Our current study is the first to demonstrate the combined treatment of advanced retinoblastoma tumors in the LHBETATAG mouse model with angiogenic inhibitors and glycolytic inhibitors avoiding chemotherapy. We have previously demonstrated the effectiveness of 2-DG alone in decreasing tumor hypoxia and in tumor control. However, when 2-DG was combined with the angiogenic inhibitor, AA, advanced tumor control was significantly increased, as the combined treatment resulted in a 23% decrease in tumor burden compared to 2-DG alone. Importantly, the enhanced response of tumors to the combination treatment occurred in the group that received 2-DG 1 day after AA treatment, whereas the group treated 1 week after AA did not differ significantly from 2-DG alone. Mice were treated at 16 weeks of age, when eyes harbor advanced tumors. As shown in prior studies, these advanced tumors consist of heterogeneous vasculature, with a greater proportion of mature versus immature vessels. As a result, AA would be expected to have a more robust effect the earlier tumors were treated, as these earlier tumors have a greater proportion of immature vessels. Moreover, advanced tumors have a greater degree of hypoxia, thus maximizing the effect of 2-DG treatment, and potentially diminishing the observed effect of the AA. Despite these limitations, combination treatment with AA and 2-DG proved to significantly enhance tumor control compared to 2-DG alone.
Solid tumors are known to consist of hypoxic regions composed of slow-growing, malignant populations of tumor cells. The cells in the hypoxic microenvironment must adapt to these harsh conditions in order to survive. Adaptations may be mediated through a variety of mechanisms, with cells altering the genetic expression of certain key pathways. Several O2
-sensitive pathways have been identified, including the mammalian target of rapamycin (mTOR), hypoxia-inducible factor (HIF), unfolded protein response (UPR), and the urokinase plasminogen activator (uPA) systems, that potentially contribute to adaptation and survival.24
Reduced oxygen tensions have been shown to upregulate the transcription factor HIF, leading to increased production of genes for glucose transporters (GLUT), glycolytic enzymes, and angiogenesis.33
Additionally, the uPA system may activate enzymes, notably gelatinases such as matrix metalloproteinases (MMP) that break down the extracellular matrix, facilitating angiogenesis and tumor invasion.36
As a result, adjuvant therapies that target key aspects of these adaptations are needed.
Glycolytic inhibitors, such as 2-DG, exploit the altered cellular metabolism that hypoxic cells utilize for survival. Unlike cells in normoxic conditions, the cells in the hypoxic regions must use anaerobic glycolysis as the sole source of energy. 2-DG competes with glucose for cellular transporters (ie, GLUT-1) and key glycolytic enzymes, such as hexokinase. As a result, 2-DG inhibits the metabolic machinery of the cells, preventing the production of energy in the form of ATP, and effectively causing cell death. Our previous studies have demonstrated that targeting hypoxic cells with 2-DG alone, as well as combining this treatment with chemotherapy, leads to a significant reduction in tumor hypoxia and tumor burden.25
2-DG has been shown to be effective at targeting hypoxic cells, as well as inhibiting endothelial cell angiogenesis.27
The spatial distribution of blood vessels and the effects of antiangiogenic agents in LHBETA
retinal tumors have been characterized previously.37
In the current study, we have shown that 2-DG significantly reduces the density of immature neovessels by 43% compared to controls. Conversely, 2-DG does not have an effect on mature tumor vasculature. Overall, total vasculature in tumors was reduced by 34%. Angiogenesis, or the formation of new vessels from existing vessels, is imperative for tumor growth and progression as neoplastic cells require nutrient delivery and waste removal. As a result of hypoxia and other factors in the tumor microenvironment, angiogenesis is stimulated, forming a haphazard network of new vessels that require stabilization by cytokines and growth factors, such as vascular endothelial growth factor. New vessels consist of proliferating ECs, while mature vessels are supported by pericytes that provide necessary stabilization. ECs have been shown to rely on glycolysis for metabolism even under aerobic conditions.38
It has been hypothesized that proliferating cells may utilize glycolysis as an adaptation to oxidative stress secondary to oxygen free radicals.39
With a greater reliance on glycolysis for endothelial proliferation, 2-DG as a glycolytic inhibitor may target ECs, effectively serving as an antiangiogenic agent.
The use of glycolytic inhibitors is a promising alternative for advanced disease. More importantly, the current study provides additional support to the mechanistic effect of 2-DG on angiogenesis. Because retinoblastoma tumors promote angiogenesis and are highly dependent on their vascular supply, a dual mechanistic effect of 2-DG targeting both the slowly proliferating hypoxic cells as well as the proliferating ECs is demonstrated.
Antiangiogenic agents, such as AA, inhibit vascular growth, effectively rendering a tumor without sufficient blood supply, leading to hypoxia and cell death. Prior studies have shown the efficacy of antiangiogenic agents in retinoblastoma tumor control in the LHBETA
As tumors undergo proliferation, an ‘angiogenic switch’ occurs, where the tumor must stimulate angiogenesis to support further development. As tumors become more advanced, these neovessels are remodeled and transformed into mature vessels. The ECs of new vessels rely on growth factors produced by gelatinase degradation of the extracellular matrix to support growth and development, while mature vessels rely on pericytes for support. Recently, we have shown that AA affects gelatinase activity, leading to a decrease in MMP-2 and MMP-9.41
Prior studies have correlated AA treatment with increased levels of plasminogen activator inhibitor 1, effectively impeding uPA activity.42
As a result, the dynamic tumor microenvironment consisting of a heterogeneous network of blood vessels can be targeted by angiogenic inhibitors. However, further strategies to modulate or target the formation of mature vessels are needed to potentially maximize tumor control.
The results of our study emphasize a key concept regarding adjuvant treatments and their effectiveness in tumor control, specifically, optimized timing. We have shown in prior studies that when AA is combined with chemotherapeutic agents, the administration of AA after completion of the chemotherapy cycle yielded more effective tumor control than treatment during the cycle.26
It is hypothesized that angiogenic inhibitors alter blood flow through vessels, potentially reducing the delivery of chemotherapeutic agents to the tumor. Additionally, we have shown that when 2-DG is combined with carboplatin, tumor control is enhanced when 2-DG is given prior to chemotherapy.25
Further, with the current study, timing of 2-DG following AA treatment led to significant differences in advanced tumor control. AA has been shown to significantly reduce total vessel density in tumors, primarily immature neovessels, as well as increase hypoxia in tumors by 28% and 17% 1 day and 1 week post-treatment, respectively.44
With reduction in blood vessels, there is a decrease in blood flow resulting in less delivery of oxygen and nutrients to highly metabolic neoplastic cells. The tumor microenvironment is altered, leading to a greater degree of hypoxia. With the increase in hypoxia following AA treatment, this is an optimal time for treatment with 2-DG which targets hypoxic cells. As a result, with optimally timed treatment using angiogenic and glycolytic inhibitors, advanced tumor control is enhanced more than with single-agent treatment. These studies emphasize that as adjuvant therapies are developed to target various aspects of the tumor microenvironment and associated genetic and molecular pathways, optimally timed delivery must be considered to not only enhance tumor control, but also to avoid antagonism between various treatment modalities.
Current therapy for retinoblastoma has improved survival and globe salvage, with chemotherapy and focal consolidation providing tumor control in 100% of tumors without seeding, to 47%–83% in more advanced tumors with diffuse retinal and vitreous seeds.6
Despite treatment success, current chemotherapy exposes children to highly toxic medications with severe side effects. Additionally, advanced disease responds poorly to treatment, often necessitating enucleation for disease control. Adjuvant therapies, such as glycolytic inhibitors, angiogenic inhibitors, and agents that target molecular and genetic pathways involved in tumor proliferation could potentially be combined with current treatment regimens. The current study shows that tumor control can be achieved with combination therapy without the need for chemotherapy. Adjuvant therapies may eliminate the need for chemotherapy or allow reduced doses or cycles of toxic agents. Finally, combined treatment may provide enhanced control of advanced tumors that previously were resistant to therapy. Although 2-DG was administered systemically through intraperitoneal injections, we are not advocating systemic delivery, secondary to potential effects on organ and vascular development. Rather, we have shown in prior studies that 2-DG and AA can be effectively delivered locally through subconjunctival injections. Additionally, with new protocols utilizing supraselective ophthalmic artery cannulation, adjuvant therapies may potentially be delivered focally by selective intra-arterial delivery.
From a recent interview with Dr Watson, a cofounder of the DNA double helix, he had this to say about the future of cancer treatment: ‘there are often many types of cancer-causing genetic “drivers” within single cancer cells … given the inherent genetic instability of most cancer cells, the use of drugs acting against single drivers would all too soon lead to the emergence of genetic variants driven by increasingly destructive second, if not third, drivers. Most anticancer drugs, then, will probably never reach their full potential unless they are given in combination with other drugs developed against second or even third drivers’.45
We feel that the future of retinoblastoma treatment echoes these sentiments, as we continue to show that adjuvant therapies that target components of the tumor microenvironment can enhance tumor control and potentially decrease the reliance on toxic chemotherapeutic agents.
In the present study, we have characterized the effect of the combination treatment with 2-DG and AA on tumor growth, hypoxia, angiogenesis, and blood vessel maturation in LHBETATAG retinal tumors. We have demonstrated that advanced tumor control is enhanced with combination therapy utilizing glycolytic and angiogenic inhibitors compared to single agents alone. The effects were significantly influenced by the temporal relationship of therapies, thus emphasizing the importance of optimally timed treatments. The current findings suggest that angiogenic and glycolytic inhibitors can be combined with a synergistic effect on tumor control. We also demonstrated the dual mechanistic action of 2-DG by targeting hypoxic cells as well as ECs, effectively reducing angiogenesis. With a greater understanding of the mechanism of adjuvant therapies such as glycolytic inhibitors and angiogenic inhibitors, treatments may be effectively combined with or without current chemotherapeutic agents. As a result, this combination may have potential value as adjuvant therapy for children with retinoblastomas.