Resistance Training in Patients with Bone Metastasis

Abstract

Bone metastasis (BM) is a complication in advanced cancer. Symptoms are pain, pathological fractures, hypercalcemia or spinal cord compression. Pain is experienced by 60–80% of patients and has a deteriorating effect on activities of daily life (ADL) and quality of life (QoL). Physical activity is an intervention recommended for early and advanced cancer patients. Resistance training (RT) offers different advantages for BM. It can improve muscle strength, bone density and QoL and prevent loss of functional activity. The question remains: how can RT be delivered in BM patients? Different approaches are possible: supervised in-patient and unsupervised training. Loading to BM sites may be avoided or implemented and certain precautions may be given. Various prerequisites are crucial before implementing RT in BM patient cohorts. This chapter will give an overview of the pathophysiology of BM and a description of various assessments. It will elaborate on the feasibility, safety and efficacy of different types of RT. It will investigate prescription details (intensity, exercise tools and additional requirements) to ensure safety.

1. Introduction

Bone metastasis (BM) is a symptom and complication of advanced cancer stages. With solid primary tumors, bone is a common location of metastasis [1]. BM is associated with relevant morbidity and a decline in survival [2]. Various cancer subtypes are inclined to spread into bony structures. Breast cancer (BC) is the most common malignancy in women [3] and prostate cancer (PC) is the most common in men [4]. These both, as well as other cancer types, for example, lung cancer (LC), myeloma and kidney cancer or melanoma, disseminate [5] by entering the circulation system and spreading to the bone marrow [3].

Therapeutical progress led to longer patient survival and shifted cancer diagnosis to chronic disease [6]. It raised the number of patients with long-term cancer [7]. This puts emphasis on health-related quality of life (QoL) during several years of cancer therapy for cancer survivors [8]. After multifaceted treatment, that is surgery, chemotherapy, radiation therapy, targeted therapy, or immune therapy, patients complain of a variety of symptoms such as muscular weakness, musculoskeletal symptoms, fatigue, restriction in range of motion and overall activity. Chemotherapy- and radiation-therapy-induced side effects, for example, polyneuropathy or fibrosis, are prevalent. In conjunction with BM, patients suffer from pain, pathological fractures with different bones, hypercalcemia, spinal cord or other nerve compressions [9], all of which decline physical activity, mobility and QoL. Fracture risk corresponds to higher mortality; loss of muscle mass increases the risk of falls [10], contributing to further fractures.

The therapeutic goal for BM is most often palliative only. It comprises surgery, radiotherapy and systemic therapy (chemotherapy, hormone therapy, immune therapy) and analgesics according to the WHO suggestions (non-opioids or opioids). Bisphosphonates and denosumab are the mainstay of medical treatment [11].

Physical activity is a type of intervention recommended for cancer patients in early as well as advanced stages [3, 12]. It comprises aerobic training, resistance training (RT) and additional individualized therapeutic goals such as balance training or relaxation therapy [13]. Adhering to physical activity gives patients the possibility to counteract therapy-related side effects and to better cope with the entire situation [12]. The goal of RT is to improve pain, muscle strength, functional activity, bone mineral density and overall QoL. During chemo- or radiotherapy, the safety of RT was already established [14]. RT in BM patient cohorts needs further research and an answer how to accomplish implementation into practice. This chapter will deal with RT in BM only, although a mixed-techniques approach is patient-centered and useful in a multiprofessional setting. Additionally, exercise programs for BM patients must be easily accessible. Barrier factors such as motivational lack or hindrances at reimbursement or travel difficulties to supervised facilities must be surmounted [15].

The question remains: how RT should be delivered in BM patients because different approaches, intensities and sophistications are possible. It will elaborate on the feasibility, safety and efficacy of different types of RT. It will investigate prescription details (intensity, exercise tools and additional requirements) to ensure safety.

2. Methods

Although this was planned as a narrative review, a structured literature search was undertaken and is presented here. The keywords “bone metastas*,” “resistance training” and “strength training” were combined with the Boolean operators AND and OR. Time limits were set for publications between 2004 and 2024 to get an insight on how the topic evolved over the last 20 years. Trials on children with cancer were excluded. To have a broad body of literature, no research design limits were set; also, no language restrictions were introduced. For details, compare Figure 1.

Figure 1.

Flow chart.

3. Bone metastasis

3.1 Incidence and types of bone metastasis

After the lungs and the liver, bone is the third most frequent location of cancer metastasis. The two most common types of cancer, that is BC in women and PC in men, tend to develop BM in the long term [8]. BM has a predictive quality for survival, limiting the range after BM detection to 1–4 years, depending on cancer type [16].

In prostate cancer, the incidence of BM is 65–75%, 69% of BM is in the spine, 41% in the pelvis and 25% in long bones such as the femur [11]. In PC, 90% of deceased investigated patients had BM with a 5-year survival of only 5%. Metastatic PC tends to affect the axial skeleton with osteoblastic, that is, bone forming metastasis, but osteolytic or mixed types are also possible [10].

During their course of disease, 70–80% of BC patients develop metastasis. In BC, most BM is of the osteolytic type [1, 3]. Seventy percent of BC patients experience BM and thereof metastatic bone destruction. Breast cancer BM is a sign of an end-stage tumor disease [9]. Both BC and PC have a predominance for BM because of their connection to sexual hormones [10].

In lung cancer, about 20–30% of patients already have developed BM at the time of their diagnosis [2]. About 35–60% of non-small cell lung cancer are diagnosed with BM at some stage of their disease [10]. The median survival of LC patients with BM is about 6–7 months [2].

In pancreatic cancer, which is highly aggressive, the prevalence of BM ranges from 5 to 20%. In this type of cancer, the liver and the peritoneum are the preferred sites of metastasis, with about 45–50% [17].

About 30% of patients with kidney cancer develop BM with an incidence of approximately four skeletal-related events (SRE) per year in this cohort. They mostly belong to the osteolytic type [18].

In an Italian cohort of melanoma patients, 83% presented with BM. The most common site was the spine (90%). Forty seven percent of those patients suffered from at least one SRE. Their mean overall survival after BM detection was less than 11 months [19].

BM can be labeled as osteolytic, osteoblastic or both, depending on cancer type [11]. BC and LC are most followed by osteolytic metastasis whereas PC predominantly leads to sclerotic, that is, osteoblastic metastasis [8]. In PC all different types can appear in one single bone and even one lesion. Hodgkin’ s lymphoma, melanoma or thyroid cancer mostly produce osteolytic lesions [16].

Apart from BM, there is cancer-treatment induced bone loss because of several surgical or systemic hormonal therapies in BC and PC such as androgen-deprivation therapy, orchidectomy or long-term tamoxifen. This reduces both testosterone and estrogen levels and comes with the side effect of a reduction in bone mineral density and an elevated risk for osteoporotic fractures [10, 20, 21]. This is another reason to adopt exercise therapy, especially RT.

3.2 Symptoms

Typical symptoms are pain, pathological fractures, hypercalcemia and spinal cord compression [8] or distal nerve compression. They are labeled skeletal-related events [11]. SREs are a widespread complication of BM. They have a negative impact on patients’ diagnosis of cancer and lead to elevated healthcare costs [22]. If spinal cord compression occurs through in spite of a multi-targeted therapy, only 50% of patients can walk, and it is doubtful if non-ambulatory will ever walk again [23]. SREs diminish QoL and increase mortality. Early diagnosis and intervention can prevent the development of SREs by evaluating biomarkers and correlating factors such as age ˃65 years and advanced disease [10].

The leading symptom is bone pain and is experienced in different degrees by 60–80% of patients. Pain can occur in advance of X-ray or otherwise detected bone changes and both pain and pathological fractures can occur earlier than cancer diagnosis [11]. We differentiate between intermittent or later constant pain. It is usually correlated with movement, cumulating at night or following palpation or touch. It can occur as seizures of “breakthrough pain,” described as acute, piercing and at a high intensity. Interestingly, approximately 25% of patients do not report pain. Pain has certainly a deteriorating effect on activities of daily life (ADL) and QoL [11].

Hypercalcemia, i.e. calcium levels above normal in cancer patients can occur through abnormal bone resorption or elevated intestinal absorption or renal excretion. Approximately 30% of patients develop hypercalcemia, which ranges from mild cases to life-threatening conditions. Patients present with renal symptoms (polyuria, increase of creatinine), constipation, nausea and vomiting. They also show apathy, fatigue and bone pain, and finally, ventricular arrhythmias and coma. The patient’s increasing immobility, coming with the loss of strength and activity, contributes to calcium resorption from the bone and the state of hypercalcemia. Depending on the cause, there are different pathways of therapy, e.g. bisphosphonates, intra-venous rehydration or calcitonin [24].

3.3 Pathophysiology

Under physiological circumstances, bone equilibrium is achieved through regulated new mineralized bone formation and bone resorption, leading to bone homeostasis [1, 2, 16, 25]. The primary tumor first invades the surrounding healthy tissue, new blood vessels are created [2]. After tumor cells have left the primary tumor and entered the circulatory system, they need to survive in this new environment. They achieve self-defense mechanisms by inhibiting the normally ongoing apoptosis [10, 11]. Research showed that e.g. BC tumor cells are capable of changing the normal bone microenvironment by secreting e.g. interleukin-1-β, a proinflammatory cytokine, which can create a so-called pre-metastatic environment [9] and prepare niches [10]. The fact that bone marrow blood vessels are fenestrated facilitates the migration of tumor cells through the vessel into bone marrow, where they remain through mechanical adhesion [16]. The tumor cells generate new blood vessels, ensuring their own supply and start to dysregulate the normal balanced bone remodeling [2]. After settling in the bone marrow, cancer cells can commence a dormancy which can last years to decades [10]. During this dormancy, they are protected against chemotherapy or immune system engagement. They may terminate this dormancy long after diagnosis and tumor treatment and cause a cancer relapse by the release of certain growth factors [10, 16]. The dormant tumor cells are characterized by slow growing but also dying, holding them stable and awaiting a reactivation through the tumor microenvironment. Also, hypoxia is discussed as an underlying factor to awaken dormant tumor cells [25].

Both osteoblasts and osteoclasts dwell in the bone marrow during their maturation process. There, they start to divide and grow, which, in BM, finally leads to bone damage. By releasing certain mediators, e.g. endothelin, tumor cells stimulate osteoblasts to start proliferation. Proliferation is followed by the formation of new bone. These osteoblasts also release RANK-L (receptor activator of nuclear factor kappa B ligand), which stimulates osteoclasts and leads to bone demineralization and bone instability [11]. The interplay of RANK and RANK-L is the dominant driver for both physiologic and pathophysiologic bone remodeling. The binding of RANK to RANK-L starts the cell differentiation and activity of osteoclasts [2]. Cancer cells can diminish the expression of the RANKL antagonist osteoprotegerin [1].

Cancer cells also can combine with macrophages, giving them osteoclastic properties [10]. BC cells can also promote osteolysis through their interplay with osteoblasts [25]. For osteoblastic metastasis, cancer cells induce osteoblastic activation by e.g. fibroblast growth factors and TGF-β [1]. Osteoclasts also release growth factors such as TGF-β, inducing further tumor growth [10]. In both types, the interaction between tumor cells and bone cells can activate a vicious cycle [10, 25]. The process enters this cycle of destruction (osteoclastic vicious cycle) and/or formation of bone (osteoblastic vicious cycle) [10]. Furthermore, it starts metastasis to other organs [9].

3.4 Imaging methods and biomarkers of bone metastasis

Early detection or prevention of BM is crucial for timely treatment and to enhance QoL and overall prognosis [9], but diagnosis on time remains difficult [16]. Although often applied, radiography has poor sensitivity. Also, bone scintigraphy, which has long been used to detect BM has a limited sensitivity and specificity [8]. Predictive quality is dependent on whether there are osteolytic or osteoblastic lesions, e.g. the sensitivity for PET-CT is better for the former [23]. CT can detect both types in the bone marrow even before X-rays are able to depict any destruction. FDG-PET-CT (fluorodeoxyglucose positron emission computed tomography) is highly sensitive and recommended for an overall skeleton assessment [2]. MRI allows for an accurate anatomical demarcation [23] and is very useful in the suspected spinal cord compression [2]. If there is an acute onset of pain both CT or MRI are recommended [5]. Bone densitometry, dual-energy X-ray absorptiometry, is another means for evaluating BM [26]. If lesions are smaller than 10 mm, which is most often the case and lack soft tissue involvement, they are mostly inaccessible for assessment, which excludes many patients with BM from new drugs in trials [27]. Biopsies would produce a definite diagnosis but are often neglected because of the patients’ risks [2].

Several cancer-type specific biomarkers, labeled bone turnover markers (bone resorption markers and bone formation markers, e.g. pyridinium crosslinks or alkaline phosphatase [1]) serve as supporting diagnostic methods [8]. They can identify high-risk patients and predict therapy response. These are protein-fragments which can be detected in both, serum and urine. They are used to predict the metastatic involvement of bony structures [1]. Elevated hypercalcemia serum level is another predictor for BM [9]. Further, a nomogram was developed for pancreatic cancer. Independent factors for prognosis were e.g. age, cancer grading, histological subtype or chemotherapy [28]. Mirels’ score (12 points max.) relies on the anatomic site, the BM size, the radiography assessment, and the pain pattern, and it is a tool to predict pathological fracture risk. A score of ≥7/12 was suggested as a cut-off value for the realization of prophylactic surgery for upper limbs and ≥ 9/12 for lower limbs [29].

Assessment methods for therapy response in BM are rather difficult. Healing only becomes apparent after a minimum of 3–6 months and is challenging to evaluate. It is further doubtful in the predominance of lytic BM [27].

3.5 Treatment

Therapy of BM is palliative, not curative [2]. Several approaches are used, which will be described in the following paragraphs. BM patients need an interprofessional team consisting of an oncologist, radiotherapist, orthopedic surgeon [10, 30], general practitioner, specialized physiotherapist and exercise physiologist [7, 30]. The latter two are the most gifted professionals to prescribe and monitor the training if they have university degrees and additional cancer expertise. At the start of their career, they should be supervised by an experienced colleague [31]. Patients want to exercise at public gyms or sports facilities, and they want to be supervised by physiotherapists but clearly need more and deeper instructions on safe therapy execution. They take part mostly because they hope to improve their pain and enhance or maintain their strength [15].

3.5.1 Pharmacological treatment

Bisphosphonates, i.e. Zolendronic acid, are bone-modifying agents and the mainstay of treatment. They are commonly used to prevent bone resorption in cancer as well as in osteoporosis. They have an apoptotic effect on osteoclasts and provide rapid pain relief and a reduction of SREs [11]. Only 50% of the dose is taken up by the bone, the rest metabolized by the kidneys [1]. Additionally, they provide an anti-tumor effect by e.g. hindering tumor-associated angiogenesis [32]. Bisphosphonates are indicated in lytic and blastic, i.e. sclerotic lesions [2].

Denosumab is a human monoclonal antibody. It effectuates RANKL inhibition, therefore preventing osteoclast maturation [1] and declining osteoclastic bone resorption. This is associated with an enhanced bone mineral density which also improves pain. Both substances mentioned can cause jaw osteonecrosis as a possible side effect if taken in the long term. This is further associated with poor oral hygiene or tooth surgery [10]. Leading symptoms of this osteonecrosis are pain, bone destruction, infections and compromised healing of the maxilla and the mandible [27]. The incidence is rather low (2% in Bisphosphonates or Denosumab) and is influenced by drug dose and cancer type. Because of the tremendous loss of QoL, physicians should be aware of this side effect and take evidence-based decisions [33]. More common, but not as grave side effects of both are nausea, fatigue and diarrhea [1].

3.5.2 Surgery

Surgery is adequate for solitary and small lesions. Accessibility depends on the location. Nailing, plates and locking screws applied by minimally invasive approaches, bone-cemented techniques or endoprosthesis are applicable [23]. If possible, surgery should be implemented as a prophylactic measure for BM. It is certainly indicated in persistent pain after radiotherapy, with the BM being ≥30 mm, if ≥50% of the cortical structure of a long bone is destroyed [27] and for impending fractures. BM surgery is correlated with poor survival with a median of 3–5 months. It is argued by preserving continence and a better functional ability [2].

3.5.3 Radiotherapy

Radiotherapy can provide sustained pain relief. So-called uncomplicated BM with pain but without fracture or spinal cord compression of every size is the target. The standard dose is a single fraction of 8 Gy [5]. Palliative radiotherapy improves QoL and lowers the incidence of SREs [10] with a positive response rate of about 70–80% and a complete alleviation of pain in about 30% of patients [27].

4. Resistance training

Physical activity is a means to first reduce the risk of developing cancer [34] and second to treat and mitigate various complaints of cancer patients and therapy-induced side effects [35] in their early and late stages. It also contributes to a certain percentage of protection against cancer relapse.

In the past, the medical staff was concerned about prescribing exercise for cancer patients with BM because of fear of bony and neurological complications. Current research has overcome some of those hesitations. Considering various safety precautions, exercise therapy has been established as an important part of the therapeutic regime for BM patients [36]. Health care staff can use e.g. the “International Exercise Guidelines for Cancer Survivors” [37] as a preliminary source, but postural alignment, controlled movements and correct technique need special attention on top of knowledge and consideration of individual bone lesions [30, 31]. Already in 2002, Courneya demanded to incorporate RT in cancer exercise programs, stating that modifications are necessary with comorbidities such as fatigue, lymphedema or metastasis [38]. Bone and muscle loss often take place simultaneously, especially in this cohort. Because of their anatomical proximity and their mechanical connection, exhibited muscle force can trigger anabolic activity in adjoining bones [39], which can be interpreted as a reason for RT.

Patients need evidence-based expertise on how to implement physical activity despite their BM diagnosis. RT offers different advantages for BM patients. RT has the potential to improve muscle strength, bone density and QoL and can prevent additional loss of functional activity [36]. The target group is very heterogeneous, so the diagnosis BM must be divided into subgroups of patients by using radiological and clinical assessments to treat them accordingly and safe. If patients with “unstable” lesions or patients with “moderate to severe bone pain” are excluded from trials and/or training prescriptions, the question arises how to handle them at all. There certainly are functional activities exhibiting similar or even greater forces than a controlled exercise prescription would do, and pain is one of the chief complaints.

Overall, a lack of advice on physical activity from oncologists could be observed, although patients are seeking advice and see their oncologist as the main source to offer it. Nonetheless, results from a survey showed that 89% of oncologic health care providers thought physical activity to be important and 82% to be safe in BM patients. On the other hand, only 43% were confident in their own prescriptions, and 85% wanted further education for doing so or expressed their need for explicit exercise guidelines to follow in the case. Another 57% would not have referred such patients to physiotherapy [7, 30]. These results show a consensus for the need how to firstly treat and guide BM patients and secondly how to meet health providers educational desires. Only recently an exercise guideline for cancer patients with BM was published [31] and awaits to be implemented in daily practice.

4.1 Exercise testing and clinical assessments

An exercise testing in advance of the training is recommended to assess for the possibility of SREs due to exercise, but this risk should not be the reason to restrain completely from exercise. Forty four percent of respondents in a survey answered they would avoid any strength testing, which places stress on the BM, another 37% would use caution therefore [31]. Several factors contribute to a patient’s personal risk: the underlying primary tumor and its prognosis, the treatment prescribed, the lesion itself and personal variables. So, exercise testing in this cohort may be neglected at all or at least the test protocol needs modification.

In detail, the health care team must gather the following information: a recent bone scan giving type, size, number and location of lesions, and a CT or MRI for structural bone quality. Furthermore, medical treatments, the use of bone stabilizing drugs or other medication, which may impair bone stability (e.g. Cortison), osteoporosis, fatigue or cachexia as a comorbidity, a diligent pain assessment (intensity, site, medication, triggers), gait/balance disturbances or recent falls, neurological symptoms, impairment of ADLs and cognitive functions and whether there is an exercise history [30, 31]. All this information can help to assess the likelihood of SREs [30]. Concerning spine BM, the Spinal Instability Neoplastic Score (SINS), based on six criteria: location, pain, type of lesion, spinal alignment, presence of vertebral compression fractures, posterolateral element involvement was suggested showing the vertebrae as “stable = 0–6,” “potentially unstable = 7–12” or “unstable = 13–18” [40].

For defining the correct intensity in every RT, exercise testing is useful [41]. Several methods have been described in the past. The one-repetition maximum test (1-RM) is defined as the maximum load a patient can manage during a specific kind of exercise just once and is frequently used in exercise testing [42]. The advantage of 1-RM is an accurate, reliable and valid testing of strength in practice settings [14]. But the average force patients muster is positively correlated to the load charged [43]. The disadvantage is obvious, as it poses several risks for different kinds of patients, e.g. with circulatory diseases or in this case patients with BM. Although some trials and protocols implemented 1-RM tests to define the individual protocol [44, 45], although eventually modified and limb-restricted [46], we should not cut out possible adverse events in weak bony structures. The second method is a multiple-repetition maximum test or submaximal strength test, where a minor load is given to the patient, followed by a calculation of the 1-RM. This is also known as the strength pyramid. Additionally, submaximal strength tests reflect the ability of ADLs better, which may be of relevance in this group.

A case report related to the 1-RM. However, non-metastatic BC for inclusion criterion, no hormone therapy and a specific warm-up, a crackling followed by pain occurred at the leg press. Ongoing assessments revealed a cover-plate compression fracture. This happened, although the patient was instructed in advance. The authors concluded that given a possible pre-exercise osteoporosis, a change to an x-RM test would enhance patient safety [47]. As we can never be sure if patients understand and process instructions correctly, an x-RM seems the right assessment. In patients with BM the risk of an SRE is unproportional to the information gathered by the 1-RM.

4.2 Intensity

The intensity is closely related to the energy the patients need while performing the training. Not only in cancer patients, RT should be preceded by a suitable warm-up (e.g. light aerobic exercise) and followed by a cool-down (e.g. stretching) [41, 48]. Exercise starts with an education on ADL and active movements without any load, progress is only slowly. Jerky, rapid and loaded full range of motion movements must be avoided [31].

Training begins with exercises and muscle groups remote to bone lesions but can move on to lesion sites under consideration of the precautions to be mentioned. As also certain ADLs put considerable load on predisposed bones, patients need education and guidance on these as well. The RT should ideally incorporate major muscle groups. Patients should classify the training intensity as moderate [41], which would be “3” using the BORG scale (0–10) or “11–13” on the 6–20 scale. Adherence rates e.g. in trials were 79 (12 SD) %, accentuating patients’ interest in exercise and its feasibility. For more details, cp. Table 1. Unsupervised settings are conceivable if patients in need are supplied with a potential check-up for re-confirmation [49]. Of course, more resources (staff and infrastructure) are necessary when considering supervision.

Modality	Weekly planning	Prerequisites	Location	Muscle activation
Open/closed kinetic chains	2–3x/week; 8–12 weeks minimum or constantly	Warm up/cool down; diligent instruction, positioning, alignment and body control	Limbs without BM; limbs with stable BM, unstable BM very cautious	Isometric
Resistance bands/ free weights	15–45 min per session, 60 min if combined with aerobic training	Medical team consultation and clearance	Cave to potential muscle overflow; shear and compression stress/strain; movement velocity and joint angle; minimal loading to area in question	Concentric 2 sec per movement
Body weight	4–8 exercises; 2–4 sets plus 1 min rest; 20 or longer sec isometric hold	Oncologist/radiologist clearance, bone scan, CT, MRI, Karnofsky Index, ECOG performance status scale, densitometry, medication, co-morbidities		Eccentric 2 sec per movement
Resistance machines/isokinetics	8–12 repetitions; slow increase of weight	Modified strength test; supervision per physiotherapist, exercise physiologist	Focus on ADL dominance; supine, prone, sitting, standing	Great muscle groups; limbs and trunk

Table 1.

Exercise modalities.

4.3 Muscle activity and positioning

Different approaches are conceivable for BM patients. First, exercises with a “no-load philosophy” for the BM sites, where only other parts of the body are exercised. Second, “isometric exercises” explicit for body regions with BM to also include and offer training feasibility for high-risk patients or prior fractures [50]. Other authors suggest a preliminary isometric trunk muscle activity, followed by concentric and eccentric limb movements [45, 51]. Slower movements, i.e. 2 seconds per one contraction [44], are advantageous for movement control. Because the faster the velocity of a contraction is, the less time is given to the body to counteract the force. The authors of a systematic review concluded with the unexpected ability of patients with BM to perform higher intensities than considered earlier and asked for further research of telerehabilitation in this area [49], therefore propagating an unsupervised approach.

Taking free weights allows for eccentric and concentric contractions at different velocities. An increase in velocity would require an enhancement of the force while using the same load. Also, the joint angle is a factor determining the resistance that is applied to a muscle [52] and needs consideration.

Both, exercises using closed or open kinetic chains, e.g. support against a wall or bridging while supine as well as moving free limbs is possible. The chosen mode of exercise significantly affects biomechanical factors such as kinetics, kinematics and muscle activity. Common exercises such as biceps curls or calf raises can be conducted. Adapted squats and lunges are possible in patients without femoral or pelvis BM involvement and are highly functional. Aside from that, upper limb exercises can train muscle strength in this region, BM absence provided.

As adjunctive tools, resistance rubber bands or dumbbells can be used to alternate resistance, but those bands come with several limitations. Firstly, it is questionable if they can always produce the overload needed for exercise-driven muscle changes [14]. Secondly, a disadvantage may be the necessity to overcome the elastic recoil in the movements’ beginning [52]. Isokinetic training facilities regulate the movement velocity during exercise. The accommodating resistance facilitates a maximal dynamic strain at every joint angle and accommodates the person’s expended strength.

As for the starting position, exercises are possible while supine, prone, sitting or standing, depending on the patient’s abilities, constitution and given contraindications for certain positions. Weaker patients can have problems lying down on or get up from the floor. In this case, a living room couch will be sufficient. Prone position e.g. will be contraindicated with adjoining heart conditions and/or breathing problems. The feasibility of e.g. an all-fours position will depend on potential humerus BM.

4.4 Repetitions, sets and intervals

Overall, 4–8 different exercises with 8–12 repetitions were chosen with 2–4 sets. The duration of the intervention was often set at 12 weeks, once at 8 weeks. Given exercises comprised upper and lower limbs or different parts of these and the trunk. Exercises were individualized according to the patients’ needs and the location of their lesions [44, 45, 51, 53, 54]. With the number of repetitions mentioned, a gain in muscle mass can be expected. The velocity of the concentric or excentric training was low, i.e. 2 seconds per movement to enhance movement control. In a real setting and knowing of ongoing muscle atrophy after terminating exercise and ongoing muscle weakening through therapy-side effects and behavior, a long-term intervention seems appropriate. Interventions must be modified in both directions of intensity if needed according to the disease’s progress and the patient’s constitution.

It has been shown that lower limb muscles exhibit greater muscle endurance than upper limbs. Therefore, the rest interval may depend on the muscle site with longer resting intervals for upper limb exercises. As the goal in these patients is achieving adaptations in muscular endurance, not maximal strength, the resting intervals may be set at approximately 1 minute. In this, oxygen consumption levels, capillary density and mitochondria are increased to enhance muscular endurance [55]. Only Yee et al. [56] reported on the interval between sets, which was fixed at 1 minute.

Depending on status and constitution, the duration of RT may be 15–45 minutes, if it is RT alone and approximately 60 minutes if RT and aerobic training are combined. RT should be performed two to three times a week on non-consecutive days. The WHO’s recommendation to spend at least 150 minutes on moderate physical activity and to work on moderate muscle strengthening involving major muscle groups can be applied in those circumstances.

4.5 Adaptations and precautions

Some researchers recommend supervised in-patient training for security reasons. A meta-analysis recommended supervised and individualized exercise for BM patients [57]. However, during the COVID pandemic arose the need of virtual and unsupervised exercises [31]. A pilot-study subjected 20 patients (BC, PC) with BM to a 3-month supervised RT with a 6 month follow-up assessment. They observed significant improvements in physical function, physical activity and QoL, the latter remained in the follow-up [48].

A combined approach was proposed by another research group. The protocol suggested three clinic-based exercise sessions per 60 min with a supplement of two sessions at home per 15 min from the third week on for 40 PC patients. A similar program was suggested for 40 BC patients with stable spinal BM: spinal isometric training combined with aerobic and flexibility exercises for 12 weeks. The training program included modifications based on BM location. Loads to the lesion were minimized as much as possible through isometric muscle activity after instruction on trunk stabilization, pelvic control and breathing technique [45, 51]. Results are not published yet. Those exercises in different positions (prone, supine, bridge, side-bridge, etc.) were ambitious. Although patients were instructed to pre-activate abdominals it remains questionable if they can perform it correctly, as such exercises are challenging even without BM in e.g. low back pain patients. Teaching patients to pre-activate abdominals does not necessarily include back muscles, but abdominal coactivity can lead to increased forces in extensors and ligaments. The passive system, i.e. ligaments can contribute to instability even at smaller compression loads (< 100 N), which is below several exercises and activities.

Positioning and limbs as a lever are crucial questions because back muscle activity is dependent on where a given load is held during ADLs, that is in front or at sides, with more activity at the latter [58]. The pelvic position can make a difference to enhance or decrease muscle forces and internal loads. It makes sense to ensure that patients know how to stabilize their pelvic region. Adding another 200 N load in front of the body increased the algorithmic calculated stability but resulted in shear and compression forces of approximately 500 and 1200 N at the lumbar spine [59]. So, educating patients on postural alignment and proper instructions is crucial. Knowledge on possible spinal loads and trunk muscle forces must be considered before designing exercise programs for BM patients. A good proportion of biomechanics is required for adequate training regimes and should be part of the desired aforementioned advanced educational programs for health care providers.

A very recent review showed that unsupervised training can also be applied without increasing unwarranted side effects. The authors concluded that both, aerobic and resistance training are safe for BM patients and bone pain is the chief complaint. Medical staff advising patients should be qualified professionals with expertise and special training in treating cancer patients [36]. A systematic review including 17 trials investigated the feasibility, safety and efficacy of different types of exercise in patients with BM (aerobic, resistance and several control interventions such as stretching or breathing therapy). The training was both supervised and unsupervised. The promising conclusion was that exercise was safe and feasible. It need not necessarily be supervised if it is preceded by a supervised instruction by qualified experts. Adverse events occurred, but in a trial with soccer training and were therefore attributed to the rough body contact during sports [49].

Physicians’ clearance should be a prerequisite to guarantee safety [31, 48, 49, 53]. The intensity should be tailored to the patient, low-level resistance is preferable. A sensible decision is the use of several assessments (e.g. Karnofsky performance status ˃70, ECOG performance status scale 0–2/5; cp. Table 1) to establish a minimum physical performance [49]. Karnovsky’s index, or Karnofsky Performance Scale, is a widely used assessment to evaluate patients’ QoL also in oncologic cohorts. It serves for the classification of functional impairments; the lower the score, the worse the QoL (0 = dead, 100 = no complaints, no evidence of disease) [60]. Score 70 accounts as “cares for self; unable to carry on normal activity or to do active work.” ECOG performance status scale differentiates into six grades (0–5; dead-fully active). It is labeled as a standard criterion to measure the impact of certain diseases on ADLs and is widely used in trials [61].

Loading to the area in question, i.e. the BM should be minimalized, overflowing pressures from the limbs to spine BM must be considered. To diminish loading of the affected sites [44], exercise selection is restricted so that lesions are subjected to minimal physical forces only [53]. Compressive loads, shear loads, and peak forces similar to walking should be chosen [46]. Using an orthopedic corset during exercise may improve feasibility and safety for the spine.

Miramini et al. reported on fracture healing by combining a model of a plate-locked tibia with a musculoskeletal model of the leg. Both lateral and vertical components of loads were considered, the latter having several detrimental effects on the healing process. The calculations under partial weight bearing of the leg in question stayed below the threshold of a 10% bone strain [62]. It is to note, that this was a calculation model without taking BM into account. Even more, the assumed atrophy of spine muscles can interrupt the former balance between a specific load and a subsequent following muscle contraction and lead to stability loss of several segments. Considering those arguments, even more caution is required in BM patients.

As compressive strains are comparably lower with lesser percent of the maximal voluntary contraction [63], this is a reason for moderate intensity. It was shown that trunk stability was best when pulling upward. Antagonistically, spinal loads peaked under forces at higher elevations and inclinations downward. Abdominals and local lumbar muscles play an important role in maintaining stability while applying horizontally oriented forces [64]. Agonists and antagonists must interact during contraction to ensure (a) driving the movement and (b) protecting the joint and giving accuracy to the movement in question.

Fifty seven PC BM patients at the spine, femur, ribs, humerus, etc., were included in a 12-week intervention program (combined aerobic, RT and flexibility) for trunk, upper and lower limbs involving large muscle groups. Three sets with 10–12 repetitions were executed (chest press, seated row, leg extension, leg curl, etc.) trying to avoid loading of individual bone lesions. The leg-extension 1-RM was omitted in femoral BM and the chest-press 1-RM in humerus/rib/thoracic spine BM. With an adherence rate of 89% and 49 patients at follow-up, no adverse events were observed and bone pain was not exacerbated [44].

Sprave et al. and presumably later Rosenberger et al. subjected the same 60 patients to free spinal stabilization exercises (“all fours, plank, swimmer, shoulder blade TheraBand”) vs. muscle relaxation. The training started with 2 weeks supervised sessions and commenced with three unsupervised months. Patients were advised to hold each position for 20 sec, and the time was subsequently enhanced, if possible. They tested the maximum plank holding time and the handgrip strength, the former showing significant differences in group comparison. No training-related SREs were reported in either group. The exercises needed modification in different percentages because of patients’ inability to perform them (13, 42, 54 and 25%, respectively). Exercises that required moving to the floor were reported to be difficult to perform but could greatly enhance patients’ self-confidence. They concluded that close supervision is indispensable in the beginning [50, 65].

Resistance training vs. passive therapy was tested in patients with spinal BM receiving radiotherapy and led to significant improvements in bone density after 3 and 6 months. Several pathological fractures occurred but could not be related to the intervention and furthermore also in the control group without a difference to the exercise group [54]. The pain was significantly lower in the intervention group after 3 and 6 months, respectively [66]. As the inclusion criteria, the intervention, and the number of patients are astonishingly similar, albeit with different outcomes (bone density, pain, QoL, fatigue, biomarkers), six papers in total, four in the same year, some sort of salami publication must be suspected in these cases.

In a scoping review, 26 studies were screened, 12 of these included patients with BM, and none reported any SRE. Eight trials measured muscle strength and found positive results. They reported that exercise adherence was comparable in supervised and home-based protocols. Pain is the most common symptom of BM and should therefore be a standard outcome [36].

The existence of motor overflow, that is involuntary muscle contractions in other body parts following a voluntary strain deserves some attention in RT for BM patients. Motor overflow is even more pronounced if the task, i.e. contraction proofs to be difficult or if weakness or fatigue is a co-factor [67]. All limitations apply to this patient cohort. Exercise instructors should bear in mind that the intended exercise may give overflowing strain to the bone lesion in question.

5. Measurements

5.1 Pain

Bone pain can be measured by Functional Assessment of Cancer Therapy Bone Pain Questionnaire (FACT-BP) with 16 items where higher scores depict lesser bone pain and better QoL. Another revised version with 20 items tried to include the potential impact of pain on daily functioning. Also, the Brief Pain Inventory (BPI), which has 11 questions on pain severity, pain site, relief and interferences, can be used as a patient-related outcome. Comparing both reveals that FACT-BP gives more details on function and includes pain-related psycho-social aspects [68]. The Visual Analog Scale can assess pain via a subjectively measured pain level, which should be noted, as well as the analgetic medication dosages taken [65].

5.2 Strength

Manual muscle testing using a 6-point scale (0–5) is commonly used in practice but lacks reliability and differentiation between moderate and normal strength. Furthermore, maximal resistance tests may be contraindicated, as has been shown. Submaximal exercise testing works with percentages of the 1-RM. In this model, eight or 12 repetitions are equivalent to about 80% or 70% of maximal strength, respectively [69]. The gold standard, isokinetic dynamometry, offers better reliability, and both, isometric and isokinetic strength can be measured. It requires of course more time and financial investments. Measuring isokinetic strength is the means to gain insight into the patient’s ability to perform ADLs [70].

The evaluation of muscle strength is possible via a handgrip dynamometer [56]. As poor handgrip strength in cancer patients is correlated with poor QoL and cancer-related fatigue, this may be a feasible and inexpensive test for clinical practice [71]. Three repetitions of the test and a calculated mean of these results is the best method. A drawback in BM patients is not only that it should not be performed with lesions in the ulna and radius, which are rare, but patients may also have problems with handgrip because of chemotherapy-induced polyneuropathy. Bioelectrical impedance can objectively inform of body composition, that is, lean mass. It is used to detect sarcopenia in cancer cohorts; it is non-expensive and quick [72].

5.3 ADL

Additional possible measurements should consider the connection between strength and functional activity. Therefore, assessments such as the 5-chair-rise test for lower limb strength and balance, the Timed-up-and-go test (TUAG) as a prediction for falls, or timed walking tests are conceivable in cancer cohorts. The TUAG can assess mobility and dynamic balance not only in older adults but also in cancer patients with similar cut-off values [73]. Although not tested in a cancer cohort, the 5-chair-rise test could inform on knee extensor strength while putting emphasis on a very functional daily activity and is correlated to mobility and potential falls [74].

The deleterious impact of BM with symptoms such as pain, limited strength and QoL leads to impaired activities and social avoidance. Tools such as the original or revised International Physical Activity Questionnaire (IPAC) can be used complementary. The revised version includes people with physical disabilities better, which may be favorable in BM patients [75, 76].

5.4 QoL

VAS can be used to assess QoL in cancer populations [77] as a stand-alone measure [78]. It provides faster results than questionnaires with good validity and excellent reliability [79]. We should not forget that a QoL assessment may be biased because of other complaints or disease progression. If using QoL questionnaires, e.g. the EORTC QLQ-C30 which is applicable in all cancer patients, a specific look into various domains is recommended.

6. Practical applications

Herman is a 74-year-old cancer patient. In 2015, he was diagnosed with prostate cancer. This was treated successfully by radical prostatectomy. In 2022, after years of unrecognized gastroesophageal reflux disease complaints, he was confronted with his second tumor diagnosis: esophageal cancer. As the tumor was unresectable, the approach was radio-chemotherapy. Despite this, the disease progressed into BM in the thoracic spine. After the prescription of home-based physiotherapy because of his weak condition and a pre-treatment assessment (anamnesis, hospital charts, vital signs, activities of daily life and overall strength), the therapy plan included resistance training, mobilization and walking exercises supplemented by a rollator. RT comprised of elbow extension (supine, shoulder in 90° flexion) with small dumbbells (1 kg) with both arms alternatively with seven repetitions and three series (cp. Figure 2).¹ After teaching him how to stabilize his trunk in an upright sitting position (90° hip and knee flexion, feet placed soundly on the ground) by activating abdominals, he was instructed to strengthen his Latissimus dorsi bilaterally by extending and internal rotating his arms with a rubber band as a resistance tool. He managed to perform five repetitions in three series (cp. Figure 3). Thirdly, in a standing position, he executed high squats while supporting his weight with his arms on a shelf and keeping the trunk straight (four repetitions, three series; approximately 30° hip and knee flexion). Between sets he paused for a minute and sat down after each squat series. The movements were slowly, all concentric muscle contractions were combined with an expiration. The choice of these exercises was guided by their relevance to keep him self-supported for transfers, to facilitate moving in bed and to/from a chair or the toilette and finally to relieve his nursing wife as much as possible. Herman was treated weekly, ten times, for 45 minutes each session. The strength of his arms was satisfactory; repetitions of elbow extension and arm extension could be slowly enhanced to 12 and 8, respectively. Conversely, leg strength declined, and it was decided to suspend the squat exercises. Despite radiation therapy, tumor growth of the spine BM led to the development of an incomplete paraplegia of the lower limbs. Consequently, he could not stand any more nor move his legs actively while sitting or lying. After 2 months of home-based physiotherapy, he was admitted to a palliative ward because his wife could not bear the responsibility of home nursing any longer. This was a shared decision made by himself, his wife, his daughter and the physiotherapist.

Figure 2.

Triceps resistance training through elbow extension.

Figure 3.

Latissimus dorsi resistance training through arm extension/internal rotation.

7. Conclusion

A key message is that patients with BM should regularly engage in exercise, and RT can also be recommended. To achieve this, they need the support and guidance of health care staff. Before starting the exercise, possible benefits must be counterweighed against theoretical SREs. Exercise prescriptions and RT are safe in individuals with BM, but adaptations to regular RT are needed.

For future trials and for prescribing and implementing safe environments for BM patients several adjustments are necessary. SREs must be a target outcome, and information must be given on whether or not SREs are related to the intervention. Pain as the chief complaint should be an outcome. The effect on outcomes such as pain, pain medication or bone density needs to be proved in future trials because existing studies often lacked power to elaborate this. It is not possible to exclude several confounders e.g. pain medication or radiation therapy, therefore, results need to be interpreted with caution. Bone density as an outcome needs time to change, therefore, longer follow-ups are needed. Specifics on existing BMs must be thorough (location, number, size, status, screening and imaging, treatments). Adaptations to testing and exercise must be patient-centered and individualized. Exercise prescriptions must include the necessary details to copy them for one’s own patients. The graveness of the diagnosis and the disease’s course calls for some “protocol violation,” as exercises, positions and strength training components must be held flexible to meet patients’ needs and to fulfill safety requirements. So, even future trials may be more methodologically pragmatic than explanatory.

For trials, a substantial loss to follow-up must be considered because of the course and severity of the primary disease. To counteract this unavoidable fact, patient numbers must be increased, and trials must be multi-centered to reach sufficient power.