Review of the endocrine organ–like tumor hypothesis of cancer cachexia in pancreatic ductal adenocarcinoma

Pancreatic ductal adenocarcinoma (PDAC) is one of the most fatal types of solid tumors, associated with a high prevalence of cachexia (~80%). PDAC-derived cachexia (PDAC-CC) is a systemic disease involving the complex interplay between the tumor and multiple organs. The endocrine organ–like tumor (EOLT) hypothesis may explain the systemic crosstalk underlying the deleterious homeostatic shifts that occur in PDAC-CC. Several studies have reported a markedly heterogeneous collection of cachectic mediators, signaling mechanisms, and metabolic pathways, including exocrine pancreatic insufficiency, hormonal disturbance, pro-inflammatory cytokine storm, digestive and tumor-derived factors, and PDAC progression. The complexities of PDAC-CC necessitate a careful review of recent literature summarizing cachectic mediators, corresponding metabolic functions, and the collateral impacts on wasting organs. The EOLT hypothesis suggests that metabolites, genetic instability, and epigenetic changes (microRNAs) are involved in cachexia development. Both tumors and host tissues can secrete multiple cachectic factors (beyond only inflammatory mediators). Some regulatory molecules, metabolites, and microRNAs are tissue-specific, resulting in insufficient energy production to support tumor/cachexia development. Due to these complexities, changes in a single factor can trigger bi-directional feedback circuits that exacerbate PDAC and result in the development of irreversible cachexia. We provide an integrated review based on 267 papers and 20 clinical trials from PubMed and ClinicalTrials.gov database proposed under the EOLT hypothesis that may provide a fundamental understanding of cachexia development and response to current treatments.

Pancreatic ductal adenocarcinoma (PDAC) is one of the most fatal types of solid tumors, associated with a high prevalence of cachexia (~80%). PDAC-derived cachexia (PDAC-CC) is a systemic disease involving the complex interplay between the tumor and multiple organs. The endocrine organ-like tumor (EOLT) hypothesis may explain the systemic crosstalk underlying the deleterious homeostatic shifts that occur in PDAC-CC. Several studies have reported a markedly heterogeneous collection of cachectic mediators, signaling mechanisms, and metabolic pathways, including exocrine pancreatic insufficiency, hormonal disturbance, pro-inflammatory cytokine storm, digestive and tumor-derived factors, and PDAC progression. The complexities of PDAC-CC necessitate a careful review of recent literature summarizing cachectic mediators, corresponding metabolic functions, and the collateral impacts on wasting organs. The EOLT hypothesis suggests that metabolites, genetic instability, and epigenetic changes (microRNAs) are involved in cachexia development. Both tumors and host tissues can secrete multiple cachectic factors (beyond only inflammatory mediators). Some regulatory molecules, metabolites, and microRNAs are tissuespecific, resulting in insufficient energy production to support tumor/cachexia development. Due to these complexities, changes in a single factor can trigger bidirectional feedback circuits that exacerbate PDAC and result in the development of irreversible cachexia. We provide an integrated review based on 267 papers and 20 clinical trials from PubMed and ClinicalTrials.gov database proposed under the EOLT hypothesis that may provide a fundamental understanding of cachexia development and response to current treatments. KEYWORDS pancreatic ductal adenocarcinoma (PDAC), cachexia, muscle wasting, tissue wasting, endocrine organ-like tumour (EOLT)

Introduction
For Pancreatic ductal adenocarcinoma (PDAC) is currently the fourth most common cause of cancer-related deaths worldwide and is projected to become the second most common cause of cancer-related deaths by 2030 (1). Due to its aggressiveness and poor prognosis, mortality remains alarmingly high among patients diagnosed with PDAC. Approximately 80%-85% of PDAC patients are diagnosed at advanced stages with unresectable or metastatic tumors, resulting in a 5-year survival rate below 10% (2). During earlystage PDAC, surgical resection is currently the only curative option, although chemotherapy and radiation therapy are also used as primary treatment options, with or without surgery. However, single-agent chemotherapies are rarely effective in PDAC (3). In general, chemotherapy regimens are not universally effective in PDAC and are associated with significant adverse effects, including the development of PDAC-derived cachexia (PDAC-CC), and cachexia occurs in 32%~71% of patients within 12 to 48 weeks of chemotherapy initiation (4).
Patients with PDAC experience a high prevalence (up to 80%) of cachexia, often with early onset (45% of PDAC patients present with cachexia at the time of diagnosis (5)), which may account for up to 30% of mortality (6). Cachexia is defined as the progressive loss of muscle mass and function (6,7) and is a catabolic multiorgan syndrome characterized by non-volitional weight loss (muscle or adipocyte loss), adipopenia, fatigue, weakness, loss of appetite, and early satiety (8,9). When muscle mass loss, it enhances chemo-toxicities and insensitivities, contributing to poor overall survival (10).
In general, tumors demand a high energy supply and can promote the wasting of peripheral tissues via hyper-catabolism. Tumors compete with other organs/tissues for energy and nutrients, resulting in elevated resting energy expenditure and inducing a negative energy balance. Energy utilization in tumors also results in increased proteolysis and lipolysis combined with decreased lipogenesis and protein synthesis (8,9,11,12). These metabolic reprogramming effects, combined with poor appetite, lead to rapid weight loss among PDAC patients and can contribute to deterioration in the overall quality of life (QoL) and overall survival (OS) (7,(13)(14)(15). The complex, multifactorial nature of the metabolic disruptions in cachexia makes effective treatment challenging. The current lack of consensus regarding how to define cachexia and a scarcity of strong evidence produced by robust, rigorous, and mechanistic studies have limited the development of effective treatments (16). In addition, most cachexia studies focus on symptoms associated with individual organs (such as tumor, muscle, or adipocyte tissues) without considering consider systemic interactions. In this review, we provide an up-to-date overview of current cachexia research in PDAC to provide insight regarding the cachexia mediators that act in different organs and explore whether the endocrine organ-like tumor (EOLT) hypothesis of PDAC can explain the development of systemic complications.

Cachexia criteria and stages
Cancer-derived cachexia (CC) is a multifactorial syndrome involving various metabolic changes in several tissues and organs (8,9,12,(17)(18)(19). Although patients with pancreatic cancer show a wide range of nutritional alterations, the primary symptom is progressive weight loss due to the loss of skeletal muscle mass, with or without the accompanying depletion of adipose tissue (6,(19)(20)(21)(22)(23). Other PDAC-CCrelated clinical manifestations include inflammation (24-26), anorexia (27, 28) and metabolic reprogramming (9,29,30) etc., Numerous studies also focus on exploring new PDAC-CC cachectic mediators, corresponding metabolic functions, and the collateral impacts on wasting organs. A systematic review also suggested a network of cytokines (interleukin [IL]-6, tumor necrosis factor-alpha [TNFa], and IL-8) that may be associated with cachexia development (31). Sah et al. (19) suggested that PDAC-CC can be categorized by three distinct metabolic phases: Phase 1 represents the earliest metabolic change, characterized by new-onset hyperglycemia; Phase 2 is associated with a greater than 5% reduction in body weight with pre-cachectic symptoms (appetite loss and impaired glucose metabolism), suggesting the initiation of cachexia; and Phase 3 is associated with dramatic reductions in all monitored metabolites, lipids, subcutaneous fat, and muscle, except fasting glucose.
Traditionally, a Body Mass Index (BMI) < 18.5 kg/m 2 was accepted as a marker of being cachectic. However, sarcopenic obesity can be observed in CC, suggesting that weight loss might not be a defining factor (32). According to the most common consensus, published by Fearon et al. (33), the current standard diagnostic criterion for cachexia is represented by percentage of weight loss, BMI values and metabolic changes (29, 33, 34). Simply, CC were classified into three stages: precachexia, cachexia, and refractory cachexia (Table 1). This classification currently did not fully applicable in clinics but is rather to be considered as a proposal under evaluation. Additional parameters (Table 2) have been developed to improve diagnosis, such as food intake measures, albumin l e ve ls , a nor e x i a as se s sm e nt , mar ke r s of sy st e m i c inflammation (CRP >10 mg/L), muscle mass measurements, the Skeletal Muscle Index (SMI), bioelectrical impedance analysis (BIA), the Fat-Free Mass Index and cachexia index (CXI). Although these diagnostic measurements did not include in the latest consensus, they suggested that several effective parameters could more accurately identify cachexia. A recent systematic review by Paval et al. described the betweenstudy inconsistencies in grouping criteria as a major hindrance to the conduct of meta-analyses for cachexia (31). Refined CC-criteria is critical for evaluating the response to cachexia/ antitumor therapy. Because early-onset PDAC-CC can present concomitant with the detection of the primary tumor burden, but cachexia can continue even after the tumors have been surgically removed or effectively treated (15). Patients received either preoperative surgery or chemotherapy/ chemoradiation; unintended weight loss coupled with muscle wasting can often be observed, contributing to poor outcomes in PDAC (10, 13, 15, 29). There is no effective strategy to mitigate refractory PDAC-CC. Therefore, the early and precise identification of PDAC-CC is needed to estimate prognosis and prevent progression to the refractory cachexia. More practical, longitudinal definitions of cachexia remain necessary that consider all aspects of the cachexia phenotype.

EOLT hypothesis in PDAC-CC
The EOLT hypothesis was proposed to explain how tumor tissues drive disease progression, including CC (31). The EOLT hypothesis states that the tumor acts as an endocrine organ, resulting in dynamic bi-directional communications between the tumor microenvironment (TME) and various organs, leading to the regulation of macroenvironmental changes.
PDAC-CC results in systemic wasting and involves multiple organ dysfunction (Figure 1), accompanied by symptoms including poor appetite, fatigue, depression, muscle wasting, fat wasting, malabsorption, and constipation (Table 3). Tumors secrete cachexia-inducing factors and stimulate host-tumor interactions involve cancer-organ metabolic reprogramming and interorgan signal crosstalk in tumor progression and cachexia development (21, 32). For example, tumor-derived cytokines induce systemic inflammation, stimulating the release of neuropeptides that lead to poor appetite, and the resulting anorexia exacerbates tissue wasting (6,14,26). With cachectic environment, adipose and muscle tissues can act like paracrine/endocrine organs in response to cachectic factors, providing reciprocal regulation of energy expenditure and cachexia process (8,9,17,33). Cachexia is a wasting disease that represents metabolic disruptions, mainly catabolisms, driven by systemic inflammation and is characterized by skeletal-muscle p r o t e o l y s i s , a d i p o s e t i s s u e l i p o l y s i s a n d h e p a t i c 1. Systemic metabolic changes, often associated with KRAS mutations (genetic instability).
The high prevalence of cachexia in PDAC is associated with distinct metabolic effects mediated by tumor created environments, including KRAS mutations, pro-cachexia mediators, and alteration in pancreas and liver. The present review summarizes the current understanding of PDAC-CC according to the EOLT hypothesis.

Metabolic alterations and high energy demands in tumors
PDACs are characterized by high energy demands within a nutrient-deprived microenvironment. Aggressive PDAC is characterized by increased glycolysis and glutamine metabolism, closely associated with downstream anabolic pathways in the tumor's hypoxic desmoplastic environment (8,9,27,29,63). The deprivation of glucose and glutamine and lactic acidosis promote glycolytic and glutaminolysis activity (61, 82, 83). Metabolic alterations are hallmarks of PDAC and PDAC-CC, particularly the dysregulation of glucose and glutamine metabolism (8,9,14,19,29,62). However, PDACs under different oxygen and nutrition conditions show distinct and heterogeneous metabolite profiles associated with aerobic glycolysis (the Warburg effect), OXPHOS (oxidative phosphorylation; also known as the reverse Warburg Frontiers in Oncology frontiersin.org effect), lipid dependence, autophagy, and glutaminolysis ( Figure 2). Metabolic alterations are positively correlated with high-grade pancreatic intraepithelial neoplasia (PanIN-3). However, earlyonset cachexia also develops independent of PDAC, occasionally presenting in the pre-diagnostic PDAC stage (4). Cachexia is a metabolic disorder involving several nutrient scavenging pathways, including autophagy, micropinocytosis, glycolysis, lipid oxidation, and micropinocytosis ( Figure 2: upper panel). PDAC survives and thrives in relatively hypoxic and nutrient-poor niches, driven by [1] reprogramming intracellular nutrient metabolism, including glucose, amino acids, and lipids; [2] scavenging and recycling nutrients; and [3] promoting metabolic crosstalk (Figures 2: lower panel and Figure 3) (8,9,62). PDAC-CC exacerbates metabolic reprogramming, promoting the deterioration of muscle and adipose tissue (Figure 3), further supporting the energy and nutrient needs of the tumor tissue.
glucose transporters (GLUT1, encoded by SLC21A) in tumor tissues facilitate glucose uptake for aerobic glycolysis. Increased glycolytic flux in response to host-tumor interactions in cachexia results in the production of high lactate levels, leading to lactic acidosis. To address lactic acidosis, PDACs robustly express monocarboxylate transporters (MCT1 and MCT4, encoded by SLC16A1 and SLC16A3, respectively) to coordinate glucose utilization and lactate mobilization (65, 84). Other glucose metabolism pathways are also altered in PDAC, such as the upregulation of the rate-limiting enzyme of the HBP, glutamine: fructose-6-phosphate amidotransferase-1 (GFPT1) (64). Many other mediators regulate glucose metabolism in pancreatic cancer cells. Under hypoxic conditions, hypoxiainducible factor-1 (HIF-1) can promote glycolysis and upregulate the expression of HBP-related enzymes, such as GFPT2, an isoform of the HBP rate-limiting enzyme GFPT1 (63).
In a model of PDAC-CC, athymic mice injected with highglycolytic MiaPaCa2 cells showed evidence of cachexia, such as weight loss, fat depletion, and muscle proteolysis (66), suggesting that glycolysis may be involved in PDAC-CC development. Glycolysis was associated with inefficient inter-organ substrate shuttles, as assessed by the lactate-to-pyruvate utilization ratio, LDH activity, and MCT1 expression, which was correlated with cachexia-related weight loss (63). The upregulation of GLUT1 and MCT1/4 promotes glucose utilization and improves the lactate-to-pyruvate utilization ratio in tumor tissue (63,65). Paradoxically, OXPHOS, also known as the reverse Warburg effect, occurs in muscle, resulting in an increased lactate-to-pyruvate production ratio, providing a potential lactate supply for tumor use and supporting tumor progression and consequent atrophy (85). Additionally, the tumor secretes interferon-gamma (IFN-g), which mediates the development of insulin resistance via reduced glucose and fatty acid [FA] uptake, leading to enhanced lipolysis in WAT (86). Inefficient interorgan substrate shuttles are regarded as hallmarks of EOLT ( Figure 3).
Aerobic glycolysis occurs more commonly in tumor tissues than OXPHOS, which requires a sufficient oxygen supply, although OXPHOS is more efficient for ATP generation (30, 87). Nutrient depletion forces tumors to adapt by inducing nutrient scavenging mechanisms to support cancer progression, which can lead to CC (8,9,88). OXPHOS occurs in the mitochondria and is sensitive to stress conditions, as the respiratory complexes in stressed mitochondria produce high levels of reactive oxygen species (ROS) (62). Autophagy is a stress response induced by ROS to remove damaged mitochondria that overproduce ROS, promoting mitochondrial metabolism (89) (Figure 2: lower panel). In cachectic patients, increased OXPHOS and dysfunctional autophagy are associated with increased muscle wasting ( Figure 3) (90). Autophagy is an important proteolysis pathway activated during PDAC-CC and muscle wasting (91, 92).
Tumors supported by an adequate blood supply can perform aerobic metabolism and tend to exhibit the reverse Warburg effect. In tumors, aerobic metabolisms may utilize intermediates, such as lactate, to fuel the TCA cycle ( Figure 2: lower panel), decreasing their dependence on glucose. The low uptake of glucose and the enhanced uptake of intermediate metabolites by tumors under aerobic conditions could protect these tumors from competing with hypoxic tumor regions (such as desmoplastic tumors) for glucose. In addition to OXPHOS, tumors able to perform aerobic metabolism can also utilize glutaminolysis as an alternative energy production pathway requiring activated mitochondrial metabolism (93). Glutamine is the most abundant and versatile nonessential amino acid (NEAA), found in both the blood and the cell cytoplasm, and can be used by both the glutamine-dependent pyruvate cycle and the TCA cycle (30, 94). In contrast, hypovascularization and desmoplasia often occur in PDAC; studies also found that Metabolism alterations in PDAC-CC. The upper panel shows that pancreatic ductal adenocarcinoma (PDAC)-derived cachexia (PDAC-CC) arises from the multi-stage progression of precursor lesions, known as pancreatic intraepithelial neoplasia (PanIN). PanINs are characterized by a continuum of increasingly stroma features (from low-grade dysplasia developing to high-grade desmoplasia). A desmoplastic response induced a fibro-inflammatory microenvironment, stimulating an aberrant metabolic response that is associated with cachexia. During early-stage PDAC, histology can be used to identify several distinct types of precursor lesions. The most common types are microscopic PanIN, low-grade (PanIN-1 and PanIN-2), and high-grade (PanIN-3). The progression to PanIN and to PDAC is associated with cachexia development. Multiple metabolic alterations follow the progression of cachexia, resulting in the reprogramming of glucose, amino acid, and lipid pathways. Metabolic alterations include nutrient scavenging pathway), such as glycolysis glutaminolysis, autophagy, proteolysis, lipid oxidation, and micropinocytosis (Most of them were upregulated during the development of cachexia). However, early-onset cachexia can arise independent of the PDAC stage, occurring in the pre-diagnostic PDAC stage. More than one-third of cancer patients were malnourished before chemotherapy, implying that the cachexia occurred early and followed a poor response to chemotherapy. Interestingly, some of cachexia occurs after the chemotherapy. The lower panel: The metabolic alterations, including increase glycolysis, glutaminolysis, lactate transport and autophagy … etc, in PDAC cell associated with PDAC-CC, primarily due to promote the expression in key enzymes (HK1/2, GFPT1, and LDHA) and transporters (GLUT1, MCT1/ 4, SLC7A5, and SLC1A5). The metabolic shift from the tricarboxylic acid (TCA) cycle and oxidative phosphorylation (OXPHOS) to aerobic glycolysis is tightly regulated. HK1/2, hexokinase; GFPT1, glutamine fructose 6-phosphate amidotransferase 1; LDHA, lactate dehydrogenase A; GLUT1, Glucose transporter 1; MCT1/4, monocarboxylate transporter 1/4; SLC7A5 (LAT1), neutral amino acid antiporter; SLC1A5, glutamine transporter.
HIF-1a (hypoxia-inducible factor-1a) stabilization promotes glycolytic enzymes to shift the metabolism by repressing OXPHOS (95, 96). Supposing that if the function of glycolysis is weakened, OXPHOS and glutamine-based processes will serve as alternative energy generation mechanisms in glucose-limited tumors (87) (Figure 2: lower panel).

Amino acids
Altered amino acid (AA) metabolism is a frequent feature in CC. Branched-chain amino acids (BCAA: leucine, isoleucine, and valine) act as important carbon sources and are useful for FA biosynthesis. High BCAA levels in plasma are associated with early PDAC and are often derived from increased protein breakdown in muscle and other body tissues ( Figure 3) (97,98). The utilization of BCAAs by PDAC can result in plasma BCAA depletion during late-stage PDAC. Similar observations have been reported for glutamate, in which the plasma levels of glutamate and the glutamine/glutamate (Q/E) ratio are significantly reduced in cachectic patients and animal models compared with their healthy counterparts (83, 99). Glutamine metabolism is a primary source of nitrogen and carbon, contributing to macromolecular synthesis and redox balance (83). Glutaminase 1 (GLS1) converts glutamine to glutamate, after which glutamate dehydrogenase (GDH) catalyzes the conversion from glutamate to a-ketoglutarate (a-KG). However, GDH is repressed in PDAC, and cytoplasmic aspartate transaminase (GOT1) is upregulated (83) (Figure 2: lower panel). Cachexia is associated with more aggressive forms Tumor-muscle-adipocyte crosstalk. Pancreatic ductal adenocarcinoma (PDAC)-derived cachexia occurs due to a feedback circuit among tumor, adipocyte, and muscle tissues. In PDAC, tumor-derived factors, including interleukin (IL)-1, IL-6, IL-8, proteolysis-inducing factor (PIF), lipid mobilization factor (LMF), and tumor necrosis factor-a (TNFa), enhance proteolysis, lipolysis, and the catabolic state of muscle and adipocytes, leading to adipose and muscle wasting. Tumor cell-triggered metabolic reprogramming in muscle and adipocytes releases metabolic products, such as branched-chain amino acids (BCAA) and free fatty acids (FFAs), to fuel tumor growth. Lipolysis and proteolysis are the two main processes causing adipose and muscle wasting in cachexia.
of PDAC, which may reflect the increased access of tumor cells to nutrients derived from protein breakdown and systemic changes in glucose metabolism (97,100). Higher circulating BCAA levels may arise from the impaired catabolism of AAs that are commonly found in muscle (Figure 3). Muscle wasting is characterized by decreased muscle mass, increased proteolysis, and reduced protein synthesis, changes which are mediated by the proteasome, nuclear factor kappa B (NF-kB), and the mammalian target of rapamycin (mTOR) pathways. The phosphoinositol 3-kinase (PI3K)/AKT/mTOR pathway is a nutrient-sensing mechanism stimulated by decreased glucose availability in the muscle. A higher mTOR activity induced by KRAS mutation in PDAC which is positively correlated to higher circulating BCAA levels (12). mTOR activation is responsible for the uptake of BCAA in tumor tissue. Circulating BCAA also can affect subcutaneous adipocyte AA dysmetabolism. Both NF-kB and AKT/mTOR signaling are involved in proteolysis. NF-kB regulates the ATP-dependent ubiquitin-proteasome proteolytic pathway, including muscle-specific E3 ubiquitin ligases (such as muscle atrophy F box protein [MAFbx/atrogin-1] and muscle RING finger-containing protein 1 [MuRF1]), which promote proteolysis and contribute to muscle atrophy (22). Cachexia is the end result of convergent metabolic adaptations induced by tumors to satisfy their metabolic requirements. A B FIGURE 4 Metabolic remolding is influenced by genetic instability in PDAC. (A) Genetic mutations: PDAC is affected by high frequencies of aberrations and mutations in KRAS, P16/CDKN2A, TP53, and SMAD4. KRAS is involved in the RAF/mitogen-associated protein kinase pathway and the phosphoinositol 3-kinase (PI3K) pathway. P16/CDKN2 mediates the PI3K/AKT/mTOR pathway. TP53 influences the transforming growth factorbeta (TGF-b)/Smad4 pathway. (B) Mutated forms of KRAS, TP53, P16/CDKN2, and SMAD4 promote glucose (Glc) uptake and enhance glycolytic flux, including the production of lactate (Lac). KRAS and TP53 can both reprogram glutamine (Gln) metabolism to balance cellular redox homeostasis. Pancreatic cancer induces metabolic shifts, including increased glycolysis, lipogenesis, glutaminolysis, and autophagy, which are related to cachexia.

Lipids
In addition to glucose and amino acid metabolism, metabolic alterations in cachexia can include lipid metabolism. Approximately 93% of triacylglycerol FAs used by tumors are synthesized de novo by the mitochondria and cytosolic acetyl coenzyme A (CoA). Enzymes that participate in de novo FA and cholesterol synthesis are upregulated in PDAC, such as FA synthase (FASN) and 3-hydroxy-3-methylglutaryl coenzyme A reductase (HMGCR). Under pancreatic inflammatory conditions, wasting adipocytes release FAs into the plasma, increasing plasma concentrations of saturated (SFAs), monounsaturated (MUFAs), and polyunsaturated FAs (PUFAs) (101). SFAs and MUFAs promote PDAC progression (102). However, a study performing transcriptomics and metabolomics suggested that lipase and a panel of FAs were significantly decreased in PDAC, and the presence of two SFAs (palmitate and stearate) inhibited tumor cell proliferation (103). Therefore, the roles played by FAs in PDAC appear complicated and remain unclear. PDAC patients present with distinct phenotypes associated with cachexia development, such as adipose tissue loss prior to skeletal muscle wasting or the loss of adipose tissue alone (104). A recent report indicated that soft tissue changes are initiated in PDAC before skeletal muscle loss (19), and the significant loss of visceral adipose tissue has been observed in PDAC-CC (18). In a retrospective cohort study, PDAC-CC was associated with the accumulation of oleic acid in plasma, resulting from malnutritional compensatory mechanisms triggered by the lack of oleic acid uptake into tissue (105). In a pre-cachexia model, increased FA oxidation occurs before muscle mass reduction, suggesting that FA may serve as a dominant energy source in PDAC-CC (18, 106). Adipose tissue lipolysis contributes to circulating FAs and subsequent FA uptake and lipid accumulation in the muscle and tumor tissue, leading to eventual metabolic derangement and muscle wasting after a period of metabolic adaptation. Several lipolytic enzymes are elevated, such as adipose   • TGF-b is a potent inducer of muscle atrophy, weight loss, and fat loss (increasing catabolism: proteolysis and lipolysis).  Pro-cachectic mediators of catabolism in PDAC-CC. Cachexia signals induce tissue catabolism by modulating gene expression profiles related to protein synthesis and degradation in muscle, lipid depletion, and tumor progression, primarily via the nuclear factor kappa B (NF-kB) and Janus kinase (JAK)-signal transducer and activator of transcription (STAT) pathways. In tumors, multiple receptors, including the Toll-like/IL1 receptors (TIRs), tumor necrosis factor (TNF), transforming growth factor-beta (TGF-b), and interleukin receptors (IL-6R being the best-studied), utilize overlapping and distinct signal transduction mechanisms to affect cellular outcomes, including increased cytokine production, proliferation, survival, migration, autophagy, and resistance to chemotherapy and immune surveillance. In addition to cytokines in tumors, circulating cytokines affect muscle and adipocytes, resulting in various metabolic alterations. For example, myostatin/activin A binds to type II receptors (ActRIIB), leading to Smad2/3 phosphorylation and the recruitment of Smad4, which results in muscle wasting. Simultaneously, myostatin/activin A signaling inhibits AKT activity and suppresses FoxOs phosphorylation, activating the ubiquitin-proteasome and autophagylysosome systems. IL6 binds to receptors to activate JAK/STAT3 signaling, increasing protein degradation. TNFa and IL1 signaling activates the IkB kinase (IKK)-NF-kB axis to initiate proteasome-mediated protein degradation. Higher levels of tumoral and stromal IL-1b expression result in a feedback circuit that attributes to cancer progression and cachexia development.
triglyceride lipase (ATGL) and hormone-sensitive lipase (HSL), further suggesting the occurrence of enhanced lipolysis. Increased ATGL and HSL activity correlate with tumor growth and WAT loss in cachexia (107) (Figure 3). Tumor progression is also associated with the shift from WAT to BAT, known as adipose tissue browning. BAT is a metabolic hallmark mediated by uncoupling protein 1 (UCP-1). In KPC and Kras LSLG12D/ + Trp53 f/f mice, adipose tissue browning is associated with increased UCP-1 expression and occurs prior to the onset of fat wasting, consistent with clinical observations (108). A few studies have suggested that fat loss is an early and precipitating event prior to muscle loss in PDAC-CC, even in the absence of muscle wasting (18, 19, 21, 33, 104, 105, 107). Clinical studies suggested that fat loss may serve as a driving force for cachexia mortality, emphasizing the important roles of adipocytes in PDAC-CC and supporting the need to monitor adipose in patients with CC (20, 21). Tumors hijack organ and tissue function, causing muscle and adipocyte wasting. Enhanced glycolysis in tumors and the upregulation of lipolysis and proteolysis in wasting tissue counterbalance the reductions in muscle and fat under cachectic conditions (Figure 3). Wasting muscle and adipocytes are among the convergent metabolic adaptations induced by tumors to satisfy their metabolic requirements. Patients with PDAC and PDAC-CC exhibit distinct and heterogeneous metabolic changes. Tumor, muscle, and adipocyte tissues all act as endocrine organs involved in the regulation of metabolic homeostasis, consistent with the EOLT hypothesis. In addition to metabolic alterations, bi-directional feedback occurs between tumor tissues and other organs, driven by the oncogenes and mediators (8, 12, 21, 32, 62).
Besides promoting high levels of glycolysis, KRAS upregulates glutaminolysis, allowing glutamine to be used as an additional carbon source for the TCA cycle (114). NEAAs, such as alanine, are alternative carbon sources that can support altered energy metabolism in PDAC (115). KRAS mutations increase glycolysis and the metabolism of amino acids, such as alanine and glutamine, activating downstream catabolic pathways, including proteolysis and lipolysis. Genetic mutations promote the recycling of wasting tissues to fuel cancer growth.
KRAS activation leads to the loss of p16, accelerating NADH oxidation and supporting increased glycolysis through the production of NAD + to support tumor growth (118). However, PDAC cells lack nutrient sensors, and mTOR complex 1 (mTORC1), a nutrient-sensing mechanism, is bypassed in PDAC. Bardeesy et al. proposed that autophagy is driven by the elevated expression of the microphthalmia transcription factor (MiTF) family members MiTF, TFE3, and TFEB in PDAC (119). The loss of SMAD4 is another frequent event associated with PDAC progression, identified in approximately 50% of PDAC cases. SMAD4 is a central component in the transforming growth factor (TGF-b) signaling cascade, and SMAD4 loss enhances glycolysis by altering the expression of the glycolysis enzyme phosphoglycerate kinase (PGK) (120). The loss of TP53 alters metabolism in PDAC by inhibiting mitochondrial respiration and simultaneously stimulating glycolysis. A recent study demonstrated that TP53 rewires glucose and glutamine metabolism in malignant PDAC by preventing the nuclear translocation of glyceraldehyde 3phosphate dehydrogenase (GAPDH) and stabilizing its function (121). Loss of TP53 can reduce the expression of fructose-2,6-bisphosphate to promote the glycolysis cycle (122). Therefore, KRAS, P16, P53, and SMAD4 have counterintuitive effects that promote tumorigenesis, further highlighting the complexity of interactions between genes and metabolisms in cancer progression and cachexia development.

FIGURE 6
Mechanistic role of miRNAs in PDAC cachexia.  Table 6 in detail). The most common genes, such as IL-6R, FOXO1, PDK4, and ZIP14, had been associated with muscle wasting in cachexia. In adipocytes, specific microRNAs may mediate the transcription factors C/EBPb/d, C/EBPa, and PPARg, resulting in adipogenesis.
which a portion of a resected human pancreatic tumors are surgically attached to the mouse pancreas; and [4] genetically engineered mouse models (GEMMs) of PDAC. Up to 85% of PDAC patients suffer from CC, and approximately 30% of PDAC patients succumb to cachexia rather than tumor burden (137,138). PDX and murine allograft models have been applied to study cachexia, resulting in the identification of Toll-like receptors 7/9 (TLR7/9), MyD88, and TGF-b as mediators of cachexia in PDAC (129,137,139,140). Most PDAC-CC studies focus primarily on weight loss, muscle wasting, and the analysis of mRNA markers. A comparison analysis of subcutaneous, IP, and orthotopic PDAC animal models indicated that the implantation site is crucial when attempting to study PDAC-CC (137). Both IP and orthotopic implantation models develop more severe cachexia symptoms (such as muscle wasting, anorexia, and a decrease in locomotive activity) than the subcutaneous implantation model. The orthotopic animal model is histologically similar to PDAC patients, mimicking the TME associated with intact tumors, suggesting that the TME may be involved in cachexia induction. Studies in PDAC animal models have demonstrated that tumor-associated macrophages mediate muscle wasting via the activation of signal transducer and activator of transcription (STAT)3 signaling (134, 141).
Most preclinical studies of CC use C-26 (colon cancer) and LLC (lung cancer) mouse models. However, these two models are associated with limitations (1): a limited interval between the onset of CC symptoms and animal death leaves only a small therapeutic window, and (2) translatability to humans may be limited, as the gene expression profiles in these mice did not correlate with those in human cancer tissue biopsies (130). However, GEMMs offer slower cachexia progression and early development than other cancer models, and PDAC-CC animal models are more translatable to humans than models using other cancer types. Therefore, animal models of PDAC-CC are clinically relevant. Preclinical PDAC murine models may be useful for understanding cachexia progression and evaluating therapeutic options for mitigating PDAC-CC. Establishing a model able to fully mimic the human condition remains necessary. Animal models can contribute to improving our understanding of the mechanisms driving tissue wasting for translation into new anti-cachexia therapies.

Pro-cachectic mediators and microRNAs 3.3.1 Pro-cachectic mediators
Endocrine organs and cells synthesize biologically active compounds that are released directly into the circulation and interact with other cells. Cachexia-associated inflammation is influenced by numerous bioactive molecules, such as TNFa, IL-1, IL-6, and IL-8 (Table 5 and Figure 5). Cachexia-affected organs can act as autocrine or paracrine organs, releasing factors into the bloodstream to promote systemic crosstalk. These cytokines have multifactorial effects, triggering a hypercatabolic feedforward loop between the tumor, adipose tissue, and muscle mediated by the NF-kB and Janus kinase (JAK)/STAT pathways (29, 174, 175) ( Figure 5). NF-kB and JAK/STAT activation enhance lipolysis, downregulate lipogenesis, and stimulate the catabolism of lean body mass (12).
A salient feature that distinguishes PDAC from other KRASmutant cancers is an extensive fibro-inflammatory stroma, which accounts for 80%-85% of the tumor bulk. These stromal cells are recruited and reprogrammed by PDAC cells during cancer progression and cachexia development. Secreted factors ( Table 5) enable these cells to communicate with PDAC, creating a dynamic feedback circuit associated with intrinsic KRAS signaling in PDAC cells ( Figure 5).
Various circulating pro-inflammatory cytokines have been implicated in PDAC-CC, including IL-6, IL-1, IL-8, TGF-b, and TNFa …etc (detailed in Table 5) (12, 21, 29, 123). These cytokines likely derive from various sources and result in systemic effects (123,176). Oncogenic RAS drives the expression of multiple inflammatory cytokines, including IL-1 and IL-6 (21, 146). Inflammatory cytokines released by tumor cells may be prominent cachexia-associated factors that regulate autocrine and paracrine function, promoting tumor progression and cachexia development. For example, IL-6 plays autocrine roles in supporting tumorigenesis in vivo and induces weight loss and inflammation in cachexia via a paracrine manner (21, 68, 152). Other cytokines, such as IL-1, IL-8, TNFa, and INF-g, have also been associated with weight loss and poor survival in PDAC (148, 177, 178). TNF-a and IL-1 can induce anorexia, producing both hypercatabolic and anorexigenic effects (Figures 1 and 5). Circulating IL-1b promotes NF-kB activation in the hypothalamus, enhancing glucocorticoid production and resulting in catabolic effects in both muscle and adipose tissue. TGF-b1 can induce proteolysis through the E3 ligase atrogin-1 in animal models. TGF-b inhibition improved muscle wasting in the KPC model (129). Activin A and myostatin belong to the TGF-b superfamily and are associated with muscle wasting through the activation of the Smad2/3 pathway, which decreases AKT-mTOR-mediated protein synthesis and enhances ubiquitin ligase-mediated proteolysis (18, 128). These cytokines drive diverse catabolic processes across multiple cells and organs, forming a catabolic feedforward loop (Figure 1).
In muscle tissue, the JAK-STAT and NF-kB pathways are the dominant catabolic pathways activated by circulating IL-6 and TNFa in muscle wasting (174). IL-6 induces NF-kB activation, which can also upregulate ubiquitin-mediated proteasomal degradation in wasting (179). Proteolysis is a prerequisite for muscle wasting, and both lipolysis and adipopenia may occur prior to muscle loss. Lipolysis results in increased circulating FFAs, triggering the secretion of Atrogin-1 and MuRF1, which induce muscle atrophy.
In adipocytes, lipolysis plays a substantial role in increasing the catabolism of stored fat. ATGL and HSL act to reduce fat to its component FAs, leading to the loss of body mass. Secreted IL-6 can trigger browning by inducing UCP-1 expression in adipocytes (18,180). The catabolic effects of IL-6 on WAT in vitro are mediated through the JAK/STAT3 and NF-kB pathways (21, 181).

Epigenetics modulation (miRNAs)
Genetic instability and epigenetic changes are both involved in pancreatic oncogenesis and cachexia development (Table 6). Recently, miRNAs, small non-coding RNAs 19-25 nucleotides in length, have been identified in an increasing number of biological processes, including KRAS signaling and the JAK-STAT, PI3K-AKT, notch, and TGF-b signaling pathways ( Figure 6A). These influences contribute to the control of several cancer-related processes in PDAC, such as tumor growth, apoptosis, metastasis, drug resistance, and the immune response. In addition to roles in oncogenesis and tumorigenesis, aberrant miRNA expression may affect cytokine production or directly alter metabolic processes, resulting in a metabolism remodeling that facilitates PDAC progression and cachexia development (see Figure 6 and Table 6).
Recent studies have detected miRNAs in serum, plasma, tissue, and tumors (Table 6). Studies indicates that miRNAs are commonly found in various EVs, such as exosomes, apoptotic bodies, microvesicles (MV), and lipoproteins, allowing them to target cells and contribute to intercellular signaling through endocrine, paracrine, and autocrine pathways (182,183,198,199,(215)(216)(217). Losses in muscle proteins and fat mass are the most important signatures of cachexia and can result in the generation of microvesicles containing miRNAs (see Figure 6B: bottom panel). Most studies have identified miRNAs expressed in tumor cells; however, some miRNAs are tissue-specific or tissue-enriched, involved in either the active or passive stimulation of metabolic changes and inflammatory responses ( Table 6). In PDAC-CC, miR-21, miR221/222, miR27a and miR155 are commonly correlated with muscle and adipocyte wasting through the transcription E3 ubiquitin ligases (mediated by FoxO3 or FoxO1 in muscle), JUN-STAT3, or TGF-b; ( Figure 6B: bottom panel). Given the roles of miRNAs in gene expression and the regulation of inflammatory responses and metabolic reprogramming, additional study of miRNAs remains necessary. An ongoing observational trial (NCT05275075) aims to analyze the miRNA profiles in patients with PDAC-CC. The causal roles of miRNAs and molecular mechanisms in cachexia remain under debate and require further discussion and study; however, therapeutic approaches for modifying multiple targets have been suggested (218, 219). The study of miRNAs could also contribute to the development of diagnostic or prognostic biomarkers and new targets for cachexia prevention or treatment.

Conclusion
The high prevalence of PDAC-CC may be associated with the unique genetic background (KRAS mutations) and modulators in PDAC, which mainly exacerbate metabolic disruptions, leading to cachexia development. Systemic metabolic alterations mediated by pro-cachectic factors, systemic inflammation, and epigenetic changes, highlighting that PDAC is a systemic disease rather than a single-organ defect. PDAC can induce metabolic disruptions in organs beyond the pancreas. We applied the EOLT hypothesis (33) to emphasize the systemic effects of PDAC, leading to tissue wasting in PDAC-CC. Currently, no FDA-approved agent is able to treat cachexia, although potential treatments are listed in Table 7. However, further studies remain necessary to generate foundational knowledge for the development of additional therapies or understanding the molecular mechanisms of PDAC-CC.

Publisher's note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.