensive neuropsychological tests (Rosas et al., 2008; Duff et al., 2010; Tabrizi et al., 2013; Paulsen, Smith & Long, 2013; You et al., 2014). It should be noted that the MoCA test composition encompasses psychometric properties to detect cognitive impairment (Vogel et al., 2015); therefore, this test is particularly useful for HD because the patients typically develop notably heterogeneous cognitive deficits with different degrees of impairment that weigh the presence of the global cognitive decline.

MoCA global performance correlations analysis demonstrated extensive cortical areas associated with cognitive performance in HD patients. In this sense, our findings are consistent with previous reports in HD that identified specific cognitive deficits –attention, working memory, and executive functions– correlating with cortical changes including cortices of the superior fronto-parietal motor circuits, frontal cognitive control centers, temporal auditory and semantic processing hubs, and occipital visual centers (Rosas et al., 2005, 2008; Harrington et al., 2014; Coppen et al., 2018; Martinez-Horta et al., 2020). It is important to mention that the MoCA test involves the assessment of several cognitive functions associated with different neurological substrates; therefore, it is understandable that the atrophy of discreet areas within the whole cerebral cortex contributed to the deterioration of MoCA global performance of the HD patients.

It is well known that striatal atrophy is the neuropathological hallmark of HD (Vonsattel et al., 1985; Aylward et al., 2000) being the caudate volume differences considered as a biomarker of disease progression (Aylward, 2014). Since the striatum is strongly associated with cognitive functioning (Graff-Radford et al., 2017), it was expected that the caudate volume differences correlated strongly with MoCA total score. Therefore, our results strengthen the previous evidence showing that caudate atrophy is associated with cognitive impairment evaluated by different neuropsychological tests (Peinemann et al., 2005; Rosas et al., 2008; Aylward et al., 2013; You et al., 2014; Kim et al., 2017). In this respect, it would be sensible to suggest that the influence of striatal atrophy on cognitive impairment is driven first by the progressive death of striatal spiny neurons; and secondly by the disconnection of this nucleus with the frontal cortex as a result of axonal neurodegeneration (Rosas et al., 2010, 2018; Poudel et al., 2015) affecting cortico-striatal tracts highly connected to regions within the motor and associative networks (Rosas et al., 2010).

Conversely, the MoCA scores also correlated with the volume differences of the left amygdala, which has been of interest because of the evidence that the amygdala is affected in presymptomatic and symptomatic HD stages (Douaud et al., 2006; Ahveninen et al., 2018; Tang et al., 2018). Although the amygdala has been strongly associated with psychiatric symptoms (Adolphs, 2000; Fine & Blair, 2000), recently, it has been shown that smaller amygdala volumes are associated with worse visuomotor skills, slower processing speed, and emotional recognition (Kipps et al., 2007; Ahveninen et al., 2018). It is important to note that in the healthy control group, we did not identify a significant correlation between MoCA global performance and cortical and subcortical gray matter differences; therefore, we suggest that the cognitive-atrophy associations found for HD patients are specific to the neuropathology of disease.

The MoCA total score also showed significant strong correlations with the clinical measures TFC and UHDRS-TMS; this is in line with previous reports showing that cognitive impairment is associated with the decline of functional and motor performance in HD patients (Hamilton, 2003; Nehl & Paulsen, 2004; Jacobs et al., 2016). In addition, the degree of cognitive impairment does increase proportionally with the HD stage progression determined by functional and motor status, which is reflecting a common behavioral phenotype of HD (Toh et al., 2014; Jacobs et al., 2016). Furthermore, it has been recently suggested that cognitive and motor symptoms share a common neurobiological basis (Garcia-Gorro et al., 2019), though these alterations may be affected at different levels and progress distinctively. Finally, even though CAG repeat length drives age of onset and severity of disease, and it has been correlated with volume differences of the striatum and motor cortex (Rosas et al., 2001; Langbehn et al., 2019) as well as with cognitive and motor deficit progression (Rosenblatt et al., 2006, 2012), here we did not find a significant correlation with MoCA global performance. Further studies would be necessary to identify if MoCA decline presents a CAG repeat length–dependent trajectory.

To the best of our knowledge, this is the first study that evaluates the correlation of the MoCA global performance with the cortical brain atrophy of a neurodegenerative disease such as HD. However, our study had some limitations to consider. (1) It would be ideal to reproduce the correlation analysis with a larger sample size to corroborate the consistency of the significant clusters of cortical atrophy patterns and subcortical volume correlations. (2) For the future development of the project, it will be relevant to complement the clinical diagnosis of HD not only with the TFC and UHDRS-TMS performance but also with the Diagnostic Confidence Level (DCL), which is the standard measure used for clinical diagnosis in at-risk individuals and is based solely on the motor evaluation. (3) All HD patients were in a prodromal or early disease stage; therefore, we recommend implementing this analysis in presymptomatic, prodromal, and moderate manifest HD patients to clarify the atrophy-progressive pattern among different clinical disease stages, always considering the potential motion artifacts produced by a hyperkinetic condition present at manifest stages. (4) An extensive neuropsychiatric evaluation was not included. Therefore, we recommend carrying out a neuropsychiatric evaluation during the cognitive assessment session to incorporate it as a possible covariate in further analyses. (5) We also did not consider medication HD-related on cognitive analysis because most of the patients were not under specific treatment with a controlled and stable dosage given the prodromal/early disease stage (see Supplemental Information–Table S4).

Conclusion

In summary, the MoCA test captures not only cognitive status but also structural brain atrophy associated with the HD cognitive impairment; particularly, the atrophy of the caudate and the superior-lateral frontal areas as well as posterior regions such as temporoparietal and lateral occipital areas. This finding may result in a more detailed understanding of the neural bases of the specific cognitive deficits of HD patients.

Overall, the MoCA test could be used as the first approach in clinical practice given its ability for screening the cognitive impairment of HD patients; then, cognitive impairment detected may be addressed with other comprehensive cognitive batteries for further characterization. Moreover, the MoCA test is a suitable screening tool due to its relationship with the brain atrophy pattern exhibited by prodromal/early HD patients, which may help to establish adequate neuropsychological rehabilitation programs or to introduce individualized disease-modifying treatment plans according to the cognitive deficits detected and the stage of the disease.

Supplemental Information