Introduction

Muscle strengthening is among the main pillars of standard of care to prevent falls in older people but its efficacy is limited by multiple challenges inherent to the target population. The current development of drugs intended to increase muscle mass and power offers an unprecedented opportunity to overcome these limitations and further reduce the risk of falls and injuries. This paper first reviews the importance of the unmet need for fall prevention. It then summarizes the published evidence supporting the concept that increasing muscle power should result in a lower risk of falls and related injuries. Then, after describing the efficacy and limitations of the standard of care to prevent falls, the manuscript offers a brief overview of the drugs that are being developed to increase muscle mass and power and could therefore be candidates for development in the indication to prevent falls and related injuries. Finally, it concludes with considerations about the population tailoring strategy for clinical development in such an indication and about some of the risks associated with this development.

Consequences of Falls in Older People

The Burden of Falls in Older People

Falls represent a major public health problem in older people, predominantly due to the resulting injuries which lead to progressive disability, immobilization and associated comorbidities, dependency, institutionalization, and death [1–3].

Thirty percent of community-dwelling persons older than 65 years fall at least once per year, and 15 % fall at least twice per year [4]. This incidence doubles in people beyond 75 years of age, as well as in older people living in residential aged care [2].

In addition to falling frequently, older people also have a high susceptibility to injury due to slowed reflexes, impaired protective responses when falling and comorbidities such as osteoporosis [2]. It is estimated that 20 % of falls result in injuries requiring medical attention [3]. In 2013, in the US, 2.5 million non-fatal falls among older adults were treated in emergency departments and more than 734,000 of these patients were hospitalized [5]. In 2006, injurious falls accounted for 1 in 10 emergency department visits among patients aged 65 years and older in the US [6]. The most common injuries were fractures (41.0 %), superficial contusion/injuries (22.6 %), and open wounds (21.4 %) [6], while other significant consequences include joint dislocations and potentially lethal traumatic brain injuries [3, 7]. In 2013, the direct medical costs of falls, adjusted for inflation in the US were $34 billion [8].

Fear of falling also has a major impact on quality of life in older people [9–12]. Older persons often reduce their physical activity after falls, which leads to physical deconditioning, social isolation, and reduced quality of life [10, 13]. In fact, falls appear to be the largest single cause of restricted activity among older adults, accounting for 18 % of restricted activity days in this group (National Health Interview Survey [2].

Due to all these factors, falls and, to a greater extent, recurrent falls represent a strong predictor of long-term admission to residential aged care facility [14]. Ultimately, unintentional injuries are the 5th leading cause of death in older adults (after cardiovascular disease, cancer, stroke, and pulmonary disorders), and falls constitute two-thirds of these deaths, representing the 6th leading cause of death in the elderly [2]. The death rates from falls among older men and women have risen sharply over the past decade; in 2013, about 25,500 older adults died from unintentional fall injuries in the US [5].

How Important are Falls to Fractures?

A fracture occurs when the energy conveyed to a bone (by a fall or another trauma or activity) exceeds bone strength (Fig. 1). This means that preventing a fracture could focus on increasing bone strength, reducing the risk of falls, or reducing the energy transmitted to bone by falls, or ideally all three of the above. Over the last decades, fracture prevention has focused primarily on increasing bone strength in people with osteoporosis rather than on preventing falls. As pointed out by Jarvinen et al., it is time to “shift the focus in fracture prevention from osteoporosis to falls” [15]. Here are the main reasons:

Fig. 1
figure 1

Respective roles of fall energy and bone weakness contributing to a fracture. A fracture occurs when the load generated by a fall (or another trauma or activity) exceeds bone resistance

Full size image

Osteoporosis drugs prevent only a minority of the non-vertebral fractures (NVFx), even in patients with osteoporosis. In pivotal trials, anti-osteoporosis drugs reduced the risk of NVFx by ~15–50 % (20–30 % for the most widely used bisphosphonates), and vertebral fractures (VFx) by ~35–70 % (~50 % for the most widely used bisphosphonates), although these numbers vary according to the drug considered [16]. This means that 50–85 % of NVFx still occur despite anti-osteoporosis treatment, even in a context of pivotal trials where treatment compliance is much higher than in routine practice [17, 18].

While it was previously considered that bone mineral density (BMD) accounts for the majority of the risk for NVFx [19], evidence collected from large cohorts suggests the opposite, at least above the age of 65 years [20]. In the Study of Osteoporotic Fractures, only 15 % of low trauma NVFx were attributable to osteoporosis as defined by a hip or spine BMD T score ≤−2.5, and only 25 % of NVFx were attributable to a T score below −1.5 [21]. A more recent study performed in 1126 women older than 70 years showed a population attributable risk of NVFx of ~30 % for a T score BMD below −1 [22].

The overwhelming majority of NVFx result from a fall: i.e. 96–100 % of wrist fractures, 95–97 % of humerus fractures, and 92–96 % of hip fractures [23–25]. This very strong link between NVFx and falls is especially important when considering that (1) the vast majority of fractures are peripheral in the elderly, and (2) NVFx have the greatest consequences on health and health care costs. Indeed, almost all NVFx are symptomatic as compared to only 22–33 % of VFx [26–29]. In addition to reducing quality of life for several months, NVFx lead, especially in older people, to immobilization-related comorbidities (thrombo-embolism, pneumonia, urinary sepsis, etc.) which account for the high rate of death after hip fracture especially [30–32]. Hip fractures alone were estimated to cost 20 billion USD in the US in 2003 [33], a figure excluding other costly NVFx such as fractures of the wrist, humerus, ribs, and pelvis [34, 35].

These dramatic figures underline the importance of addressing falls to reduce the risk of fracture and to extend fall and fracture prevention efforts beyond osteoporotic populations.

A fall from standing height should not be considered as a minor trauma since it delivers an amount of energy that substantially exceeds the force required to break a femur or distal radius [23, 36–39]. Despite this, only ~5 % of the falls in older people result in a fracture [4]. It has been postulated that the following four characteristics play a key role in whether a fall will or will not result in a fracture [23]:

  • Location of impact (driven by direction of the fall and sometimes voluntary movements).

  • Energy generated by the fall (combination of mass and velocity at impact).

  • Absorption of energy (protective response, soft tissue over the bone, and ground hardness).

  • Bone strength (geometry, quality, and density).

Because muscle strength is not only an important determinant in the occurrence of falls (hence to the resulting energy) but also of the direction of falls (hence to location of impact) and of the efficacy of protective responses (hence to absorption of energy), increasing muscle strength in older people at high risk may reduce not only the incidence of falls but also the incidence of related injuries including fractures.

Observational Studies: Muscle Weakness is Consistently and Strongly Associated with the Risk of Falls and Related Injuries

A high number of causes and risk factors for falling have been identified. These include older age, history of falls, muscle weakness, dementia, neurologic disorders, visual impairment, medication use (especially sedative drugs and antidepressants), orthostatic hypotension, environmental hazards, fear of falling, gait characteristics (reduced stride length and velocity), and lower limb pain [2, 40–42].

Muscle weakness has been measured by various means. These include measurement of specific muscle groups (i.e., grip, knee extension, or hip abductor muscles) or multiple muscle groups (leg press). Muscle performance has also been assessed under isometric and dynamic (isokinetic) conditions and measured as strength (peak force) or power (rate of force development). Finally, strength and power have been inferred from functional assessments such as repeated chair rise and stair climbing speed.

Moreland et al. [43] conducted a systematic review and meta-analysis of studies evaluating muscle weakness as a potential risk factor for falls (studies published in up to 2002). For lower extremity weakness, the combined OR was 1.76 (95 % confidence interval 1.31–2.37) for any fall and 3.06 (95 % CI 1.86–5.04) for recurrent falls. For upper extremity weakness, the combined OR was 1.53 (95 % CI 1.01–2.32) for any fall and 1.41 (95 % CI 1.25–1.59) for recurrent falls [43]. Since then, other studies have confirmed the importance of muscle weakness as one of the main risk factors for falls [2, 44–53]. This consistent association between muscle weakness and the risk of falls is demonstrated not only for single and recurrent falls, but also for injurious falls and for fall-related fractures [23, 54–63]. Falls are also predicted by low muscle or lean mass [64, 65] but the relationship between fractures and muscle/lean mass is less firmly established [66–68].

Muscle Strength and Power Play Central Roles in Balance Recovery

In order to maintain balance when it is threatened or simply during activities of daily living, we assess our position relative to the environment through the sensory (afferent) system (visual, vestibular, and proprioceptive functions) and adjust our body position and muscle tone accordingly with the efferent system (Fig. 2) [45, 69, 70]. The efferent system comprises the peripheral nerves that link with the muscular system to generate postural adjustments including corrections required after a perturbation to stance or gait [45]. Therefore, although muscle weakness is only one of numerous causes for falls, it plays a crucial role in determining whether or not a fall will occur following a balance perturbation [45, 69, 71]. For instance, muscle weakness can predispose to falls by preventing an effective correcting step when slipping or by impairing foot clearance when negotiating an obstacle [71, 72].

Fig. 2
figure 2

Central role of muscle power in the process of fractures or other injuries. Lack of muscle power increases the risk of injurious falls mainly by impairing: (1) dynamic balance (ability to recover and prevent a fall when balance is threatened), and (2) protective responses intended to minimize and/or avoid injury upon falling

Full size image

An illustration of the critical role of muscle strength and power in balance comes from the very high propensity to falls observed in disorders affecting mostly or exclusively the muscular system such as inclusion body myositis [73], muscular dystrophy [74, 75], and myotonic dystrophy [76, 77].

Various perturbations have been used in biomechanical studies to elucidate balance recovery mechanisms. These perturbations have included trips on obstacles [71], slips on slippery floor surfaces [69], floor translations [78, 79], level changes during walking [80], and tether release from a leaning angle [81–83]. Several of these studies confirmed the important role of lower limb muscle strength and power in the ability to recover successfully from a balance challenge [69, 71, 78, 81, 84, 85]. It has also been shown that trunk muscle strength [86] and arm strength [87–93] play complementary significant roles in balance recovery.

Muscle Power is Particularly Critical to Balance Recovery and Decreases Rapidly with Age

Muscle power is the rate of force development, whereas as strength is the production of a maximal contraction independent of time. Muscle power has been shown to be more important than muscle strength not only for physical function [94, 95] but also for dynamic balance [96], which is not surprising since the time needed to take corrective actions when balance is threatened is critical to the ability to conduct an appropriate corrective strategy, such as a protective step [97].

Lower limb muscle strength and power decrease progressively from early adulthood to end of life [98–101]. There is no commonly accepted definition for “muscle weakness” based on leg strength, but the prevalence of muscle weakness based on hand grip strength (FNIH definition) increases ten-fold between the ages of 60–79 years (2 %) and 80 years and over (19 %) [102]. Causes for decreased muscle strength/power include any cause of muscle atrophy as well as disorders of the peripheral and central nervous systems. Age-associated muscle atrophy is considered as multifactorial, with neurologic (central nervous system atrophy, degeneration of neuromuscular junctions, and motor unit loss/reorganization) [103], endocrine (decreased production of growth hormone and insulin growth factor-1, hypogonadism, and insulin resistance), inflammatory (high level of inflammatory cytokines), nutritional (denutrition), hereditary, and behavioral origins (reduced physical activity) with some comorbidities playing a key role in this process as described later in this paper [104, 105].

Age-associated muscle wasting involves a decrease in the size and number of type I and II fibers, with a differentially higher loss of type II fibers. Type II muscle fibers are fast twitched, mostly anaerobic, and have a larger cross-sectional area than type I fibers which are mostly aerobic and contract more slowly [106]. Therefore, type II fibers are better suited for movements requiring explosive muscle power (such as regaining posture after a balance perturbation, rising from a chair, or climbing steps) while type I fibers are more prominent in muscles involved in endurance. As a result, the age-associated decrease in muscle power is even larger than the decrease in isometric strength, especially between the 7th and 9th decades of life [97, 99, 107–109].

Therefore, while age-associated muscle wasting impacts various aspects of muscle performance, it effects particularly (although not exclusively) muscle power which is key to avoiding fall-related injuries through dynamic balance and protective responses. As a result, when assessing the mechanism of action of potential candidate drugs to be tested for fall prevention, it will be important to determine their respective effects on muscle power versus strength.

Safe Landings: Muscle Strength and Power are Key to the Ability to Develop Protective Responses When Falling

Muscle weakness (of both the lower and upper limbs) is predictive of whether the impact of a fall will result in an injury [110–112]. This association is likely due to weaker people having less efficacious protective responses during a fall. Thus, they are unable to use either their upper limbs or body movements for avoiding direct impact on critical or fragile parts of the body such as the greater trochanter or the head [113, 114]. Such protective responses explain why only 5 % of falls result in a fracture, despite the high amount of energy delivered by a fall from a standing height [2, 3, 23, 37].

Protective responses have been studied in multiple settings that are often specific to one fall direction (forward, sideways, or backward) since this direction determines not only the nature of the resulting injuries but also the strategy developed by older people to avoid them. Breaking falls with movements of the upper limbs is a common strategy irrespective of the fall direction and requires substantial muscle power to be effective in preventing fractures [37, 114–117]. In the Study of Osteoporotic Fractures, it was observed that fallers who broke a fall by hitting or grabbing an object were three times less likely to have a hip fracture compared to those who did not [112]. During sideways falls, a quick rotation of the trunk enables fallers to avoid a direct shock on the hip [37, 79] which is a critical determinant of whether or not a fall will result in a hip fracture [38, 112, 118]. During backward falls, it is suggested that pelvis velocity at impact is substantially reduced by an eccentric contraction of lower limb muscles which absorb the energy of the fall [37, 119] and that a decrease in lower limb muscle strength reduces the effectiveness of this protective response and increases impact severity [119].

Role of Muscle Wasting in the Downward Spiral of Falls and Related Injuries

Older people suffer more serious injuries as a result of a fall [120], and they recover more slowly which results in longer periods of reduced activity compared to younger people [2]. Reduced activity, often aggravated by fear of falling, leads to deconditioning, muscle atrophy, and weakness, which in turn further increases the risk of subsequent falls and injuries [10, 121–123]. This step by step downward spiral (Fig. 3) is well recognized and explains partly why history of falls is a strong independent risk factor for future falls [2, 10, 122, 124]. Besides falls and related injuries, this downward spiral of deconditioning often involves other events such as a sepsis, a cardiac event, or a chronic obstructive pulmonary disease (COPD) exacerbation, which also result in bed rest, decreased physical activity, further disuse atrophy and muscle weakness. Therefore, even if falls constitute only one of these possible events on the path to motor impairment and frailty, they represent not only a cause but also a consequence of muscle atrophy and weakness.

Fig. 3
figure 3

Downward spiral of falls and related injuries. Older people have an increased fall frequency, high susceptibility to injury, and slow recovery. These characteristics contribute to a vicious circle through which injurious fall induce deconditioning and predispose to further falls and associated harm. In addition, fear of falling may arise at various levels of the disablement process and reduce physical activity even in the absence of previous falls or fall injuries

Full size image

In addition, inability to rise from the floor is common in older people [125], contributes to complications after a fall [126], and is a significant risk factor for fall-related injury [127–129]. The contribution of muscle weakness to this inability to get up from the floor [129–131] emphasizes another important objective of muscle strengthening in frail elderly.

Effects of Progressive Resistance Training (PRT) on Balance, Falls, and Injuries

Resistance exercise training is the current standard way to increase muscle strength and power in older people with reduced muscle performance [132]. Its magnitude of efficacy varies from one study to another, with indications that high-velocity PRT is particularly efficacious in improving muscle power [133]. The overall findings, however, are consistently positive [134, 135].

Effects of PRT on Measures of Balance

As of 2008, 29 clinical studies had evaluated the effects of PRT on various measures of balance and were the topic for a detailed Cochrane review [136]. The vast majority of these studies assessed balance as a secondary or exploratory endpoint, with a statistical power that was most often below 20 % according to a posteriori computations [136]. Despite this limitation, PRT significantly improved balance in roughly half of these studies. In addition, the effect was most prominent in the studies that (1) had a higher statistical power, (2) used a higher intensity training program, (3) focused on frail elderly people, and (4) used a dynamic rather than a static balance test [136]. This latter part could be expected since dynamic balance (the ability to execute a response following a balance perturbation) is more demanding in terms of muscle power than static balance. Other studies published since 2008 have provided somewhat heterogeneous results, but the findings of the majority are consistent with those from earlier studies [137–141]. In addition to these effects on various measures of balance, biomechanical studies with older adults have shown that PRT reduces slip propensity and severity [142] and improves the ability to recover from a leaning position [143].

Effects of PRT on the Rate of Falls

More than 150 clinical trials have evaluated exercise training in the prevention of falls [144]. While overall, exercise reduces the risk of falls, the vast majority of trials have included combinations of various exercise types, including not only muscle strengthening procedures but also balance, endurance, and flexibility training. The systematic review by Sherrington et al. concluded that balance training was the most essential component for efficacious exercise fall prevention [145]. This likely indicates that balance training generalizes readily to activities of living skills required by older people. However, the systematic review was based on brief descriptions of exercise content of published trials, and it proved difficult to dissect out balance training from muscle strengthening or other types of exercise in many studies. Eight published studies have investigated the effects of PRT alone (i.e., without any other intervention such as balance training) on the risk of falls in samples with average age of at least 75 years [146–153]. This list excludes studies that have focused on specific diseases (cardiac cachexia, COPD, diabetes, and cancer cachexia) which may not respond homogeneously to such intervention. These trials testing PRT alone were underpowered for falls as an outcome, but have showed trends for a decrease in fall rate in the PRT arm versus control, despite the fact that none of these trials specifically targeted weak subjects. In addition to improving muscle strength and power, PRT has demonstrated benefits for several physiological systems, as well as for mental health [154] and cognition (especially executive functioning) [155]. Such benefits may constitute ancillary mechanisms for preventing falls and improving mobility in older people.

Effects of PRT on the Rate of Fall-Related Injuries

No studies have published findings on the effects of PRT alone on the risk of fall-related injuries. However, a meta-analysis of fall prevention trials using various combinations of exercise showed stronger efficacy on the most severe forms of injuries compared to less severe forms [156]. Specifically, in this analysis, the authors concluded that exercise reduced the risk of all injurious falls by 37 % and of falls requiring medical care by 30 %, compared with 43 % for severe injuries and 61 % for fractures [156]. These results suggest that exercise training reduced not only the risk of falling (by improving dynamic balance) but also the risk of a resulting injury when falling (by improving the efficacy of protective responses). Similarly, with a treatment strategy based upon an increase in muscle power, improving both dynamic balance and protective responses would be expected to have a stronger preventive effect toward more severe falls.

Unmet Need in Fall Prevention: Exercise Training with Muscle Strengthening is One of the Main Pillars of Fall Prevention but it is Rarely Implemented in Routine Practice Due to Multiple Challenges Inherent to the Target Population

Despite the multifactorial nature of falls, the evidence summarized in the first parts of this paper supports the hypothesis that increasing muscle strength and power should reduce the risk of falls and related injuries.

Although a robust demonstration of this concept will only come from a dedicated, appropriately sized, clinical trial, this evidence is considered strong enough that the current standard of care to prevent falls and related injuries in older people at high risk for falls relies essentially on (1) exercise training intended to increase lower limb muscle strength/power and to improve balance, both of which are frequently impaired in elderly fallers, and (2) customized interventions intended to correct individual risk factors according to each patient’s needs: improving vision, decreasing the number of psychotropic drugs, treating orthostatic hypotension, removing home hazards, etc. [157].

Despite the efficacy of these interventions, falls and related injuries remain a major health issue for elderly people, mainly for three reasons. First, although clinical practice varies across countries, use of physical rehabilitation or enrollment in a fall prevention program is rare, even after an ER admission for a fall [158–160]. Multiple reasons have been identified for the very low uptake of fall prevention strategies especially in terms of exercise training: presence of mobility impairment, transportation challenges, recurrent relocations, painful musculoskeletal conditions, unstable comorbidities, denial of fall risk, etc. [161–164]. Second, exercise (group or home-based exercise programs) reduces the rate of falls and the risk of falling by approximately 15–30 % according to published studies [144], which means that 70–85 % of the falls still occur despite such interventions. Third, the long-term sustainability of the benefits of exercise training is unknown.

Some alternatives to exercise training have been suggested as strategies for preventing falls. These include whole body vibration training, electrical stimulation of critical muscle groups, perturbation training, and nutritional supplementation. Systematic review evidence indicates that the evidence for whole body vibration therapy in preventing falls is inconclusive [165]. However, a recent large RCT found low-magnitude high-frequency vibration was effective in improving muscle strength and balance and reducing falls over an 18-month period [166]. There is only preliminary evidence for electrical stimulation as a fall prevention strategy in older people, with pilot studies indicating electrical stimulation can improve measures of muscle mass and mobility [167]. Several small trials have examined the effects of perturbation training on balance recovery and two large trials have demonstrated single session perturbation (slip or trip recovery) training have lasting benefits for fall prevention [168]. Finally, large epidemiological trials have revealed poor nutritional status is a risk factor for immobility and falls [169, 170], but systematic review evidence from three studies indicates nutritional supplementation as a single fall prevention strategy does not reduce the risk of falls [144]. The above strategies present some potential opportunities for fall prevention, but each has limitations regarding contraindications, acceptability, accessibility and reach.

Therefore, a pharmacologic treatment for increasing muscle strength and power may fulfill a significant unmet need in the care of the elderly, by reducing the rate and severity of falls as well as their devastating consequences. A number of molecules are being developed that have the potential to increase muscle strength and power. These include androgens, SARMs, and drugs targeting the myostatin pathway.

Drugs that are Currently Being Developed to Increase Muscle Mass and Strength Could be Tested for Reducing the Risk of Fall-Related Injuries

Testosterone has been shown to increase muscle mass and strength in several clinical trials, including in older men with mobility limitations [171]. However, for safety reasons, its use is likely to remain confined to men with testosterone deficiency. Several selective androgen receptor modulators (SARMs) are being developed to increase muscle strength with the hope that they would not have the same safety liabilities as testosterone. The most advanced molecule in this field is enobosarm, for which there is published phase 2 data showing a statistically significant increase in stair climbing power in a population of patients with cancer [172]. However, this efficacy was not confirmed in phase 3, and it is not yet known whether this latter result was due to the molecule itself or to the fact that the high number of sites in phase 3 led to unacceptably high variability of the outcome measure [173]. Other SARMS have entered clinical stages of development [174]. Finally, a number of drugs are being developed that target the myostatin pathway. These include several anti-myostatin antibodies, one of which has been shown to improve power intensive measures of physical function in 102 elderly weak fallers compared with 99 patients receiving placebo [175], and an antibody directed to the receptor for myostatin/activin which has been shown to increase 6 min walking distance in 11 patients with inclusion body myositis versus 3 placebo patients [176].

While all these molecules may have the potential to reduce the risk of falls and related injuries, they would have to fulfill two pre-requisites: (1) to have a clean safety profile since the target population is large and has multiple comorbidities, and (2) to increase muscle power sufficiently for the effect on falls to be clinically meaningful and to be detected in a reasonably sized clinical trial.

What Could Make Muscle Strengthening Fail to Reduce the Risk of Falls and Related Injuries?

Falls are Too Multifactorial

Falls are so multifactorial in nature that the effect size of a muscle drug (single intervention) might be too small to be clinically meaningful and/or to be detectable in a clinical trial. For instance, it is unclear whether increasing muscle power would reduce the risk of falling in a severely demented older person whose falls are driven by hazardous behaviors. Therefore, for such a pharmacologic treatment to work on falls, the population should be carefully tailored to the expected mechanism of action. This means not only including older people with muscle weakness, but also excluding patients with major confounders, i.e., those with major causes for falls independent of muscle weakness (end stage Parkinson disease, dementia, etc.). The feasibility of recruiting such a population has been demonstrated by a clinical trial testing an anti-myostatin monoclonal antibody versus placebo in older weak fallers [175].

Another design feature that could help reduce the influence of confounders is the choice of the outcome measure for falls. The Kellogg International Working Group on the Prevention of Falls in the Elderly defined a fall as “unintentionally coming to the ground or some lower level and other than as a consequence of sustaining a violent blow, loss of consciousness, sudden onset of paralysis as in stroke or an epileptic seizure” [177]. By excluding falls due to major intrinsic events or overwhelming environmental hazards, it allows to focus on the type of falls that is amenable to interventions such as those improving muscle power and dynamic balance. The PROFANE definition [178], on the other hand, includes falls from all causes, and is therefore more appropriate for other situations such as studies that are not focused on a specific mechanism of falling, or when details of falls are unrecorded (routine surveillance data/accident records), or where a high proportion of subjects cannot provide reliable information about their falls (i.e., those with delirium and cognitive impairment).

Also single interventions, and in particular exercise interventions, have been shown to be as effective as multifactorial interventions in preventing falls in older people [179]. Further, this would not be the first multifactorial event that could be prevented by a drug targeting only one out of multiple risk factors. Another classical example is the reduction of risk of major cardiovascular events (MACE) with cholesterol lowering drugs including statins. Risk factors for MACE include not only high LDL cholesterol but also low HDL cholesterol, high blood pressure, smoking, heredity, diabetes, high C-reactive protein, etc., and many patients have multiple risk factors. Despite this multifactorial nature, it has been possible to demonstrate repeatedly that statins decrease the risk of MACE, both in secondary and in primary prevention, and without even excluding participants with other major MACE risk factors independent of cholesterol levels [180, 181].

Increasing Muscle Strength and/or Power May Increase the Risk of Falling

It could be speculated that a higher muscle power could increase physical activity, which in turn could increase older people’s exposure to the risk of falling. While this possibility cannot be ruled out without an appropriately sized clinical trial, it is important to note that the first assumption enabling this hypothesis is not demonstrated: as of today, we are not aware of any published data showing that an increase in muscle strength or power could result in increased physical activity. In fact, some data suggest the opposite: a clinical trial using progressive resistance exercise led to significant improvements in physical function without any increase in physical activity in functionally limited older adults [182].

Expected Differential Effects on Muscle of PRT Compared to a Drug Increasing Muscle Strength and/or Power

One could fear that a drug increasing muscle mass does not have all the benefits of PRT which is thought to not only increase muscle mass and power but also train a functional skill. It appears that several drugs increasing muscle mass do improve measures of lower limb physical function such as stair climbing performance in the absence of any systematic exercise training [171, 175]. In addition, a drug that is to be administered systemically may increase muscle power in all muscle groups as opposed to PRT which is not feasible in routine for more than a few muscles. Published studies suggest that muscles involved in balance recovery and protective responses are numerous and include not only lower limb muscles such as the quadriceps, hamstring, gluteal medius, gastrocnemius, and ankle dorsiflexors and plantarflexors, but also multiple trunk and upper limb muscles [45, 86, 183]. This difference represents a potential advantage for a muscle drug over PRT for the purpose of preventing falls and related injuries. Another hypothetical issue is that the positive effect of exercise on the risk of falls might result partially from a positive effect on bone strength (through increased loading for instance), which would not necessarily be obtained with a drug increasing muscle power. As of today, the evidence available does not clearly support this hypothesis [184].

Should Exercise Training and/or Nutritional Supplementation be Expected to Work Synergistically with a Drug Increasing Muscle Mass and Power?

Older malnourished people might not have the nutrients necessary to accrue muscle mass hence benefit optimally from a muscle anabolic treatment. Therefore, in such a population, a protein-enriched diet may enhance the effect of such a drug and result in synergy. Whether this would be also true for older people with a normal nutritional status is debatable. While an additive effect is likely, the authors are not aware of published evidence supporting true synergy (i.e., a total muscle mass/power increase that is superior to the sum of the increases induced by the drug and the protein enrichment.).

The scarce information published about the combination of exercise with a muscle anabolic treatment versus none or either of these two does not suggest synergistic effects either [185, 186]. However, this absence of published evidence does not definitely exclude the possibility of a synergy. The trial by Bhasin et al. [185] tested supra-physiologic doses of testosterone versus placebo with or without resistance exercise training in healthy young men. It reported additive effects on muscle mass as well as muscle strength, but did not involve really functional measures such as climbing stairs or rising from a chair. The trial by Hildreth et al. [186] tested testosterone supplementation versus placebo with or without resistance exercise in older men with low-normal testosterone; this one failed to detect any positive effects of testosterone on strength or function (maybe due to the small sample size), so the absence of observed synergy should not necessarily be taken at face value here.

Exercise training is very complex and costly to standardize in late phase drug development, so the decision to include it in a drug trial should rely on solid evidence. In addition, drug developers are facing conflicting needs from various stakeholders including prescribers, regulators, and payers. Comparing an investigational drug (head to head) to standard of care (exercise training) may not be acceptable for regulatory approval due to the impossibility of a fully blinded design. Therefore, and in the absence of any drug approved for fall prevention in older weak fallers, a comparison to placebo is the most logical choice. Should this comparison be performed on background standard of care, i.e., in patients who all perform exercise training? In this case, the trial would not be representative of routine practice since this high risk population hardly performs any fall prevention exercises in the real world. On the other hand, performing such trials on background usual care (i.e., allowing investigators to choose how they want to prevent falls in their patients) may not be acceptable to the stakeholders who want to know whether the investigational drug adds anything to exercise. The ideal solution would be to test both designs, as long as resources are available to do so.

Conclusion

In conclusion, the considerable evidence supporting the hypothesis that a muscle strengthening approach could reduce the risk of fall-related injuries contrasts with the absence of formal demonstration of such an effect.

The ongoing development of drugs expected to increase muscle power offers a new opportunity to address this major unmet need. The intent here is not to replace exercise training with drugs but rather to offer a pharmacologic alternative to older people who are at high risk of fall-related injuries and (1) who cannot perform resistance exercise training due to musculoskeletal pain or other reasons, or (2) who continue falling despite standard of care including resistance training. Potentially, a pharmaceutical could also be offered in combination with balance training or task specific strength training in weak older people at increased fall risk.

Because of the multifactorial nature of falls, the intervention would need to be targeted to the population that would benefit from the posited mechanism of action, i.e., weak patients, excluding those with major causes for falls independent of muscle weakness. Provided such a targeted strategy was followed, the central role of muscle power in dynamic balance and protective responses suggests that a muscle anabolic may indeed address this major unmet need.