This article is part of our October 2020 special issue. Download the full issue here.
Abstract
Angiogenesis is a normal physiological process involving the production and organization of blood vessels, a process required for wound healing and general health maintenance. In the case of cancer, however, the angiogenic process becomes an enabling characteristic that allows tumors to grow and metastasize more effectively. As 1 of the hallmarks of cancer, angiogenesis is also a vulnerable point for malignancy that is susceptible to influence by external modulators. Conventional treatments have the capacity to help hinder cancer growth and spread through antiangiogenic agents; however, they also have associated toxicity that can create side effects that limit their application. A number of natural health products have shown promise in clinical application against cancer through antiangiogenic mechanisms. In this paper we explore a few of these agents that have some of the strongest related research for application. We discuss some of the clinical evidence that currently exists and, importantly, how to enable effective research to further explore this topic.
Introduction & Background
Angiogenesis Process
Similar to normal tissue, malignant tumors require a sustained delivery of nutrients and removal of waste to subsist. Delivery and removal are primarily achieved through tumor neovascularization via a process called angiogenesis, which involves the formation of new blood vessels from existing vasculature. Unlike normal tissue, malignancies circumvent the ordinarily quiescent and well-organized nature of angiogenesis and exploit the process to facilitate unimpeded malignant growth.1,2
The process of cancer-related angiogenesis is heavily influenced by mediators in the tumor environment and is accelerated by localized tissue breakdown. Amongst other angiogenic mediators, vascular endothelial growth factor (VEGF) is the primary signal protein implicated in the process of angiogenesis.1,3 As a ligand, it stimulates VEGF receptor 2 (VEGFR-2), which is highly expressed by endothelial cells engaging in angiogenesis.3 Binding results in up-regulation of genes that mediate proliferation and migration of endothelial cells, while promoting survival and permeability of the vasculature.3
Angiogenesis in Modern Cancer Care
In the early 1970s, Dr Judah Folkman theorized that angiogenesis could serve as a potential therapeutic target,4 with the first approved pharmaceutical agents coming out in the early 2000s.3-6 Bevacizumab, a monoclonal antibody that binds circulating VEGF, was the first FDA-approved angiogenesis inhibitor, which when combined with standard chemotherapy, improved survival for patients with metastatic colorectal cancer.7 Similarly, patients with advanced non-small-cell lung cancer (NSCLC) experienced a survival benefit with bevacizumab in combination with standard chemotherapy.8 Overall, pharmaceutical VEGF inhibitors have shown modest benefit when added to standard care and have proven to be a valuable addition to standard care in certain circumstances.3
Based on 33 randomized controlled clinical trials (RCTs) (N= 17,396 NSCLC participants), the addition of approved angiogenesis inhibitors significantly improved progression-free survival (PFS; HR: 0.81, 95% CI 0.76–0.85, P<0.001), overall survival (OS; HR: 0.95, 95% CI 0.92–0.98, P= 0.004), objective response rate (ORR; RR: 1.54, 95% CI 1.37–1.73 P<0.001), and disease control rate (DCR; RR: 1.18, 95% CI 1.10–1.27, P<0.001) compared to non-angiogenesis inhibitor treatments.9 Approved angiogenesis inhibitors have also demonstrated improved survival outcomes for patients with ovarian10 and gastric11 cancers. In contrast, a meta-analysis of 7 RCTs (N=1,322 participants) exploring use of angiogenesis inhibitors in small-cell lung cancer (SCLC) found angiogenesis inhibitors did not significantly improve prognosis,12 indicating the potential for cancer-type specificity related to clinical impact.
The clinical value of pharmaceutical angiogenesis inhibitors is tempered by notable side effects. Possible serious adverse events include hemorrhage, hypertension, neutropenia, thromboembolic events, and impaired wound healing.13 Balancing positive effects on prognosis against patients’ quality of life and side-effect burden is an essential part of the evidence-based medicine model and needs to be factored into clinical decision making.
Natural Health Products & Angiogenesis
Compounds found in nature exist that possess antiangiogenic properties when ingested.14 An effort to both research and develop agents from phytochemical sources to address a hallmark1 of cancer progression—angiogenesis—could yield additional options to accompany conventional cancer care. Scrutiny of available evidence to select promising phytochemicals for further investigation would foster efficient and collective efforts that could support the identification of new therapeutic options for patients with cancer. This review presents certain phytochemicals with some of the most robust current evidence for clinically meaningful angiogenic-inhibitory effects. We suggest that if natural health products are to be used for this objective (antiangiogenic effects), these agents deserve priority consideration and future investigations through more rigorous clinical research.
Angiogenesis-Inhibiting Compounds Found in Nature
Notable phytochemicals with evidence for antiangiogenic effects through different pathways include resveratrol, green tea catechins, curcumin, silymarin/silibinin, castanospermine, sanguinarine, brucine, tylophorine, colchicine, vinblastine, ginsenosides, taxol, and triphala churna.14 These compounds have been found to act on different angiogenesis targets and are classified into 3 major families: polyphenols, alkaloids, and terpenoids/tannins.14 Of these, certain compounds have more extensive research, making them better poised for immediate clinical investigation and judicious clinical consideration on a case-by-case basis. For each of the compounds below, the curated preclinical evidence, bioavailability, and clinical trial data provide a comprehensive review of each phytochemical, which may also be utilized for future investigation.
Green Tea
Produced from the common tea plant (Camellia sinensis), green tea has been extensively studied for its effects on health and human physiology, which are thought to be largely due to its catechin content, specifically (-)-epigallocatechin-3-gallate (EGCG). Epidemiological studies have found that green tea consumption is associated with a reduced risk of prostate cancer,15 liver cancer,16 breast cancer,17 and cardiovascular disease.18 At the cellular level, EGCG has been observed to inhibit VEGF expression and activity in certain cancer cell lines including gastric cancer,19 head and neck squamous cell carcinoma, and breast20 cancer, among others.
Preclinical Evidence of Antiangiogenic Activity
Green tea catechins have been extensively researched in preclinical models to identify and understand their antiangiogenic properties. A green tea extract product containing 65% EGCG inhibited invasion of MDA-MB231 breast cancer cells by up to 40% compared to controls in a concentration-dependent manner, and inhibited neovascularization in vivo in C57BL/6 mice.21 Intraperitoneal injection of EGCG inhibited gastric cancer growth in nude mice by 60.4%, with reduced tumor micro-vessel density, VEGF-induced endothelial cell proliferation, tube formation, and migration.22 EGCG treatment of human cervical carcinoma and hepatoma cells inhibited hypoxia and serum-induced hypoxia-inducible factor (HIF) 1-alpha accumulation, resulting in significant VEGF-expression reduction.23 Treatment with EGCG dose-dependently inhibited VEGF expression and platelet-derived growth factor in human vascular smooth muscle cells.24
Bioavailability
While in vitro studies show antiangiogenic activity, achievable EGCG serum concentrations in humans are hindered by poor bioavailability, likely explaining inconsistencies in epidemiological research.25 Following oral administration in humans and animals, peak catechin plasma levels are often in the sub- or very low micromolar range, falling short of the effective concentration range of 1 to 100 μmol/L.25 Human oral bioavailability studies have yielded plasma levels that are 5 to 50 times less than those in preclinical models.26 Rat studies have shown that <5% of orally administered tea catechins present in systemic circulation, and in humans about 0.16% present in plasma.25 A small portion of ingested tea catechins undergo extensive phase II enzyme metabolism in the small intestine (primarily absorbed through passive diffusion), with the remainder passing to the colon where microorganism degradation likely occurs.25
Several approaches have been tested to increase the systemic absorption of oral EGCG. One randomized, controlled crossover trial26 assessed the effect of fasting states with differing catechin dosing in healthy volunteers. A 3.5-fold increase was observed for the average maximum plasma concentration (free EGCG) during a fasting state compared to a fed state, but no significant differences for total plasma levels.26 Other areas of active research to improve bioavailability include the formation of nanostructure carriers, molecular modification, and coadministration with bioactive components.25
Clinical Research of Antiangiogenic Effects
An open-label, single-arm phase 1 clinical trial of 26 males with prostate cancer awaiting radical prostatectomy found that 800 mg of EGCG for 6 weeks resulted in a significant reduction in plasma serum levels of VEGF compared to baseline (P<0.03), with 6 out of 25 participants experiencing ≥25% reduction.27 A double-blind, placebo-controlled, multicentered study of 33 participants with bladder cancer, split into 3 arms (800 mg EGCG, 1,200 mg EGCG, or placebo), found no significant differences between groups for VEGF tissue levels.28 A randomized, placebo-controlled, double-blind “split-face” dermatological study of 4 volunteers with significant facial erythema and telangiectasia found that an EGCG cream (2.5% weight for weight) significantly reduced VEGF levels in biopsy samples compared to control (P<0.005).29
Curcumin
Isolated from the common spice turmeric (Curcuma longa), curcumin is estimated to have been used in traditional medicine for thousands of years.30 Contemporary research has shown that curcumin is tolerated in high doses (up to 12 grams), and exerts antioxidant and anti-inflammatory effects.30 Further research has yielded evidence of VEGF-pathway alteration in models of breast cancer,31 human intestinal microvascular endothelial cells,32 and hepatocellular carcinoma.33 While traditionally turmeric as a whole has been implemented therapeutically, recent investigations have focused primarily on curcumin itself at levels beyond what would be typically consumed, in hopes of achieving drug-like actions.
Preclinical Evidence of Antiangiogenic Activity
A gastric cancer cell line study (AGS and SGC-7901 cell lines) showed that curcumin exerts antilymphangiogenic activity via inhibition of HMGB1/VEGF-D signalling.34 In a hepatocellular cancer model, curcumin significantly induced H22 cell apoptosis at the 40 µM and 80 µM concentrations compared to untreated controls (P<0.05 and P<0.01, respectively),35 and in mice treated with 50 mg/kg curcumin, it significantly reduced H22 tumor growth (P<0.05).35 The mechanism of observed effects in both the cell line and mouse model was the significant inhibition of VEGF expression and PI3K/AKT signalling.35 A colorectal cancer model study found that in vitro, curcumin inhibited CT26 cell proliferation and migration and exerted apoptotic activity, while inhibiting tumor growth in vivo through VEGF signalling modulation.36 Nude mice injected with A549 lung cancer cells and given 100 mg/kg curcumin intraperitoneally experienced reduced tumor weight and size, with suppression of VEGF expression.37
Bioavailability
Pure curcumin absorbs poorly in humans, resulting in low achievable serum concentrations.38 Pharmacokinetic studies have shown that curcumin undergoes both extensive phase I and II liver metabolism, and is further metabolized by microbiota.38
In 25 participants with high-risk/premalignant lesions, oral daily doses of 1,000, 2,000, 4,000, 8,000, and 12,000 mg/day curcumin were found to cause serum curcumin peaks at 1 to 2 hours, with the average concentrations ranging from 0.51 +/- 0.11 µM to 1.77 +/- 1.8 µM (dose-dependently).38,39 These concentrations are notably lower than the controlled environments created in experimental preclinical models. A turmeric extract dose of 0.4 to 2.2 g (containing 36-180 mg of curcumin) given to 15 patients with chemotherapy-refractory colorectal cancer yielded metabolites that were undetectable in both blood and urine.38,40 These studies, among others, present a notable bioavailability issue with unaltered oral curcumin, resulting in novel approaches to circumvent this problem.
Unlike normal tissue, malignancies circumvent the ordinarily quiescent and well-organized nature of angiogenesis and exploit the process to facilitate unimpeded malignant growth.
The problem of absorption has been addressed extensively, with patented formulations being produced. The contrast of 2 similar controlled studies describes the progress of altered forms of curcumin influencing bioavailability and achievable serum concentrations.41,42 A 2008 study that used 8 grams (8,000 mg) of curcumin orally (simple plant extract with no composition alteration) in patients with pancreatic cancer reported low peak levels of 22 to 41 ng/mL, which remained relatively constant over 4 weeks.41 A 2013 study assessed a novel curcumin formulation (Theracurmin®) in a similar population in order to investigate bioavailability changes.42 This formulation is composed of microscopic particles contained in a colloidal agent. In 10 participants with either pancreatic or biliary tract cancer resistant to chemotherapy, an escalating dose of 200 to 400 mg of Theracurmin resulted in a median peak plasma level of 342 ng/mL (range 47-1,029 ng/mL) at the 200-mg dose, and 440 ng/mL (range 179-1,380 ng/mL) at the 400-mg dose.42 Authors report that no unexpected adverse events occurred, and it was safely continued in 3 surviving patients for over 9 months.42
In addition to Theracurmin, other patented formulations such as Meriva®43 and BCM-95®,44 along with additive approaches such as piperine45 products combined with curcumin, are an active field of study pertaining to optimizing absorption.
Clinical Research of Antiangiogenic Effects
One randomized, controlled, double-blind, crossover study involving 30 obese participants found that 1 gram of a curcumin-complex product for 4 weeks significantly reduced VEGF (P=0.01) compared to control.46
Resveratrol
A unique type of polyphenol called a stilbene, resveratrol is primarily found in the skins of berries and grapes. At therapeutic doses, it exerts antiangiogenic effects in experimental tumor models.14 In comparison to other natural health products, resveratrol has had extensive human clinical investigation via supplementation trials for a number of non-cancer-related conditions, including dyslipidemia,47 metabolic syndrome,48 and hypertension,49 with overall mixed results. Preclinical angiogenesis resveratrol studies primarily focused on ovarian cancer models.
Preclinical Evidence of Antiangiogenic Activity
A 3D ovarian cancer cell aggregate model (ie, an organoid) treated with increasing doses (10, 20, or 30 μM) of resveratrol or a derivative (acetyl-resveratrol) reduced cell growth and suppressed VEGF secretion.50 An earlier, similar ovarian organoid model study found that resveratrol or acetyl-resveratrol suppressed VEGF secretion in a dose-dependent fashion.51 A lung cancer model (A549 cells) found that treatment with resveratrol reduced both interleukin (IL-6) and VEGF secretion.52 Hepatocarcinoma cells (HepG2) treated with escalating concentrations of resveratrol (ranging from 0-40 μM) experienced inhibition of VEGF gene expression, resulting in inhibited proliferation.53
Bioavailability
While resveratrol has low water solubility and absorbs primarily via passive diffusion, hindering absorption, its capacity to form a wide range of complexes increases intestinal permeability.54 While it is fairly well established that distribution of resveratrol in tissues is very low and overall it has low bioavailability, resveratrol appears to show efficacy in vivo; this is thought to be due to metabolic conversion and recirculation of metabolites that may have biological activity.54 Multiple approaches are being investigated to improve the bioavailability of resveratrol, including lipid nanocarriers, liposomes, nanoemulsions, micelle solutions, polymeric nanoparticles, solid dispersions, and nanocrystals.55 It should be noted that while resveratrol absorption is a concern for clinical efficacy, there is risk of toxicity with higher dosing, indicating a possibly narrow therapeutic window.56 The dosing approach is not linear, in the sense that higher doses don’t necessarily equate to more favorable outcomes, and future investigations should not only focus on increasing absorption but also, more importantly, regulating the dosing to offset toxicity.
Clinical Research of Antiangiogenic Effects
A triple-blind, placebo-controlled, randomized study of 61 polycystic ovary syndrome participants receiving 800 mg/day resveratrol for 40 days reported significant reduction in VEGF expression in recovered granulosa cells (P<0.0001).57 A randomized, double-blind placebo-controlled study of 72 participants undergoing peritoneal dialysis found that those receiving high-dose trans-resveratrol (450 mg/day) had significantly lower rates of VEGF in retrieved effluent compared to the low-dose and placebo groups.58
Ginsenosides
Triterpene saponins concentrated in the roots of red ginseng (Panax ginseng), ginsenosides have been found to exert antiangiogenic effects.14 Ginsenosides are classified using an “Rx” system, with the “R” representing the root and the “x” describing the chromatographic polarity (alphabetical order).59 Over 30 ginsenosides have been identified, with researchers adopting standardized products via specific extraction processes to improve consistency.59 A vast amount of research exists for ginsenosides’ antiangiogenic activity in multiple cancer models and using many different forms, of which Rg3 is the most rigorously studied.
Preclinical Evidence of Antiangiogenic Activity
In a dual in vitro and in vivo study, Rg3 inhibited migration and invasion in 4 thyroid cancer cell models and suppressed pulmonary metastasis in nude mice, with inhibition of VEGF expression observed.60 In a gastric cancer cell model study, where hypoxia was induced to stimulate vascular growth factors, levels of VEGF were significantly lower in the Rg3-treated group (P<0.05).61 Under induced hypoxic conditions, an esophageal cancer cell model study found that Rg3 significantly inhibited proliferation and reduced VEGF messenger RNA (mRNA).62 In an endometriosis model utilizing allotransplantation in rats, Rg3 was found to inhibit growth of ectopic endometrium in a dose-dependent fashion via downregulation of VEGF, p-Akt, and p-mTOR pathways.63
Bioavailability
Ginsenoside absorption across intestinal mucosa has not been clearly described; however, it appears to cross via energy-dependent transport and appears nonsaturable.59 One in vivo study using A/J mouse intestinal models determined that certain bacterial families are involved in the conversion of ginsenosides into active secondary metabolites.64 A comparative mouse study (Walker 256 tumor-bearing mice vs normal mice) using 50 mg/kg of oral Rg3 discovered that Rh2 ginsenosides were found in plasma, likely due to glycosylation hydrolysis.65 One diabetes clinical trial using ginsenoside Re found that oral administration resulted in no detectable plasma amounts at the 30-minute and 6-hour point.66 The absorption, biotransformation, and overall bioavailability of ginsenosides appear complex and variable between different forms, requiring further exploration.
Clinical Research of Antiangiogenic Effects
A randomized, placebo-controlled study of Rg3 supplementation in 20 patients with acute leukemia for 2 months in conjunction with chemotherapy significantly reduced serum VEGF levels.67 A randomized controlled trial of 71 postoperative participants with advanced gastric cancer found that those who received Rg3 + mitomycin C/tegafur, compared to chemotherapy alone, after 14 weeks experienced VEGF levels within normal range, with the median survival being 40 and 25 months, respectively (P=0.047).68 A 3-armed, randomized, controlled study (Rg3 vs Rg3 + chemotherapy vs chemotherapy alone) of postoperative patients with NSCLC found no significant survival differences over 3 years between groups (P>0.05), with VEGF levels not associated with outcomes.69 A study of 60 patients with advanced esophageal cancer receiving either Rg3 + gemcitabine/cisplatin or chemotherapy found that the intervention group had significantly lower posttreatment VEGF levels (P=0.002).70
Clinical Implications & Closing Remarks
Green tea catechins (especially EGCG), curcumin, resveratrol, and ginsenosides are phytochemicals that currently hold extensive foundational antiangiogenesis research. These 4 compounds possess preclinical, bioavailability, and preliminary clinical research, poising them for rigorous clinical investigation, compared to other phytochemicals that still require further preliminary investigation prior to being considered candidates for extensive clinical investigation. The next step would be to select 1 or a combination of phytochemicals and perform a formal systematic search to identify all relevant research, including on safety, clinical data, and pharmacokinetics/dynamics. In addition, an accompanying review of the literature should be performed in order to determine the most accurate and feasible ways to assess the effects of an intervention on angiogenesis in homine.
It is notable that while the identified compounds possess an ample amount of preclinical evidence to indicate antiangiogenic effects, data of this sort cannot be relied on alone to predict effects in patients, especially in highly vulnerable patient populations, such as those with cancer. The shortcomings of preclinical data translating to human effects are twofold, with consistent evidence showing that often: 1) preclinical data are a poor predictor of effect (often no significant benefit/effect is found);71 and 2) human subjects often experience unanticipated harms even when the intervention appears safe in preclinical models, as they are also poor predictors of toxicity.72,73 Collectively, while reviewed information provides a strong rationale for subsequent clinical trial investigation, caution is warranted for extrapolating data further directly to clinical care.
Based on available evidence, a future well-designed clinical trial, with the objective of addressing malignant angiogenesis in cancer patients via phytochemical supplementation, would do best by using an agent with human-level and preclinical data showing VEGF effects, and by choosing a highly bioavailable form of that agent. By fast-tracking a select few phytochemicals that already have essential pieces of evidence into a rigorous clinical trial, rather than investigating a number of agents lacking foundational research, we can execute more efficient trials and hopefully identify meaningful associations.