1. Exercise and Bone Metastasis
1.1. Safety of Exercise
The National Cancer Institute defines an adverse event (AE) as an “unfavorable and unintended sign (including an abnormal laboratory finding), symptom, or disease temporally associated with the use of a medical treatment or procedure that may or may not be considered related to the medical treatment or procedure”
[49][1]. Based on this definition, AEs in studies testing exercise can be categorized as non-exercise-related AEs, i.e., unrelated to exercise intervention, and exercise-related AEs, i.e., occurred during the exercise sessions. One investigation did not measure safety
[42][2]. Whereas eight trials did not record any AEs during the exercise period
[26,27,28,29,30,31,36,39,43,45,48[3][4][5][6][7][8][9][10][11][12][13][14],
50], two investigations reported exercise-related AEs
[33,34,35,44][15][16][17][18], two described non-exercise-related AEs
[38[19][20],
41], and four recorded both exercise and non-exercise-related AEs
[37,40,46,47][21][22][23][24]. Among them, only two studies reported serious AEs related to exercise training
[33,34,40][15][16][22]. Concerning non-exercise-related AEs, the number of reported side effects appears similar among patients who engage in an exercise intervention compared to the controls
[37,38,40,41][19][20][21][22]. On the other hand, most AEs occurring during the exercise sessions were classified as non-serious, such as fatigue, back pain, dizziness, and muscle strain, whereas only two studies have associated exercise with serious SREs
[33,34,35,40,41,44][15][16][17][18][20][22]. For instance, Uth and colleagues, in their trial which tested soccer in a sample of 57 patients with advanced prostate cancer (19.3% of them with bone metastasis), reported two fibula fractures and one partial rupture of the Achilles tendon during the training
[33,34,35][15][16][17]. Nevertheless, whether these AEs occurred in patients with or without bone metastases is unclear. A second investigation on 214 men with prostate cancer (19% with bone metastases) described two ruptures of the Achilles tendon associated with exercise
[40][22]. In this case, however, a sub-analysis revealed that those side effects occurred in patients without skeletal metastasis, thus excluding the possible association with bone disease
[46][23]. Of note, focusing on the type of exercise, no serious SREs were observed in trials investigating resistance training as part of the exercise sessions.
Overall, the available data support the safety profile of exercise in patients with cancer affected by bone metastases, even in those interventions which included resistance training, an activity traditionally considered at high risk for fracture. However, some considerations are mandatory. Firstly, patients included in the current investigations might be highly selected and, thus, not fully representative of the entire cancer population with bone metastases. In this sense, most studies are addressed to patients with prostate and breast cancers, whereas limited or no information for other cancer types, such as lung or kidney, is available. Additionally, inclusion criteria for selecting patients with bone metastases rarely report detailed information and often exclude the frailest patients, such as those with bone pain or unstable metastases
[26,27,33,39,44,45][3][4][10][12][15][18]. Secondly, the adopted criteria for monitoring and reporting AEs are sometimes not specified or heterogeneous across the studies. The introduction of a standardized classification may help to improve the accuracy of AE monitoring, which is fundamental to adequately assess safety, while preserving patient’s safety within a clinical trial. Future investigations should address these gaps in order to definitely consolidate the safety profile of exercise in patients with cancer affected by metastatic bone disease.
1.2. Effect of Exercise on Bone Health
In healthy subjects, physical exercise is a recognized lifestyle component able to maximize bone development, and improve and preserve bone health across the lifespan
[51,52][25][26]. On the other hand, whether or not exercise may harbor the same benefits in patients with bone metastasis is still a significant subject of debate.
The available studies in this setting show mixed results
[29,34,35,41,46,47][6][16][17][20][23][24].
Bjerre and colleagues found no significant differences in total hip and spine bone mineral density after 6 months of soccer training in 41 patients affected by prostate cancers with skeletal metastases
[46][23]. Another similar investigation has explored bone adaptation to soccer training in 57 patients with advanced or metastatic prostate cancer (19.3% of them with bone metastases). Whereas post-intervention evaluations did not detect improvements in total body and leg bone mineral density, the bone mineral content of the leg (mean difference 13.8 g, 95% CI: 7.0 to 20.5 g) and total (mean difference 26.4 g, 95% CI: 5.8 to 46.9 g) statistically increased in the experimental group compared to the controls, thus suggesting a possible response in bone tissue after an exercise intervention
[34][16]. Beyond the systemic impact on bone quality, exercise may directly affect the bone lesion, potentially contributing to its remineralization. In this sense, a randomized controlled trial testing isometric resistance exercise did not show significant differences in the density of the metastatic bone or pathological fracture rate in 60 patients affected by unstable spinal metastases and undergoing palliative radiotherapy. However, this entry is characterized by a short survival in both groups (mean 4.4 months), leading to a high dropout rate (73% in the experimental group and 63% in the controls), which makes it difficult to know if the lack of results in bone outcomes are attributable to the small sample size or to exercise ineffectiveness
[43][11]. Another similar investigation has compared the effect of exercise on metastatic bone density during radiotherapy in patients with stable spinal metastases
[29][6]. Sixty patients were randomized to receive passive muscle therapy (controls), or isometric resistance training performed five days per week over two weeks and then three times per week until six months. Compared to controls that remained stable, the experimental group reported an improvement in bone density in all spine metastases, which significantly increased by 28.3% and 80.3% after three and six months, respectively. A sub-analysis by metastasis types revealed that, while no differences emerged from osteoblastic lesions, osteolytic metastases seemed to benefit more from exercise, increasing their density by about 88.8% and 179.3% after three and six months
[29][6]. Moreover, biochemical evaluations found significant enhancements in bone turnover markers, especially pyridinoline and C-terminal cross-linking telopeptide of type I collagen, in the experimental group
[32][27], further strengthening the hypothesis that exercise might be an adjunctive strategy able to produce a synergistic effect on radiotherapy to improve the recalcification of metastases.
1.3. The Overall Effect of Exercise
Across the studies including patients with bone metastases, other outcomes, such as physical function, treatment-related side effects, and quality of life, have been investigated. Most of the investigations reported improvements in cardiorespiratory fitness and muscle strength
[26[3][9][10][12][18][19][21][24],
36,37,38,39,44,45,47], whereas the results on body composition appear more debated
[33,34,36,38,39,44,46,47][9][10][15][16][18][19][23][24]. For instance,
Cormie et al., in a randomized controlled trial in patients with bone metastatic prostate cancer, observed that 12 weeks of resistance training at moderate intensity twice a week was able to improve muscle strength, aerobic capacity, and lean body mass, whereas no effect in fat mass was detected
[26][3]. On the contrary, a similar study combining aerobic and resistance training for three months in patients with metastatic bone disease confirmed a positive increase in strength and cardiorespiratory function but did not find any significant changes in lean and fat mass
[36][9].
Regarding patient-reported outcomes, more than half of the studies did not report improvement in quality of life, distress, and fatigue levels, nor did they report negative effects
[26,36,37[3][9][11][18][19][20][21][22],
38,40,41,43,44], while other investigations suggest a possible positive impact on these outcomes
[27,39,42,45,46][2][4][10][12][23]. Intriguingly, pain level has also been monitored.
Rief and colleagues, in their trial assessing resistance training in patients with spinal bone metastasis, observed that exercise was able to relieve pain levels and reduce the oral morphine dose, as well as the concomitant non-opioid analgesics over six months
[53][28]. Another three-arm randomized controlled trial including 516 patients with mixed cancer types (51.3% with bone metastases) has compared controls (arm 1) versus telerehabilitation (arm 2) (composed of walking-based program and resistance activities) and telerehabilitation plus pharmacological pain management (arm 3). After six months, compared to controls, both interventions exhibited equal effectiveness in improving pain interference (arm 2, −0.4: 95% CI: −0.78 to −0.09; arm 3, −0.4: 95% CI: −0.79 to −0.10) and intensity (arm 2, −0.4: 95% CI: −0.78 to −0.07; arm 3, −0.5: 95% CI: −0.84 to −0.11). Additionally, the total hospital days (335 days for arm 1 vs. 213 days for arm 2 vs. 284 days for arm 3) and the length of stay (7.4 days for arm 1 vs. 3.5 days for arm 2 vs. 5.0 days for arm 3) were lower in experimental groups than the control arm
[42][2]. Since pain is one of the most impactful consequences of bone metastases, seriously affecting patients’ independence and quality of life, exercise may be considered a non-pharmacological adjunctive therapy with a potential analgesic effect in this setting.
2. Exercise and Bone Loss
Different studies have investigated the role of exercise in both patients with non-metastatic disease at high risk of losing bone and in those with a recognized bone fragility condition, i.e., affected by osteopenia or osteoporosis.
2.1. Safety of Exercise
Although most investigations have not assessed the presence or absence of AEs
[54,55,57,58,59,62,63,66,68,70,72][29][30][31][32][33][34][35][36][37][38][39], the reported findings support the safety profile of exercise
[56,61,62,64,65,67,69,71,73,74,75][34][40][41][42][43][44][45][46][47][48][49]. In trials including patients with cancer at high risk of accelerated bone loss, e.g., those undergoing chemotherapy, endocrine therapy, or in postmenopausal status, the majority did not find any serious AEs
[56[34][40][42][43][45][46][48],
62,64,65,69,71,74], while three registered mild side effects
[61,74,75][41][48][49]. For instance,
Nikander and colleagues, in their randomized controlled trial, which consisted of a 12-month exercise intervention involving patients with breast cancer undergoing endocrine therapy, recorded 4 moderate AEs
[61][41]. The reported injuries were related to overuse, such as joint/muscle pain and muscle stiffness. However, these side effects were transient, and patients fully recovered in a few days
[61][41]. Considering the studies including patients with bone health impairments (e.g., osteoporosis or osteopenia), no AEs were registered
[55,67,73][30][44][47]. Notably, no skeletal fractures have occurred neither in interventions involving high-impact training, such as that of
Taaffe et al., which proposed for patients with prostate cancer (50% with osteopenia, 4% with osteoporosis) initiating ADT a six-month supervised aerobic nor during resistance training at high impact
[73][47]. Considered comprehensively, exercise appears safe in this population; however, given the inconsistency in the collection and reporting of the AEs across the investigations, the abovementioned considerations made for the metastatic bone disease are relevant here as well.
2.2. Effect of Exercise on Bone Health
Exercise has been hypothesized as a strategy able to counteract the acceleration of bone loss due to cancer and its treatments. In this sense, a meta-analysis, including 26 randomized controlled trials, has demonstrated that exercise may produce significant improvements in bone-related outcomes, such as whole body, hip, trochanter, and femoral neck bone density among patients with cancer
[25][50]. Analyzing the trials, which included patients at high risk of losing bone, some reported the inability of exercise to preserve bone in patients with cancer
[56,65,66,68,70,71][36][37][38][40][43][46]. On the other hand, different investigations found improvements in bone mineral density among patients at high risk of losing bone tissue, even if considerable heterogeneity regarding the skeletal sites has been observed
[59,60,62,64,69,74,75][33][34][42][45][48][49][51]. For instance, a 12-month randomized controlled trial, including 498 patients with breast cancer treated with chemotherapy and/or radiotherapy and/or undergoing endocrine treatments, has explored the impact of a supervised weekly aerobic or circuit training plus home-based, vigorous-intensity aerobic activity 2–3 times per week on bone tissue. Post-intervention evaluations revealed that compared to usual care, women in premenopausal status who performed the experimental intervention reported preservation in femoral neck bone mineral density (−0.2%, 95% CI: −0.9 to 0.6 vs. −1.4%, 95% CI: −2.1 to 0–07;
p = 0.01), but not in the lumbar spine
[59][33]. On the contrary,
Winters-Stone and colleagues reported that 12 months of combined aerobic and resistance exercise intervention was able to improve lumbar spine body mass density (0.41 vs. −2.27;
p = < 0.01), but not that of the femoral neck (−1.37 vs. −2.06;
p = 0.27) in postmenopausal patients with breast cancer
[60][51]. Focusing on studies that included patients with cancer and a diagnosed osteopenia or osteoporosis condition, only one investigation did not report improvements in terms of bone outcomes
[67][44]. A 6-month exercise intervention, composed of a supervised and home-based aerobic training program performed 5 days per week, was shown to maintain bone mineral density in 75 postmenopausal women with breast cancer, 11% affected by osteopenia
[55][30]. Another trial investigating 12 months of supervised and unsupervised strength training twice a week did not produce significant effects in terms of the bone mineral density of the spine and hip. However, the subtle changes in bone tissue were sufficient to produce a shift in the distribution of bone categories favoring the experimental group over the controls: a major number of women allocated in the usual care group became osteopenic at the spine compared to patients who performed the exercise program
[63][35]. However, two main factors seem to influence the effectiveness of exercise in bone enhancement: adherence to exercise training and the timing of starting the exercise program with respect to endocrine therapy. For instance, a randomized controlled trial has investigated the effect of 24-month strength training on bone mineral density, in addition to calcium, vitamin D, and risedronate, in 249 patients with breast cancer affected by osteoporosis or osteopenia. The intention-to-treat analysis did not find significant differences in bone health improvement compared to controls that received medication alone. Per-protocol analysis revealed that those patients who attended at least ≥50% of the exercise sessions were less likely to lose bone than controls. In particular, in this subgroup of subjects, only 1.2% and 12.3% lost total hip and femoral neck bone mineral density, respectively, in contrast to controls, in which 8.6% and 26.7% reported a decrease in bone in the same skeletal sites
[58][32]. Regarding the optimal timing for exercise initiation according to endocrine therapy, a study involving 104 patients with prostate cancer has explored if it is more efficacious to prevent bone loss using exercise from the start of ADT rather than trying to recover bone health initiating training after 6 months of endocrine therapy
[73][47]. In this sense, a group was allocated to an immediate six-month supervised aerobic and resistance training, while the other was assigned to usual care followed by six months of the same training. Although total hip and whole-body bone mineral density declined similarly between the 2 groups, the spine bone mineral density was largely preserved in patients who engaged early in exercise (−0.4% vs. −1.6%), thus suggesting that exercising since the time of treatment may be more efficacious to prevent or attenuate the development of treatment-related side effects
[73][47].
2.3. The Overall Effect of Exercise
Beyond the impact on bone health status, exercise may confer several other benefits to patients with cancer in this setting. Although not all the studies reported positive results on other outcomes, and most found no changes or even an increase in fat tissue
[56[36][40][41][51],
60,61,66], exercise may improve physical parameters, as well as patients’ psychological status and quality of life
[62,65,66,71,72][34][36][39][43][46].
Cormie and colleagues proposed a supervised combined exercise program involving aerobic and strength sessions for 63 patients with prostate cancer scheduled to undergo ADT. After three months, compared to men allocated in the controls, those in the experimental arm experienced significant preservation in appendicular lean mass (mean difference 0.4, CI. 0.1 to 0.7,
p = 0.01), a decrease in fat mass (mean difference −1.4, CI: −2.3 to −0.6,
p = 0.001), and an increase in cardiorespiratory fitness (mean difference 1.1, CI: 0.4 to 1.9,
p = 0.004) and strength. Additionally, the exercisers experienced improvements in treatment-related symptoms, fatigue, sexual activity and function, psychological status (distress and depression), and total cholesterol
[65][43]. Similarly, another investigation on 100 patients with breast cancer in postmenopausal status, which tested 16 weeks of aerobic and strength training thrice a week, found similar results, e.g., improvements in cardiorespiratory fitness, muscle strength fatigue, depression, and quality of life
[71][46]. However, most of the data come from studies that excluded patients affected by bone fragility conditions (osteopenia or osteoporosis), thus necessitating an expansion of research on the impact of exercise in these populations in the future.
3. Mechanisms by Which Exercise Improves Bone
Bone is a dynamic tissue that continuously undergoes remodeling throughout life, thanks to the constant activities of renewal and repair
[76][52]. In this sense, bone homeostasis is strictly regulated by the well-balanced actions of osteoclasts, responsible for bone resorption, and osteoblasts involved in the formation of new bone. Whereas these two processes, if stable, guarantee a constant amount of bone, some conditions may impair the regulatory pathways shifting the balance towards an accelerated bone turnover (e.g., osteoporosis, bone metastases) and/or an increase bone production
[76][52]. The main determinant of bone remodeling is represented by the mechanical stress (and, thus, the obtained tissue deformation—strain) induced by the loads carried by the bones. This system, known as “
mechanostat theory”, involves bone cells that, if stimulated above a certain threshold of strain, react to strain, shifting the balance toward an increase in bone formation
[77][53]. On the other hand, if the strain produced is lower than the homeostasis threshold, bone loss occurs
[77][53].
In this context, exercise may produce an adequate load stimulus able to enhance bone formation, with deposition predominating over resorption
[78][54]. However, not all the stimuli generated by exercise are similar and produce the same effects on bone turnover. For instance, activities with low/absent mechanical load, such as swimming and cycling, are unable to generate an adequate signal to shift the balance toward bone formation
[79][55]. On the contrary, weight-bearing training, such as walking, stair climbing, and jogging, has been shown to have a great degree of load and, therefore, a greater capacity to induce osteogenesis. Bone modifications are site-specific and not systemic, in other words, a better anabolic response occurs in those skeletal sites subjected to a greater load
[79][55]. Moreover, evidence states that bone mechanical loading is more effective if dynamic rather than static. In addition, the rate of applied strain affects the osteogenic capacity of exercise, i.e., bone responds better if loads are applied at a high rate
[79][55]. In practice, exercises with high impact, e.g., those which include jumping, should be preferred to build bones, even if safety issues regarding these types of activities should always be kept in mind, especially in frail and elderly populations
[79][55]. Finally, bone cells acquire desensitization to the mechanical loading immediately after a few repetitions; thus, inserting rest periods between exercises is the best way to maximize the anabolic response in bone
[79][55].
From a closer perspective, the load produced by exercise is usually perceived by ion channels, cell adhesion/cytoskeletal molecules, and G protein-related molecules, which are classified as mechanoreceptors in bone cells and translate the mechanical
stimuli into biological signals
[79][55]. Subsequently, a series of biochemical signaling have been identified as potential pathways to propagate the
stimuli within cells and thus activate osteogenesis. In this sense, it has been found that mechanical stimulation activates the prostaglandin G/H synthase (or cyclooxygenase [COX])-prostaglandin E2(PGE2) and nitric oxide (NO) pathways, as well as the OPG/RANKL/RANK signaling pathways which, in turn, have been related to the suppression of bone resorption, and to enhancement in bone formation, thus favoring bone anabolic response
[79][55]. PGE2 has been found to stimulate osteoblasts proliferation and differentiation. The mechanical load may increase osteocyte-derived PGE2 release and the expression of COX-2, the key enzyme involved in PGE2 production
[78][54]. Conversely, NO exhibits dual effects on osteoblasts activity, depending on its concentration. A high dosage of NO induced by cytokine-stimulated cells inhibits bone formation by reducing osteoblasts’ proliferation, enhancing their apoptosis, and increasing the osteoclast-mediated resorption
[80][56]. On the contrary, a low amount of NO, released by mechanically stimulated osteoblasts and osteocytes, has been shown to increase osteoblasts’ proliferation
[80][56]. Exercise may enhance bone by regulating bone morphogenic proteins (BMP). BMP are members of the transforming growth factor beta (TGFβ) superfamily and are directly implied in osteoblastogenesis. The mechanical strain induced by exercise has been shown to upregulate several types of BMP, such as BPM-2, and BMP-7, which enhance the osteoblasts’ differentiation
[81][57]. Moreover, the activity of osteoclasts is highly modulated by the OPG/RANKL/RANK signaling pathways. RANKL is a mediator produced by osteoblasts that can bind RANK, a specific receptor expressed on osteoclast progenitor cells and mature osteoclasts, which in turn enhances the transformation of mononuclear precursors into mature osteoclasts. The OPG, on the other hand, binds RANKL before its interaction with RANK, thus preventing osteoclast differentiation. Exercise acts on this pathway by increasing the level of OPG and reducing the expression of RANKL, finally resulting in an inhibition of osteoclasts’ differentiation and activity
[82][58]. Another pathway triggered by exercise load and suggested as the major contributor to bone cell mechanotransduction is the Wnt signaling pathway
[83][59]. The Wnt pathway modulates the expression of osteoblastic factors which, through the stimulation of the mesenchymal stem cells, promotes the proliferation and differentiation of osteoblast precursors. Moreover, Wnt signaling is also implied in the downregulation of osteoclastic activity and osteoclastogenesis, slowing down bone resorption
[83][59]. The activity of the Wnt signaling is highly modulated by sclerostin, a protein produced by the SOST gene, which inhibits the pathway, thus reducing osteoblastogenesis and bone formation
[83][59]. Mechanical loading can downregulate the sclerostin expression in bone, allowing for the subsequent activation of the Wnt pathway, thereby increasing bone formation and decreasing the resorption through the inhibition of osteoclast activity
[84][60].
In a more indirect manner, exercise may favor bone anabolic response through the modulation of the inflammatory status. Indeed, emerging evidence suggests that inflammation may elicit a direct impact on bone turnover. The effect of inflammatory processes on bone has been described in several chronic inflammatory diseases, such as periodontitis, rheumatoid arthritis, aseptic prosthesis loosening, and chronic obstructive pulmonary disease
[85][61]. Tumor-promoting inflammation is a hallmark of cancer, making cancer a full-fledged inflammatory disease
[86][62]. Although the exact mechanisms by which inflammation may regulate bone remodeling remain to be elucidated, several cytokines and growth factors have been shown to regulate the osteoblasts’ and osteoclasts’ activity
[87][63]. Some inflammatory mediators, including IL-1, IL-6, and IL-11, may act through the OPG/RANKL/RANK pathway, upregulating the RANKL expression and thus stimulating osteoclastogenesis
[88][64], while others, such as TNF-α, may impair bone remodeling through the disruption of osteoblasts’ differentiation
[85][61]. On the other hand, other cytokines, such as IL-4 and IFN-gamma, have demonstrated an inhibitory effect on osteoclasts’ differentiation, even if it is often overshadowed by those which promote osteoclasts’ activation
[85][61].