There is a wide array of nutraceuticals (drugs derived from foods) and herbal products with action against cancer cells in vitro. It is impossible to know from in vitro studies alone the effects of these products on a tumour in a living body. In vivo studies, usually using mice or rats implanted with cancer cells, often involve administration of therapeutic drugs via intraperitoneal injection, which doesn’t mimic human oral ingestion of the substance being studied. In vivo studies using long-established cancer cell cultures are also often unreliable due to their genetic homogeneity, which does not mimic the diversity of cells found in a spontaneous human tumour. In this section I will focus on non-prescription products that have shown efficacy against brain cancer in high-quality animal studies, clinical trials, and convincing anecdotal case reports in humans.

To join a discussion about the supplements described on this page, and about multi-agent approaches for brain tumors, visit the new blog community, called our brain tumor cocktails and stories.

blogger icon

Note: for information on the immune-stimulating properties of supplements such as Maitake D-fraction and curcumin, see the Re-educating the Immune System page.

Ashwagandha (Withania somnifera)

The immunostimulatory effects of Withania have been described on the Re-educating the Immune System page. A team from India published a study [35] showing anti-tumor effects of ashwagandha water extract when given orally to Wistar rats bearing intracranial C6 rat gliomas. The preparation used in this study consisted of sun-dried and ground dry ashwagandha leaf powder, which was suspended in distilled water and stirred overnight, followed by filtration. After air-drying this extract, the yield was 14.7% (1470 mg dried extract was recovered from 10 grams of dried leaf powder).

Wistar rats bearing orthotopic C6 rat gliomas were given the ashwagandha water extract orally for 21 days starting on day one after tumor cell implantation. At day 21, animals were sacrificed and their brains were investigated. Most of the untreated tumor-bearing rats showed weight loss and motor impairment by day 19. In contrast, most (8/10) of the ashwagandha-treated rats maintained normal functioning. At day 21, average tumor volume was reduced by 43.6% in the ashwagandha-treated group in comparison with tumors from untreated rats. Histologically, the control tumors were large and invasive, while tumors from ashwagandha-treated rats were smaller with clearer boundary between tumor and normal brain, suggesting an anti-invasive effect of the treatment. When tumors were analyzed for various markers, the ashwagandha-treated tumors showed markedly lower expression of cyclin D1, phosphorylated Akt, bcl-xl, and VEGF, proteins involved in the cell cycle, cell survival, proliferation, and angiogenesis. Nuclear presence of the p65 subunit of nuclear factor kappa beta (a pro-tumor transcription factor) was significantly reduced in tumors of the ashwagandha-treated rats.

Studies with mice showed that withanamides from ashwagandha crossed the blood-brain barrier after intraperitoneal injection of Withania extract.


Pterostilbene is a naturally-occurring resveratrol analog, found especially in blueberries. It may be purchased commercially as a nutraceutical from various suppliers. It has far better oral bioavailability than resveratrol and has been detected in mouse and rat brains after oral administration.

A Taiwanese study [33] published in early 2015 demonstrates that pterostilbene reduces cell viability as well as cell migration and invasion of two GBM cell lines (GBM8401 and U87) at a concentration ~1 micromolar (uM). At an even lower concentration of 0.5 uM, pterostilbene inhibited GBM8401 sphere formation by 29%. Pterostilbene also sensitized U87 to 5 Gy of radiation, and both cell lines to 10 Gy of radiation. GRP78 is a stress-resistance protein, and is increased in GBM cells, especially in the treatment-resistant stem-like cells. A 2 uM concentration of pterostilbene significantly increased levels of microRNA-205, a tumour-suppressor, in GBM8401. This led to decreased levels of GRP78 and the tumour-promoter C-Myc. C-Myc is active not only in glioblastoma tumours, but also in 65% of IDH-mutant low grade gliomas and 88% of IDH-mutant high grade gliomas [34].

To validate these findings in vivo, mice were subcutaneously implanted with GBM8401 tumor spheres. The mice were then injected with either a small dose of pterostilbene 3 times weekly, or given a single dose of 10 Gy radiation, or both of these treatments combined. Pterostilbene (PT) or irradiation alone significantly slowed tumour growth, while the combined treatment gave the largest effects.

2015 Huynh Pterostilbene in vivo


Berberine is a plant alkaloid derived from species such as Goldenseal and Oregon Grape. It has an effect quite similar to the diabetic drug metformin (discussed in the Repurposed Drugs section) and both drugs have the same cellular target (mitochondrial respiratory complex 1) [17]. Like metformin, berberine also suppresses the manufacture of glucose by the liver (gluconeogenesis). Though the drugs are not related, they may be considered to have equivalent functions. Though berberine is poorly absorbed, a dose of roughly 1 gram berberine per day may have a blood glucose lowering effect equivalent to the same dose of metformin. However, this dose of berberine may be associated with gastrointestinal symptoms such as constipation.

A 1990 study [18] showed that berberine may also have a direct effect against experimental glioma in rats. However, in this study berberine was injected into the rats intraperitoneally, bypassing its very poor oral bioavailability. Very little further testing has been done with berberine as an agent against gliomas in vivo.

A further use of berberine of much potential interest in cancer therapy is its apparent protection of liver function against damage by cytotoxic chemotherapies. As the liver is the primary filter through which all orally ingested agents must pass before entering the systemic circulation, the functioning of this organ is often adversely affected by the accumulation of ingested drugs and toxins. In at least five rodent studies, low to medium doses of berberine given before toxic doses of other drugs proved to effectively protect against chemical-induced liver toxicity. Berberine protected against liver damage from toxic doses of acetaminophen [26], carbon tetrachloride [26, 27, 28], and two chemotherapy drugs: doxorubicin [29] and cyclophosphamide [30]. Berberine was given orally to rats in three of these studies, and intraperitoneally injected into mice in two studies. The proposed mechanism of liver protection in these studies included inhibition by berberine of toxic metabolite generation by CYP enzymes [26], inhibition of TNF-alpha, COX-2, and iNOS [27, 30], and increased superoxide dismutase and other antioxidant activity in the liver [28, 30]. Liver damage is often detected clinically as elevated serum transaminases, such as alanine transaminase (ALT) and aspartate transaminase (AST), and alkaline phosphatase (ALP). The rise of these enzymes in the blood following hepatoxic drug treatment was prevented or significantly reduced by berberine pretreatment, indicating greatly reduced liver damage.

December 17, 2014
A study [31] just published by a group from Shandong University School of Medicine provides the best evidence to date for the use of berberine as an anti-glioma agent. Testing was performed in vitro and in a nude athymic mouse model. In vitro, the lowest concentration of berberine tested was 15 uM (micromolar), which is far higher than levels achievable in blood plasma after oral dosing. At this concentration, berberine caused a significant increase in the percentage of senescent (non-active) cells in U87 and U251 GBM cultures. This senescence was shown to be caused by downregulation of EGFR and the downstream kinases ERK and MEK.

To test the activity of berberine against glioma cells in vivo, human U87 GBM cells were implanted into the brains of athymic mice. The mice were then treated orally with moderate doses of berberine (50 and 100 mg/kg per day) for five weeks, followed by sacrifice of all the mice. Upon sacrifice, tumours were found in the brains of all untreated mice, in 4 out of 5 mice on the lower dose of berberine, and in only 2 of 6 mice receiving higher dose berberine. According to a graph shown in the paper, tumours in the control group measured between 100 and 350 (median ~150) cubic millimetres. Tumours in the lower dose treatment group all measured under 100 cubic millimetres (median less than 50), and the two tumours from the higher dose treatment group are not distinguishable from zero on the graphic shown. Similar to the in vitro results, the tumours from the berberine treated mice showed extensive signs of cell senescence, meaning those cells are no longer in active cell cycle. Staining for Ki-67, a marker of cell proliferation, was also greatly reduced in the tumours of treated mice, as was staining for EGFR, one of the most commonly upregulated drivers of glioblastoma pathology. No evidence of reduced body weight or toxicity was seen in the berberine-treated mice.

This study constitutes the best current proof favoring berberine as an oral anti-glioma agent. The oral administration of non-toxic doses of berberine, use of an orthotopic (rather than flank-injected) model, and the pronounced inhibition of tumour growth in this study are impressive. The results would have been even more impressive had a patient-derived xenograft (PDX) model been used, as U87 has been continuously cultured for over 40 years, and has likely been altered by genetic drift over that time.

In summary, all of these activities (blood glucose control, liver protection against damage by toxic agents, and direct anti-glioma activity) make berberine a very interesting candidate for supplementation. I would recommend 500-1000 mg daily, in divided doses. Ingesting with coconut oil may improve absorption.

Boswellia serrata (Indian frankincense)

Boswellia serrata is a widely available herbal anti-inflammatory agent which acts as a 5-LOX inhibitor. 5-LOX is an enzyme that catalyzes the conversion of arachidonic acid into inflammatory leukotrienes, which are prevalent in glioma tissue and contribute to inflammation.

A clinical trial testing the ability of Boswellia to reduce cerebral edema in brain tumor patients undergoing radiation proved the efficacy of this herbal supplement [1]. 20 patients were given 4200 mg of Boswellia daily, the equivalent of 12 large capsules. Another 20 patients received placebo. Many patients found it difficult to swallow 12 large Boswellia/placebo capsules daily. Some of the patients in each group required additional dexamethasone to control edema. However, by the end of radiotherapy, 60% of the patients in the Boswellia group had edema reduced to less than 25% of baseline, while only 26% of patients in the placebo group achieved such a reduction. Average volume of edema during radiation in the Boswellia group was less than half that of the placebo group (45.7 mL versus 97.4 mL).

Boswellia was also tested in vivo on Wistar rats implanted with C6 rat glioma cells [2]. Only the highest dose of oral Boswellia was sufficient to significantly prolong survival. This group was given 720 mg per kilogram body weight oral Boswellia daily, which equates to nearly 7 grams for a 60 kg human, using a conversion based on body surface area. These animals showed the side-effect of slight hair loss, and it is likely that this dosage in humans would cause significant gastrointestinal side effects, in addition to the sheer quantity of capsules that would need to be consumed. It must also be noted that the C6 rat glioma cell line is not considered to be a reliable model for spontaneous human tumours.

In summary, while very high doses of Boswellia may have an inhibiting effect on tumour progression, its beneficial effect on cerebral edema and inflammation have been demonstrated in clinical trials.

Cannabinoids: Cannabidiol (CBD) and THC

There is currently much excitement surrounding the use of cannabinoids to fight cancer, including malignant gliomas. In vitro studies abound, yet these studies often use drug concentrations unachievable in a living body. Also, cannabinoid biology in the nervous system is very complex, and cannabinoids may dose-dependently either suppress or increase cell proliferation. Different results may be obtained depending on which of the two cannabinoid receptors is activated, the density of cannabinoid receptors in the cells being studied, and the drug dosage and duration of exposure. Furthermore, most or all in vivo studies testing the cannabinoid THC against intracranial tumours in mice, use direct injection into the peritumoral area, as opposed to systemic application (oral or intravenous) which would be more relevant to patient usage. Finally, THC is also widely documented to suppress Th1 (T helper 1) type immunity, the same component of the immune system responsible for destroying malignant cells, and high-dose THC injection into immunocompetent tumour-bearing animals has increased tumour growth rate through an immunosuppressive mechanism [24].

On the other hand, there are several anecdotal reports of tumour shrinkage following oral cannabis administration (raw juiced leaves, or concentrated oil extracts), specifically several pediatric cases (see case reports below).

Cannabidiol (CBD)

The second most prevalent cannabinoid in the cannabis plant, cannabidiol (CBD), has been tested in a mouse model of glioblastoma via intraperitoneal injection [8, 32]. In the most recent of these studies, published online in January 2015, researchers from the California Pacific Medical Research Institute (first author Eric Singer, corresponding authors Liliana Soroceanu and Sean McAllister) tested cannabidiol in two patient-derived xenograft (PDX) glioblastoma mouse models. At day 9 after intracranial tumour cell injections, cannabidiol was administered to the athymic mice at a dose of 15 mg/kg mouse body weight by intraperitoneal injection for 5 days per week until the end of the experiment. Although statistically significant, this relatively high dose CBD treatment prolonged mouse survival by only several days.

Figure 2 Singer 2015

At day 22 of the experiment, tumour growth was strongly inhibited by the CBD treatment, though by day 29, treated tumours had resumed rapid growth through resistance mechanisms. Further investigations revealed that cannabidiol-mediated tumour inhibition is dependent on reactive oxygen species (ROS) generation, and that in response to this increased oxidative stress, GBM cells upregulate expression of antioxidant response genes such as the cystine/glutamate transporter SLC7A11 (also known as xCT). Cystine is imported into the cell by this transporter in order to provide cysteine for glutathione synthesis, providing increased protection against oxidative damage by ROS. The group attempted to combine cannabidiol treatment with pharmacological inhibition of xCT by sulfasalazine (SAS), though due to limited solubility and potency of the drug, this strategy was unsuccessful in vivo. Attention was then shifted to Erastin and piperazine erastine, a novel class of system Xc inhibitors. In vitro, these novel inhibitors synergistically inhibited glioma stem cell viability when combined with cannabidiol. However, neither of these new molecules crosses the blood-brain barrier, so no attempt was made to use this strategy in the intracranial glioma model.

This study showed that cannabidiol as a single agent may have limited direct anti-tumour efficacy due to resistance mechanisms involving tumour cell upregulation of antioxidant programs including increased cystine import (implying increased glutathione synthesis). Importantly, the combination of CBD plus system Xc inhibitors led to synergistic decrease in tumour cell viability in vitro. A clinically relevant system Xc inhibitor that can sufficiently cross the blood-brain barrier remains to be identified.

In August 2014, Insys Therapeutics received orphan drug designation from the FDA for cannabidiol as a potential treatment for glioblastoma. In addition to any direct effects on tumour cells, cannabidiol is increasingly recognized as a valuable anti-seizure medication, and the cannabidiol formulation Epidiolex (GW Pharmaceuticals) has received orphan drug status for childhood epilepsy. Orphan drug designation facilitates drug development, though clinical trials still need to be carried out before the candidate drug is FDA approved for these indications.

See below for evidence of potent radiosensitizing effects of combined CBD and THC in a sygeneic mouse glioma model.

CBD and THC blend

November 19, 2014.
Systemic treatment with CBD and THC slows growth in the GL261 mouse model of malignant glioma.
By and large, most studies testing cannabinoids in mouse or rat models of glioma have utilized direct injection into the tumour or peritumoral area as the mode of drug administration. This unfortunately does not tell us much about the efficacy of these agents when taken systemically (orally or injected into a site distant from the tumour). The exception has been the work done by Liliana Soroceanu and colleagues at the California Pacific Medical Center in San Francisco, who have studied the effects of cannabidiol injected intraperitoneally into mice bearing intracranial gliomas. One such study [32] is described above.

Another group from the University of London in the UK published research in November 2014, in which low doses of pure THC and CBD were injected into mice intraperitoneally, with or without irradiation [25]. The GL261 cell line was used, which is a high-grade mouse glioma grafted into immunocompetent wild-type mice. Mice were divided into four groups: untreated control mice; mice receiving a single dose of 4 Gy irradiation to the head on day 9; mice receiving 2 mg/kg each of pure THC and pure CBD on days 9, 13, and 16; and mice receiving both irradiation and THC/CBD. All mice were sacrificed at day 21 and tumour volumes were compared across the groups. The GL261 xenograft tumours were apparently resistant to a single dose of irradiation, which made no difference to the average tumour volume at day 21. Cannabinoid treatment alone slowed, but did not stabilize, tumour growth in the mice. Dramatically, cannabinoids combined with irradiation almost completely halted tumour growth, amounting to a 90% growth inhibition.

Figure 4 Scott 2014

The low doses of cannabinoids used in this study would translate to around 10 mg each of THC and CBD for an adult human of average weight. However, when taken orally larger doses might be required to have the same effect as a lower dose injected ip, due to metabolism of the drugs in the liver. It’s also noteworthy that the mice were dosed with cannabinoids only once every 3-4 days. A greater effect might have been observed had they been treated every day. The major significance of this study is showing a dramatic radiosensitizing effect of cannabinoids administered concurrently with and several days after irradiation. It would be interesting to determine if systemically administered cannabinoids might also have a sensitizing effect combined with tumor treating electric fields via Optune (formerly called NovoTTF). Research at the California Pacific Medical Center is also being conducted to determine anti-glioma effects of cannabidiol combined with TMZ as well as combinations of cannabidiol and small molecule inhibitors of antioxidant response mechanisms.

Importantly, GW Pharmaceuticals, the company which developed and markets Sativex, announced in November 2013 that they are conducting a Phase I/II trial of THC/CBD added to temozolomide for recurrent glioblastomas, with a primary endpoint of 6 month progression free survival. The trial is expected to recruit twenty patients. This will be the first formal trial using orally administered cannabinoids for glioma patients [9].

Case reports of pediatric brain tumour responses to cannabinoids

Case 1: This case was described in a video (posted online July 2011) on the health benefits of juicing raw cannabis leaves. The story is told by the child’s mother and a hospice worker. Two-year-old Amber had multiple brain tumours, considered terminal. She underwent all conventional therapies (surgery, radiation, chemotherapy), which did not stop the tumours from growing. The parents took the child home, and proceeded to feed the child a shot glass full of juiced cannabis fan leaves every day. After a month, when the hospice worker called the parents, she was informed that the recent CT scan showed that the tumours had shrunk and were fewer in number. The child was doing well and was no longer on hospice care. This anecdote begins at minute 8 of the video. We aren’t told in the video how long this response lasted.

Case 2: An 8-month old baby was diagnosed with a large inoperable optic pathway glioma in August 2011. The father managed to treat the baby at home foregoing any conventional treatment. This home treatment consisted of cannabinoid oil (3.8% CBD) placed on the baby’s pacifier twice daily. The November 2011 MRI showed clear tumour shrinkage, the January 2012 MRI showed the tumour had disappeared and the December 2012 MRI was still clear. Medical marijuana doctor William Courtenay described this case on HuffPost Live, and further details were given in the winter/spring 2013 edition of O’Shaughnessy’s, “the journal of cannabis in clinical practice”.

Case 3: Also reported in O’Shaughnessy’s is another case of pediatric optic pathway glioma responding to cannabinoid treatment. Tracy Ryan describes the treatment of her two-year old daughter Sophie with cannabis oil concurrently with chemotherapy at Kaiser Permanente Medical Center in Los Angeles. The parents started Sophie on smaller doses and increased the dose to 1 gram per day, divided into four doses three hours apart. The ratio of THC to CBD is approximately 2:1. The parents had been told that this type of tumour does not respond well to chemo, the best that could be hoped for would be temporary disease stabilization, and that she would probably be blind in one eye. Chemo started in October 2013 and Sophie has also been given THC/CBD cannabis oil since that time. The December 2013 MRI showed tumour shrinkage with necrotic spots. By June 2014, in month 9 of a 13 month chemo protocol, the child’s quality of life is “great”, with no visual impairment, and the tumour is described as having decreased “very significantly” (doctor), and being “almost gone” (mother). A youtube video shows the neurologist comparing the September 2013 MRI to the June 2014 MRI. Though it is difficult to separate the effects of chemotherapy from the effects of the cannabinoid treatment in this case, the child’s well-being and tumour shrinkage have surpassed all expectations of both the doctors and parents.

Comment: case reports such as these show that cannabinoid treament may be effective for some pediatric brain tumour patients. Without clinical trials, we cannot know what percentage of cases, or which specific types of tumours will respond. As mentioned below, there is one clinical trial underway testing Sativex (1:1 THC and CBD) added to temozolomide for recurrent glioblastoma. Given the case reports that are accumulating, including the two cases of pediatric optic pathway glioma responding to cannabinoids described above, further clinical trials are certainly warranted.

Cannabinoid sources

Sativex, a cannabidiol/THC blend in a 1:1 ratio has been approved in Europe and Canada for certain medical conditions. Medical marijuana dispensaries may be found in larger cities in States which have legalized it.


Curcumin, the active medicinal ingredient in the spice turmeric, seems to display a limitless variety of mechanisms by which it fights cancer. Notably, it has recently been found to be a DNA methyltransferase inhibitor [3, 4], and therefore may have some value as a novel epigenetic agent in G-CIMP tumours (see discussion in IDH1 mutation section).

In a syngeneic glioma mouse study [7], curcumin was administered to mice mixed in with their diet at a concentration of 0.05% by weight. Mice were pretreated with curcumin for 7 days prior to implantation of syngeneic glioma cells. Two different glioma cell lines were tested (Tu-2449 and Tu-9648). 15% and 38% of the mice in the two treatment groups remained tumour-free, while all mice in the untreated control groups perished by around day 40 (with one exception). The 0.05% curcumin diet translates into around 100 mg curcumin per kilogram mouse body weight per day. This in turn translates into around 500-600 mg curcumin for a 60 kg adult.

F7.large Weissenberger 2010

Unformulated curcumin has very low bioavailability, and is usually undetectable in the blood following oral ingestion. Typically only curcumin metabolites, rather than free curcumin, are detected in the blood. Probably the most notable specific curcumin product for those with brain cancer is the special lipidated formulation called Longvida (Verdure Sciences) [5]. This formulation was developed by UCLA scientists working on Alzheimers disease, and was specifically designed to enter the bloodstream as free curcumin and accumulate in brain tissue to higher concentrations than any other curcumin formula. In testing on mice, brain tissue concentrations of over 5 micromolar were achieved after two weeks using modest dosing of a lipidated curcumin formula similar to the final Longvida product [6]. The best source of practical information on using Longvida formulated curcumin is from the UCLA Alzheimer Translation Center.

Epigallocatechin gallate (EGCG, green tea extract)

While EGCG and other tea catechins, powerful antioxidants, are regarded as anti-cancer agents, little direct testing has been done either in glioma trials or in vivo studies. One in vivo study with mice showed that a modest dose of oral EGCG alone had little to no effect on the growth of U87 and U251 (glioma) xenografts. However, when the EGCG was administered along with oral temozolomide, survival time was improved in both models above temozolomide alone [19].


Melatonin is a hormone secreted by the pineal gland which regulates the sleep cycle. For cancer therapy, the dosage of 20 mg is often used, though this dosage may be reduced if it causes excessive grogginess upon waking in the morning. Melatonin can stimulate the anti-cancer immune system [12]. It may also stimulate platelet production, which is useful for those undergoing platelet suppressing chemotherapies [13, 14, 15]. Importantly, melatonin should only be administered in the dark part of the day, before going to sleep, as its effects may be altered if taken during daylight hours. Some of the benefits of melatonin may be related to its connection to sleep regulation.

There has been one clinical trial testing the efficacy of melatonin added to radiotherapy for glioblastoma patients [16]. Only 14 patients received melatonin-radiotherapy and were compared with a control group of 16 patients receiving radiotherapy only. The statistical methods in this study are questionable, as far more patients would be required to do a statistically valid comparison utilizing an internal control group. We would learn more from this study by comparing the outcome of the patients given melatonin to historical controls from the same era. The melatonin treated patients in this study had a median survival of about 11 months (according to the Kaplan-Meier chart appearing in the published study) and this is about what one would expect for glioblastoma patients in the early 1990s treated with radiotherapy.

It is likely that melatonin is beneficial due to its ability to stimulate anti-cancer immunity as well as its ability to protect blood counts and healthy tissues during chemotherapy and radiation regimes. It is disappointing that the trial mentioned above was never followed up with more rigorous testing in glioma patients.

October 30, 2015
A study [36] by Brazilian researchers correlates an index of two genes involved in melatonin biosynthesis and metabolism with grade and prognosis in gliomas. First, the aggressive, invasive GBM cell line U87 was found to express low levels of ASMT (an enzyme involved in melatonin biosynthesis) and high levels of CYP1B1 (which converts melatonin to an inactive metabolite than can be excreted in the urine), with reduced levels of melatonin being found in the U87 culture medium compared to two other glioma cell lines. When a low physiological level of melatonin (1 nM) was added to the cell cultures, the proliferation of U87 cells was significantly inhibited by melatonin in the U87 cell line, with less inhibition in the other two lines.

The researchers then sought to correlate RNA levels of these two genes (ASMT and CYP1B1) with tumor grade and prognosis in 351 glioma samples from The Cancer Genome Atlas (TCGA). ASMT (the enzyme involved in melatonin biosynthesis) expression was significantly decreased in GBM samples versus grade 2 gliomas. GBM samples also had elevated CYP1B1 compared to grade 2 and 3 gliomas. A two-gene index was then devised, which is the ratio of ASMT:CYP1B1. This index correlated with tumor grade, with GBM samples scoring lower on the ASMT:CYP1B1 index. A lower ASMT:CYP1B1 score also correlated with higher expression of NFκB target genes, such as the anti-apoptotic BCL2A1, and pro-proliferation genes. Finally, the ASMT:CYP1B1 index was found to be a significant prognostic indicator, considering all patients together, high grade glioma (grade 3 to 4) only, or grade 3 gliomas and grade 4 gliomas as separate group. Increased ASMT:CYP1B1 was correlated with improved survival in all groups.

These findings may be considered support for melatonin supplementation, especially for high grade glioma and glioblastoma.

Vitamin D3 and calcitriol

In vitro

We often hear about the anticancer effect of vitamin D, especially with regard to common cancer such as breast, prostate, and colon cancer. The evidence for brain cancer is not so clear, and much of this evidence comes from in vitro laboratory studies. These in vitro studies, often showing a glioma-inhibiting effect of calcitriol (1,25-dihydroxyvitamin D3 – the hormonally active form of vitamin D), may be misleading, as these studies often use calcitriol concentrations which far exceed the levels found in human serum. Calcitriol levels in human serum usually fall between 40 and 130 picomolar (pM), or about 100 trillionths of a mole per litre of serum. In contrast, lab studies testing the efficacy of calcitriol may use concentrations such as 100 nanomolar (nM), or 100 billionths of a mole per litre – 1000 times higher concentration than is likely to be found in serum.

One study [21] tested nine different glioblastoma cell lines with various concentrations of calcitriol (1,25-dihydroxyvitamin D3) every other day for a total of six days. At the physiologic concentration of 100 picomolar (100 picomoles per litre) of calcitriol in a reduced serum medium, proliferation was stimulated in three cell lines and inhibited in five. Only one GBM cell line was inhibited to less than 80% compared to untreated controls. In serum-free medium, two GBM cell lines proliferated, one was inhibited, and one line had no response to physiologic levels of calcitriol. When exposed to physiologic levels of calcidiol (25-hydroxyvitamin D3 – the predominant circulating form of vitamin D), three GBM cell lines proliferated, two showed little response, and four were inhibited to around 70% compared to untreated controls.

Therefore, as we’ve seen with retinoic acid, glioma cell response to vitamin D metabolites may be dependent on the specific cell culture (or tumour), as well as the dose. In the study described above [21], 20 GBM biopsy samples were compared to normal brain tissue. 19 out of 20 glioblastoma samples had reduced expression of the vitamin D receptor (VDR) compared to normal brain. In contrast, 16 out of 20 samples had higher levels of CYP27B1 – the enzyme which converts calcidiol to hormonally active calcitriol. In this study, the cell line which showed the most notable proliferation in response to calcitriol had a CYP27B1 level 60 times higher than normal brain, but a vitamin D receptor level only 0.1 times that of normal brain. A reasonable speculation might be that this particular cell line is metabolizing vitamin D in a manner not dependent on the vitamin D receptor, resulting in proliferation rather than inhibition. Although the more common response to physiologic levels of calcitriol in this study was inhibition of cell proliferation, the fact that a few cell lines showed increased proliferation (including one out of nine which showed markedly increased proliferation) should give us pause for consideration before making a decision regarding high-dose vitamin D supplementation.

Vitamin D Receptors and prognosis

In contrast to the above study, another study [22] published online in the Journal of Neuro-Oncology (Online First Articles, February 2014) tested 61 human glioblastoma samples for expression of vitamin D receptor (VDR), comparing them to 8 non-malignant brain samples. Both the percentage of samples positive for VDR (75% versus 37.5%) and the VDR immunoreactivity scores (median 3 versus median 0) were higher in the glioblastoma samples compared to normal brain. VDR positivity was also associated with a statistically significant increase in survival in the glioblastoma cases, with a median survival of 389 days for the VDR positive group versus 230 days for the VDR negative group. In vitro, silencing the VDR gene in T98 glioblastoma cells increased cell survival and migration. Calcitriol (hormonally active vitamin D) also had growth inhibitory effects in cell culture, though the concentration used in this study was one micromolar, or 10,000 times higher than the physiologic level of around 100 picomolar. This study leads one to conclude that vitamin D effects in brain tumours are likely determined by the vitamin D receptor levels, with higher VDR levels leading to positive outcomes.

For those with IDH-mutant tumours, please see the favorable evidence for vitamin D3 as a Hedgehog pathway inhibitor on the Exploring Strategies for IDH1 Mutated Glioma page.

A small pilot trial of the vitamin D analogue alfacalcidol is discussed on the Repurposed Drugs page. This trial apparently demonstrated an impressive benefit of vitamin D supplementation (in the form of alfacalcidol) in 2 out of 10 glioblastoma patients.

Complementary therapy and survival in glioblastoma

May 7, 2015
A new study published in Neuro Oncology Practice evaluated the effects of complementary therapy in a cohort of 470 GBM patients, 77% of whom used complementary therapies. Adjusted for age, KPS, and extent of surgery, use of supplemental vitamin D was associated with better survival which was marginally significant (hazard ratio=0.74, p=0.09). In contrast vitamin E was associated with marginally higher mortality (hazard ratio=1.54, p=0.09). In this study, multivitamins and omega-3 fatty acid use had no significant effect on survival. “These exploratory analyses suggest no mortality association with the use of multivitamins or omega-3 fatty acids. Associations observed with vitamins D and E merit further investigation.

The unreliability of in vitro evidence and the lesson of quercetin

Quercetin is a dietary flavonoid found in many different edible plant species. One of its suggested uses is as an anti-cancer supplement, and its use in glioma is supported by several in vitro studies which demonstrated suppression of glioma cell cultures when applied in concentrations of over 5 micromolar.

It has also been demonstrated that the effects of various cell toxins are dose-dependent, with growth promotion at lower doses, and growth suppression at higher doses. This phenomenon is called hormesis, or the biphasic dose response. This is one of the reasons that in vitro studies are often unreliable in terms of their clinical value: in vitro studies may be designed without consideration of the actual concentrations of drug achievable in living tissue. Secondly, in vitro studies do not account for the complex drug metabolism which occurs when taken in by a living body, nor the multitude of other factors that come into play in a living body, such as the functioning of the immune system.

A study published online in 2013 in the journal Food and Chemical Toxicology provides a perfect example of the potential unreliability of in vitro studies [23]. This study was done by a group which had previously published in vitro work with quercetin and glioma, showing a growth suppressive effect. The researchers wanted to provide more solid proof by doing in vivo studies with rats, using the same glioma cell line, the rat glioma C6. Rats were injected intraperitoneally with quercetin at the daily dose of 50 mg per kg body weight. Quercetin injection began at day 5 after glioma cell implantation into the brains of the rats. In the first experiment, quercetin was administered every day for 10 days, after which time no reduction in tumour volume was observed. In the second experiment, the rats were treated with quercetin for 15 days. The tumours of the quercetin-treated rats were statistically significantly larger than the control rats after 15 days. Concentrations of quercetin in the rat brains were found to be about 530 nanomolar (nM) or 0.53 micromolar, which is a significantly smaller concentration than used in the prior in vitro studies. T-cell proliferation was slightly reduced in the treated mice, though not enough to provide a full explanation of the increased tumour size in the quercetin treated rats. The authors conclude with a note of caution about the use of compounds such as quercetin on the basis of in vitro evidence alone. Other plant compounds, such as resveratrol and the green tea catechin EGCG, as well as cannabinoids such as THC have all shown hormetic, dose-dependent effects in vitro, with growth stimulation at lower concentrations and growth suppression at higher concentrations. These experiments still do not model the complex interactions involving drug metabolism or the immune system. In conclusion, while in vitro evidence should be regarded as the first preliminary evidence on the road to clinical utility, this example of quercetin shows that in vitro evidence alone is not a sufficiently reliable grounds on which to base clinical decisions, especially regarding substances with a proven biphasic dose response, or potential immunosuppressive properties.

In summary, plant compounds hold much promise for cancer patients looking for complementary therapies. However, attention to details such as dosing, especially when it comes to interpreting in vitro tests, is required. Unfortunately, the system as it exists today is not particularly interested in performing clinical trials on these non-patentable substances.

  1. Boswellia serrata acts on cerebral edema in patients irradiated for brain tumors: a prospective, randomized, placebo-controlled, double-blind pilot trial. Kirste et al. 2011.
  2. Boswellic acids inhibit glioma growth: a new treatment option? Winking et al. 2000.
    READ ABSTRACT (email me for PDF copy)
  3. Development of curcumin as an epigenetic agent. Fu et al. 2010.
  4. Curcumin down-regulates DNA methyltransferase 1 and plays an anti-leukemic role in acute myeloid leukemia. Yu et al. 2013.
  5. Optimized curcumin, myth versus fact.
  6. Curcumin structure-function, bioavailability, and efficacy in models of neuroinflammation and Alzheimer’s disease. Begum et al. 2008.
    READ SOURCE DOCUMENT (discussion of Longvida precursor solid lipid curcumin particles on page 12 of the PDF version, near the end of the study Discussion)
  7. Dietary curcumin attenuates glioma growth in a syngeneic mouse model by inhibition of the JAK1,2/STAT3 signaling pathway. Weissenberger et al. 2010.
  8. Id-1 is a key transcriptional regulator of glioblastoma aggressiveness and a novel therapeutic target. Soroceanu et al. 2013.
  9. GW Pharmaceuticals commences phase 1b/2a clinical trial for the treatment of glioblastoma multiforme (GBM). Press Release, November 11, 2013.
  10. Maitake D fraction: healing and preventive potential for cancer. Nanba. 1997.
  11. Oral administration of soluble beta-glucans extracted from Grifola frondosa induces systemic antitumor immune response and decreases immunosuppression in tumor-bearing mice. Masuda et al. 2013.
    READ ABSTRACT (email me for PDF copy)
  12. Melatonin: buffering the immune system. Carrillo-Vico et al. 2013.
  13. Treatment of cancer chemotherapy-induced toxicity with the pineal hormone melatonin. Lissoni et al. 1997.
    READ ABSTRACT (email me for PDF copy)
  14. Chemoneuroendrocrine therapy of metastatic breast cancer with persistent thrombocytopenia with weekly low-dose epirubicin plus melatonin: a phase II study. Lissoni et al. 1999.
    READ ABSTRACT (email me for a PDF copy)
  15. Thrombopoietic properties of 5-methoxytryptamine plus melatonin versus melatonin alone in the treatment of cancer-related thrombocytopenia. Lissoni et al. 2001.
    READ ABSTRACT (email me for PDF copy)
  16. Increased survival time in brain glioblastomas by a radioneuroendocrine strategy with radiotherapy plus melatonin compared to radiotherapy alone. Lissoni et al. 1996.
    READ ABSTRACT (email me for PDF copy)
  17. Berberine and its more biologically available derivative, dihydroberberine inhibit mitochondrial respiratory complex I: a mechanism for the action of berberine to activate AMP-activated protein kinase and improve insulin action. Turner et al. 2008.
  18. Laboratory studies of berberine used alone and in combination with 1,3-bis(2-chloroethyl)- 1-nitrosourea to treat malignant brain tumors. Zhang et al. 1990.
    READ SOURCE DOCUMENT (you may or may not be able to get the PDF here. email me for scans of the article, roughly 2.5 Mb)
  19. Green tea epigallocatechin gallate enhances therapeutic efficacy of temozolomide in orthotopic mouse glioblastoma models. Chen et al. 2011.
  20. Green tea polyphenol (-)-epigallocatechin 3-gallate inhibits MMP-2 secretion and MT1-MMP-driven migration in glioblastoma cells. Annabi et al. 2002.
    READ ABSTRACT (email me for PDF copy)
  21. Vitamin D(3) metabolism in human glioblastoma multiforme: functionality of CYP27B1 splice variants, metabolism of calcidiol, and effect of calcitriol. Diesel et al. 2005.
  22. Vitamin D receptor expression is associated with improved overall survival in human glioblastoma multiforme. Salomon et al. Journal of Neuro-Oncology, Online First Articles, February/March 2014.
    READ ABSTRACT (Email me for a PDF copy)
  23. Quercetin promotes glioma growth in a rat model. Zamin et al. 2014.
    READ ABSTRACT (Email me for PDF copy)
  24. Delta-9-tetrahydrocannabinol inhibits antitumor immunity by a CB2 receptor-mediated, cytokine-dependent pathway. Zhu et al. 2000.
  25. The Combination of Cannabidiol and Delta 9-Tetrahydrocannabinol Enhances the Anticancer Effects of Radiation in an Orthotopic Murine Glioma Model. Scott et al. 2014.
    READ ABSTRACT Email me for a PDF copy
  26. Studies on preventive and curative effects of berberine on chemical-induced hepatotoxicity in rodents. Janbaz et al. 2000.
  27. Hepatoprotective activity of berberine is mediated by inhibition of TNF-α, COX-2, and iNOS expression in CCl(4)-intoxicated mice. Domitrović et al. 2011.
    READ ABSTRACT Email me for a PDF copy
  28. Hepatoprotective effects of berberine on carbon tetrachloride-induced acute hepatotoxicity in rats. Feng et al. 2010.
  29. Protective Effects of Berberine on Doxorubicin-Induced Hepatotoxicity in Mice. Zhao et al. 2012.
  30. Berberine mitigates cyclophosphamide‑induced hepatotoxicity by modulating antioxidant status and inflammatory cytokines. Germoush et al. 2014.
    READ ABSTRACT Email me for a PDF copy
  31. Berberine induces senescence of human glioblastoma cells by downregulating EGFR-MEK-ERK signaling pathway. Liu et al. 2014.
    READ ABSTRACT Email me for a PDF copy
  32. Reactive oxygen species-mediated therapeutic response and resistance in glioblastoma. Singer et al. 2015.
  33. Pterostilbene suppressed irradiation-resistant glioma stem cells by modulating GRP78/miR-205 axis. Huynh et al. 2015.
    READ ABSTRACT Email me for a PDF copy
  34. cMYC expression in infiltrating gliomas: associations with IDH1 mutations, clinicopathologic features and outcome. Odia et al. 2013.
  35. Withania somnifera Suppresses Tumor Growth of Intracranial Allograft of Glioma Cells. Kataria et al. 2015.
    READ ABSTRACT Email me for a PDF copy
  36. Melatonergic system-based two-gene index is prognostic in human gliomas. Kinker et al. 2015.
    READ ABSTRACT Email me for a PDF copy