Biomarkers of sarcopenia in clinical trials—recommendations from the International Working Group on Sarcopenia

Since the advent of the term “sarcopenia” in 1989, there has been a dramatic increase in publications in this area and clinical interest in this condition [6]. Originally described as the age-related decrease in skeletal muscle mass [7], until very recently, there has been a lack of consensus on the operational definition of sarcopenia without clinically appropriate correlates for this syndrome. In the past 2 years, a number of academic societies have put forward operational definitions of sarcopenia [8‐11]. Although each consensus definition has some distinct features, there is general agreement among these groups on the definition of sarcopenia. A summary of consensus sarcopenia definitions is presented in Table 1. The characteristics of sarcopenia highlighted in these reports include an objective measure of muscle or fat free mass using dual energy X-ray absorptiometry (DXA) or computed tomography (CT), a reliable measure of muscle strength, and/or an objective test of physical functioning. Although the sequence of events and specific recommendations somewhat differ, the general approaches proposed require that patients be identified with measured deficits in physical function for which sarcopenia may be the cause and subsequently quantification of muscle strength and mass to definitively confirm the diagnosis.

A biomarker is defined as “a characteristic that is objectively measured and evaluated as an indicator of normal biological processes, pathogenic processes, or pharmacologic responses to a therapeutic intervention” [12]. Hence, biomarkers support the diagnosis, facilitate the tracking of changes over time, and help clinical and therapeutic decision-making processes. Taking this definition into account, the functional, biological, or imaging-related parameters considered in the present document will be hereby generally referred to with the term “biomarker.”

There are currently numerous parameters that are potentially able to track the age-related skeletal muscle decline. Depending on the parameter chosen to define sarcopenia, different information might be obtained. Such variability depends on the specific characteristics of each parameter and the mechanisms measured by the parameter. The intrinsic (e.g., accuracy, specificity, sensitivity) and extrinsic (e.g., cost, availability, time to be performed) properties of each biomarker will largely drive its use in research trials, making it more suitable for screening, baseline evaluation, and/or definition of outcomes (Table 2).

The use of biomarkers in a given study must be “fit for purpose.” Thus, several different biomarkers may be required to support different aspects of the development of a therapeutic intervention. For example, biomarkers for detection and diagnosis may not be the same as those that ideally track disease progression. Likewise, for new therapeutic agents, a single assay may not suffice as a biomarker reflecting both target engagement and the pharmacodynamic effects of a drug.

Although muscle mass can objectively define the presence of sarcopenia, several components of skeletal muscle function are not adequately captured by simply measuring mass or cross-sectional area. It is now clear that there is a certain degree of divergence between changes in muscle mass and alterations in muscle performance. The well-described decline in skeletal muscle mass in older adults is a critical determinant of age-related weakness, which is defined as a reduction in maximal voluntary joint torque or power. Yet, it is now clear that the relationship between force production capability and muscle size in older adults is less robust than it is in young people [13]. Indeed, longitudinal studies have demonstrated that the age-related decline in muscle strength far exceeds the observed changes in muscle mass or size, particularly in weight-stable individuals [14, 15]. Furthermore, longitudinal studies indicate that maintenance or even gain of muscle mass may not prevent weakness in older adults [15, 16]. In addition, a number of age-related changes in force production capability are not readily explained by a reduction in muscle mass, including decreased specific force (force per cross sectional area) [17, 18] and slower rate of isometric force production (expressed relative to peak torque or to body weight) [19, 20]. Furthermore, voluntary weight loss leads to reductions in muscle mass/size with no declines in muscle strength [21]. It is also noteworthy that pharmacologic interventions that increase muscle mass/size do not necessarily improve voluntary strength. Similarly, physical activity interventions that increase muscle strength do not necessarily augment muscle size [22, 23]. Noticeably, gains in muscle strength secondary to increased physical activity generally precede measurable changes in skeletal muscle mass/size.

The progressive muscle atrophy with aging is associated with a loss of overall muscle force and changes in force and power generation of the remaining muscle fibers [24]. However, several additional physiological mechanisms that accompany the phenomenon of sarcopenia may directly influence muscle function and force production with advancing age. Recent evidence has shown that adipose tissue accumulation around and between muscle fibers concomitant with reductions in muscle cross-sectional area occurs with aging, and that this skeletal muscle attenuation is inversely associated with muscle performance [18, 25]. Age-related changes in the nervous system may also play a substantial role in the decline in muscle power generation [26]. These include loss of motor neurons and concomitant remodeling of motor units through collateral reinnervation [27], impairment of neuromuscular activation observed as decreased maximal motor unit firing rates [28‐30], and uncoordinated patterns of intermuscular neural activation [31]. Finally, changes in individual muscle fiber composition and intrinsic contractile properties may influence the decline in muscle force among older adults. For instance, cross-sectional observations suggest that reductions in muscle torque may be related to changes in fiber composition and, in particular, to the preferential atrophy of type II (fast-twitch) fibers with aging [32]. Specific changes in the intrinsic ability of aged muscle to generate force have also been observed [33]. Decreases in specific force (force normalized per cross sectional area) and unloaded shortening velocity in type I and IIA fibers have been reported in older males compared with young controls [32, 34]. Conversely, recent longitudinal data have demonstrated that, despite reductions in whole muscle cross-sectional area, single muscle fiber contractile function is preserved with advancing age as existing fibers may compensate and partially correct these deficits, therefore maintaining optimal force-generating capacity [14].

Although precise and valid measures of muscle mass are important components of sarcopenia assessment, these gross measures of muscle size do not adequately account for the dynamic components (force, power, activation) of muscle function that are responsible for performing activities of daily living. Future trials on sarcopenia adopting clinically meaningful endpoints should evaluate these key biomarkers of muscle function through the use of state-of-the-art methodologies.

The bidimensional definition of sarcopenia simultaneously includes a functional parameter (i.e., muscle performance) and a quantitative index (i.e., muscle mass). Therefore, techniques aimed at capturing the objective amount of skeletal muscle mass are required. Multiple methodologies are currently available to accomplish this task [35].

DXA is the most commonly used imaging technique for several reasons: first of all, because it is commonly available in clinical and research settings, being relatively inexpensive, sufficiently precise, and well-accepted by older persons. Second, the initial operative definition of sarcopenia proposed by Baumgartner and colleagues [3] was based on appendicular lean mass measured by DXA. Later on, DXA was used to provide alternative definitions of sarcopenia based on the fat-adjusted residual method [36]. Nevertheless, it cannot be ignored that the first operative definition is dated more than 10 years, and during this time, several steps forward have been made in refining imaging techniques as well as understanding the sarcopenia phenomenon.

The identification of the “gold standard” for the quantitative evaluation of muscle mass in clinical trials (which is currently lacking) should be based on criteria of accuracy (i.e., the degree of conformity of a measure to a standard or a true value), precision (i.e., the degree of refinement with which an operation is performed or a measurement stated), reproducibility (i.e., the quality of being reproducible under the same operating conditions over a period of time or by different operators), sensitivity to change (i.e., the degree of being modified by interventions), and accessibility (i.e., its usual availability in research and clinical centers).

DXA currently represents the more accessible technique for body composition assessment. It may accurately provide estimates of lean, fat, and bone tissues in the entire body or in specific regions. Moreover, it is inexpensive and quick to be performed. The radiation exposure associated with DXA is low and highly acceptable (about 1 mrem, a quantity similar to that of a 3-day background). The main limitations of this imaging approach reside in some analytical differences across manufacturers and models, and the risk of biased results due to the low differentiation between water and bone-free lean tissue.

CT accurately measures a direct physical property of the muscle (e.g., cross-sectional area and volume). It also allows the evaluation of muscle density (a parameter related to intramyocellular lipid deposits) as well as subcutaneous and intramuscular adipose tissue deposition. The radiation exposure associated with this technique is higher (i.e., about 15 mrem) than with DXA.

Magnetic resonance imaging (MRI) presents a high agreement with CT and provides similar measures. It does not involve radiation exposure and also has the additional capacity of multiple slice acquisition, thus rendering 3D volumetric estimates. The lack of radiation exposure makes MRI the method of choice for many studies where ethics committee or national authority approval is more difficult to obtain for CT. The major limitations of this methodology reside in the higher technical complexity and costs, and in the inapplicability to subjects with older models of implanted metal devices (e.g., joint prostheses, pace-makers, etc.). Both CT and MRI may be limited in the ability to accommodate very obese individuals.

Finally, it needs to be emphasized that imaging provides information only about one of the two sarcopenia dimensions. As discussed earlier, changes in muscle function and quantity do not necessarily follow similar trajectories with aging [37]. Therefore, interventions able to increase lean mass may not necessarily produce parallel gains in strength and vice versa [38]. To overcome this issue and include the two components of sarcopenia in the same variable, it has been proposed to compute an index of skeletal muscle quality derived from the ratio between strength and mass [15, 39, 40].

One of the most recently developed techniques which might find larger application in the near future for the evaluation of sarcopenia is the electrical impedance myography (EIM) [41]. This is a noninvasive, painless approach based on the surface application and measurement of a high-frequency, low-intensity electrical current applied to specific muscles. EIM detects changes in the conductivity and permittivity of skeletal muscle caused by alterations in muscle composition and structure. EIM is repeatable and sensitive to skeletal muscle changes in patients with amyotrophic lateral sclerosis [42]. Moreover, its changes over time may also have clinical relevance as they are predictive of survival in animal models of amyotrophic lateral sclerosis [43]. Finally, it is also noteworthy that the EIM phase shows a consistent inverse relationship with age [44].

An alternative method to measure skeletal muscle size is by ultrasonography. This technique has shown to be a valid (versus MRI-based measurements) and highly reliable way for assessing cross-sectional areas of large individual human muscles [45]. It is particularly useful in mobility-impaired subjects who cannot easily be transported to scanners such as CT or MRI machines.

Also remarkable is the development of mass isotopomer distribution analysis based on the evaluation of protein and proteome synthesis rate obtained by heavy water labeling [46, 47]. Although this technique can still be considered suitable mainly for research settings, its flexibility and the large amount of information it provides about a wide spectrum of proteins make it extremely promising.

Other techniques are also available to detect sarcopenia, but their limited validation, low accuracy, and difficult large-scale implementation discourage their use. For example, bioelectrical impedance analysis (BIA) is a popular, very simple, and low-cost technique, but its results are far from being accurate. The BIA technique is based on the notion that tissues rich in water and electrolytes are less resistant to the electrical passage than adipose tissue. The BIA is therefore based on a single body resistance parameter (not a direct measure of skeletal muscle), and its results can be easily altered by fluid retention and health status in general. For these reasons, a recent consensus paper by the Society of Sarcopenia, Cachexia and Wasting Disorders has discouraged the use of BIA for the assessment of sarcopenia [9].

There is still resistance to accept sarcopenia as a clinical condition despite its well-established relationship with major health-related negative events (in particular, mobility and physical disability) [8]. This issue might (at least partly) be explained by the current lack of clinically relevant thresholds that distinguish normal from abnormal values of skeletal muscle mass.

Several approaches can be adopted to identify critical cut-points. A paradigmatic example potentially lending support to the operative definition of sarcopenia might be provided by the approach previously adopted to identify osteoporosis on the basis of bone mineral density. In fact, approaches that have been developed for bone and osteoporosis may serve well for skeletal muscle and sarcopenia. The clinical definition of a specific condition (which will consequently lead to the indication for treatment) might be based on:

In summary, the presence of sarcopenia might be determined by (1) relying on a clinical diagnosis closely related to skeletal muscle decline (e.g., mobility disability) after exclusion of secondary causes, (2) a representative scientific committee identifying a critical threshold for a biological parameter directly representative of skeletal muscle health, and/or (3) developing a risk index to guide treatment.

Given the syndromic nature of sarcopenia, intervention strategies aimed at preventing/treating its process might need to target multiple risk factors. In this context, several biological markers have been shown to be associated with skeletal muscle mass, strength, and function, thus representing potential markers for the effect of the studied interventions. Such a list is quite long, and each biomarker identifies a specific mechanism contributing the age-related skeletal muscle decline, although they are not specific to muscle and many are likely to turn out to be only weakly associated with clinically relevant outcomes. The most common markers are inflammatory biomarkers [e.g., C-reactive protein [52, 53], interleukin-6 [52‐54], and tumor necrosis factor-α [52, 54]], clinical parameters [e.g., hemoglobin [55, 56], serum albumin [57, 58], and urinary creatinine [59]], hormones [e.g., dehydroepiandrosterone sulfate [60], testosterone [61], insulin-like growth factor-1 [62], and vitamin D [63‐65]], products of oxidative damage [e.g., advanced glycation end-products [66], protein carbonyls [67, 68], and oxidized low-density lipoproteins [69]], or antioxidants [e.g., carotenoids [70, 71] and α-tocopherol [70]].

Other promising biomarkers have been identified in the last years and may represent useful parameters to more directly explore sarcopenia because they are closely related to skeletal muscle changes. For example, plasma concentrations of procollagen type III N-terminal peptide (P3NP) represent an interesting marker of skeletal muscle remodeling [72, 73]. P3NP is a fragment released by the cleavage of procollagen type III to generate collagen III (a protein produced in soft connective tissues, skin, and muscle). Preliminary studies have also suggested an interesting role played by biomarkers specifically linked to the neuromuscular junction in evaluating skeletal muscle modifications [74, 75].

Ultimately, the goal of clinical trials for sarcopenia treatments will require the evaluation of clinical benefit. In fact, clinical measures can also be considered as biomarkers as they reflect the impact of the pathological process of sarcopenia on the patient's health. The assessment of measures of muscle strength (e.g., hand grip), muscle power (e.g., leg extension power), and physical performance [e.g., Short Physical Performance Battery [4] and gait speed tests] comprise important indices of the individual's physical function. In addition, functional outcome measures will need to be developed in order to help understand the impact of any treatment-related quantitative gains in performance on the person's daily life.