Avenues of research in dietary interventions to target tumor metabolism in osteosarcoma

The caloric restriction (CR) diet typically refers to a continuous reduction of normal daily recommended calorie intake by approximately 30%, with proportional reduction of all macronutrients, while maintaining adequate amounts of vitamins and minerals, without developing malnutrition. It has been widely studied and applied to prevent or reverse obesity and aging [73, 87, 88]. However, there are concerns regarding patient tolerance to this diet, especially considering that it causes weight loss and cancer patients may be prone to developing cachexia. Therefore, clinical studies that utilize this diet in cancer patients always include careful nutritional monitoring [54, 87].

The main mechanism of action for its antitumor effects is through inhibition of IGF-1. As a result, the IGF-1/PI3K/Akt signaling pathways are repressed, thus inducing apoptosis and reducing angiogenesis, cell cycle progression, and metastasis [73, 89, 90]. Additionally, AMP-activated protein kinase (AMPK) is activated, which, in turn, also inhibits the PI3K/Akt-mTOR pathway, and, consequently, the anabolic pathways [87].

Although dietary interventions against cancer have not been extensively explored, the antitumor effects of CR have been confirmed in a number of preclinical investigations—none, however, for OS. In mice models for cancer, colon cancers showed reduced tumor volumes when compared to controls [91], while prostate [92], breast [93] and pancreatic [94] cancers were prevented. In a rat model for breast cancer, not only did CR prevent tumor development, but it also prevented its progression, with the demonstration of an inhibitory effect on CSCs [95]. In addition, CR has been studied as an adjuvant to standard chemotherapy and radiation therapy, demonstrating the ability to improve treatment results. The addition of CR to ganitumab (an anti-IGF-1R drug) improved tumor reduction in prostate cancer mice when compared to the drug alone, with increased apoptosis and reduced cell proliferation [96]. In a triple-negative breast cancer mouse model, CR reversed treatment-induced inflammation from cisplatin through IGF-1 modulation, thereby reducing drug resistance [97]. CR was also shown to improve radiation therapy treatment in two aggressive breast cancer models, delaying metastasis and tumor growth, and increasing survival in comparison to the radiation therapy alone [98]. Another study using a murine triple-negative breast cancer model demonstrated that CR was capable of decreasing the number of intratumoral TILs and increasing the number of effector cells after radiation therapy [99].

Given the increased levels of IGF-1 and the strongly activated PI3K/Akt-mTOR pathway observed in OS [60, 72, 75], CR could potentially be effective against this tumor. It would be, therefore, interesting to study the effects of CR on development and metastasis of OS.

Fasting is a complete deprivation of nutrients during a specific period of time, with no food intake—only water. However, there are different versions of clinically-feasible fasting. The most common are the fast-mimicking diets (FMDs), which can be performed with varied protocols. For instance, FMDs can be followed by means of cycles of restricted access to food (very low caloric intake—typically 300 to 1100 kcal a day—, for example, five consecutive days once a month, for 3 months) [100], or by means of intermittent fasting (IF), which involves a specific number of hours of complete nutrient deprivation cycled with a period of unrestricted access to food (for example, 16–18 h daily) [100‐102].

The rationale for using fasting as an antitumor therapy is to limit the amount of glucose and amino acids (AA) available to cancer cells, thereby disturbing their metabolism. Mechanistically, the effects of fasting regimes seem to be similar to those of CR, in that IGF-1 is reduced and AMPK is activated, inhibiting the PI3K/Akt-mTOR pathway. Additionally, due to the restriction of glucose, aerobic glycolysis, on which tumors rely for proliferation and immunosuppression, is also inhibited [17, 101, 103]. Moreover, the limitation of AA availability also impacts cancers, given that in general, they require large amounts of AA for cell proliferation and survival. Although they are capable of synthesizing non-essential AA intracellularly, they still require exogenous sources of both essential and non-essential AA, as the produced amount is insufficient for the cell requirements [104, 105]. Therefore, disturbing AA metabolism is yet another mechanism of action of fasting.

Fasting has successfully inhibited cancer cells in a few preclinical studies, although none of these studies focused on OS. For instance, an astrocytoma mouse model received an IF protocol (24 h in alternate days) and exhibited significantly smaller tumors with significantly higher survival rates, relative to controls [106]. Similarly, a colorectal cancer mouse model received an IF regime and showed significantly slowed tumor growth. Interestingly, the authors verified that the diet suppressed M2 polarization of TAMs, pointing to a more favorable immune response switching [107]. On the other hand, IF (24 h, twice a week) did not exert antitumor effects on a prostate cancer mouse model [108]. A colorectal cancer mouse model was also used to evaluate the effect of an FMD, with two weekly cycles, where each week was divided into three parts: day 1 with 50% of normal caloric intake, days 2–3 with 10% of normal caloric intake, and days 4–7 with normal food intake. The results showed, once again, a slower progression, with inhibition of proliferation and demonstration of a significantly reduced glucose intake [109]. Another study demonstrated that two weekly cycles of two-day fasting regimes inhibited colon cancer growth in mice in comparison to controls, while also demonstrating that cells switched from a glycolysis-based metabolism to an oxidative phosphorylation-based metabolism. In this same study, the authors also tested the effect of oxaliplatin—a standard chemotherapeutic agent—alone and in combination with fasting and verified that the combination therapy had the best results [103]. Other studies also evaluated fasting in combination with other treatments. For instance, in a lung cancer mouse model, animals underwent two cycles of 48 h-fasting and one cycle of 24 h-fasting in combination with a PD-1 immune checkpoint blockade. Results indicated that the combination treatment had a synergistic effect, inhibiting tumor progression and metastasis more effectively than untreated controls or either diet or anti-PD-1 alone [110].

Importantly, short-term fasting (48–60 h pre- and post-chemotherapy or radiation therapy) was demonstrated to induce differential stress resistance both in animal and human studies. By means of this mechanism, normal cells acquire a protected and slow-division state against the toxic effects of chemotherapy or radiation therapy, or other toxic agents, while cancer cells remain susceptible. Based on this lower toxicity to the host’s cells, in many instances, fasting allows the employment of higher and, thus, more effective, treatment doses [101, 111‐116].

As mentioned, studies about the effects of fasting on development and treatment for OS are missing. Yet, given that OS relies heavily on glycolysis, as well as on the IGF-1/PI3K/AKT signaling pathways [60, 66, 67, 75], fasting approaches could also be a promising tool against OS. Moreover, the metabolism of many AA was shown to be upregulated in OS, especially alanine, aspartate, glutamate, arginine, proline, cysteine and methionine [66]. Fasting could, therefore, also impact AA metabolism in OS cells, contributing to the other anticancer effects. Future preclinical studies are needed to evaluate the effectiveness of this approach for treatment of OS.

The ketogenic diet (KD) is defined as a high-fat, moderate-protein, and low-carbohydrate diet, typically with a 4:1 ratio of fat to protein and carbohydrates. Systemically, the low glucose concentrations inhibit insulin and activate glucagon secretion, and the body is forced into a ketogenic state, whereby liver cells oxidize fatty acids, producing ketone bodies—namely acetoacetate, β-hydroxybutyrate and acetone. These enter the circulation and are distributed to other tissues, serving as fuel for cell metabolism, after entering the TCA cycle. This diet was initially formulated to treat epilepsy and is currently also being used to treat obesity [117‐119]. The KD is usually well-tolerated, although it may lead to mild side-effects, such as lethargy and nausea; however, if adequately monitored, they are easily avoided [54, 120, 121].

The very low amount of glucose ingestion due to KD blocks the cells’ ability—including cancer cells’—of using the glycolytic pathway. This hinders the cancer’s main source of building blocks for proliferation, thus leading to a lower proliferative rate. The limitation of the glycolytic pathway also impacts one of the most important sources of antioxidants, leading to higher oxidative stress, as cancer cells produce high amounts of ROS, thus potentially causing apoptosis [117, 122, 123]. Similar to the previously mentioned diets, the KD is also capable of reducing IGF-1 levels, which induces activation of AMPK, inhibiting the PI3K/Akt-mTOR pathway [119, 124].

A possible limitation of reducing glucose intake is the fact that it could further hamper the host’s defense, as effector cells require glucose to initiate an immunologic response. However, it has been demonstrated that effector cells are still able to mount an immune response despite low amounts of glucose. Although the reason behind this is unknown, it was hypothesized that these cells could utilize ketone bodies instead of glucose and still maintain their metabolic activities [125].

Compared to CR and fasting, there are more animal studies on the antitumor effects of KD, especially for brain tumors. However, none of these studies were performed on OS. Because of the blood–brain barrier, brain tissue cannot metabolize fatty acids; thus, the brain relies on glucose as its main source of energy. In the absence of glucose, brain cells are able to metabolize ketone bodies, while brain tumors likely have defective ketone metabolisms and rely mostly on a glycolytic metabolism, therefore rendering them more sensitive to a KD [126, 127]. In glioblastoma mouse models, KD significantly slowed tumor progression and increased survival rates when compared to normal diet [125, 128, 129]. Another study using a glioma mouse model interestingly found that the KD was able to influence the TME, by downregulating hypoxia-related markers, which favor tumor progression, in addition to downregulating angiogenesis-related factors [130]. KD was effective in reducing tumor growth and improving survival rates in neuroblastoma mice [131]; however, it was unsuccessful in eliciting antitumor effects in medulloblastoma mice [132]. Lussier et al. analyzed the immunologic effects of the KD’s impact on the TME of gliomas in a mouse model and found that it was able to promote phenotypic changes in the TILs. Tumors treated with KD showed an increased CD4+ T cell population and a reduced regulatory T cell population when compared to controls, while reducing expression of PD-1 on CD8+ T cells, thus opposing immunosuppression [125]. The KD produced significantly smaller tumors in a squamous cell carcinoma mouse model [133], as well as in breast [123, 134], prostate [135], colorectal [136, 137], pancreatic [80], and gastric [138] cancer mouse models, relative to controls. While a KD also produced these antitumor effects in a liver cancer mouse model [139], it failed to do so in another study by the same group, where the dietary intervention took place when tumors were in a later stage, suggesting that the KD, in this case, had a more preventive than therapeutic action [140]. The beneficial effects of KD may also be cancer-specific, since studies on a mouse model of melanoma showed that KD enhanced tumor growth [141]. In other studies, KD was evaluated in combination with other treatments. For instance, KD synergized with radiation therapy [142] and chemotherapy [143, 144], leading to significantly smaller gliomas and prolonged survival in mice in comparison to stand-alone treatments and controls. Similar effects were observed in neuroblastoma [145] and breast cancer [146] mouse models treated with KD and chemotherapy, as well as in a lung cancer mouse model treated with KD, chemotherapy and radiation therapy [147].

Fewer studies analyzing the effects of KD have been conducted in humans, most of which had the primary focus of determining feasibility and safety. These studies concluded that KD is overall safe and feasible for different cancer patients, in most cases improving quality of life; however, none showed antitumor effects of KD as standalone treatment [148‐156]. An interesting find was a significant decline in the amount of lactate in head and neck cancers of patients treated with KD compared to those with normal diets, which may confirm the potential to counteract immunosuppression [157].

As with the aforementioned diets, the KD has yet to have its antitumor effects tested on OS. However, it does have potential to be effective in this regard, as the reduced amount of glucose impacts the glycolytic pathways, as well as the IGF-1/PI3K/AKT signaling pathways, which have been shown to be enhanced in OS [60, 66, 67, 75]. It could, therefore, be interesting to develop new preclinical studies to test this diet on OS.

Dietary supplements could also be used in addition to, or in alternative to, dietary modifications [158, 159]. However, no dietary supplements have been approved as anticancer therapies in the United States. This is due, in part, to the observational nature of the available studies about dietary supplements, as well as the lack of regulation for their use [159, 160]. Among the various dietary supplements available, quercetin—largely used worldwide and found in fruits, vegetables, tea, and wine—has shown some potential. Quercetin is a bioactive flavonoid with antioxidant, antiestrogenic, and antiproliferative effects that may be at the basis of its therapeutic properties [161]. Several in vitro and in vivo preclinical studies have been conducted regarding its anticancer effects, as reviewed in [162, 163] and shown in [164]. Although the exact mechanisms of Quercetin anticancer properties are not clear, they seem to occur via modulation of VEGF, apoptosis, PI3K/Akt/mTOR, and Wnt/β-catenin signaling pathways [162, 163]—all of which are affected in OS. Indeed, quercetin has consistently shown effectiveness against OS in vitro and in vivo. For instance, studies have shown that Quercetin induces inhibition of proliferation, cell cycle arrest, and apoptosis of MG-63 [165, 166], U2OS, Saos-2 [166], HOS [167], and 143B [168] cells. Additionally, Wu et al. showed that quercetin induces autophagy of MG-63 cells in vitro and in vivo [169]. Lan et al. demonstrated that quercetin attenuated cell migration and invasion, with downregulation of HIF-1α, VEGF, MMP2, and MMP9 expression on HOS and MG-63 cells, when compared to vehicle control. They also verified reduction in the lung metastases in OS xenograft models [170]. While these studies indicate that Quercetin may have some effects on OS, additional preclinical studies and randomized clinical trials are needed to confirm its anticancer efficacy. Considering the significantly high doses, well above the normal dietary intake, that may be required to achieve the anticancer effects, safety considerations should be carefully assessed [171, 172].