The Etiology and Impact of Muscle Wasting in Metastatic Cancer

Abstract

Metastasis arises when cancer cells disseminate from their site of origin and invade distant organs. While cancer cells rarely colonize muscle, they often induce a debilitating muscle-wasting condition known as cachexia that compromises feeding, breathing, and cardiac function in metastatic cancer patients. In fact, nearly 80% of metastatic cancer patients experience a spectrum of muscle-wasting states, which deteriorates the quality of life and overall survival of cancer patients. Muscle wasting in cancer results from increased muscle catabolism induced by circulating tumor factors and a systemic metabolic dysfunction. In addition, muscle loss can be exacerbated by the exposure to antineoplastic therapies and the process of aging. With no approved therapies to alleviate cachexia, muscle health, therefore, becomes a key determinant of prognosis, treatment response, and survival in metastatic cancer patients. This review will discuss the current understanding of cancer-associated cachexia and highlight promising therapeutic strategies to treat muscle wasting in the context of metastatic cancers.

Muscles make up nearly 50% of our total body mass and play a key role in normal physiological activities such as breathing, chewing and swallowing food, and locomotion. Yet, often unrecognized, these basic functions are severely compromised in 80% of advanced cancer patients due to a muscle-wasting syndrome known as cachexia (Fearon et al. 2012; Baracos et al. 2018). Cachexia involves a dramatic loss of skeletal muscle mass and function that drastically decline the rate of survival and quality of life of metastatic cancer patients. Importantly, cachexia is not caused by the metastasis of cancer cells to muscle. Instead, tumors secrete factors into the circulation that perturb normal physiological and metabolic programs in the host and negatively impact the functions of skeletal and cardiac muscle. Moreover, since tumors compete with their host for available nutrients, a limited nutrient supply can also cause energy deprivation and trigger muscle catabolism (Fearon et al. 2012). Unlike starvation, however, dietary supplementation and parenteral nutrition fail to reverse cachexia, which underscore that cachexia does not emerge merely due to failure of a nutritional supply-and-demand balance, but perhaps an inability of host tissues to utilize available nutrients to fulfill their energy and metabolic needs.

Muscle wasting has dire consequences on vital physiological activities of the cancer patients (Kalantar-Zadeh et al. 2013). For example, when jaw muscles deteriorate, chewing, swallowing, and feeding responses are compromised. With declining muscle mass and function in limb muscles, patients experience restricted movement and lose their functional independence. Due to the dysfunction of the diaphragm and cardiac muscles, cancer patients die prematurely from respiratory distress and cardiac failure, respectively (Baracos et al. 2018). Cachexia also reduces tolerance to chemotherapy and increases the risk of postsurgical complications, which in turn compromise anticancer treatment outcomes and negatively impacts survival (Fearon et al. 2013). The discovery of effective treatment options for cachexia is therefore expected to restore tolerance to anticancer drugs, improve quality of life and prolong survival in advanced cancer patients.

SPECTRUM OF MUSCLE-WASTING STATES IN CANCER PATIENTS

Multiple factors contribute to muscle catabolism during cancer progression and treatment ( Figs. 1 and 2). Chemotherapy-induced muscle damage can exacerbate cachexia. (Dewys et al. 1980; Barreto et al. 2016a). Sarcopenia, an aging-associated decline in muscle mass, is often observed in elderly patients (Fielding et al. 2011; Bowen et al. 2015) and results in the reduction of tolerance to chemotherapy, latency before cancer progression, and life expectancy compared to nonsarcopenic cancer patients (Prado et al. 2009a). Not surprisingly, the aggravation of existing muscle-wasting conditions by anticancer therapy leads to a further decline in cancer patient survival (Prado et al. 2007, 2011a). Moreover, muscle wasting can often be masked by obesity and thereby remain undiagnosed for extended periods of time. For instance, breast cancer patients rarely show overt cachectic symptoms such as body weight loss and even gain weight during cancer treatment. However, body composition analysis by computerized tomography (CT) often reveals muscle mass loss in breast cancer patients can be compensated by increased adipose tissue mass (Martin et al. 2013). It is important to note that muscle depletion or lean tissue loss is associated with higher dose-limiting toxicities from antineoplastic therapies (Mourtzakis et al. 2008; Prado et al. 2009a, 2011a). Knowledge of the underlying mechanisms that drive muscle wasting, therefore, has the potential to eliminate the myriad complications that cancer patients endure with debilitating muscle-wasting conditions during both active treatment and subsequent recovery.

Figure 1.

Spectrum of muscle-wasting states in metastatic cancer patients.

Figure 2.

An outline of mediators of muscle atrophy highlighting the multifactorial regulation of muscle wasting in cancer.

CANCER-INDUCED MUSCLE WASTING

Cachexia is frequently associated with cancers that originate in the pancreas, gastrointestinal tract, and lung and manifests as an energy-wasting syndrome that disrupts muscle protein homeostasis (Baracos et al. 2018). The efficient cellular protein homeostasis depends on tight regulation of protein biosynthesis and breakdown pathways. It is estimated that 280 g of protein is synthesized and degraded each day in a 70-kg adult human (Mitch and Goldberg 1996). Through the ubiquitin-mediated proteasomal degradation pathway, the covalent attachment of ubiquitin molecules by E1, E2, and E3 ligases to proteins leads to their degradation by the 26S proteasome. However, in cancer patients, this normal process is often disrupted, which induces the breakdown of muscle proteins, sometimes coupled with reduced synthesis (Mitch and Goldberg 1996; Cohen et al. 2015; Zhao and Goldberg 2016). The increase in muscle protein degradation occurs primarily through the ubiquitin-proteasome and autophagy pathways (Mitch and Goldberg 1996) and, coupled with the additional loss of organelles and cytoplasm from muscle cells, results in muscle atrophy (Cohen et al. 2015; Sandri 2016). In particular, the E3 ligases MuRF1, MAFbx/atrogin-1, MUSA1, TRAF6, and Fbx031 are transcriptionally up-regulated in atrophic muscles under multiple catabolic conditions including denervation, immobilization, sepsis, glucocorticoid exposure, and cancer, leading to the increased turnover of sarcomeric and other muscle proteins (Bodine et al. 2001; Gomes et al. 2001; Paul et al. 2010; Sartori et al. 2013; Milan et al. 2015; Wang et al. 2018). MuRF1 ubiquitinates sarcomeric proteins, myosin heavy chain and actin (Polge et al. 2011) while atrogin-1 degrades the myogenic regulatory factor MyoD in muscle cells (Tintignac et al. 2005), and importantly, loss of Murf1, Mafbx, and Musa1 in mice protects muscles from denervation-induced atrophy (Bodine et al. 2001; Paul et al. 2010; Sartori et al. 2013). However, there is a paucity of loss-of-function genetic studies designed to specifically analyze the role of these E3 ligases in cancer-induced muscle wasting.

The delicate balance between protein synthesis and degradation is maintained by a cross-talk between their regulatory pathways (Fig. 2). For instance, the IGF1/PI3K/AKT pathway induces hypertrophy in muscle cells via activation of protein synthesis while simultaneously blocking induction of the catabolic atrophy mediators MuRF1 and MAFbx by forkhead box O (FOXO) transcription factors (Sandri et al. 2004; Stitt et al. 2004). AKT phosphorylates all three mammalian FOXO family members (FOXO1, FOXO3, and FOXO4), which blocks their nuclear function by sequestering them in the cytoplasm. Nonphosphorylated FOXO factors, on the other hand, are free to enter the nucleus and induce the expression of MuRF1 and MAFbx. Under atrophy-inducing conditions, PI3K/AKT pathway activity diminishes in differentiated muscle cells (myotubes) leading to FOXO3-mediated induction of MAFbx (Sandri et al. 2004). Consistently, the expression of IGF-1, activated PI3K or activated AKT is sufficient to inhibit the up-regulation of MuRF1 and MAFbx during dexamethasone-induced muscle atrophy in a FOXO-dependent manner (Stitt et al. 2004). Moreover, the transcriptional activity of FOXO1 and FOXO3a increases under muscle-atrophy-inducing conditions (Reed et al. 2012), and transgenic overexpression of FOXO1 in otherwise-normal skeletal muscle results in decreased muscle mass and reduced muscle function (Kamei et al. 2004). The inhibition of FOXO activity in muscles has also been shown to prevent cancer-induced muscle atrophy as well as reduce the expression of atrophy signature genes such as Murf1, Mafbx, and Bnip3 (Reed et al. 2012). The IGF1/PI3K/AKT and FOXO pathways, therefore, play a pivotal role in muscle fiber size regulation.

It is unclear whether cancer-induced muscle wasting exists to benefit tumors. Skeletal muscle comprises 75% of the protein of the human body (Wolfe 2006) and thereby serves as the primary reservoir for amino acids that sustain both protein synthesis and energy production if other energy sources become depleted (Parry-Billings et al. 1991; Chen et al. 1993; Souba 1993; Luo et al. 2014; Argilés et al. 2016). As such, skeletal muscle proteolysis represents a vast energy resource for a developing tumor (DeBerardinis and Cheng 2010), which may initiate muscle catabolism to meet its high energy requirements. Yet, human tumors comprising <1% of the total body mass can trigger muscle catabolism (Fearon et al. 2012), and while increased nutritional intake or intravenous feeding partially reverses fat loss in cancer patients, neither is able to reverse muscle wasting (Nixon et al. 1981; Popp et al. 1981; Cohn et al. 1982; Nixon 1982). Since muscle catabolism also persists in the presence of hypercaloric feeding (Kotler 2000), it may be driven by mechanisms that are independent of the energy needs of a growing tumor. An alternative hypothesis to explain cancer-associated muscle wasting is that muscle catabolism is merely an unintended consequence of a cancer-induced disruption in muscle energy utilization. The possibility of ineffective energy utilization in the muscle is supported by the frequent development of insulin resistance in the muscles of cachectic cancer patients as well as an increase in energy-inefficient processes driven by the liver, such as gluconeogenesis and the acute phase response (APR) (Fig. 2; Argilés et al. 2014). Additional experiments are still needed to clarify whether cancer-induced muscle wasting evolved as a means to support the growth and survival of cancer cells.

New insights from Drosophila models have shed light on the connection between mediators of insulin resistance and muscle atrophy (Figueroa-Clarevega and Bilder 2015; Kwon et al. 2015). In a Drosophila cancer cachexia model generated by the Bilder group, secretion of the insulin growth factor binding protein homolog ImpL2, an antagonist of insulin signaling, leads to insulin resistance in peripheral tissues as well as muscle wasting (Figueroa-Clarevega and Bilder 2015). ImpL2 also causes systemic metabolic changes and muscle wasting when secreted from an overproliferating gut (Kwon et al. 2015). Whether an analogous pathway drives human cancer cachexia remains to be elucidated, yet insulin resistance is indeed common in cancer patients and animal models with cachexia (Asp et al. 2010, 2011; Honors and Kinzig 2012). It can accelerate muscle proteolysis through suppression of the PI3K/AKT pathway and activation of the ubiquitin-mediated proteasome pathway (Wang et al. 2006). Interestingly, an insulin tolerance test revealed that the C26 cancer cachexia model exhibits a blunted blood glucose response to insulin before the onset of weight loss (Asp et al. 2010). In this mouse model, insulin resistance also correlates with reduced AKT phosphorylation and increased expression of genes involved in proteolysis and autophagy, such as Murf1, Mafbx, and Bnip3I, in the muscle (Asp et al. 2010). Furthermore, insulin resistance leads to an increase in hepatic gluconeogenesis, an energy-wasteful process in which glucose is synthesized from substrates such as lactate, alanine, glycerol, and glutamine in the liver (Hatting et al. 2018). Amino acids released from muscle protein breakdown during cancer cachexia are used as substrates for this process (Argilés et al. 2016). Insulin resistance, therefore, contributes to the negative energy balance observed in cachectic cancer patients (Honors and Kinzig 2012; Hatting et al. 2018) and may be an early event during the development of cachexia.

The liver initiates an APR where large quantities of tissue injury response proteins are synthesized at the expense of structural proteins (Kushner 1993; Kotler 2000; Argilés et al. 2018). This is similar to the pathological response involving the innate immune system that is normally triggered during infection. The APR is defined by a change in the concentration of certain plasma proteins that reflect either a 25% increase (termed “positive APR proteins”) or 25% decrease (termed “negative APR proteins”) (Gabay and Kushner 1999) in their levels (Fearon et al. 1999). Examples of positive APR proteins are C-reactive protein, serum amyloid A (SAA) and fibrinogen while negative APR proteins include albumin and transferrin. C-reactive proteins are opsonins that bind to denatured lipopolysaccharides/proteins and nucleic acids and lead to complement activation and macrophage-mediated phagocytosis (Kushner 1993; Cray et al. 2009) whereas APR SAA proteins are involved in immune cell chemotaxis (Cray et al. 2009; Lee et al. 2019). Under inflammatory conditions triggered by exposure to cytokines such as interleukin-6 (IL-6) and IL-1, by glucocorticoids (Baumann et al. 1987; Mackiewicz et al. 1992) or in the context of cancer, structural protein synthesis gives way to increased production of APR proteins (Falconer et al. 1995; Fearon et al. 1999). Synthesis of many of the plasma proteins not involved in host defense, such as albumin, is down-regulated (Kushner 1993; Cray et al. 2009). In fact, clinical studies in pancreatic cancer patients show that the plasma concentration of APR proteins can be altered by 40% at the time of diagnosis and up to 80% at the time of death (Falconer et al. 1995; Fearon et al. 1999). Since the APR alters host metabolism and reduces appetite, it is thought to contribute to the negative energy balance in cachectic patients by both reducing energy intake and increasing energy expenditure. The APR, therefore, represents yet another factor that contributes to increased resting energy expenditure (REE) and hypermetabolism in cancer cachexia (Falconer et al. 1994). Whether the APR actually benefits the host or represents a futile, energy-intensive response to cancer remains debatable; however, it is clear that the chronic activation of the APR that occurs in the context of cancer leads to a continuous subversion of amino acid resources and has adverse effects on cancer patients.

Increased energy expenditure is a common feature of cachexia across many cancer types (Cao et al. 2010; Xu et al. 2012). White adipose tissue (WAT) loss and adipose tissue browning contribute to this phenomenon and thereby promote a negative energy balance in patients with cancer cachexia. Genetic studies show that ablation of the adipose tissue triglyceride lipase gene (Atgl) prevents loss of both white adipose cells and muscle protein, suggesting an important cross-talk between adipose tissue and muscle during cachexia development (Das et al. 2011). An increase in energy expenditure also arises from the browning of WAT to form beige adipose cells. WAT specializes in storing energy in the form of lipids while brown adipose tissue participates in energy expenditure (Cypess et al. 2009). WAT browning can be triggered by exposure to a number of stimuli including cold temperatures (Sidossis et al. 2015), the tumor-derived polypeptide parathyroid hormone-related protein (PTHrP) (Kir et al. 2014), lactate (Carriere et al. 2014), chronic inflammation and IL-6 (Petruzzelli et al. 2014), or the muscle-derived cytokine irisin (Boström et al. 2012). As such, each of these events has the potential to exacerbate the negative energy balance observed in cachectic cancer patients.

The oxidative capacity of a muscle cell is largely determined by its mitochondrial function (Vitorino et al. 2015). During cancer-induced muscle wasting, mitochondrial function is compromised (Brown et al. 2017). For example, cachectic muscles from multiple cancer models exhibit distorted mitochondria with swollen and fragmented cristae and sarcomeric disintegration (Shum et al. 2012; Tzika et al. 2013; Brown et al. 2017). Moreover, the ratio of mitochondrial DNA to nuclear DNA is reduced in the Apc(Min/+) cancer cachexia model along with decreases in the protein levels of cytochrome c and cytochrome c oxidase complex subunit IV (White et al. 2011a). A similar reduction in complex IV activity and mitochondrial oxidative capacity was observed in cachectic rats with peritoneal carcinosis (Julienne et al. 2012). This occurred without either mitochondrial uncoupling or changes to the efficiency of ATP synthesis. In contrast, a separate study in cachectic muscles from mice bearing Lewis lung carcinomas showed a reduction in both ATP synthesis rate and TCA cycle flux, which is indicative of mitochondrial uncoupling (Tzika et al. 2013). While it remains unclear whether the difference in outcome between these two studies is due to the specific model, cancer type or degree of cachexia, it is likely that mitochondrial dysfunction in muscle cells also contributes to the negative energy balance observed in patients with cancer cachexia.

Studies performed over the past several decades have identified the proinflammatory cytokines tumor necrosis factor-alpha (TNF-α), interferon-gamma (IFN-γ), and IL-6 as important mediators of cancer cachexia. A comprehensive review on the role of these cytokines function during cancer cachexia is available elsewhere (Fearon et al. 2012). For example, cachexia can be induced in nude mice when inoculated with Chinese hamster ovary tumor cells overexpressing either TNF-α or IFN-γ (Oliff et al. 1987; Matthys et al. 1991). Blockade of TNF-α signaling in mice reduces tumor-induced muscle wasting (Llovera et al. 1998a,b), and loss of IL-6 in the Apc(Min/+)-intestinal and C26 colon-cancer cachexia models (McCart et al. 2008) protects against muscle atrophy (Strassmann et al. 1992; Fujita et al. 1996; Tsujinaka et al. 1996; White et al. 2011b). However, clinical trials have shown that blocking these cytokines does not effectively reverse cachexia, which could be due to their varying expression levels in cancer patients or the existence of multiple mediators of cachexia. Indeed, TNF-α acts in concert with IFN-γ to inhibit myosin heavy chain transcription in muscle cells through reduction of MyoD expression (Acharyya et al. 2004), and the TNF/TNFR superfamily exhibits functional redundancy among its family members (Dogra et al. 2007; Bhatnagar et al. 2012; Johnston et al. 2015). IL-6 family members also exhibit redundancy with respect to their ability to induce cachexia (Bonetto et al. 2011); ciliary neurotrophic factor (CNTF) and leukemia inhibitor factor (LIF) can induce muscle wasting by activating the same downstream STAT3-mediated signaling pathway as IL-6 (Henderson et al. 1994; Espat et al. 1996; Bonetto et al. 2011; Kandarian et al. 2018). Thus, blocking STAT3 might represent an effective alternative strategy for treating cachexia (Bonetto et al. 2012; Zimmers et al. 2016). The existence of functional redundancy among these different cytokine families ultimately highlights the importance of targeting multiple mediators to effectively reverse cachexia.

Several members of the transforming growth factor-beta (TGF-β) superfamily, including myostatin, have been implicated in cancer cachexia. Myostatin plays a key role in regulating muscle size. Mice lacking myostatin or growth and differentiation factor 8 (GDF8) develop profound muscle hypertrophy (McPherron et al. 1997), while systemic overexpression of myostatin induces severe cachexia (Zimmers et al. 2002). Myostatin binds to the activin type II receptors ACVR2 and ACVR2B and thereby triggers the phosphorylation of activin type I receptors and activation of the downstream SMAD2/3 signaling pathway (Benny Klimek et al. 2010). Importantly, systemic administration of soluble forms of ACVR2B, either ACVR2B-Fc secreted from implanted CHO cells (Benny Klimek et al. 2010) or the decoy receptor sActRIIB (Zhou et al. 2010), inhibits cachexia and prolongs survival in mouse models (Zhou et al. 2010), thus providing promising translational opportunities.

The TGF-β signaling pathway has another important function in mediating cachexia specifically in the context of metastatic cancers (Waning and Guise 2014; Waning et al. 2015; Regan et al. 2017). The release of TGF-β from bone initiates a vicious cycle of bone destruction and metastatic tumor growth as well as muscle weakness in mice with bone metastases (Waning et al. 2015). In this model, TGF-β up-regulates NAPDH oxidase 4 (NOX4) in muscle cells, which increases the oxidation of ryanodine receptor/calcium release channel (RYR1) on the sarcoplasmic reticulum (Waning et al. 2015). Excess RYR1 oxidation promotes leakiness of calcium channels and thereby compromises muscle force generation. As such, muscle weakness can be reversed by inhibiting either TGF-β, TGF-β receptor I kinase, NOX1/4, or the RYR1 Ca²⁺ release channel stabilizer RYCAL, providing new avenues of intervention for cancer cachexia. Recent studies from our laboratory showed that TGF-β up-regulates the metal-ion transporter ZRT- and IRT-like protein 14 (ZIP14) in muscle cells, which functions as a mediator of cachexia in metastatic cancers (Wang et al. 2018). ZIP14 up-regulation causes an increase in muscle-cell zinc uptake that induces myosin heavy chain loss in mature muscle cells and blocks muscle differentiation in muscle progenitors (Wang et al. 2018). Therefore, blocking TGF-β signaling, inhibiting the function of ZIP14, and modulating dietary zinc intake represent additional potentially effective strategies for preventing or treating cachexia in metastatic cancer patients.

THERAPY-INDUCED MUSCLE WASTING

Many antineoplastic agents give rise to adverse side effects such as reduced appetite and premature satiety, which can impact food intake and body weight in cancer patients (Kayl and Meyers 2006). Skeletal muscle catabolism and loss of lean muscle mass (Gilliam and St Clair 2011; Prado et al. 2011a; Gilliam et al. 2013; Barreto et al. 2016a; Pin et al. 2018) are further consequences of antineoplastic therapy and can be readily measured by body composition analysis. Longitudinal CT scans provide precision, clarity, and specificity with respect to the type, location and quantity of adipose and skeletal muscle tissue loss in cancer patients (Mourtzakis et al. 2008). Consequently, studies have shown that cancer patients with reduced lean muscle mass develop greater toxicities from a number of drugs, such as 5-fluorouracil, capecitabine, and sorafenib (Prado et al. 2011a). In fact, the loss of lean muscle serves as an independent predictor of toxicity in cancer patients exposed to antineoplastic therapy (Prado et al. 2007, 2009a, 2011b, 2014; Antoun et al. 2010a,b). Patients with reduced lean muscle tissue are also more likely to require a dose reduction of, delays in, or early termination of their anticancer treatment and to be excluded from participating in clinical trials (Prado et al. 2011a). This has a significant negative impact on cancer patient survival. Therefore, it is critically important to identify the underlying mechanisms that drive therapy-induced muscle loss and to design effective strategies to prevent or reverse these devastating side effects of anticancer therapy.

Chemotherapy is the mainstay of treatment for metastatic cancers (Schakman et al. 2009), and while it effectively kills cancer cells, it also causes extensive damage to normal tissues such as the gut, bone marrow, kidney, and nervous system (Kayl and Meyers 2006). Fatigue and weakness represent yet another prominent side effect of most chemotherapy agents that continues even after cessation of treatment (Tavio et al. 2002). Fatigue is largely due to loss of muscle mass and function, and is responsible for the morbidity and poor quality of life of chemotherapy-treated patients (Tavio et al. 2002). Indeed, cancer patients with low muscle mass have significantly lower rates of survival and increased treatment-induced toxicities (Prado et al. 2009a; Blauwhoff-Buskermolen et al. 2016; Guigni et al. 2018). For instance, when adjusted for variables such as sex, age, and scope of metastases, colon-cancer patients with chemotherapy-induced muscle loss of ≥9% had a 7% 1-yr survival rate compared to patients with <9% muscle loss showing 49% survival (Blauwhoff-Buskermolen et al. 2016). Still, it remains unclear whether chemotherapy-induced muscle wasting is similar to or distinct from cancer-induced muscle wasting, which pathways mediate chemotherapy-induced muscle wasting and whether different chemotherapy agents cause unique muscle-damaging effects. Over the past decade, new studies have begun to address these questions and are briefly discussed below.

The TNF-α signaling pathway and its effector NF-κB are commonly activated in muscle cells in response to both the presence of cancer and treatment with chemotherapy (Li et al. 1998; Guttridge et al. 2000; Judge et al. 2007; Damrauer et al. 2018). Muscle-specific activation of NF-κB was originally shown to promote muscle catabolism and muscle mass loss (in the absence of cancer and chemotherapy) through up-regulation of the E3 ligase MuRF1 and activation of the ubiquitin-dependent proteasome pathway (Li and Reid 2000; Cai et al. 2004). In one study, Cisplatin treatment (3 doses of 5 mg/kg body weight given intravenously over 2 wk) was shown to activate NF-κB signaling in the limb muscles of healthy mice and to elicit features of cachexia without inducing MuRF1 or activating the ubiquitin-proteasome pathway (Damrauer et al. 2018). In contrast, a different study showed that cisplatin treatment of healthy mice (2.5 mg/kg body weight given intraperitoneally daily over 10 d) leads to the muscle-specific up-regulation of E3 ligases, atrogin-1 and to a lesser extent MuRF1, and activation of the ubiquitin-proteasome pathway (Chen et al. 2015). Cisplatin treatment also triggered a number of additional events in this study that may contribute to muscle loss, including the increased expression of p38 and myostatin and decreased expression of Akt. Disruption of the Akt pathway, as well as hyperactivation of the proteasome and autophagy pathways, may also mediate cisplatin-induced muscle wasting of differentiated muscle cells (Fanzani et al. 2011). Importantly, the administration of ghrelin, an endogenous ligand of the growth hormone secretagogue receptor (GHSR), inhibits cisplatin-induced muscle toxicity and improves muscle function (Chen et al. 2015). The mechanisms driving cisplatin-induced muscle loss, as well as the influence of cisplatin dosage and duration on the severity of the associated phenotypes, remain to be further investigated.

Diaphragm weakness and dyspnea (shortness of breath) are common in chemotherapy-treated cancer patients (Morrow et al. 2002; Gilliam and St Clair 2011). Treatment with doxorubicin induces diaphragm and limb muscle weakness in healthy mice (Gilliam et al. 2011a) and leads to an increase in the expression of TNF receptor 1 (TNFR1) and its localization to the membrane (Gilliam et al. 2009, 2011a). This causes a depression of diaphragm-specific force, which can be rescued by etanercept, a soluble form of the TNF receptor that blocks the activity of circulating TNF, or by the genetic loss of TNFR1 (Gilliam et al. 2009, 2011b). Similar to cisplatin (Fanzani et al. 2011; Chen et al. 2015), doxorubicin treatment in healthy mice (a single dose of 15 mg/kg body weight given intraperitoneally) up-regulates the expression of atrogin-1 and activates the proteasome in skeletal and heart muscles (Yamamoto et al. 2008) suggesting common pathways driving muscle wasting between these two anticancer agents.

Muscle-wasting phenotypes vary with the type of chemotherapy agent, mode of action and metabolism. For instance, a single dose of the alkylating agent cystemustine (20 mg/kg body weight) leads to a reduction in body weight and loss of muscle mass within 5 d of treatment in healthy mice, which recover gradually but never back to the levels observed in untreated mice (Samuels et al. 2001; Tilignac et al. 2002). In mice bearing C26 colon adenocarcinomas that develop cachexia, the same dosing schedule of cystemustine showed an additive effect with the cachexia-induced muscle wasting and completely eliminated the tumors. Interestingly, body and muscle mass in these mice were restored to levels observed in cystemustine-treated, nontumor-bearing mice, possibly as a direct result of treatment with this particular type of chemotherapy agent. Moreover, cystemustine treatment eventually led to a reduction in ubiquitin-dependent proteolysis and protein breakdown in both control and tumor-bearing mice to levels below those observed in untreated healthy mice, an effect that was mimicked by the combination of the alkylating agent ifosfamide with cisplatin (Tilignac et al. 2002) and associated with the down-regulation of proteasome subunit expression. Thus, while treatment with alkylating agents initially leads to muscle wasting, it also appears to support a compensatory muscle-rebuilding response after complete tumor elimination in the C26 model that has not been observed with other anticancer agents (Barreto et al. 2016b). Some chemotherapeutic drugs disrupt muscle function through mechanisms that are independent of proteasome pathway activation. Folfiri (5-fluorouracil, leucovorin and CPT-11), a drug combination frequently used in treating metastatic colon cancers, significantly depletes skeletal muscle mass and strength (Barreto et al. 2016b) through a combination of decreased muscle anabolism (via reduced Akt phosphorylation) and activation of catabolic pathways (ERK1/2 and p38-MAPK) (Barreto et al. 2016b). Notably, these folfiri-induced effects could be rescued by blocking MAPK signaling in differentiated muscle-cell cultures. Muscle deterioration, therefore, represents a common side effect of chemotherapies in general, and a better understanding of the various mechanisms that drive this process is likely to improve the quality of life of cancer patients with muscle wasting.

Mitochondrial dysfunction has emerged as another feature associated with cancer- and chemotherapy-induced muscle wasting, although the phenotypes vary with tumor type, therapy regimen, and animal model. In the muscles of patients with cancer cachexia, bioenergetic insufficiency likely arises due to an altered mitochondrial structure characterized by swollen vesicles and fragmented cristae (Shum et al. 2012), a reduced oxidative capacity arising from a decrease in complex IV activity (Julienne et al. 2012), and/or a decreased rate of ATP synthesis (Tzika et al. 2013). Treatment with chemotherapies such as doxorubicin has been shown to inhibit mitochondrial respiration and disrupt mitochondrial energy metabolism (Gilliam and St Clair 2011; Gilliam et al. 2013) and redox balance (Gilliam et al. 2013). Moreover, a single dose of doxorubicin administered to rats increases mitochondrial hydrogen peroxide emission and impairs mitochondrial calcium retention in single muscle fibers (Gilliam et al. 2013). In response to folfiri, mice show decreased mitochondrial size and number along with a significant reduction in the expression of genes that regulate mitochondrial metabolism in muscles (Barreto et al. 2016b). Moreover, cyclical treatment of doxorubicin and dexamethasone every 3 wk for four cycles (doxorubicin intraperitoneally at 10 mg/kg body weight and dexamethasone subcutaneously at 2.5 mg/kg body weight) significantly impairs mitochondrial function in muscles and degrades muscle mass and function (Gouspillou et al. 2015). In contrast to the folfiri regimen (Barreto et al. 2016b), the doxorubicin–dexamethasone treatment fails to change the expression of most key mitochondrial metabolism genes. Instead, it induces a significant increase in reactive oxygen species in muscle-cell mitochondria accompanied by a sustained impairment of mitochondrial respiratory capacity (Gouspillou et al. 2015). Interestingly, mitochondrial dysfunction and long-term muscle damage are also observed in cardiac muscles in response to chemotherapy (Ewer and Ewer 2010), but the underlying mechanisms are thought to be distinct. Long-term, chemotherapy-induced cardiotoxicity is driven by oxidative-stress-induced loss of mitochondrial DNA and reduced mitochondrial volume (Yamada et al. 1995; Palmeira et al. 1997; Serrano et al. 1999; Lebrecht et al. 2003, 2007, 2010), neither of which are observed in skeletal muscles (Gouspillou et al. 2015). The disruption of muscle-cell mitochondria, therefore, appears to be a key mediator of the many debilitating features of chemotherapy- and cancer-induced cachexia.

Experimental and clinical studies indicate that cancer- and chemotherapy-induced muscle damage and compromised physical function persist as a chronic problem in cancer survivors (Scheede-Bergdahl and Jagoe 2013; Kurk et al. 2019). Recent studies have also begun to address whether there are distinct mechanisms that drive muscle wasting in response to either cancer or chemotherapy and whether other anticancer therapies induce muscle damage. Bonetto and colleagues revealed unique alterations in metabolic flux through the tricarboxylic (TCA) cycle and β-oxidation pathways between cancer- and chemotherapy-induced muscle wasting. They found that low-density lipoprotein particles (LDL-Ps) increased in cancer-induced, but not chemotherapy-induced, muscle wasting (Pin et al. 2019) and that plasma levels of branched-chain amino acids, which normally comprise 14%–18% of total amino acids and stimulate protein synthesis, decrease during cancer-induced muscle wasting but increase in response to chemotherapy (Pin et al. 2019). However, whether these metabolic alterations have a causative role in the development of muscle wasting in response to either cancer or chemotherapy is unclear. Muscle damage also occurs with targeted anticancer agents in both animal models and patients (Blumenschein et al. 2009; Chen and Wang 2018; Huot et al. 2019). For example, treatment of mice with kinase inhibitors such as sorafenib or regorafenib reduces cardiac and skeletal muscle mass, and induces muscle weakness, through the ERK1/2 and GSK3 pathways (Huot et al. 2019). Clinical studies show that 2.1 kg of skeletal muscle is lost every 6 mo in patients treated with sorafenib compared to placebo-control-treated patients who show stable body weight during the same time interval (Antoun et al. 2010a). Whether additional targeted inhibitors or immunotherapies induce muscle wasting has yet to be explored. Interestingly, a targeted MEK1/2 inhibitor fails to elicit a strong antitumor response but surprisingly, increases skeletal muscle mass in biliary tract cancer patients (Bekaii-Saab et al. 2011; Prado et al. 2012a; Talbert et al. 2017). The analysis of these findings in cancer cachexia mouse models showed that MEK inhibition prevents muscle wasting even in mice bearing a MEK-inhibitor-resistant tumor clone in which tumor growth is unaffected by MEK inhibition (Quan-Jun et al. 2017; Talbert et al. 2017). These experiments indicate that the ability of MEK inhibition to preserve muscle is independent of its antitumor functions. Furthermore, the combination of both a MEK inhibitor and a PI3K/AKT inhibitor significantly reduces tumor burden and preserves skeletal muscle mass (Talbert et al. 2017), providing a viable means of simultaneously treating both cancer and associated cachexia. A comprehensive analysis of anticancer therapies for their ability to inhibit or promote muscle loss, and for their associated long-term sequelae, is therefore expected to better inform treatment strategies to improve drug tolerance, prognosis, and quality of life in cancer patients.

AGING-RELATED MUSCLE WASTING

Sarcopenia is a muscle-wasting state that develops as part of the natural aging process and involves a degenerative loss of skeletal muscle mass, quality, and strength that often leads to functional impairment and disability (Argilés et al. 2015). Muscle mass loss typically begins around 30 yr of age and continues gradually such that by 80 yr of age, 30% of the original muscle mass is lost (Hughes et al. 2002). Interestingly, 40%–50% of newly diagnosed cancer patients exhibit sarcopenia (median age of 65 yr) compared to 15% of healthy individuals of similar age (von Haehling et al. 2010; Prado et al. 2012b). The consequences of sarcopenia are severe for cancer patients, who experience a higher incidence of toxicity from chemotherapy, poor surgical outcomes, a shorter time to cancer progression and a lower rate of survival (Prado et al. 2009a, 2016; Barret et al. 2014; Joglekar et al. 2015). There is a misconception that sarcopenia is inversely associated with body mass index (BMI), but body composition studies have shown that overweight or obese patients with a high BMI can have severe muscle depletion, which is termed “sarcopenic obesity” (Prado et al. 2016). Moreover, since the calculation for chemotherapy drug dosage is based on body surface area (determined by height and weight), lean tissue content and muscle mass are rarely considered despite a large body of clinical evidence demonstrating that lower muscle mass correlates with an increased likelihood of dose-limiting drug toxicities (Prado et al. 2007, 2009a, 2013; Antoun et al. 2010a). A comprehensive review of the link between body composition analysis and outcome in cancer patients is available elsewhere (Mourtzakis et al. 2008; Prado et al. 2009b). Since many aspects of aging-related muscle loss can be counteracted by regulating dietary intake and exercise (Penna et al. 2011; Ballarò et al. 2019), these findings make a compelling case for body composition analysis before administering cancer treatment.

The underlying causes of sarcopenia, which is characterized by loss of both muscle size and number, are multifactorial (Fielding et al. 2011) and range from an age-related decline in neurons, hormones and growth factors to nutritional deficiencies and reduced physical activity (Fig. 2; Campbell et al. 2001; Visser et al. 2003). Unlike tumor-induced muscle wasting in which REE and systemic inflammation are high, sarcopenia involves a low REE and a low level of chronic inflammation. Whether decreased muscle protein synthesis and/or increased protein degradation drives sarcopenia remains controversial at present. For instance, the IGF1/AKT1/mTOR signaling pathway has been shown to induce muscle protein synthesis and suppress protein breakdown through FoxO regulation (Milan et al. 2015). While muscle-expressed IGF1 declines with age in rodents and its reactivation can reduce sarcopenia (Musaro et al. 2001), this phenomenon was not observed in humans (Sandri et al. 2013). Instead, a compensatory but futile attempt at maintaining muscle mass, involving increased activation of biosynthetic pathways such as those driven by mTOR, occurs with aging in both rodents and humans (Kimball et al. 2004). During cachexia, muscle protein breakdown is increased through hyperactivation of the ubiquitin-proteasome and autophagy pathways; however, it either remains unchanged or is sometimes suppressed during sarcopenia (Welle et al. 2004). For instance, E3 ligases such as MuRF1 and MAFbx are up-regulated in many muscle-wasting states, including cancer cachexia, but not in aged muscles (Whitman et al. 2005; Sandri et al. 2013). Thus, the nature and degree of muscle protein breakdown represent a major difference between sarcopenia and cachexia. It has been postulated that autophagy may be an important mechanism driving sarcopenia (Carnio et al. 2014). Autophagy is required for the clearance of dysfunctional organelles and aberrant protein in cells, and its inhibition by muscle-specific deletion of Atg7 results in a premature aging phenotype in mice in which the development of sarcopenia is accelerated by mitochondrial dysfunction, increased oxidative stress and inhibition of neuromuscular structure and function (Carnio et al. 2014). Interestingly, alterations in neuromuscular junction genes were reported in a global muscle transcriptomic analysis in aged rats with sarcopenia that indicated a functional denervation of the neuromuscular junction (Ibebunjo et al. 2013). Autophagy-related genes are suppressed in the muscles of older, frail women (Drummond et al. 2014), suggesting a potential link between an autophagy defect and sarcopenia development. A comprehensive review of the mechanisms driving sarcopenia is available elsewhere (Larsson et al. 2019).

Sarcopenia is driven in part by a disruption of mitochondrial energy metabolism, which involves a decrease in mitochondrial mass, DNA content, oxygen consumption, and oxidative phosphorylation, as well as changes to mitochondrial morphology (Welle et al. 2003; Short et al. 2005; Gouspillou et al. 2014a; Leduc-Gaudet et al. 2015; del Campo et al. 2018). Muscles from sarcopenic rats exhibit repression of genes that regulate mitochondrial fission and fusion, oxidative phosphorylation, and the TCA cycle, along with a depletion of myofibrillar protein, increased apoptosis and declining muscle mass (Ibebunjo et al. 2013). Aged muscles were shown in a different study to be associated with dysfunctional mitochondria, impaired mitophagy, increased apoptosis, and lower mitochondrial ADP affinity (Gouspillou et al. 2014a,b). The increase in muscle-cell apoptosis appears to be triggered by the generation of reactive oxygen species from damaged mitochondria (Dirks et al. 2006; Barbieri and Sestili 2012), and since fusion and fission regulate mitochondrial morphology (Farmer et al. 2018), the reduced expression of the mitochondrial-fusion regulator optic atrophy protein 1 (OPA1) observed in sarcopenic patients may account for the morphological defects (Tezze et al. 2017). Indeed, Opa1 deletion in adult muscle results in smaller mitochondria with dilated cristae and increased oxidative stress, which culminates in muscle atrophy (Tezze et al. 2017).

Exercise and orally administered therapies show promise with respect to reversing sarcopenia. Clinical studies have demonstrated that regular endurance exercise can partially reduce aging-related mitochondrial dysfunction (Lanza et al. 2008) and that high-intensity resistance training counteracts physical frailty and improves muscle strength and size in elderly individuals (mean age of 87 yr) (Fiatarone et al. 1994). Moreover, long-term oral administration of nicotinamide mononucleotide, a key NAD⁺ intermediate, in aging mice has been shown to enhance mitochondrial respiratory capacity in skeletal muscle, to improve insulin sensitivity and to act as an overall antiaging intervention (Mills et al. 2016). Interestingly, the aging-related decline of the bone-derived hormone osteocalcin has been shown to correlate with reduced muscle mass (Mera et al. 2016a). Karsenty and colleagues showed that administration of exogenous osteocalcin is sufficient to increase muscle mass in aged mice and that deletion of osteocalcin, or the gene encoding its only known receptor GPRC6A, inhibits muscle protein synthesis and decreases muscle mass (Mera et al. 2016b). Osteocalcin signaling in muscle fibers is also necessary for adaptation to exercise by promoting uptake and catabolism of glucose and fatty acids, and exogenous osteocalcin is sufficient to restore the exercise capacity of 15-mo-old (aged) mice to that of 3-mo-old (young) mice (Mera et al. 2017). This bone-to-muscle endocrine axis could potentially be exploited to improve muscle function and exercise capacity in patients with sarcopenia.

CONCLUDING REMARKS

The importance of preserving lean muscle mass and function in cancer patients is often overlooked in the study of metastatic disease. Skeletal muscle wasting can be driven by aging, tumor, and/or anticancer therapy. Metastatic cancer patients can indeed experience a spectrum of these muscle-wasting states during the course of their disease and subsequent lifetime. The loss of lean muscle mass is almost invariably detrimental to patient outcome, treatment tolerance, quality of life, and survival and therefore represents a powerful predictor of morbidity and mortality (Baracos et al. 2018). It is important to acknowledge that a thin and emaciated physique cannot be relied upon as an identifying feature of cachexia or muscle wasting in obese cancer patients, where cachexia is often masked by fat gain accompanied by minimal changes to total body weight or BMI (Mourtzakis et al. 2008; Prado et al. 2016). Indeed, CT-based imaging analysis has significantly improved the diagnosis of lean body mass changes over time (Prado et al. 2016); however, this is not routinely practiced in oncology clinics. The challenge following the diagnosis of muscle wasting is devising strategies to reverse it. There is a significant gap in our understanding mechanisms of cachexia and as such, there are currently no approved treatments for cachexia. Several roadblocks are likely responsible (Penna et al. 2016). First, cachexia appears to be driven by multiple mediators, so single therapies targeting pathways individually may not be effective. Second, cachexia results from a systemic metabolic dysfunction, which is poorly understood and difficult to target therapeutically without adversely affecting normal physiology. Third, the limited success of prospective cachexia therapies in clinical trials is likely due to an insufficient understanding of the mechanisms that drive human cancer cachexia, the lack of clinical validation of experimental studies using patient samples, and the paucity of preclinical models of cachexia in the context of advanced cancer. However, as highlighted in this review, advances are being made in each of these areas with the development of promising targeted agents. It is important to generate awareness among the medical community that cachexia is not simply a problem of reduced appetite and it cannot be solved by excess dietary supplementation. It is also crucial to identify the antineoplastic agents that induce muscle wasting so that cancer treatment strategies can be devised to enhance success rate and reduce muscle toxicities and fatigue. As such, therapies that combine strong antitumor activity with the preservation of lean muscle mass currently hold the most promise for improving quality of life and survival in patients with metastatic cancer.

COMPETING INTEREST STATEMENT

Both authors declare no conflicts of interest and are not aware of any affiliations, funding, memberships, or financial holding that might be affecting the objectivity of this review.