Cachexia is a progressive loss of body weight, accompanied by loss of appetite, which affects as many as 80% of cancer patients. The current focus of anti-cachexia research is blockade of tumour-derived circulating factors at the level of fat and muscle. However, given that the brain is the master regulator of metabolic control, targeting the brain to alter metabolic outcomes in cachexia is an attractive idea. Ghrelin is a peptide hormone which rises in response to fasting, drives eating behaviour, decreased energy expenditure and growth hormone release. The primary target neurons of ghrelin are the Agouti related peptide/neuropeptide Y-containing neurons in the arcuate nucleus of the hypothalamus (AgRP neurons). Therapies using ghrelin analogues have been trialled as a way to target the brain to treat cachexia. We used a mouse model of pancreatic ductal adenocarcinoma (PDAC) to assess ghrelin action in vivo. After the onset of PDAC-induced anorexia, PDAC-carrying mice ate significantly less than control mice in response to injected ghrelin, indicating ghrelin resistance is present in cancer cachexia even before noticeable wasting has occurred.
To circumvent the observed ghrelin resistance, we used targeted chemogenetics (DREADDs) to chronically artificially activate AgRP neurons during cancer cachexia in PDAC-bearing mice. AgRP neuronal activation rescued fat and skeletal muscle mass loss, decreased brown fat thermogenesis and slightly but significantly increasing locomotor activity in PDAC mice. We measured circulating levels of the pro-cachexia factors, activin A and B. PDAC-bearing mice with or without AgRP neuronal activation showed a similar, significant elevation in activin A and B levels, compared to non-PDAC bearing mice. Importantly, AgRP neuronal activation protected mice from the wasting effects of elevated activins, as the levels seen in this model are sufficient to drive significant wasting. This provides early evidence that targeting central feeding circuits may improve outcomes in cancer cachexia.