Cancer cachexia can be divided by underlying cause into two basic categories: primary and secondary. Secondary cancer cachexia is caused by reduced nutrient intake resulting from the physical effects of the tumor itself or the therapies used to treat it. It is intuitively obvious that many tumors have a negative impact on host nutritional status simply because of their location or size: for instance, large intraoral masses may prevent normal food intake, and diffuse neoplastic infiltration of the small bowel may significantly disrupt normal digestion or absorption of nutrients. It also is well established that decreased food intake may occur secondary to various anticancer therapies. Cancer treatment may be associated with important abnormalities in nutrient intake, digestion, and absorption because of the nausea, vomiting, mucositis, and diarrhea caused by radiation therapy or chemotherapy.
In contrast, primary cancer cachexia is not understood nearly as easily. It probably is best defined as a paraneoplastic syndrome in which various metabolic abnormalities lead to inefficient energy utilization by the host, which results in accelerated loss of both lean body mass and adipose stores. The presence of these underlying metabolic abnormalities means that, unlike secondary cancer cachexia, primary cancer cachexia cannot be reversed simply by increasing nutrient intake. Published work suggests that primary cancer cachexia is a complex systemic inflammatory syndrome caused by alterations in proinflammatory mediators that have wide-ranging effects on intermediary metabolism. Interleukin-1α (IL-1α), IL-1β, IL-6, tumor necrosis factor-α (TNF-α), and interferon-γ all have been implicated in the pathogenesis of this disorder, as have a number of eicosanoids. Changes in the production of these substances also may lead to neuroendocrine abnormalities, including variations in catecholamine and cortisol secretion and changes in the relative and absolute concentrations of plasma insulin and glucagon. The specific changes that occur vary from affected individual to individual but ultimately cause characteristic perturbations in the metabolism of all three energy substrates—carbohydrate, protein, and lipid. Classically, abnormalities in carbohydrate metabolism have been described most often and have included hyperlactatemia, increased rates of whole body glucose turnover and disposal, increased rates of gluconeogenesis from lactate and amino acids, abnormal glucose tolerance curves, hyperinsulinism, and insulin resistance. Documented changes in protein metabolism consist of altered serum amino acid profiles, increased rates of whole body protein turnover, decreased protein fractional synthetic rates in skeletal muscle, and increased protein fractional synthetic rates in liver. Finally, abnormalities in lipid metabolism include accelerated fat oxidation, increased lipolysis, hypertriglyceridemia, decreased lipoprotein lipase activity, increased synthesis of triacylglycerols and very low-density lipoproteins, and increased plasma concentrations of nonesterified or free fatty acids and ketone bodies (acetoacetate and β-hydroxybutyrate).
Theoretically, host energy expenditure should be increased as a consequence of futile cycling and the changes in flux through these varied metabolic pathways, with host weight loss being the ultimate result. Many investigators have attempted to prove this hypothesis by using indirect calorimetry to measure energy expenditure in tumor-bearing subjects. Unfortunately, the results of these studies are variable and difficult to interpret. Energy expenditure seems to vary with tumor type and stage of disease. It appears to be increased in some studies, to be apparently unchanged in others, and actually to be decreased in still others. One valid conclusion must be that significant variation exists in energy expenditure and the manifestations of cancer cachexia, not only among tumor types but also among individuals with the same tumor, depending on the stage of the disease. However, one must also consider several additional explanations for these findings. First, the methodology involved in indirect calorimetry is complex, and results may be difficult to reproduce consistently. Second, selection of the best controls for such studies is problematic: some authors compare their cachectic cancer patients with young, healthy, weight-stable control subjects, whereas others insist that the only appropriate comparison is between weight-losing cancer patients and weight-losing patients with nonmalignant disease. Finally, the analysis of results is complicated further in animals because the true energy requirements of even healthy cats and dogs are controversial and incompletely defined. Much more work is needed to define thoroughly the changes in energy expenditure that occur in people and animals with cancer.
Although cancer-associated weight loss has been studied most extensively in people and rodent models, work by veterinary investigators suggests that many of the metabolic changes typical of primary cancer cachexia also are present in dogs with naturally occurring neoplastic disease. Dogs with lymphoma have been studied in greatest detail, and this tumor has been proposed as a model of primary cancer cachexia. However, a convincing clinical association between documented metabolic abnormalities, actual weight loss, and poor prognosis has yet to be demonstrated in dogs with any type of cancer. Recent work shows that clinically relevant hyperlactatemia is uncommon in dogs with all types of neoplastic disease, including lymphoma. When hyperlactatemia is present, it almost always is related to decreased tissue perfusion and not to increased production by tumor cells (Touret et al, 2010). Furthermore, weight loss appears to be less common in dogs with cancer than in people with neoplastic disease. One study found that only 4% of dogs evaluated or treated for neoplastic disease by a large referral oncology practice were cachectic as defined by body condition scoring, whereas 29% were obese (Michel et al, 2004). Similarly, another group of investigators found that even though dogs with malignant tumors were somewhat less likely to be overweight than dogs without neoplastic disease, there was no difference in the proportion of underweight or very thin dogs in the two groups (Weeth et al, 2007). There are several possible explanations for the apparent discrepancy between the prevalence of weight loss in dogs and in people with cancer. The most common types of cancer diagnosed and treated in dogs and in people are not the same, and those treated most often in dogs (e.g., lymphoma) are relatively less likely to result in weight loss in people. In addition, it seems probable that people receive aggressive anticancer therapy despite facing a poor prognosis more often than do dogs, and this type of patient would be expected to be at increased risk of weight loss. Finally, dogs with cancer routinely are euthanized when their quality of life declines, and this is likely to decrease further the prevalence of cancer-associated weight loss in dogs compared with people. Regardless, the potential differences between dogs and people with respect to the prevalence of cancer-associated weight loss are intriguing and merit further confirmation and study.
It is interesting to note that published work suggests that the relationship between weight loss and the tumor-bearing state may be different in cats than in dogs. In one study, the effect of body condition score on prognosis was examined in cats with neoplastic disease (Baez et al, 2007). Almost half of the cats evaluated were underweight or very thin, and over 90% of them had evidence of muscle wasting. Body condition score was strongly correlated with survival time and prognosis: cats with decreased body condition scores had markedly shorter survival times. Other investigators also have shown that weight loss is a negative prognostic indicator for some cats with lymphoma (Krick et al, 2011). Further work is needed to determine if the relatively lower body condition that apparently exists in many cats with cancer is related specifically to the presence of underlying neoplastic disease or whether it represents a more generic feline response to illness. It seems possible that sick cats are simply more likely to experience weight loss than sick dogs, regardless of whether they have neoplastic disease or not.
