Neurodegeneration is described as progressive neurological dysfunction and loss of neurons; it is a hallmark of aging and neurological disorders, such as Alzheimer’s disease, multiple sclerosis (MS), and Parkinson’s (Mayne et al., 2020). Neurodegeneration is a complex process, as the exact cause is not entirely understood, and rates vary within and between diseases/disorders (Przedborski et al., 2003). While neurodegeneration is a common hallmark, the resulting physiological and functional consequences differ markedly between diseases, reflecting variation in both rate and severity of decline (Ahmed et al., 2016). For example, neurodegeneration in the peripheral nervous system leads to unstable neuromuscular transmission, resulting in impaired or inconsistent function of the neuromuscular junction and reduced reliability in muscle fiber activation (Hepple and Rice, 2016), while in the central nervous system, it can cause hyperexcitability, characterized by increased spontaneous firing and diminished inhibitory control within motor pathways (Bouteloup et al., 2009). Additionally, neurodegeneration in the hippocampus and motor cortex has been associated with reduced long-term memory (Roux et al., 2021) and slower reaction time (Keleman et al., 2020), respectively. Importantly, cognitive and motor functions, such as working memory, thinking skills, walking, and grasping, arguably represent the most prominent and patient-centered behavioral outcomes in neurodegeneration (Pasko et al., 2022). Exploring such impacted behavioral outcomes in conjunction with underlying physiological mechanisms will promote understanding and, subsequently, refine intervention strategies to help slow neurodegeneration.

Cardiorespiratory fitness (CRF) is associated with a plethora of health benefits, including but not limited to improved heart and respiratory function, enhanced cognitive performance and mental health, and the delayed onset of several chronic diseases, such as osteoarthritis and Type 2 Diabetes (Ruegsegger and Booth, 2018). Previous work has explored CRF as a protective factor in neurodegenerative models. More specifically, studies suggest that maintaining CRF can slow the decline in total gray matter, protect hippocampal subfields, and improve functional connectivity in neurodegeneration (Ahlskog et al., 2011; Bonanni et al., 2022; Bray et al., 2023a; Dougherty et al., 2021). Cross-sectional studies in MS further suggest that higher CRF levels are associated with improved functional outcomes and favorable Transcranial Magnetic Stimulation (TMS) measures (Chaves et al., 2020; Madsen et al., 2019; Zhu et al., 2014). However, other research has shown CRF to be associated with neuropsychiatric burden and have no benefits to cognition in people suffering from neurodegeneration (Ciria et al., 2023; Forbes et al., 2015; Steinberg et al., 2009). Regular participation in physical exercise, which can improve CRF, can facilitate motor performance, particularly in upper extremity tasks for older adults (Hübner and Voelcker-Rehage, 2017; Ploughman et al., 2015). Conversely, other studies report inconsistent changes in motor outcomes with exercise (Bray et al., 2020). Concurrently examining the degree to which CRF may influence physiological processes and behavioral outcomes in neurodegeneration could provide valuable insight into dose-dependent prescriptions, a current goal of exercise physiology research (Barha et al., 2022).

Several cross-sectional studies have investigated the cognitive and motor consequences of CRF in people suffering from neurodegeneration, but we are unaware of any longitudinal work that has done so, with the exception of post and follow-up assessments following an exercise intervention (Chaves et al., 2020). As a result, no study has also looked to understand the underlying physiological changes that may occur concurrently. Therefore, we aimed to investigate whether CRF predicts longitudinal change in behavioral (manual dexterity and cognition) and neurophysiological (corticospinal excitability measured using TMS) outcomes in individuals experiencing neurodegeneration, specifically persons having stable multiple sclerosis (MS). We hypothesized that participants with higher CRF at the first time point would have better physiological and behavioral outcomes at the second, 2 years later.

We aimed to determine whether CRF protects against longitudinal changes in community-dwelling persons living with stable MS. Indeed, higher levels of CRF and engagement in exercise have been linked to lower relapse risk in MS (Proschinger et al., 2022). Importantly, participants who relapsed just prior to (i.e., < 3-months) or during the study period were excluded from our study/analysis. While this may mean our sample represents a somewhat healthier subgroup of people with MS, this design allowed us to specifically examine progression independent of relapse activity (PIRA) (Tur et al., 2023). Taken together, although this approach may limit the generalizability of our findings to the full spectrum of people with MS, it provides a unique opportunity to isolate neurodegenerative changes occurring outside of relapses, which is increasingly recognized as a primary driver of long-term disability accumulation in people with MS. Within this framework, we show that CRF at T1 (baseline) did not predict T2 (follow-up) scores in any behavioral or neurophysiological variable. Instead, with the exception of the number of days between visits for CSP, the T1 score for each outcome was the sole significant predictor of its corresponding T2 score. Importantly, we also observed considerable variability in change scores between T1 and T2, including increases, decreases, and no change, with the most striking being a significant improvement in non-dominant 9HPT time and MoCA score. Together, these results suggest that T1 performance and not CRF predict change in MS, but that MS is not synonymous with continuous and linear decline in behavioral and neurophysiological function.

CRF is widely regarded as neuroprotective in aging and neurodegeneration due to its positive associations with brain structure and function, enhanced cerebral blood flow, and improved metabolic and inflammatory profiles – all of which contribute to maintaining cognitive and motor function (Blackmore et al., 2009; Honea et al., 2009; Smith et al., 2021). However, in our sample, CRF at T1 did not predict hand dexterity (i.e., dominant or non-dominant 9HPT time), global cognition (i.e., MoCA), or corticospinal excitability (i.e., TMS outcomes of AMT, CSP, and asymmetry ratios). We believed CRF would predict and ultimately protect against neurodegeneration-induced decline in behavioral and neurophysiological variables based upon prior cross-sectional studies in healthy and clinical populations demonstrating CRF to be positively associated with such outcomes (Buragadda and Ploughman, 2025; Hübner and Voelcker-Rehage, 2017; Ploughman et al., 2015; Sandroff et al., 2019; Spirduso et al., 1988); meaning, higher CRF has been typically associated with better performance in hand dexterity, global cognition, and corticospinal excitability. The absence of a relationship between T1 CRF and T2 outcomes may reflect sample heterogeneity; that is, variable disease progression, a common occurrence across individuals suffering from MS, likely diluted any consistent predictive signal of CRF. To this end, and as a reminder, some of our participants demonstrated declines in performance (i.e., > –5% change), while others remained stable (i.e., < –5% change) or even improved (i.e., > +5% change). Importantly, heterogeneity in MS progression is well-documented; while some individuals experience rapid declines, others remain stable (Baecher-Allan et al., 2018; Engelhardt et al., 2022) or, at least in our sample, demonstrate modest but statistically significant improvements, as per non-dominant 9HPT (i.e., hand dexterity) and MoCA (i.e., global cognition) score. It is possible that compensatory neuroplasticity, stable disease pathology, and/or ongoing treatment contributed to the observed improvements (Holm et al., 2022; Robertson and Moreo, 2016). Interestingly, some individuals diagnosed with mild cognitive impairment, a prodromal or transitional stage between “normal” aging and a dementia syndrome, also experience an improvement in cognition, although the underlying mechanisms remain unclear (Canevelli et al., 2016).

Variability in disease progression may be attributed to several complex and interconnected factors. Firstly, disease progression reflects a complex interplay between age, genetics, lifestyle, and environmental exposures, among other factors (Van Hijfte et al., 2022; Zeydan and Kantarci, 2020). In addition to MS, Alzheimer’s, cancer, and HIV/AIDS are just some of the other diseases that demonstrate heterogeneity in progression (Bray et al., 2023b; Jamal-Hanjani et al., 2015; Lukashov and Goudsmit, 1998). Second, MS disease progression is characterized by periods of relapse and remission, when individuals experience a significant worsening or improvement in symptoms, lasting for a few days to several weeks or months (Iuliano et al., 2008). Third, various disease-modifying therapies are approved for treatment in MS (Robertson and Moreo, 2016). Of course, all disease-modifying therapies aim to help with MS symptoms, but none ultimately cure the disease. Further, their effects are not uniform and individuals experience a variety of side effects, even when using the same medication (Robertson and Moreo, 2016). Along these lines, individuals may be engaged in several other pharmacological and non-pharmacological interventions, including newer, more effective therapies, all of which could impact disease progression and symptom management (Torkildsen et al., 2016). Finally, one more factor to consider relative to disease variability is whether our participants truly reached V.O₂max or were only able to achieve a V.O₂peak during their CRF testing. V.O₂peak is the highest recorded oxygen consumption during a test, but it does not necessarily reflect a true physiological maximum (i.e., V.O₂max); failure to achieve a true V.O₂max may be attributed to a variety of overlapping factors, one being localized muscle fatigue, which is prominent in MS populations (Bray et al., 2023b; Jamal-Hanjani et al., 2015). Ultimately, our CRF assessments may have captured peak oxygen uptake rather than true V.O₂max, limiting the accuracy of our fitness measurements and predictive ability for some participants (Jamal-Hanjani et al., 2015). In this regard, it is important to highlight that the heterogeneous nature of MS was also reflected in the wide range of CRF levels observed in our sample. In particular, individuals with lower EDSS scores (< 2) displayed fitness levels ranging from very poor to superior (Ferguson, 2014), while those with EDSS scores > 2 generally exhibited a downward trend in fitness with advancing disability.

A discussion surrounding outcome variability is incomplete without considering participants’ baseline or starting point. We found that every outcome’s T2 score was significantly predicted by their T1 score. Such findings suggest that initial levels of impairment, and not CRF, play a critical role in shaping long-term trajectories; this emphasizes the importance of early identification and intervention in persons living with MS as it provides an opportunity to minimize disease progression and maximize health. For example, it is possible that baseline disability level moderates the extent to which individuals benefit from physical activity; at the very least, there may be an interaction. Specifically, those with lower disability (i.e., better function) may have a more intact neurophysiological system, conferring a greater propensity for adaptive neuroplasticity in response to exercise. At the same time, those with lower disability (i.e., better function) are likely capable of engaging in more difficult or demanding physical exercise, thereby inducing greater physiological changes that translate to reduced or slowed disease progression (Ksiazek-Winiarek et al., 2015; Pilutti et al., 2015). A similar perspective exists in other neurodegenerative diseases. For example, mild cognitive impairment is considered a more optimal period for intervention than later disease stages, such as mild or moderate Alzheimer’s (Albert et al., 2011). Similarly, individuals with pre-frailty have more capacity and can therefore engage in more demanding physical exercise than their counterparts with more severe frailty (Bray et al., 2016). Importantly, the early or transitional stages of MS are an active area of research. For example, our group recently explored the relationship between objective and subjective hand dexterity in people living with mild MS (i.e., score of 3 or less on the EDSS). We found that robotic-based measures of hand dexterity, but not the gold-standard 9HPT, differentiated between those that did and did not report subjective hand impairments (Bray et al., 2024).

Despite our findings and the value added to the literature, our study is not without limitations. The time between the T1 and T2 was approximately 2 years, but this varied between participants, largely due to the disruptions caused by the COVID-19 pandemic. Although we controlled for this variability, such delays introduced temporal confounding, which may have made it harder to track consistent change patterns over time. Further, it is possible that a 2-year follow-up was too short to capture meaningful neurodegenerative change or to observe the protective effects of CRF, particularly in a condition as heterogeneous as MS. Another limitation is the wide age range of participants (i.e., 19–65 years). CRF may have had stronger predictive ability in behavioral and physiological outcomes if we focused exclusively on younger or older adults. Importantly, aging is associated with an increased risk of various other diseases and syndromes, such as pain, hearing loss, and polypharmacy, all of which can individually and synergistically have a negative impact on health (Livingston et al., 2024; Maher et al., 2014; Palmer et al., 2023). Along these lines, we previously demonstrated that individuals with MS performed hopping—a muscle power activity—at a level comparable to people 30 years older but without MS (Kirkland et al., 2022); this highlights the phenomenon of accelerated physiological aging in MS and underscores the complexity of interpreting age-related fitness measures in this population. In regard to both aging and MS, previous work has demonstrated that both experiences are sex-specific; unfortunately, our sample size precluded us from individually analyzing males and females. Additionally, Magnetic Resonance Imaging (MRI) data were not included due to variability in clinical imaging schedules, accessibility, and cost. The absence of MRI-based markers limits our ability to directly relate functional and neurophysiological findings to underlying structural disease activity. Finally, our study was geographically limited to Newfoundland and Labrador, Canada, which is known for its high prevalence of chronic disease (Statistics Canada, 2023) and unique genetic mutations (Merner et al., 2008; Rahman et al., 2003). As such, the generalizability of our findings to the broader MS population cannot be assured.

Our longitudinal study in community-dwelling persons living with stable MS demonstrated that CRF does not predict longitudinal changes in behavioral and physiological variables, including hand dexterity, global cognition, and corticospinal excitability. In fact, we found that the strongest predictor of follow-up performance was baseline score and that participants experience heterogenous changes over 2 years, including decline, no change and even improvement. Overall, our study provides important insights into the longitudinal relationship between CRF and MS-related outcomes while highlighting the complexity and variability in MS progression.