The Nutritional Challenges of Cancer Cachexia

Abstract

Cancer cachexia, or progressive weight loss often despite adequate nutrition, contributes greatly to cancer morbidity and mortality. Cachexia is metabolically distinct from starvation or protein malnutrition, though many patients with cancer and cachexia exhibit lowered appetite and food consumption. Tumors affect neural mechanisms regulating appetite and energy expenditure, while promoting wasting of peripheral tissues via catabolism of cardiac and skeletal muscle, adipose, and bone. These multi-modal actions of tumors on the host suggest a need for multi-modal interventions. However, multiple recent consensus guidelines for management of cancer cachexia differ in treatment recommendations, highlighting the lack of effective, available therapies. Challenges to defining appropriate nutritional or other interventions for cancer cachexia include lack of consensus on definitions, low strength of evidence from clinical trials, and a scarcity of robust, rigorous, and mechanistic studies. However, efforts to diagnose, stage and monitor cachexia are increasing along with clinical trial activity. As well, preclinical models for cancer cachexia are growing more sophisticated, encompassing a greater number of tumor types in organ-appropriate contexts and for metastatic disease in order to model the clinical condition more accurately. It is expected that continued growth, investment, and coordination of research in this topic will ultimately yield robust biomarkers, clinically useful classification and staging algorithms, targetable pathways, pivotal clinical trials, and ultimately, cures. Here we provide an overview of the clinical and scientific knowledge and its limitations around cancer cachexia.

Cancer cachexia, an unmet clinical need

Cancer associated cachexia is an underestimated consequence of many cancers. Up to 80% of cancer patients with advanced cancer develop cachexia with a subsequent one-year mortality rate that can reach 20-60%¹. According to the current consensus definition, “cancer cachexia is a multifactorial syndrome characterized by an ongoing loss of skeletal muscle mass that cannot be fully reversed by conventional nutritional support and leads to progressive functional impairment”². Skeletal muscle wasting, low skeletal muscle mass, often termed “sarcopenia”,³ and weight loss are independent prognostic factors for overall survival of cancer patients^{4, 5}. Cancer patients with weight loss suffer from more frequent and more severe chemotherapy-associated toxicities compared to patients with stable weight. Consequently, they ultimately receive less chemotherapy and demonstrate lower survival rates⁵. Cachexia also associates with poorer outcomes from surgery, radiotherapy, and immunotherapy and increased cost of care.⁶ Across cancer types and interventions, cachexia decreases patients’ quality of life (QOL) as reflected by increased fatigue and decreased physical and social functioning⁷. Currently there is no effective, approved therapy for cancer cachexia in most of the world, underscoring both the need and the opportunity for further research and development in this condition.

Challenges for intervention in cancer cachexia

There are no targeted therapies nor standardized supportive care regimens for cancer cachexia, despite its outsized impact on patient quality and length of life. The disappointing state of the field can be attributed to lack of clarity on clinical definitions, lack of high-quality data for nutritional or other supportive care interventions thereby hampering treatment consensus and reimbursements, and lack of clinical data supporting novel pharmaceutical interventions. As well, the pre-clinical data defining key mechanisms and targetable pathways are generally small and not rigorously tested in gold-standard models of site-specific cancer and anti-cancer therapy. We consider these challenges in turn here.

Diagnostic criteria of cancer cachexia

According to the framework developed by Fearon et al ², the cachexia syndrome includes three stages of severity—pre-cachexia, cachexia, and refractory cachexia. At the pre-cachectic stage, the patient exhibits simple weight loss (<5%) accompanied by clinical and metabolic symptoms like anorexia and glucose intolerance. The patient is considered cachectic given the presence of one or more of these symptoms: 1) weight loss >5% over past 6 months, 2) weight loss >2% and BMI <20, or 3) weight loss >2% and sarcopenia as detected by mid upper-arm muscle area (MUMA) (men <32 cm², women <18 cm²), appendicular skeletal muscle index by dual energy x-ray absorptiometry (DXA) (men <7·26 kg/m²; women <5·45 kg/m²), lumbar skeletal muscle index by computed tomography (CT) (men <55 cm²/m²; women <39 cm²/m²), or fat-free mass index without bone by bioelectrical impedance (BIA) (men <14·6 kg/m²; women <11·4 kg/m²). Refractory cachexia represents the terminal stage corresponding to limited self-care or complete disability on World Health Organization performance status and life expectancy less than 3 months. At this point, patients are not expected to benefit from any weight management therapy and therapeutic intervention is limited to alleviating cachexia-associated complications,² thus highlighting the importance of early detection.

While the consensus definition associates cachexia with mortality⁸ and has propelled substantial research in the field, garnering nearly 1,200 citations to date, the current criteria are insufficient to make a distinction between no-cachexia and pre-cachexia⁹ and often difficult to implement in clinical practice. In this definition, cutoffs for weight loss and muscle mass lacked rigorous evidence to be predictive of clinical outcomes. As well, weight loss can be masked by fluid retention and ascites or overlooked due to measurement error or patients’ inaccurate recall. Measurements of muscle mass is also challenging in the clinic. Mid upper-arm muscle area measurement is accessible but requires special training to ensure measurement consistency and minimize inter-personnel error. DXA and BIA are not readily available at cancer centers or typically ordered or reimbursed for cancer patients. Muscle mass can be determined from diagnostic CT scans, often readily available in oncology practice. However, CT measurements of muscle require both specialized software and training.¹⁰ Another complicating factor is the large discrepancy among different measurement methods in detecting low muscle mass. Blauwhoff-Buskermolen et al. (2017) reported that prevalence of cachexia in the same group of cancer patients varied from 37% with MUMA as a measurement method to 48% with BIA¹¹. This underscores the need for revisiting current cutoffs for each method. Finally, suitability of these cutoffs for different ethnicities and races requires further investigation¹². Functional assessments might ultimately complement or supersede body composition measurements. Grip strength has been shown to correlate with sarcopenia, inflammation, and survival¹³, for example, and stair climb power and upper body strength associate to cachexia.¹⁴ However, robust data to support specific functional assessments in diagnosis, staging, or monitoring of cachexia are currently lacking.¹⁵

More recent efforts to define and classify cachexia include a grading system incorporating both body mass index (BMI) and history of weight loss¹⁶, criterion values for food intake impairment and the inflammatory biomarker C-reactive protein (CRP)¹⁷, the combination of weight loss, BMI, and mid-upper-arm muscle area ¹⁸, a combination of routine biochemistry with food intake, weight loss and performance status, among others ¹⁹, or even the abridged patient-generated subjective global assessment ²⁰. Against this backdrop of new studies, an expert panel is currently engaged in review of these and other grading systems to develop a new consensus framework for diagnosis and staging of cancer cachexia.

Manifestations and mechanisms of cancer cachexia

Anker et al. estimated that 527,100 US patients in 2104 and 800,300 European Union patients in 2013 suffered from any kind of cancer cachexia.²¹ Findings differ, but generally cachexia is most prevalent and severe in cancers of the upper gastrointestinal tract, with incidence of cachexia in pancreatic and liver cancers reaching 70-85%.^{22, 23} Over half of patients with gastro-esophageal and head and neck cancers will suffer cachexia, along with 50% of patients with non-small-cell lung cancer. Cachexia is significant in patients with breast (25%) and prostate (15%) cancer also. It is believed that weight loss severity increases with disease progression, such that most patients with metastatic disease develop cachexia.²⁴ Magnitude of weight loss generally follows prevalence among disease conditions, with greatest average weight loss in gastrointestinal cancers and least in breast and prostate cancers.

Importantly, while patients with severe cachexia become emaciated, cachexia is distinct from starvation. Anorexia is a major, though not a solo player in cachexia. Large numbers of advanced cancer patients report loss of appetite, early satiety, or both leading to weight loss and malnutrition^{25, 26}. Multiple factors contribute to cancer associated appetite changes notably depression and anxiety due to diagnosis and/or therapy. Nausea and vomiting are common side effects of chemotherapy and both radio and chemotherapy can cause stomatitis and esophagitis as side effects leading to a further decrease in food intake²⁶. Factors produced by tumors such as lactic acid or due to host-tumor interaction such as inflammatory cytokines also cause appetite suppression through direct action on the hypothalamus²⁷. However, the limited effectiveness of nutritional support and dietary counseling in improving cancer cachexia^{28, 29} highlights the role of metabolic dysregulation in weight loss in cachexia.

Despite the shared outcome of weight loss with starvation and protein malnutrition, cancer cachexia manifests unique metabolic reprogramming and behavior, as reviewed by Olson and colleagues (2020)³⁰. Under starvation, resting metabolic rate decreases to minimize energy expenditure with a preferential utilization of fat stores over lean mass to meet energy demands. However, cachectic patients show an increase in resting metabolic rate compared to pre-cachexia followed by a decrease in late stages of cachexia probably due to depletion of energy stores. Behaviorally, starved subjects exhibit increased appetite and foraging activity searching for a food source. Patients with cancer cachexia, however, display decreased appetite and feeding disinterest. Even though there is an increase in catabolism of muscles in protein malnutrition, possibly to compensate for lacking amino acid/s, it is much milder than what happens in cachexia. Like starvation, protein malnourished individuals exhibit an increase in foraging activity and appetite, however, in this case to protein-rich food.³⁰.

Muscle and fat wasting are prominent hallmarks of cancer cachexia. In cachexia, muscle and lipid catabolism are estimated to rise by 40-60% and 30-80% respectively. Muscle wasting is mediated by increase in protein degradation rate and suppression of protein synthesis^{31, 32}. Both the ubiquitin proteasome system and autophagy play a role in protein degradation^33-35. The decrease in protein synthesis can, at least partially, be attributed to suppression of mTORC1 signaling that induces protein synthesis³⁶. In addition, a recent study has pointed out a possible role for decreased ribosomal capacity in downregulation of muscle protein synthesis through downregulation rDNA³⁷. Cardiac atrophy in advanced cancer patients, involving a reduction in both size and number of cardiac muscle fibers, has been documented from many years³⁸, and heart failure due to cardiac muscle wasting is hypothesized to play a role in high mortality among cachectic patients³⁹. White adipose tissue (WAT) wasting is an early step in cancer cachexia. The importance of adipose tissue wasting is indicated by a study showing that ATGL lipase knocked out mice were not only protected from WAT loss but also had attenuated muscle loss⁴⁰. In addition to lipolysis, at least in rodents, WAT undergoes a process called browning characterized by shifting from energy storing phenotype (white) to energy spending phenotype (brown) through upregulation of UCP1 protein that uncouples electron transport chain from ATP synthesis. This in turn contributes to increase in energy expenditure seen in cachectic animals⁴¹. Recent evidence indicates that inflammation-induced lipolysis from adipose tissue in cancer cachexia contributes to myofiber atrophy and is additive to inflammation-induced protein catabolism in muscle, suggesting that adipose loss can drive muscle loss in cancer.⁴²

Circulating mediators of cachexia

Cachexia results from the host response to the tumor. As such, inflammatory cytokines produced at the tumor-host interface are major mediators of cancer cachexia⁴³. Major players in this interaction include the cytokines Interleukin-6 (IL-6) and Tumor Necrosis Factor-alpha (TNF). IL-6 is over-expressed in cachectic compared to non-cachectic cancer patients^{44, 45}. IL-6 promotes lipolysis and browning of WAT⁴⁶. Treatment of myotubes with IL-6 induces autophagy⁴⁷, and inhibits protein synthesis through suppression of the mTORC1 pathway³⁶. In models of cancer cachexia, inhibition of IL-6 and downstream Stat3 signaling alleviates weight loss and muscle wasting ^{34, 48, 49}. Some studies have even reported favorable response to anti-IL-6 antibody therapy in cachectic cancer patients making it a promising drug target^{50, 51}. TNF-α inhibits expression of GLUT-4, thus inhibiting glucose uptake in adipose and promoting lipolysis. TNF and other inflammatory cytokines can also activate NF-KB signaling in muscles which in turns inhibits myogenic differentiation and upregulates expression of genes involved in the ubiquitin-proteasome system⁴³.

The Transforming Growth Factor – β (TGF- β) family members Activin, Myostatin and Growth/Differentation Factor-15 (GDF-15) have been interrogated in cancer cachexia to the point of clinical trials. Exogenous administration of each of these induces cachexia.^{52, 53} Activin is produced by tumor and/or host in experimental systems of pancreatic, ovarian, colon and other cancer cachexia, and its inhibition with neutralizing antibodies or a soluble receptor trap mitigates wasting in the absence or presence of anti-tumor therapy.^54-58 Unfortunately, a phase 2 trial of an anti-Activin antibody, bimagrumab, failed to show promise in patients with cancer cachexia due to lung or pancreatic cancer (ClinicalTrials.gov NCT01433263); bimagrumab is actually now under investigation to promote fat loss in diabesity.⁵⁹ Targeting of the related factor, Myostatin, which plays an essential role in limiting muscle mass and binds the same receptors as Activin, also failed to confer clinical benefit in a phase 2 trial of patients with pancreatic cancer.⁶⁰ GDF-15 is over expressed in most inflammatory disease conditions, and has been shown to promote anorexia through nausea and emesis.^{61, 62} GDF-15 is increased in human and experimental cancer cachexia.⁶³ Neutralization of GDF-15 signaling through GFRAL⁶⁴, reduces nausea and vomiting, improves food intake and restores body weight in rodent models of cancer cachexia and cancer therapy.⁶⁵ Clinical studies are underway to determine efficacy of GDF-15/GFRAL neutralization in patients with cancer cachexia.

Along with these well-described molecular targets, pre-clinical studies have led to a considerable expansion of causal mediators of cancer cachexia of late. Rodent studies have implicated tumor-derived Fn14⁶⁶, extracellular HSP70 and HSP90⁶⁷, and leukemia inhibitory factor⁶⁸, along with host-derived Ataxin-10⁶⁹, VEGF-A⁷⁰, lipocalin 2⁷¹, as well as tumor/host IL-8⁷², among others.

Management of cancer cachexia

Multiple recent consensus guidelines for management of cancer cachexia highlight the struggle to find effective therapies for cancer cachexia. In 2017, The European Society for Clinical Nutrition and Metabolism (ESPEN) recommended 1) screening for nutritional risk regardless of BMI or history of weight loss, 2) assessment of food intake, body composition, inflammation, energy expenditure and physical function, and 3) treatment with individualized multi-modal therapy to improve nutritional intake, lessen inflammation and metabolic stress, and increase physical activity.⁷³ In contrast, the 2020 American Society of Clinical Oncology (ASCO) recommended only dietary counselling and appropriate use of corticosteroids and progesterone analogs.⁷⁴ Both ESPEN and ASCO stressed that artificial feeding (enteral and parenteral) should only be applied for patients with impaired oral intake or intestinal dysfunction. They advise against non-conditional tube feeding given no proven benefit and to avoid non-necessary complications such as diarrhea, infection, refeeding syndrome, etc⁷³. In 2021, the European Society of Medical Oncology (ESMO) confirmed the need for 1) nutritional screening, and 2) comprehensive assessments of the patient’s clinical, psychological, and social condition along with medications and tumor status, to 3) arrive at a tailored intervention addressing nutrition, social support and exercise, recommending for corticosteroids, olanzapine, and progestins to improve appetite, while cautioning risk of serious side-effects, including thromboembolism, for the last.⁷⁵

Malnutrition is prevalent among cancer patients due to disease/ therapy associated gastrointestinal dysfunction or anorexia⁷⁶. Malnutrition decreases QOL and increases infection risk, treatment toxicity, and mortality^{77, 78}. Countering malnutrition in cancer patients requires early and continuous nutritional screening, regardless of patient weight/ BMI⁷⁹, to identify patients at risk of malnutrition and ensure early intervention. At diagnosis, all cancer patients should receive nutritional counseling to educate them about their calories, protein, micronutrient needs and practices to maintain lean mass. Screening tools should check for signs of anorexia, changes in appetite, and physical function in addition to symptoms that may affect food intake e.g. constipation, nausea, etc^{73, 80}. When possible, using body composition analysis technique can facilitate early detection of sarcopenia and cachexia that may be masked in obese patients⁷⁹. Measures of systemic inflammation level, like Glasgow Prognostic Score ⁸¹ that depends on serum levels of C-reactive protein and albumin, can help predicting patients at risk of malnutrition and cachexia⁸². Calorimetry measures can enhance accurate estimation of calorie and protein needs⁸⁰, notably that REE is enhanced in some cachectic patients²⁹. When needed, personalized nutrition management plan should be developed, and suitable nutritional supplements can be provided^{73, 80}.

Progesterone analogs are synthetic versions of female hormone progesterone that were shown to improve appetite through unknown mechanism/s. Megestrol acetate (MA), a widely investigated one, has been approved by FDA in 1993 as a therapy for anorexia, cachexia, and unexplained weight loss in AIDS patients⁸³. Clinical trials have shown positive effect of MA on cachectic cancer patients considering appetite and non-fluid weight gain^{84, 85}. However, reported effects of MA treatment on QOL were inconsistent probably due to variation in tools used to assess QOL, patient’s inclusion and exclusion criteria, and/ or used dosage. Major side effects include edema, impotence in male patients, and to a lesser extent thrombosis^{83, 86}.

Corticosteroids are synthetic analogues of steroid hormones, more precisely glucocorticoids. Like glucocorticoids, corticosteroids have an anti-inflammatory effect⁸⁷. They are frequently used in palliative care to improve anorexia, fatigue, pain, and QOL^{88, 89}. They are also established treatment for chemotherapy associated nausea and vomiting⁹⁰. Clinical trials showed that corticosteroids improve anorexia and fatigue in cancer patients^{91, 92}. Many mechanisms may mediate steroid positive effects: maintaining physiological levels of glucocorticoids that are necessary general well-being, anti-inflammatory effect, inhibition of prostaglandins and substance P production; both are implicated in vomiting response⁹³. Treatment usually lasts 7 to 14 days likely to avoid notorious side effects of steroids. Major side effects include hyperglycemia, immune suppression, and adrenal suppression⁹⁴. However, adverse effects are usually observed with higher doses and/ or long-term treatment.

Pre-clinical modeling of cancer cachexia

Challenges to defining appropriate nutritional or other targeted interventions for cancer cachexia include low strength of evidence from clinical trials, as reflected in the consensus guidelines, but also a scarcity of robust, rigorous, and reproduced studies in pre-clinical models. While much of what we know about cancer cachexia has been learned through rodent models, most mechanisms have been inferred from a small number of models with rather low fidelity to human tumor biology. Such mechanisms are generally extrapolated to diverse cancer types and treatment contexts without robust direct experimental testing. While such studies have pointed to important biological underpinnings and revealed potential targets, almost no studies of interventions for cancer cachexia have been tested in the context of both a tumor and a clinical therapy. Thus, the current state of cancer cachexia research is akin to early days of cancer research, where tumors, mechanisms, and treatments were conflated. Hence, there is an opportunity to improve the pre-clinical modeling of this condition. With the intent of spurring additional research in mechanisms of cancer cachexia, particularly among nutritional experts, we review here the more historical models as well as newer models as the field grows in sophistication.

Lewis lung carcinoma.

Lewis lung carcinoma cells were isolated from spontaneous carcinoma in the lung of a C57BL mouse. In 1992, Ohira et al reported that implantation of LLC cells transfected with IL-6 cDNA in C57BL/6 mice induces a cachexia like phenotype⁹⁵. The model improved over time and wild type LLC cells are now implanted intramuscularly to induce cachexia. After 2 to 3 weeks of implantation, this model exhibits a decline in physical activity, increased systemic lipid oxidation, and subsequent fat and muscle wasting without appreciable reduction in calorie intake compared to normal control.⁹⁶ Muscle loss correlates to tumor burden and is due to increase in protein degradation rate, mainly through ATP dependent proteolysis ^{35, 97} without change in protein synthesis rate. This leads to a decline in muscle and about 5% loss in bone density³⁵, as well as cardiac atrophy.⁹⁸ Recent studies reveal alterations of diaphragm function, hepatic metabolism, and the gut microbiome in LLC cachexia. This model has revealed important roles for host-derived initiators of cachexia as well as potential therapeutics, in part due to its compatibility with C57BL/6 strains of genetically modified mice. However, limitations include the ectopic tumor location, low tumor resemblance to human lung carcinomas with little complexity in the micro-environment, high tumor size heterogeneity, and short time course.

C26 colon carcinoma.

The C26 colon cancer model, heavily used in cancer cachexia, was originally developed by Tanaka et al (1990)⁹⁹ and later was molecularly, cellularly, and physiologically characterized¹⁰⁰. This model is generated by subcutaneous or intramuscular injection of C26 murine colon cancer cells into CD2F1 or Balb/c mice. The mice usually show weight loss, with fat and muscle wasting within 15 days of implantation without significant reduction in food intake. This is accompanied by hyperglycemia, an acute phase response, and increase in serum levels of inflammatory cytokines⁹⁹, notably IL-6^{48, 101}, which was shown to be downstream of the related cytokine, Leukemia Inhibitor Factor. The mice also exhibit decrease in muscle strength and increased fatigability¹⁰⁰. By the time of euthanasia, weight loss reaches up to 40% in carcass weight compared to controls⁹⁹. Investigators have used intrasplenic injection of C26 cells to produce liver metastases to model metastatic colorectal cancer, resulting in greater cachexia severity than the original model.

Genetic model of intestinal cancer.

Apc^min mice are heterogenous for a truncating mutation in APC (adenomatous polyposis coli) gene¹⁰². This gene encodes a tumor suppressor protein that regulate Wnt signaling pathway¹⁰³, and it is frequently mutated in both hereditary and sporadic colorectal adenomas and carcinomas^{104, 105}. This model develops spontaneous intestinal adenomas and tumors detectable at 4 weeks of age¹⁰⁶. Early cachexia is observed around age of 16 weeks³⁴. The mice exhibit both fat and muscle loss without change in food intake compared to wildtype controls. They show increase in muscle protein degradation rate initially through ATP dependent ubiquitination system and then by ATP independent autophagy as well at more advanced stages³⁴. Also, plasma IL-6 is elevated in this model and has been shown to mediate muscle wasting in this model³⁴. These mice develop cardiac atrophy¹⁰⁷; however cardiac muscle loss appears to be mediated by autophagy and not the ubiquitin-proteasome system as in skeletal muscle. Both skeletal and cardiac muscle wasting is accompanied by decrease in activation of mTOR signaling pathway that activates protein synthesis^{34, 107} As in humans, polyp burden in Apc^min is influenced by diet¹⁰⁸ and genetic factors. Thus, traditionally they are kept at C57Bl/6J background. While much has been learned about mechanisms of cancer cachexia in this model, including sex specificity of cachexia mechanisms and phenotypes^{109, 110} and bone loss¹¹¹, the prevalence of intestinal polyps over colon tumors limits the utility of this mouse for modeling colorectal cancer cachexia.

Genetically engineered models of pancreatic cancer.

Tumor suppressor Trp53 and proto-oncogene Kras are mutated in more than 50 and 80% respectively of pancreatic ductal adenocarcinoma (PDAC) patients. Using this knowledge, Hingorani et al employed the Cre-Lox system to target a Kras activating mutation and Trp53 dominant negative mutation into mouse pancreatic cells, generating the first genetically engineered KPC PDAC model¹¹². Disease burden reaches significance levels in these mice as early as age of 10 weeks. These tumors are highly metastatic with preference of liver and lungs. Most of animals develop cachexia and achieve 100% mortality by 12 months. These and related strains also show both muscle and white adipose tissue wasting.^113,114. Spontaneous occurrence of tumors in this model and relatively long survival of tumor bearing mice allows study of cachexia throughout different progression stages.

Orthotopic pancreatic tumor models.

Several groups have isolated cell lines from PDAC lesions of KPC mice and implant them into the pancreata of recipient mice as syngeneic allograft giving rise to KPC orthotopic model^{42, 115, 116}. Such models exhibit anorexia reflected by decreased food intake around 8 days after implantation. They also show both lean mass and fat loss and decreased locomotor activity. IL-6 and IL-6 trans-signaling via soluble IL-6 receptor seems to play a significant role in this model of cachexia⁴². Usually, mortality is reached 14-21 days after implantation¹¹⁵. A clear limitation of this model is that the observed phenotypes depend largely on the cell line used, which varies among different labs. Use of highly characterized lines, specifically multiple isolates of the several defined molecular phenotypes in PDAC, across laboratories would enable replication and improve rigor and reproducibility of results.

Other cancer types and metastatic disease.

In addition to those above, novel models used in pivotal mechanistic studies have included cell implantation of variant isolates of lung adenocarcinoma¹¹⁷, renal cell carcinoma¹¹⁸, ovarian cancer¹¹⁹, and other primary tumors (reviewed in Tomasin et al.¹²⁰) as well as genetic models of lung cancer^{121, 122} and others, but there exist only a handful of studies for each to date. Given the prevalence of cachexia in patients with metastatic disease, investigators have also sought to model cachexia in mice with induced metastasis. This has included the MMTV-PyMT transgenic model of metastatic mammary cancer¹²³ as well as intra-tibial¹²⁴ and intra-cardiac injection¹²⁵ to induce osteolytic breast tumors, along with the previously mentioned intra-splenic injection of colon cancer to induce seeding of tumors in the liver¹²⁶. Implantation of tumor cells followed by resection and relapse led to development of multiple metastatic cancer cachexia models by Wang et al.¹²⁷ However, overall use of metastatic models for cachexia research has been limited to a handful of labs.¹²⁰

In general, cancer cachexia models that use cell lines to induce cancer share common disadvantages. For example, tumor in these models can reach up to 10% of body weight¹⁰⁰ which typically does not happen in patients with cachexia-associated tumors. The location of implanted cancer cells, ectopic vs orthotopic, is also likely to affect tumor microenvironment and hence cancer cell behavior. Orthotopic versus ectopic implantation of PDAC cells has been shown to produce different cachexia phenotypes¹¹⁵. A study has shown that difference in cells storage condition affects plasma IL-6 levels and degree of decline in voluntary wheel running activity as a measure of fatigue in C26 model¹²⁸. Thus, possible effects of passage number, number of implanted cells, and suspension buffer (Matrigel, PBS, media) on observed phenotype need further investigation. Furthermore, fast disease progression does not allow study of early stages of cachexia. Finally, it is not known how cachexia in different patients, disease contexts, or therapies is driven by different mechanisms. Increasing sophistication of models should begin to address these limitations and opportunities for discovery.

Conclusions

Despite considerable effort by a rather small number of dedicated researchers around the globe, cancer cachexia remains a major unmet medical need. This is to be expected, however, given the relatively low research volume in cancer cachexia to date. As of August 2021, the 4,433 primary publications on “cancer cachexia” in PubMed constituted only 0.12% of all 3,718,027 papers on “cancer”. The 192 interventional clinical trials in cancer cachexia registered on ClinicalTrials.gov were only 0.28% of the 68,551 interventional trials for cancer. This contrasts with 7,040 interventional trials for lung cancer, a disease in which substantial progress has been made in the past few years. Importantly, such statistics hide the current state of exponential growth in research and trials in cancer cachexia (Figure), in addition to the growth in investment from funding agencies and pharmaceutical companies and considerable activity in looking for pathways to regulatory approvals.²¹ It is expected that continued growth, investment, and coordination of research in this topic will ultimately yield robust biomarkers, widely-accepted classification and staging algorithms, targetable pathways, pivotal clinical trials, and ultimately, cures.

Figure: Growth in cancer cachexia research.

Primary publications by year returned when searching for “cancer” (left axis) versus “cancer cachexia” (right axis) in PubMed, accessed August 2021.