Mechanisms of Ovarian Cancer-Associated Cachexia

Abstract

Cancer-associated cachexia occurs in 50% to 80% of cancer patients and is responsible for 20% to 30% of cancer-related deaths. Cachexia limits survival and treatment outcomes, and is a major contributor to morbidity and mortality during cancer. Ovarian cancer is one of the leading causes of cancer-related deaths in women, and recent studies have begun to highlight the prevalence and clinical impact of cachexia in this population. Here, we review the existing understanding of cachexia pathophysiology and summarize relevant studies assessing ovarian cancer–associated cachexia in clinical and preclinical studies. In clinical studies, there is increased evidence that reduced skeletal muscle mass and quality associate with worse outcomes in subjects with ovarian cancer. Mouse models of ovarian cancer display cachexia, often characterized by muscle and fat wasting alongside inflammation, although they remain underexplored relative to other cachexia-associated cancer types. Certain soluble factors have been identified and successfully targeted in these models, providing novel therapeutic targets for mitigating cachexia during ovarian cancer. However, given the relatively low number of studies, the translational relevance of these findings is yet to be determined and requires more research. Overall, our current understanding of ovarian cancer–associated cachexia is insufficient and this review highlights the need for future research specifically aimed at exploring mechanisms of ovarian cancer–associated cachexia by using unbiased approaches and animal models representative of the clinical landscape of ovarian cancer.

Cachexia, a multifactorial and multiorgan comorbidity of cancer, negatively impacts quality of life, reduces treatment tolerance and drives higher morbidity and mortality rates (1-5). A major component of cachexia is the wasting of skeletal muscle tissue (often referred to as sarcopenia), which also correlates with poorer survival. Cachexia is generally underexplored, especially in a setting of ovarian cancer (OC), which is the deadliest of gynecologic malignancies, thereby representing a major gap in the understanding of this deadly syndrome. Despite difficulties in diagnosing changes in muscle mass due to ascites formation and bloating, sarcopenia plays a critical role as a prognostic factor for survival in patients with OC (6-8). Here we review the current understanding of OC-associated cachexia by summarizing the current clinical and preclinical literature and by highlighting pathophysiologic mechanisms and relevant signaling pathways involved in promoting the onset of this deadly condition.

Cachexia Epidemiology and Clinical Relevance

Definition, Diagnosis, and Treatment of Cancer-Associated Cachexia

Cachexia is a wasting condition associated with a variety of diseases such as cancer, chronic obstructive pulmonary disease, AIDS, and cystic kidney disease. In cancer, cachexia was originally defined as “a multifactorial syndrome characterized by an ongoing loss of skeletal muscle mass (with or without loss of fat mass) that cannot be fully reversed by conventional nutritional support and leads to progressive functional impairment” (9). Due to its multifactorial nature, cachexia can be difficult to diagnose, especially in individuals with cancer. In an attempt to fill this gap, a framework was established for diagnosing cachexia in subjects affected with cancer and was based on 3 criteria: (1) weight loss >5% over 6 months; (2) body mass index <20 and any degree of weight loss >2%; or (3) presence of sarcopenia (males <7.26 kg/m², females <5.45 kg/m²) (9). Given the elevated average age of patients with cachexia and co-occurrence of obesity and/or large tumors that can hide changes in overall body weight and composition, use of measures such as skeletal muscle radiation attenuation, which also take into account muscle quality, are increasingly being applied as tools to interrogate aspects of skeletal muscle biology (10, 11) and contributed to establishing diagnostic criteria that could determine presence or absence of cachexia in patients, often retrospectively. The same measures could be used to assign a stage to the condition of cachexia, thereby leading to the distinguishing of stages such as precachexia, cachexia, and refractory cachexia (9). It is worth noting that, given the absence of reliable biomarkers for early diagnosis of cachexia, these stages are currently defined essentially on the basis of a patient's clinical history and prognosis. Research is underway to determine other metrics clinically available and feasible that may be informative for cachexia diagnosis and survival outcomes.

Despite recent progress, to date there are no approved therapeutic interventions for the treatment of cachexia, with the only exception of Japan, which approved the ghrelin receptor agonist anamorelin for use in certain cancers in 2020 (12). Recently, the American Society for Clinical Oncology provided evidence-based guidance for adults with advanced cancer, suggesting, in certain contexts, dietary counseling and progesterone or corticosteroid use to improve appetite or weight gain (13). However, given a lack of robust preclinical and clinical data, pharmacologic agents directly targeting molecular mechanisms of the condition are not recommended.

Prevalence, Mortality, Morbidity

It is estimated that cachexia affects over 80% of patients diagnosed with cancer worldwide, and as many as 20% to 30% of cancer-related deaths are directly attributed to the development of this syndrome (14). Based on traditional diagnosis, cachexia is more prevalent in certain cancer patient populations, including patients with pancreatic, lung, and colon cancer (15). Regardless of specific disease or cancer type, cachexia associates strongly with poor prognosis and reduced quality of life. In patients with cancer, cachexia also predicts worse response to chemotherapy, compounded by the fact that certain anticancer therapies independently exacerbate symptoms and mechanisms underlying the wasting condition (16-18). Despite recent progress in diagnosis and treatment, cancer-associated cachexia remains a major unmet medical need worldwide.

Phenotype and Mechanisms of Cachexia

Traditionally, cachexia was characterized as a wasting syndrome mainly resulting from muscle and fat depletion. There is now general agreement on the idea that cachexia is rather a multifactorial and multiorgan syndrome that promotes systemic effects and impacts the entire body (Fig. 1). Here we summarize the phenotype of cachexia and highlight the underlying pathophysiologic mechanisms.

Figure 1.

A summary of the cachexia phenotype. Cachexia is induced during cancer by tumor burden and anticancer therapy, impacting a variety of tissues including skeletal muscle, adipose tissue, bone, brain, liver, and stomach. In these tissues, various structures and functions are deregulated, generating the cachectic phenotype. This figure was created using BioRender.com

Skeletal Muscle

Skeletal muscle remains a primary focus of cancer-associated cachexia research since muscle mass, quality, and function can all determine outcomes such as response to therapy and even overall survival. Indeed, cachexia leads to several derangements in skeletal muscle, including increased inflammation, disrupted metabolism, dismantling of neuromuscular junctions, fibrosis, intramuscular fat accumulation, and skeletal muscle atrophy which precipitate dysfunction. Gene expression analysis of skeletal muscle during cachexia reveals increased expression of a variety of genes involved in protein catabolism including the ubiquitin-mediated proteasome (UPS) and autophagy–lysosome system, consequently termed atrogenes (19).

Historically, UPS has been recognized as the main driver of muscle protein degradation and involves ubiquitin-activating enzyme, ubiquitin-conjugating enzyme, ubiquitin ligase, protease, and their substrates. Muscle-specific ubiquitin ligase ligases Atrogin-1 and MuRF1 have been shown to be strongly involved in skeletal muscle atrophy in patients and preclinical cancer-associated cachexia models. These atrogenes are promoted by several signaling pathways including signal transducer and activator of transcription 3 (STAT3), nuclear factor kappa-light-chain-enhancer of activated B cells, fork head box O (FOXO), mothers against decapentaplegic 2/3 (SMAD2/3), and mitogen-activated protein kinase cellular signaling (14, 20-24). These pathways are stimulated by a variety of factors including interleukin (IL)-1) (25-27), IL-6 (28-35), IL-8 (21, 36-39), and tumor necrosis factor alpha (TNFα) (40, 41), which may be released by the tumor itself or by the host in response to the tumor.

Autophagy is another mechanism utilized for removal of damaged organelles and dysfunctional proteins, making it critical for normal tissue homeostasis. During cancer-associated cachexia, this system is often overactive in skeletal muscle and along with several major signaling pathways including FOXO, mammalian target of rapamycin (mTOR), p38, SMAD1/5/8, and mitogen-activated protein kinase (42). FOXO3 is suggested to be a critical regulator of autophagy during cancer-associated cachexia given its inhibitory effects on phosphoinositide 3-kinase/AKT (Protein Kinase B/PKB) signaling and well-documented overexpression in skeletal muscle during muscle wasting (43).

While protein catabolism is stimulated in muscle during cancer-associated cachexia, anabolism is often inhibited through several distinct mechanisms. One of the main pathways involved in anabolism is mTOR and adenosine monophosphate–activated protein kinase signaling. mTOR contains 2 distinct complex mTORC1 and mTORC2. mTORC1 regulates protein anabolism and is critical for muscle autophagy (44). mTORC1 also stimulates mitochondrial biogenesis by activating eukaryotic translation initiation factor 4E binding protein 1 and increasing peroxisome proliferator-activated receptor gamma coactivator 1-alpha expression (45, 46). Inhibition or deletion of mTORC1 can lead to skeletal muscle wasting, although it appears dispensable for maintenance of skeletal muscle in adult animals (47, 48). In early stages of cancer-associated cachexia, protein synthesis is impaired, and this is correlated with reduced insulin-like growth factor (IGF1)–mTOR signaling (49). During late stages of cancer-associated cachexia, mTORC1 can be inhibited by IL-6 due to adenosine monophosphate–activated protein kinase activation (50). Despite this, activity of mTORC1 may contribute to sarcopenia since mTORC1 signaling is elevated in muscle during advanced age in both mice and humans (51). Further, overexpression of mTORC1 in skeletal muscle causes a sarcopenic molecular signature, suggesting it may contribute to muscle wasting that occurs with age (48). Therefore, mTORC1 could be a potential therapeutic target for treatment of age and/or cancer related muscle wasting.

During cancer-associated cachexia, IGF1 levels are often lower and skeletal muscle can become insulin resistant (52-55). IGF1 and insulin are factors that bind to the IGF1 receptor and activate phosphoinositide 3-kinases stimulating protein synthesis via AKT–mTOR. Further, activation in these pathways can diminish FOXO activity, reducing protein degradation (56-58). IGF1 is primarily produced by the liver, whereas insulin is primarily produced by the pancreas. Decreased insulin abundance is linked to reduced protein synthesis in cancer patients likely due to decreased glucose tolerance and insulin sensitivity (54).

The transforming growth factor β (TGFβ) superfamily of ligands is known to have both beneficial and detrimental effects on skeletal muscle size (59). For instance, compared with healthy controls bone morphogenic protein (BMP) was found to be reduced in patients with cancer and linked to reduced muscle mass. Factors that become elevated during cancer such as IL-6 and activin A can stimulate expression of BMP inhibitor Noggin in skeletal muscle (60). The reduction in BMP signaling in muscle and motor nerves was found to cause disruptions at the neuromuscular junction, denervation, and eventually muscle wasting, which we reported in models of cancer- and chemotherapy-driven cachexia (61). BMP7 also has similar functions where its overexpression contributes to AKT–mTOR activation and inhibition can stimulate muscle wasting (62, 63). Both BMP and BMP7 activate Smad1/5/8 signaling. Conversely, myostatin and Activin A are also TGFβ family members and are negative regulators of muscle growth (63-66). Deletion of myostatin causes muscle growth and overexpression leads to skeletal muscle atrophy (63, 65). Both myostatin and activin A stimulate Smad2/3 signaling by engaging with activin receptor type 2A. Therefore, this family of proteins exerts both hypertrophic and atrophic signaling and regulating TGFβ family signaling has shown promise as a countermeasure to muscle wasting during cancer.

Adipose Tissue

Adipose tissue wasting and browning of white adipose tissue (WAT) occur alongside muscle wasting during cancer-induced cachexia (67). Adipose tissue can be divided into 2 distinct types, WAT and brown adipose tissues. WAT browning promotes lipolysis and alters the secretome of adipose depots to induce protein catabolism (68). During WAT browning, uncoupling protein 1 is upregulated which increases thermogenesis leading to elevated energy expenditure (69, 70). Lipolysis usually occurs prior to WAT browning in cancer-associated cachexia and both processes are enhanced by inflammatory factors (eg, IL-6) (34, 35, 71).

Bone

The musculoskeletal system is compromised during cancer, and new research highlights the role of bone remodeling in cancer-associated cachexia where it contributes to muscle wasting, frailty, and risk of injury (72-74). In fact, given the close proximity of muscle and bone, numerous myokines and osteokines have been identified that contribute to crosstalk between bone and muscle (75). Cancers that metastasize to the bone induce muscle weakness due to bone-derived factors (76). Interestingly, cancers without bone metastasis can cause bone weakness and remodeling (5, 77, 78). Several bone-derived molecules, termed osteokines, have been shown to exert catabolic and anabolic impact on skeletal muscle. For instance, bone-derived Ihh and osteocalcin have been shown to have positive impacts on skeletal muscle size and function whereas TGFβ, activins, FGF23, and FGF12 drive muscle dysfunction. Alternatively, muscle-derived factors, termed myokines, have similar impacts. For instance, IGF1 and FGF2 produced by the muscle can stimulate bone formation (75). Certain factors, including receptor activator of nuclear factor kappa beta ligand (RANKL), from the tumor can also impact both muscle and bone further dysregulating their relationship whose contributions to muscle and bone dysfunctions in cancer will be discussed later in this review (79).

Gut Dysfunction

The gastrointestinal system and microbial communities that inhabit it are strongly affected during cancer and can contribute to the onset of cachexia. This is proposed to occur through several mechanisms including gut barrier dysfunction and alterations to microbiome composition (80, 81). Indeed, gut barrier dysfunction has been detailed in several preclinical cancer-associated cachexia models likely due to disruptions of tight junction, increased gut permeability, and local immunosuppression (67, 82-84). Consequently, gut microbiota is likely a source for Inflammation during cancer due to dislocation of pathogen associated molecular patterns. Interestingly, lipopolysaccharide can influence muscle cell catabolism via Toll Like Receptor 4. Finally, the microbiome composition can change during cancer leading to outgrowth of opportunistic bacteria such as Klebsiella oxytoca (85). Microbial composition has been reported to be altered in several models of cancer-associated cachexia and there is some evidence in mouse models that prebiotics and probiotics may be useful at preventing muscle and fat wasting while lengthening cancer survival and this was attributed to modulation of the gut microbiota composition (86).

Brain and Anorexia

The brain is a critical mediator of systemic health and exerts many functions that regulate systemic metabolism. Hypothalamic inflammation has been shown to drive lost appetite or nausea that results in lower food intake or anorexia (87). Anorexia is a major comorbidity of cachexia in cancer and contributes to the systemic energy deficit that stimulates wasting (88). The hypothalamus controls appetite by integrating signals from various organs in the body with hypothalamic centers. There are 2 different neuronal populations that mediate appetite including the anorexigenic pro-opiomelanocortin and cocaine and amphetamine-regulated transcript expressing neurons, as well as orexigenic neuropepetide Y/agouti-related peptide neurons (89). For instance, ghrelin, a gut-derived hormone that has been widely investigated as a therapeutic target, can increase appetite by suppressing pro-opiomelanocortin and stimulating neuropepetide Y/agouti-related peptide neurons (90). During cancer-associated cachexia the orexigenic axis is dysfunctional and anorexigenic axis is overstimulated leading to appetite suppression (91). Although decreased food intake is an important feature of cachexia, clinical trials have demonstrated that nutritional interventions to address underlying anorexia in cancer patients was insufficient to reverse the wasting (13). The hypothalamus also has major roles in regulation of energy expenditure and a large portion of cancer patients have elevated resting energy expenditure likely due to inflammatory factors such as TNFα that contribute to muscle and fat wasting (92). The hypothalamic pituitary is also disrupted during cancer by proinflammatory factors leading to an increase in glucocorticoid production by the adrenal gland. Glucocorticoids are known to be elevated during cancer and directly cause skeletal muscle wasting (92-94).

Liver

The liver is a major metabolic organ that is impacted by tumor burden and can contribute to cancer-associated cachexia through several mechanisms. Many proinflammatory molecules upregulated due to tumor burden (eg, serum amyloid A, C-reactive protein, fibrinogen) and involved in activation of acute phase response (APR) (95) are known to influence inflammatory signaling in the liver leading to lost insulin and glucose homeostasis (96, 97), circadian homeostasis (98), fat accumulation (99), and bile transport (100), further disrupting the tissues function. The liver is also responsible for acute phase response to tissue injury and inflammation. Muscle breakdown to free amino acids to facilitate APR activity results in lower muscle mass and this is thought to be mediated by IL-6-STAT3 (95). Activation in APR therefore results in increased resting energy expenditure. Ketogenesis has also been reported to be aberrant in livers from cancer-associated cachexia models (101, 102). Ketones are produced by the liver and can be used for energy generation in a variety of tissues including skeletal muscle. Low levels of ketones have been reported in cancer models and are thought to contribute to the systemic energy imbalance and elevated glucocorticoid signaling, both of which worsen muscle atrophy by stimulating catabolic signaling and inhibiting anabolic signaling (103).

Sex Organs and Hormones

Differences between male and females including body composition, fat distribution, metabolism, and hormones have been reported in both humans and animals. It was reported that muscle depletion is more prevalent in male vs female cancer patients (104). These male patients also demonstrate higher fatigue and greater reduction in force than female patients with cancer. This is potentially due to estrogen levels in females. Estrogen has been shown to regulate muscle function and anabolism (105). Estrogen can also impact the hypothalamus resulting in altered physical activity and resting energy expenditure (106). Hormone replacement therapy can actually mitigate wasting in postmenopausal women (107). In contrast, testosterone is an anabolic agent, although testosterone can improve lean body mass but not impact overall survival in patients (108, 109). Also, certain treatments are only effective in male or female mice with cancer, and this suggests that sex may be a critical consideration for the treatment of cachexia (110). Notably, most preclinical cachexia data is produced in male mice, thereby prompting for future studies investigating the role of sex and hormone activity in cancer-associated cachexia.

Chemotherapy

While cancer itself can cause cachexia, certain anticancer treatments also drive muscle and fat wasting, acute sickness behavior, and elevate levels of inflammatory factors. Chemotherapies, in particular, are known for their dose-limiting off-target effects that complicate their clinical utility including nausea, nephrotoxicity, fatigue, and weakness. Studies have identified the mechanism responsible for chemotherapy-induced muscle, fat, and/or bone wasting. Interestingly, many mechanisms responsible for these processes in cancer seem to also contribute to chemotherapy-induced cachexia. In such regard, we and others have shown that routinely administered chemotherapeutics can drive occurrence and sustainment of a cachexia-like state, along with loss of muscle mass and function (16, 111), and can overall exacerbate cachexia in cancer hosts (17). Weight instability is also a major component of patient assessment, and failure to maintain weight often results in less treatment options and reduced treatment capacity (112). This further highlights the necessity for generating treatments that can protect against cancer and cancer-therapy induced wasting to improve overall survival and quality of life for people with cancer.

Cachexia in Ovarian Cancer

Ovarian Cancer Epidemiology and Treatment

OC comprises a group of diseases that can originate in the ovaries, within the fallopian tubes, or the peritoneum. OC is the fifth leading cause of cancer-related deaths in women and the leading cause of cancer-related deaths for all gynecologic malignancies. In the United States, it is estimated that 19 710 new cases of OC will be diagnosed, whereas 13 270 women will die from the disease in 2023 (113). OC most often affects postmenopausal women, and the majority of OCs are diagnosed at advanced stages, therefore leading to disease recurrence, chemoresistance, and poor outcomes (114). The overall 5-year survival rate for OC from 2013 to 2019 is 50.8%, though it strongly depends on the stage of disease at diagnosis and is dramatically influenced by its progression (113). While women affected with localized (∼18% of cases) or regional disease (∼20% of new diagnoses) present with 92.4% and 72.9% survival rates, respectively, in the case of distant, metastatic disease (making up to 55% of new cases, often consequential to late diagnosis and to the onset of chemoresistance) survival rates are drastically lower at 31.5% (113). Epithelial OC represents about 90% of new diagnoses and the majority are considered high-grade serous carcinoma (115), being characterized by near ubiquitous TP53 mutations, DNA repair deficiencies, high propensity for chromosomal copy numbers, and a worse prognosis (116, 117). The standard of care for OC regardless of stage involves the administration of platinum-based (eg, carboplatin, cisplatin) and taxane-based (eg, paclitaxel, docetaxel) drugs (118, 119). These chemotherapies are often given as a combination therapy and can be delivered by intravenous or intraperitoneal injection. Intraperitoneal injection can improve outcomes but also exacerbates treatment related toxicity and is not recommended unless tumors are optimally debulked (120, 121).

Clinical Assessment of Cachexia in Ovarian Cancer

Diagnosis of cachexia in OC has long been neglected and/or understudied due to the typical presentation of the disease, associated with ascites formation and bloating, often masking the underlying weight loss and muscle condition. However, recent data from clinical and preclinical research is now highlighting the burden of cachexia also in patients with OC (6-8). Despite limitations associated with the low number of studies assessing cachexia and/or sarcopenia, based on meta-analysis of clinical studies there is now general consensus on the idea that loss of skeletal muscle mass is a frequent feature accompanying the onset of OC (122-128) (Table 1).

Table 1.

Association of sarcopenia with survival in patients with ovarian cancer

Reference info	Study inclusion			Sarcopenia diagnosis			Outcomes
Author, year	OC stage	N	Age	Muscle mass assessed by CT scan at	Sarcopenic cut-off	% Sarcopenia	Survival	Notes
Ataseven et al, 2018	III-IV	323	NR	L3	SMI < 41 cm2/m2	29.40%	HR 1.14, P = .730 (OS)
					SMD < 32 HU	21.1%	HR 1.32, P = .003 (OS)
Aust et al, 2015	III-IV	140	60	L3	SMA < 39 HU	35%	HR 2.25; P = .028 (OS)	Eotaxin and IL-10 correlate with low SMA
					SMI < 41 cm2/m2	28.90%	HR 1.23; P = .565 (OS)
Bronger et al, 2017	III-IV	105	63	L3	SMI 38.5 cm2/m2	11.40%	HR 3.17; P = .012 (OS)	Change in muscle mass over time not associated with survival
							HR 2.64; P = .012 (PFS)
Bruno, 2021	I-IV	239	NR	L3	SMI < 38.9 cm2/m2	35.10%	HR 1.2, P > .05
					SMD < 21.24 HU	50%	HR 2.66, P < .05
Chae et al, 2021	I-II	82	52	L3	SMI ≤ 38.7 cm2/m2	20.7	HR 58.4; P = .0008 (OS)	BMI not associated with survival
Huang et al, 2020	III	139	54.4	L3	SMI 39.0 cm2/m2	34.10%	HR 1.08; P = .002 (OS)	SMI loss associates with worse OS and PFS
							HR 1.03; P = .04 (PFS)
Grande et al, 2021	III-IV	69	63.6	L3	SMI < 41.0 cm2/m2	29%	HR .95; P > .05 (OS)	SMA was associated with early discontinuation of chemotherapy
Kim et al, 2020	III-IV	179	57.5	L3	SMI < 39.0 cm2/m2	42.50%	HR .87; P = .636 (OS)	High fat to muscle ratio showed significantly worse OS
							HR 1.29; P = .157 (PFS)
Kumar et al, 2016	III-IV	296	64.6	L3	SMI < 39 cm2/m2	44.60%	HR .99, P = .97 (PFS)
					SMD < 36.4 HU	NR	HR 1.26, P = .0009 (PFS)
Matsubara et al, 2019	III-IV	92	55.3	L3	SMA < 92.92 cm2	50%	HR 2.186; P = .030 (OS)	Psoas major volume was superior to SMA for prognosis prediction
							HR 1.272; P = .402 (PFS)
Nakayama et al, 2019	III-IV	94	61.8	L3	SMI 30.88 cm2/m2	77%	P = .337 (OS)
Rutten et al, 2017	II-IV	216	63.1	L3	SMI 38.73 cm2/m2	32.40%	HR 1.36; P = .076	Low SMI shows trend towards association with poor OS
Rutten et al, 2016	III-IV	123	66.5	L3	SMI 41.5 cm2/m2	50.40%	HR .89; P = .613	Loss of SM during chemotherapy associates with worse OS
Staley et al, 2020	III-IV	201	63.6	L3	SMI ≤ 41cm2/m2	64%	P > .05 (OS)	Trend toward worse neutropenia was noted in the sarcopenic group
							P > .05 (PFS)
Ubachs et al, 2020	III	212	60.9	L3	SMI > 2%/100 days	58%	HR 1.41; P > .05	Patients with SMI loss demonstrate more perioperative adverse events
Wood et al, 2023	II-IV	174	64.1	L4	SMI 38.0 cm2/m2	55.70%	P = .95 (OS)	> 2% decrease in VAT per 100 days was significantly associated with a decreased OS
							P = .68 (PFS)
Yoshikawa et al, 2021	I-IV	72	62	L5	PMI < 5.4 cm2/m2	NR	HR 3.87; P = .0098 (OS)
Yoshino et al, 2020	I-IV	60	63.5	L3	SMI 39.0 cm2/m2	60% (pre), 68% (post)	HR 3.17; P = .022 (OS)	Post-IC neutrophil count correlates with SMI

Clinical retrospective and prospective studies were collected that looked at associations of OS or PFS with sarcopenia defined by SMI, SMD, SMA, or PMI. Multivariate statistics were reported when available.

Abbreviations: IC, immune complex; NR, not recorded; OS, overall survival; PFS, progression-free survival; PMI, psoas muscle index; SMA, skeletal muscle attenuation; SMD, skeletal muscle density; SMI, skeletal muscle index.

Symptoms associated with cachexia such as physical frailty, weakness, and early satiety are frequently reported by patients with OC. Further, a major component of cachexia is sarcopenia, which is muscle wasting and dysfunction that can occur independent of age and is easily assessed using computed tomography (CT)-based metrics, as we recently showed in a cohort of women affected with OC (79). The impact of sarcopenia on survival and treatment outcomes in OC is still partially controversial, with recent meta-analyses reaching different conclusions (122, 123, 128, 129), whereas other studies suggest that skeletal muscle wasting correlates with overall survival and progression free survival especially in patients with advanced disease (130-138). Some discrepancies in defining inclusion criteria and skeletal muscle condition may explain misalignment in conclusions, as in the case of recent studies, which failed to see similar associations between sarcopenia and survival outcomes (139-146). The data is further complicated by the varied definitions used to clinically define sarcopenia. Indeed, different CT-based metrics have been utilized such as skeletal muscle index (SMI), skeletal muscle density (SMD), psoas muscle index (PMI), and skeletal muscle attenuation (SMA), thus contributing to generating confusion. It is also important to note that body mass index, which is often used for the diagnosis of cachexia, does not always correlate well with markers of skeletal muscle loss and should be utilized with caution (131, 133). Current evidence also suggests a negative impact of obesity on cancer progression, though this is still being debated in the case of OC, where body composition, and in particularly the presence of high visceral adiposity (with or without sarcopenia) was found to associate with worse survival (147). The staging of OC may also be important to look at association of sarcopenia with outcomes, since pooling of higher stage studies (III-IV) was found to identify higher risk of mortality linked to low SMI (128).

Of note, several studies also highlighted considerable heterogeneity in treatment outcomes related to the extent of sarcopenia (8, 130, 134, 138, 141, 148, 149), mainly as a consequence of the fact that the loss of skeletal muscle was assessed at different stages of the disease relative to the start of anticancer therapy (eg, prior, after, or during surgery and/or chemotherapy). Nonetheless, the currently available data suggests that the loss of skeletal muscle during surgery and chemotherapy is an important indicator of outcomes in OC (134, 150). In particular, patients with OC with sarcopenia were found to be more likely to receive neoadjuvant chemotherapy (ie, chemotherapy prior to cytoreductive surgery) with cytoreduction instead of primary cytoreductive surgery alone, and this has been shown to reduce morbidity and mortality following treatment (151, 152). Similarly, a meta-analysis of 35 articles involving 6894 patients also showed that sarcopenia was present in ∼39% of patients with OC prior to initiation of chemotherapy treatments, and it was significantly and independently associated with postoperative complications, chemotherapy-induced toxicity and poor survival (153).

Lastly, systemic inflammation, despite being well known as a major driver of muscle wasting during cancer, was assessed in relation with sarcopenia in only a modest number of OC studies, though a recent one identified a clear association between malignant ascites and muscle loss after primary debulking surgery and adjuvant chemotherapy in advanced-stage OC (79, 131, 154). In a recent study, we showed that RANKL (Tnfsf11), a TNF superfamily member and master regulator of bone remodeling normally responsible for osteoclast differentiation and osteoclastic bone resorption (155), was elevated in the blood of 31 women with OC compared with women affected with benign gynecologic conditions, consistently with increases in circulating levels of C-telopeptide of type I collagen, a marker of bone turnover. Interestingly, RANKL expression is aberrantly expressed in many solid tumors, including breast, prostate, liver, gastric and ovarian (156-159), and was described to correlate with poorer clinical outcomes in patients with OC (159). Consistently, in our report we observed increased expression of RANKL in different OC histological types and showed that patients with OC with low RANKL–expressing tumors correlated with improved overall survival.

Cachexia in Preclinical Models of Ovarian Cancer

Preclinical and basic cancer-associated cachexia research has historically relied on certain tumor models that recapitulate the wasting phenotype in mice. Much of our molecular understanding of cancer-associated cachexia is gleaned from these models, which have predominately looked at colorectal, lung, and pancreatic cancer models, usually in male mice only. Because of this, research related to the OC field remains inadequate, particularly due to the limited availability of optimal preclinical models that are characterized in terms of muscle and bone phenotype or function. Indeed, until recent, only 1 OC xenograft, the mouse bearing the TOV-21G clear cell carcinoma cell line, a distinct pathologic subtype of epithelial OC only accounting for less than 5% of all ovarian malignancies, had been described and partially characterized for cancer-associated effects on skeletal muscle (160).

To date, new OC mouse models have emerged that can be utilized to shed light on mechanisms and potential targets for therapy (Table 2). We were the first to show that implantation of ES-2 OC cells caused ascites accumulation in the peritoneal cavity, accompanied by severe cachexia, as suggested by marked body weight loss, as well as by concurrent muscle and bone wasting (5, 79). It is important to note that although the ES-2 cells were originally classified as ovarian clear cell carcinoma (161), their genomic profiling followed by comparison with the most extensively studied high grade serous OC (HGS-OC) histotypes (eg, SK-OV-3, OVCAR-3, CAOV-3) recently led to reclassification of the ES-2 cells as a suitable model for the study of HGS-OC (162, 163). Our model also demonstrated that cachexia-associated factors such as IL-6 were elevated in plasma and, at a greater extent, in ascites (5, 79). Interestingly, ascites fluid accumulation has been suggested to be a contributing factor to cachexia in OC, potentially by causing early satiety and driving elevated levels of inflammatory factors (23). In a recent study, patient ascites was used to treat C2C12 myotubes in vitro. Myotubes demonstrated reduced protein synthesis in response to patient ascites, and this reduction correlated with SMI assessment in patients with OC. These data also suggest that ascites itself can promote the appearance of sarcopenia potentially by reducing protein synthesis directly in skeletal muscle (164).

Table 2.

Summary of preclinical ovarian cancer cachexia studies

	Cell culture				Animals
Authors	OC cell lines	Phenotype	Biomarkers	Outcomes	N	Mouse strain	Tumor burden (days)	cell #	Biomarkers	Muscle wasting	Muscle pathways (RNA)	Muscle pathways (WB)	Main finding
Kim-Muller et al, 2023	TOV-21G	NA	NA	NA	12-15	BALB/C	48	5 × 106 SC	↑ GDF15	↓ Lean and Hindlimb mass, ↓ TA and Gast CSA	↑ Atrogin-1, MuRF-1	NA	GDF15 Neutralization attenuates muscle wasting and dysfunction
Pin et al, 2019	ES2	↓ Myotube diameter	↑ IL-6	↑ pSTAT3/STAT3	6-10	NSG	15	1 × 107 IP	↑ IL-6	↓ TA, Gast, Quad, Heart mass; ↓ TA CSA	↑ Atrogin-1, MuRF-1, Fbxo31	↑ p-P38, pSTAT3	ES2 caused bone and fat wasting, IL-6 higher in ascites than plasma
				↓ OPA1, PGC1a, Cytochrome C								↓ AKT1/2, OPA1, Mitofusin-2, Cytochrome-C
Pin et al, 2021	ES2	↓ Myotube Diameter	↑ IL-6 and RANKL		5-7	NSG	15	1 × 107 IP	↑ IL-6 and RANKL	↓ TA, Gast, Quad, Heart mass; ↓ TA CSA	↑ Atrogin-1, MuRF-1	NA	RANKL neutralization lessens bone and muscle wasting during ES2 tumor burden
Straughn et al, 2020	A2780	NA	NA	NA	10	NSG	28	8 × 105 IP	NA	↓ Forelimb and Total grip strength,	↓ MyoD, Ern1 (Quad)	↓ P62 (Quad)	WFA attenuates muscle mass loss and activation in ALS and UPS signaling
										↓ TA mass and CSA, GA mass, QF Mass	↑ RelA, IKKB, Pax7, EIF2a, Atf4, Ddit3, Ppplr15a, Hspa5, Hsp90b1, Tnfrsf10b; Traf6, Fbx030, Fbx032, Trim63, Sqstm1, Map1lc3b, and Becn1	↑ pP65, pERK, IRE1a, Ubiquitin LC3, Beclin1 (Quad)
Ubarchs et al, 2021		↓ Protein synthesis	↑ IL-6 and IL-8 (ascites)	↑ NF-kB	15	NA	NA	NA	↑ IL-6 and IL-8 (ascites)	NA	↑ NF-kB (C2C12)	NA	6 sarcopenic patients showed significant, lower SMI which correlated to the loss of protein synthesis in the in vitro model.
Yi Luan et al, 2022	NA	NA	NA	NA	6	Pik3ca	65 and 83	NA	↑ Activin A, GDF15, IL-6, IL-1B, TNFα	↓ Lean mass, TA CSA, TA, Gast, and Quad mass	↑ Murf1, Atrogin1, Lc3, Myogenin	↑ FOXO3 and P38 phosphorylation
											↓ Pax7
Zhou et al, 2010	TOV-21G	NA	↑ Activin A	NA	12	BALB/C	NA	5 × 106 SC	↑ Activin A	↓ Lean and Gast mass	NA	NA	ActRIIB Antagonisms attenuates muscle wasting and improves survival

Preclinical studies were compiled and summarized from various mouse models of cancer cachexia. Disrupted muscle phenotype and signaling pathways as well as tumor- and host-derived biomarkers were recorded.

Abbreviations: ALS, autophagy-lysosomal system; FOXO, fork head box O; IL, interleukin; IP, intraperitoneally; NA, not available; NF-kB, nuclear factor kappa-light-chain-enhancer of activated B cell; OC, ovarian cancer; PGC1, peroxisome proliferator-activated receptor gamma coactivator 1-alpha; SC, subcutaneously; SMI, skeletal muscle index; STAT, signal transducer and activator of transcription; TNF, tumor necrosis factor; UPS, ubiquitin-mediated proteasome; WFA, withaferin A.

In a recent study employing the same model, we also showed that ES-2 OCs express high RANKL and that OC hosts present high circulating level of this factor, whereas its neutralization by means of anti-RANKL monoclonal antibodies was sufficient to mitigate muscle and bone decline in ES-2 tumor bearing mice (79). Altogether, these observations contributed to identifying RANKL as a novel therapeutic target for the treatment of musculoskeletal complications associated with nonbone metastatic OCs.

In a work by Luan et al genetically engineered mice presenting constitutive activation of phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha (Pik3ca) were shown to develop OC accompanied by ascites formation and occurrence of cancer-associated cachexia identified by muscle and fat wasting. Interestingly, in this model elevated activin A and growth differentiation factor 15 (GDF15), known procachexiogenic factors (64, 165) were also observed in serum (166). This is not the first evidence that links OC growth with the onset of activin A- or GDF15-driven cachexia. Indeed, Zhou et al previously reported that growth of TOV-21G promotes progressive body and muscle weight loss, whereas administration of ACVR2B/Fc, an inhibitor of the activin receptor 2B signaling known to preserve muscle mass and prolong survival in tumor hosts and to increase muscle and bone mass following chemotherapy treatment (167), was shown to restore muscle mass in this model (160). The therapeutic potential deriving from targeting the activin-dependent signaling was further supported by evidence from Pettersen et al, showing a functional link between the activin A and IL-6 signaling pathways in causing cachexia in OC (168). In the same model, recent observations by Kim-Muller and colleagues (169) showed that treatment with anti-GDF15 antibody mAB2 promotes body weight gain with near-complete restoration of muscle mass and markedly improves muscle function and physical performance.

Using another model, Straughn et al reported that inoculation of ovarian surface epithelium cancer cell line A2780 into female mice lead to development of cachexia (170, 171), whereas treatment of A2780 tumor-bearing mice with Withaferin A, a steroidal lactone purified from the Withania somnifera plant known for its anti-inflammatory properties and inhibitory effects on cell proliferation and invasion of OC (172), was sufficient to ameliorate cachexia via nuclear factor kappa-light-chain-enhancer of activated B cells suppression (170). Interestingly, in a follow up study, Withaferin A was then demonstrated to cause muscle hypertrophy likely due to its regulation of myogenic progenitors and proteolytic systems in skeletal muscle (171).

Conclusions

Cancer-associated cachexia remains a substantially neglected area of research and represents a significant challenge for patients affected with cancer and for their caregivers, mainly because of poorer response to anticancer therapies, worsened quality of life and increased morbidity and mortality rates. While the mechanisms responsible for the onset of cachexia are generally well studied in pancreatic, lung, and colorectal cancers, less is known about the effects of cachexia in OC, which represents a leading causes of gynecologic cancer-related deaths worldwide.

As we reported herein, this is primarily a consequence of difficulties in diagnosing changes in muscle mass in subjects affected with OC, and mainly results from ascites formation and bloating, which ultimately contribute to masking the progressive decreases in skeletal muscle mass, strength and function. Due to recent advancements, CT scan imaging is becoming widely utilized to assess changes in skeletal muscle mass and has contributed to validating sarcopenia as a powerful indicator of poor clinical characteristics, chemotherapy response and adverse prognosis also in patients with OC.

Our knowledge of initiators and regulators of cachexia, as well as the availability of reliable cachexia biomarkers are currently insufficient and have led to very limited translatability, mainly because of the lack of optimal preclinical animal models. This is true especially for OC. Until recently, most of the available data related to OC cachexia were generated in mice bearing the TOV-21G, neglecting the fact that this cell line is representative of clear cell carcinoma, hence barely accounting for less than 5% of all OC histotypes (160), and possesses few copy number alterations and a “hypermutated” genome, which sets it apart from most OC cell lines and in particular from HGS-OC lines (162). Recent attempts by our group and others have led to developing more clinically relevant and representative animal models of OC and contributed to the identification of novel mediators of musculoskeletal wasting, including RANKL (5, 79).

Nonetheless, while cachexia researchers remain focused on the longer-term objective of developing effective and novel treatments for cachexia, further research is necessary to assess the true extent of OC cachexia and associated symptoms, in both preclinical and clinical settings.

Abbreviations

APR

acute phase response

BMP

bone morphogenic protein

CT

computed tomography

FOXO

fork head box O

GDF

growth differentiation factor

HGS

high grade serous

IGF

insulin-like growth factor

IL

interleukin

mTOR

mammalian target of rapamycin

OC

ovarian cancer

PMI

psoas muscle index

RANKL

receptor activator of nuclear factor kappa beta ligand

SMA

skeletal muscle attenuation

SMAD

mothers against decapentaplegic

SMD

skeletal muscle density

SMI

skeletal muscle index

STAT

signal transducer and activator of transcription

TGF

transforming growth factor

TNF

tumor necrosis factor

UPS

ubiquitin-mediated proteasome

WAT

white adipose tissue

Funding

This work was supported by Department of Pathology and Comprehensive Cancer Center at University of Colorado Anschutz Medical Campus, and by grants from National Institute of Arthritis and Musculoskeletal and Skin Diseases (R01AR079379, R01AR080051) and American Cancer Society (132013-RSG-18-010-01-CCG) to A.B., as well as National Cancer Institute (R37CA261987) and Department of Defense (W81XWH-21-1-0382, W81XWH-21-1-0233) to B.G.B. C.S.C. was supported by a T32 Institutional Training Grant from the National Institute of Arthritis and Musculoskeletal and Skin Diseases (AR080630).

Disclosures

The authors have nothing to disclose.

Data Availability

Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.