Background
Monoclonal gammopathy of undetermined significance (MGUS) and smouldering multiple myeloma (SMM) are asymptomatic precursors to the blood cancer multiple myeloma (MM) [
1], and typically arise in older adults, affecting > 4% of adults aged > 50 years in the general population [
2,
3]. The differential diagnosis of MGUS
vs. SMM is defined by < 10% or ≥ 10% cancerous bone marrow plasma cells, and monoclonal (M)-protein in blood < 30 g/L or ≥ 30 g/L, respectively [
4]. A diagnosis of MM is confirmed with MM-associated symptoms and/or a MM-defining event [
4]. The overall risk of disease progression to MM ranges from 1 to 10% per year for MGUS and SMM, respectively [
5,
6]. Despite the risk of progression to MM, treatment with conventional anti-cancer therapies is not advocated for MGUS and SMM [
7]. Instead, MGUS and SMM are managed by ‘watchful waiting’ with routine measurements of blood biomarkers: M-protein and free light chains (FLC) [
8].
Considering the relatively non-invasive and accessible nature of disease biomarkers in blood and the absence of therapeutic intervention, MGUS and SMM represent a relevant clinical model to explore the isolated effects of exercise training on early cancer disease activity in humans. Only one study to date– a case study of one patient– has investigated the effects of exercise training on SMM disease activity [
9]. Across a two-year period of moderate training (10–20 h/week) following SMM diagnosis, M-protein decreased from 32.9 g/L to 25.3 g/L (− 23%) [
9]. Subsequently, a three-year period of supervised multi-modal training was performed for 12–20 h/week, and M-protein decreased further from 25.3 g/L to 18.4 g/L (− 27%) [
9]. The patient was aged 44 years– which is relatively young in the context of MGUS and SMM [
3]– and was a former elite handball athlete with a 20-year training history at high volumes (> 30 h/week) prior to diagnosis of SMM [
9]. Thus, the findings of this case study need validating in a demographically broader group of MGUS and SMM patients.
As with many cancers, the risk of developing MM is reduced among people reporting the highest levels of physical activity in the population [
10], however, the mechanisms by which physical activity reduces symptomatic cancer risk remain to be elucidated. Importantly, evidence from epidemiology studies demonstrates that the earliest manifestations of cancer– so called cancer ‘precursors’– do not appear to be prevented in people reporting high levels of physical activity [
11,
12], suggesting that the anti-cancer mechanism(s) of physical activity are induced after early neoplasia (e.g., MGUS) has arisen [
13]. This idea is reinforced by the previously summarised case study where the patient was an elite athlete for 16 years (1985–2001) and completed > 30 h/week training for four years (2001–2004) prior to a diagnosis of SMM in 2005 [
9].
The anti-cancer mechanism(s) of physical activity, which result in delayed tumour outgrowth, remain unknown. It is widely predicted that physical activity prevents cancer outgrowth either by suppressing endogenous systemic factors (e.g., inflammation) that induce cell division/damage/proliferation, or enhancing immune competency, as reviewed recently elsewhere [
13]. For example, one prominent hypothesis proposes that acute bouts of exercise augment cancer immunosurveillance and elimination in peripheral tissuesby natural killer (NK) cells or T cells [
14‐
16]. In the case of MGUS/SMM, this hypothetical exercise-induced immunosurveillance response would take place in the bone marrow where the tumour bulk comprised of long-lived neoplastic plasma cells resides [
17]. This immunosurveillance response to acute exercise is intensity-dependent, with a larger-magnitude mobilisation of leukocytes in response to higher/vigorous intensity exercise compared to lower/moderate intensity exercise [
16,
18].
It is thus of interest to investigate how short-term exercise training–and particularly vigorous intensity aerobic exercise– affects disease activity in MGUS/SMM. To achieve this outcome, the safety and feasibility of vigorous intensity exercise training for patients with MGUS/SMM should be confirmed. Secondly, it is important to determine whether short-term exercise training alters disease activity, as indicated by a prior case study which included high-intensity exercise modalities [
9].
Therefore, the primary aim of this pilot study was to assess the feasibility and safety of aerobic exercise progressing from moderate intensity to vigorous intensity in MGUS/SMM. Specifically, to assess the proportion of exercise sessions attended (i.e., adherence) and the proportion of exercise sessions completed as prescribed (i.e., compliance). The secondary aim was to measure pre- to post-exercise training changes to blood biomarkers of MGUS and SMM disease activity. In addition to assessing the feasibility, safety, and efficacy of vigorous intensity exercise in MGUS and SMM, a separate objective of this pilot study was to evaluate the effects of exercise training on fitness and wellbeing outcomes in MGUS and SMM. Physiological and quality of life benefits of exercise training are apparent in MM [
19,
20] but have not yet been investigated in MGUS and SMM. As such, resistance, balance, and flexibility training were included in the intervention, in line with exercise recommendations for older adults [
21]. Finally, we also assessed the effects of the exercise programme on body composition, blood immunophenotype, and biomarkers of inflammation.
Discussion
The principal finding of this pilot study was that short-term progressive exercise training was feasible and safe for patients with MGUS/SMM who enrolled in this trial and passed screening. Indeed, there were no serious adverse events, retention was high, adherence to the intervention was excellent, and compliance to the exercise prescription was high overall. However, the low rate of participant uptake indicates that a revised trial design should be implemented in a future RCT to attract more participants.
Prevalent barriers of ‘lack of time’ and ‘travel burden’ were reported by patients who declined involvement in the trial. The exercise programme was largely delivered via supervised exercise sessions within the hospital, with the aim of maximising compliance to specific exercise intensity prescriptions and providing instant access to medical assistance if required. However, this design required a total of 35 hospital visits which reduced accessibility of the trial where the catchment area of the hospital extends > 30 miles. In a future RCT, virtually-supervised exercise sessions could be trialled to improve uptake, as such an approach has been shown to be feasible in patients with MM, a more advanced disease stage than MGUS/SMM [
36,
37]. Such modifications to the trial design warrant further evaluations of feasibility and safety across a broader range of MGUS/SMM patients.
The exercise protocol used herein was purposefully designed to evaluate the feasibility of vigorous intensity aerobic exercise in MGUS/SMM. This is firstly because a prior case study linking reductions to SMM disease activity with exercise training included high-intensity exercise modalities [
9], but principally because it is commonly proposed that acute bouts of exercise augment cancer immunosurveillance and elimination in peripheral tissues, in an intensity-dependent manner [
14‐
16]. We found that vigorous intensity aerobic exercise was feasible in the MGUS/SMM patients enrolled in this study, but only at an intensity marginally above moderate intensity. Indeed, when the exercise prescription progressed from moderate to vigorous intensity (at 60–70% V̇O
2PEAK), the vast majority (> 80%) of treadmill walks were performed at the target (vigorous) intensity. However, our investigation identified that 70% V̇O
2PEAK may represent a maximum tolerated intensity when performed for three, 8-min intervals in MGUS/SMM, as compliance to 70–80% V̇O
2PEAK was considerably lower (~ 60%) and thus deemed infeasible at that intensity. However, despite poorer compliance at 70–80% V̇O
2PEAK, the average intensity achieved for all participants at this prescription was 70% V̇O
2PEAK, indicating that vigorous intensity exercise was still performed. Thus, overall, participants performed vigorous intensity exercise twice-weekly for ten weeks of the intervention.
A key secondary finding of this pilot study was that short-term progressive exercise training did not induce statistically significant or clinically meaningful changes (i.e., reductions) to biomarkers of MGUS and SMM disease activity. Indeed, changes to disease biomarkers did not exceed physiological variation observed during monitoring of stable disease [
35] and did not meet the minimum criteria for response to anti-MM therapy of a 25% reduction from baseline [
38]. The lack of change to involved FLC– which have a shorter half-life of 2–6 h compared with 5–23 days for intact immunoglobulins [
39] and thus provide near real-time monitoring of changes to disease activity– from pre- to post-exercise training mitigates any doubt that the absence of a reduction to M-protein following 16 weeks of exercise training is due to its relatively long half-life. Thus, the present study shows that a 16-week period of exercise training neither measurably improves nor worsens MGUS and SMM disease activity. Our findings contradict those of a prior case study whereby a young, former elite athlete diagnosed with SMM had a decline in M-protein of 27% during a supervised, multi-modal exercise training programme, albeit this occurred over a three-year time period [
9].
The findings of this investigation align with those of a recent study in watch-and-wait chronic lymphocytic leukaemia, where there was no reduction to tumour cells reported in those who completed a 12-week high-intensity exercise training programme [
34]. Watch-and-wait chronic lymphocytic leukaemia has similarities to MGUS/SMM, as it represents a treatment-naïve B cell lineage cancer which is clinically monitored via blood testing, and it is a disease that mostly affects older adults. Similar results have also been observed in the related model of active surveillance for localised prostate cancer [
40‐
42]. Indeed, in aerobic exercise intervention trials spanning 12 weeks to 12 months during active surveillance for prostate cancer, prostate-specific antigen (PSA) was unchanged [
42] or reduced modestly (CV 2.9–6.6%) [
40,
41]– within the physiological variation for PSA (CV ~ 7%) [
43]– in the exercise group. The clinically-insignificant reduction to PSA in these studies - and the findings from our study - suggests that existing precursor disease cannot be unilaterally eliminated - e.g., via exercise-induced immunosurveillance - arising from short-term exercise training. Additionally, we found that M-protein and FLC were unchanged by short-term exercise training even in participants with low physical activity levels prior to commencement of this trial. It should be noted, however, that subgroup analysis according to baseline physical activity level was not planned
a priori and includes only a small number of participants within each subgroup.
In this study, we did not observe any changes to blood immunophenotype in the participants enrolled in the exercise intervention. Given that the anti-cancer properties of exercise are likely derived over a longer time horizon, this is perhaps an unsurprising result. Indeed, it has been proposed that exercise elicits anti-cancer effects by preserving immune function (e.g., diversity, persistence) against cancer clones that have mutated to become immunogenic [
13]. It is tempting to suggest that this is demonstrated in Fig.
3, where there appears to be a pattern for lower disease activity in participants with a high physical activity level prior to commencement of the trial.
This study cannot exclude the possibility that exercise augmented immunosurveillance against recently mutated immunogenic cells, which may have been identified in this study were a control group included. For example, in watch-and-wait chronic lymphocytic leukaemia, there was a greater increase in disease activity– measured via enumeration of tumour cells in blood– in a non-exercise control group, compared to those completing an exercise training programme [
34]. This supports the idea that exercise augments anti-tumour responses against immunogenic cell clones, but not clones lacking immunogenicity. However, it does not confirm that this response arises as a result of acute-exercise induced immunosurveillance, and only that exercise training elicited this outcome. Regardless, longitudinal studies are required to evaluate the longer-term effects of regular exercise on disease progression from cancer precursors (e.g., MGUS and SMM) to symptomatic cancer (e.g., MM).
A strength of this study is the breadth and depth of secondary outcome measures. Firstly, we found that body fat percentage estimated by DEXA and BMI were unchanged from pre- to post-exercise training. This likely explains why inflammatory mediators were also unchanged in the present study, as adipose tissue– particularly in older age– is a key contributor to systemic inflammation [
44]. Nevertheless, the largest meta-analysis to date revealed that associations between self-reported physical activity level and clinical cancer risk withstood adjustment for BMI for nearly all cancer sites, including MM [
10]. As such, the lack of change to body composition and associated inflammation herein does not undermine the proposition that MGUS and SMM disease activity cannot be reversed by 16 weeks of exercise, as it is likely that mechanisms independent of obesity are involved in the association between physical activity and cancer risk.
In the present study, we also observed no change to lean soft-tissue mass estimated by DEXA, which may be explained by immune dysfunction in MGUS and SMM. Indeed, the repair of muscle damage induced by exercise, which results in muscle hypertrophy, is dependent on a coordinated immune response [
45] and immunoparesis– an indicator of immunosuppression– was present in 72% of participants in the present trial. Alternatively, a more likely explanation is that the exercise prescription was insufficient to induce muscle hypertrophy. Indeed, despite resistance exercise progressively increasing in load, compliance to repetition-maximum targets was consistently poor in this study. Exercise prescriptions to induce muscle hypertrophy are relevant to MGUS and SMM, as increases to muscle-derived cytokines in physically active individuals may be involved in maintaining immune competence against immunogenic cancer cells to reduce clinical cancer risk [
13]. As such, the lack of change to blood immunophenotype may be due to lack of muscle hypertrophy following the 16-week exercise programme. Future studies should include measurements of immune cells in the tumour microenvironment, which are likely to yield more insightful information regarding the anti-tumour immunological effects of exercise in comparison to measurements of immune cell phenotypes in blood [
13].
With regards to measurements of fitness, the maximum work-rate attained during CPET increased from pre- to post-exercise training, indicating an improvement in efficiency and exercise tolerance. Additionally, we detected improvements in resting heart rate and diastolic blood pressure suggestive of improved resting cardiovascular function following 16 weeks of progressive exercise training. We also observed improvements to functional fitness (e.g., sit-to-stand 30, 8ft up-and-go) from pre- to post-exercise training in the present study. In contrast, we found that V̇O
2PEAK was unchanged from pre- to post-exercise training, which has also been shown previously following exercise training in MM [
19,
20]. Disease pathophysiology of MGUS, SMM, and MM may preclude improvements to V̇O
2PEAK, as the oxygen carrying capacity of blood is a limiting factor [
46] and the expansion of clonal plasma cells compromises the production of erythrocytes in the bone marrow [
47] which may affect oxygen transport. Indeed, erythrocyte count and haemoglobin were unchanged from pre- to post-exercise training in the present study (data not shown). Another potential explanation for the lack of change to V̇O
2PEAK from pre- to post-exercise training is the higher prevalence of sub-maximal tests at follow-up compared to baseline. However, it cannot be discounted that the exercise prescription may have been insufficient to increase V̇O
2PEAK.
Global quality of life scores crossed a threshold from ‘average’ to ‘high’ life satisfaction from pre- to post-exercise training in our study, suggesting a meaningful improvement [
48]. Furthermore, improvements to physical functioning and energy/fatigue domains of health-related quality of life measured via SF-36 occurred from pre- to post-exercise training. However, fatigue measured via FACIT fatigue scale, plus PQSI sleep quality and other domains of SF-36 health-related quality of life, were unchanged. To the authors’ knowledge, this is the first study to report improvements to aspects of quality of life in MGUS/SMM patients following exercise training. This may be of clinical significance, as individuals with MGUS and SMM report similar levels of mental health-related quality of life, psychological distress, and anxiety to MM patients [
49], and quality of life scores predict overall survival in newly diagnosed MM [
50].
It should be considered that despite using a ‘within-patient control’ approach using historical disease activity data, the single-arm design of this pilot study does not capture changes to disease activity that occur– in the absence of an exercise training programme– over a 16-week period in participants during ‘watchful waiting’ usual care. Indeed, as anti-cancer therapy is not advocated in MGUS and SMM [
8] disease activity biomarkers are expected to steadily increase over time. Indeed, it is plausible that M-protein and FLC could increase within a period of 16 weeks, as guidelines recommend monitoring newly-diagnosed SMM after 8–12 weeks initially, increasing to 16–24 weeks if stable [
8]. Furthermore, the single-arm design precludes improvements to general health outcomes being definitively attributed to the exercise programme. In addition, while this pilot study captured a range of participants of different ages and sex– which greatly expands the available evidence from one patient with SMM [
9]– a larger sample is required to include diversity in ethnicity and clinical status. A larger sample size is also required for meaningful interpretation of pre-post statistical analyses, which are underpowered in the present pilot study. Additionally, the exercise programme was designed to incorporate progressively higher intensity exercise, which may not be suitable for all MGUS/SMM patients, as treadmill walking for 30 min at 70–80% V̇O
2PEAK was shown not to be feasible in this study. Lastly, the resistance exercise prescription could be optimised for muscle hypertrophy in conjunction with appropriate nutritional guidance.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.