When less may be more: calorie restriction and response to cancer therapy

CR, a chronic reduction of dietary energy intake by approximately 30% without incurrence of malnutrition, is a broadly effective dietary intervention that significantly decreases adiposity and inflammation and improves metabolic profiles in non-obese humans and rodents [1‐4]. Preclinical studies in mammalian models demonstrate that CR extends lifespan, ameliorates risk factors, and delays onset of age-related diseases, including cancer, type II diabetes, and cardiovascular and neurodegenerative diseases [4]. In response to decreased caloric intake, metabolic alterations foster health promoting characteristics, including increased insulin sensitivity and decreased blood glucose, growth factor signaling, inflammation, and angiogenesis [4]. While the impact of CR on age-related pathologies has been studied most extensively in rodent models, data from human observational and randomized clinical trials demonstrate that CR in non-obese humans results in metabolic and molecular changes similar to those observed in rodent models [5]. Within the scope of cancer research, a meta-analysis of preclinical rodent models evaluated the impact of CR across multiple cancer types and through a variety of tumor models [6]; overall, CR displayed a 75.5% reduction in tumor incidence. Longitudinal studies at the National Institute of Ageing and the University of Wisconsin showed a significant reduction in the incidence of cancers in rhesus monkeys fed a CR diet compared to a control diet [7]. While the antitumorigenic effects of CR are well established, the mechanism behind this relationship remains unclear, though it is believed that the tumor suppressive effects are mediated, in part, by enhanced apoptosis within tumors, modulation of systemic signals such as insulin-like growth factor (IGF)-1, insulin, metabolic and inflammatory pathways, as well as by reduced angiogenesis.

Exposure to an energy restricted diet results in reduced systemic glucose and growth factors such as IGF-1 [1, 8, 9]. Preclinical studies in breast, pancreatic, and colon cancer have demonstrated that modulation of IGF-1 signaling plays a major role in CR’s anticancer effects [8, 10, 11]. In alignment with this, population studies have demonstrated that the IGF-1 signaling pathway plays a significant role in the development and progression of many cancer types [12]. IGF-1 is a nutrient-responsive growth factor that activates two major signaling cascades, namely Ras/MAPK and PI3K/AKT. Activation of the Ras/MAPK pathway promotes activity of transcription factors and subsequent expression of genes involved in proliferation and cellular growth. Initiation of the PI3K/AKT pathway promotes decreased apoptosis by disrupting the BCL2-Bad complex, increases protein synthesis via mTOR activation, and increases glucose metabolism by inhibiting GSK-3β [13]. Cancer cells utilize the IGF-1 signaling pathway to redirect their metabolic investment towards proliferation and growth, and thus reduction of IGF-1 levels in CR results in decreased tumor growth and progression [8, 10, 11]. Addition of exogenous IGF-1 leads to the partial reversal of the anticancer effects of CR, further supporting the role of IGF-1 in tumorigenesis [10]. In addition, expression of signaling factors downstream of IGF-1 has been correlated with either resistance or sensitivity to several cancer therapies [14]. Nevertheless, although IGF-1 signaling is a promising anticancer target, drugs targeting the pathway have been largely unsuccessful [12].

CR also induces activation of AMP-activated protein kinase (AMPK), a molecular sensor that increases catabolism and inhibits anabolic metabolism, working in opposition to IGF-1-mediated activation of mTOR [4, 15]. AMPK activation in response to CR conditions results in increased apoptosis within brain tumors while protecting normal cells from the stress [16, 17]. AMPK induces expression of metabolic control genes, including SIRT1, resulting in increased fatty acid oxidation and glutaminolysis to provide auxiliary substrates when glucose is scarce [18]. In line with this, CR results in elevated serum glutamine [19] and ketone bodies [20]. Many tumors undergo metabolic reprogramming, including enhanced fatty acid oxidation and glutaminolysis in addition to increased glucose metabolism [21‐23]. Some cancers can therefore become autonomous, uncoupling their growth from the availability of systemic factors under normal conditions. Despite it not being clear whether administration of CR would support tumor growth in these circumstances, it is thought that transformed cells lack the metabolic adaptability to respond to altered substrate availability, having already invested heavily in metabolic reprograming and thus being more sensitive to increased pressure on ketone metabolism and fatty acid oxidation than normal cells [24].

The AMPK/mTOR axis also controls autophagy, a process through which proteins, macronutrients and organelles are enveloped in double-membraned vesicles and degraded into building blocks that can then be shuttled into synthetic pathways. Under CR conditions, AMPK activation stimulates increased autophagy to recycle cellular components and provide needed substrates for metabolism and homeostasis [4, 15]. Autophagy can act as a double-edged sword in cancer – it is thought to be tumor suppressive since defects in autophagy drive oxidative stress, mitochondrial defects, DNA damage, genomic instability, and tumor growth [25]. Conversely, it is believed to be tumor promoting because established tumors can utilize autophagy to reduce oxidative stress and increase mitochondrial function and metabolism in order to promote survival and overcome stress and low nutrient conditions [25, 26]. Due to the dual role of autophagy in cancer, autophagy inducers and inhibitors have become targets for cancer therapy [27]. Recent research in our lab showed that combining autophagy inhibition with a CR regimen reduced tumor growth more than either treatment alone [20].

Furthermore, increasing preclinical and human evidence suggests that CR reduces inflammation [11, 28, 29]. Multiple inflammatory signaling cascades can promote the growth and survival of neoplastic cells [30]. The reduction of energy intake in CR reduces the amount of adipose tissue, a major endocrine organ that secretes pro-inflammatory factors including leptin, adiponectin, monocyte chemo-attractant protein-1, tumor necrosis factor, and interleukin-6 [31]. CR in human studies is associated with reduced adiposity and a lessened inflammatory adipose secretome, as well as particularly decreased systemic levels of pro-inflammatory adipokines [32]. In addition, CR has been shown to consistently reduce the expression of the pro-angiogenic factors vascular endothelial factor [9, 33] and plasminogen activator inhibitor-1 [15], both of which induce the growth of new blood vessels to provide growing tumors with oxygen and glucose. CR has also been shown to reduce vascularization of tumors [9, 16]. In addition to altering systemic inflammatory mediators, CR has been shown reduce the expression of inflammatory genes in cancer cells, including nuclear factor kappa B [11, 34] and peroxisome proliferator-activated receptors [35], which are ligand-activated transcription factors involved in regulating inflammation, proliferation, and glucose and lipid homeostasis and often expressed in cancer cells [36, 37].

To date, much of the research into the tumor suppressive effects of CR relate to the preventative capacity of the intervention, rather than its application as an anticancer therapy. Recent attention has focused on the potential of CR as an adjunct therapy for a range of cancers in combination with traditional chemotherapy or radiation therapy [38]. While chronic CR may be well tolerated in preclinical and clinical studies of healthy individuals, patients diagnosed with cancer are at a greater risk of weight loss due to toxic cancer therapies, as well as cachexia and sarcopenia from tumor-derived signals to degrade adipose and muscle tissues, to which chronic CR may contribute. Furthermore, as CR is anti-inflammatory, chronic CR may be a concern to patients with immunodeficiency or following surgery. Intermittent CR, achieved through fasting where no calories are consumed for defined periods of time (between 24 h up to 6 days), causes similar metabolic and anti-inflammatory alterations as seen during chronic CR, and can often result in greater changes in the short term [39]. In contrast to chronic CR, fasting results in glycogen release from the liver for use as an energy source. Once glycogen stores are depleted, amino acids and fatty acids are catabolized to generate glucose and ketone bodies, respectively [39]. Short-term fasting has been shown to improve chemotherapeutic treatment with etoposide [40], mitoxantrone, oxaliplatin [41], cisplatin, cyclophosphamide, and doxorubicin [42] in transgenic and transplant mouse models of neuroblastoma, fibrosarcoma, glioma, melanoma, and breast and ovarian cancers. Alternate day fasting has also been shown to improve the radiosensitivity of mammary tumors in mice [38, 43], likely due to enhanced oxidative stress and DNA damage during short-term fasting on cancer cells. Fasting has also been shown to control circadian clock genes, the expression of which usually oscillates at specific intervals throughout the day and is coupled to processes such as oxidative stress response and DNA damage repair [44]. Chemotherapeutic treatment administered at different times of the day has been found to improve efficacy, believed to be due to circadian rhythmic control of stress responses [45]. Fasting may therefore improve the efficacy of anti-cancer therapies in part by controlling the circadian rhythm.

Elegant work by Rafaghello et al. [40] demonstrated that short-term fasting elicits differential responses to chemotherapy in normal and cancer cells, with normal cells inactivating growth signals, such as Ras, Akt and IGF-1, in response to short-term fasting protecting them from therapeutic toxicity. In contrast, cancer cells, which have arisen due to activation of these signals and evasion of senescence-inducing signals, do not undergo this inactivation, remaining vulnerable to cytotoxic treatment via chemotherapy or radiation therapy. The induction of these cancer-specific stress responses may impact mechanisms related to chemoresistance, including multidrug resistance [46].

More recently, a fasting-mimicking diet, in which mice are fed the same amount of food as control mice, albeit with a severely reduced caloric density, showed a similar reduction in tumor growth as short-term starvation and displayed synergistic therapeutic effects when combined with doxorubicin and cyclophosphamide [41, 47]. Cycles of this diet have recently been shown to improve metabolic and inflammatory biomarkers associated with cancer risk in humans [48]. Mechanistically, the synergizing effects of the fasting-mimicking diet were associated with increased autophagy in the cancer cells and reduced heme oxygenase-1 (HO-1) in the microenvironment, causing increased circulating CD8⁺ T cells and reduced T_reg cells, and resulting in enhanced immunosurveillance and clearance of tumor cells [41, 47]. Moreover, prolonged cycles of fasting have also been shown to protect immune cells during chemotherapeutic treatment [49, 50], suggesting the possibility of combining immunotherapies with traditional chemotherapy alongside such dietary interventions. Similarly, CR also maintains immunological fitness of CD4⁺ T cells during aging to enhance cancer immunotherapy, specifically OX40-agonist immunotherapy [50]. In addition, a CR regimen can reduce desmoplasia and the inflammatory microenvironment of tumors [9], previously shown to impede the delivery of therapeutic drugs to the tumor cells.

While preclinical studies are mounting on the effects of intermittent CR in combination with chemotherapy and radiation therapy, clinical studies are slow to follow, likely due to the concerns listed above. A summary of past and current clinical trials of intermittent CR, fasting-mimicking diets, and ketogenic diets in combination with anticancer therapies is included in Table 1. A small study comprising ten subjects diagnosed with malignancies, including breast, esophageal, prostate, and lung, who underwent a 48–140 h fast prior to and a 56 h post-chemotherapy fast revealed significant improvement in self-reported side effects of therapy, including nausea, vomiting, diarrhea, weakness, and fatigue [51, 52].

Larger trials are currently underway to determine the potential for short-term fasting in reducing the side effects and efficacy of chemotherapies, and will likely be the launching point for future clinical trials with intermittent CR as a potential adjuvant therapy.