Repurposing some older drugs that cross the blood–brain barrier and have potential anticancer activity to provide new treatment options for glioblastoma

Abstract

Glioblastoma is a brain neoplasm with limited 5‐year survival rates. Developments of new treatment regimens that improve patient survival in patients with glioblastoma are needed. It is likely that a number of existing drugs used in other conditions have potential anticancer effects that offer significant survival benefit to glioblastoma patients. Identification of such drugs could provide a novel treatment paradigm.

Glioblastoma: current knowledge of therapeutics

Glioblastoma (GBM) is the most common and most lethal form of primary brain tumour worldwide 1. Most often, GBM is diagnosed at between 45 and 70 years of age 2. Currently, the standard treatment for patients with GBM involves tumour resection followed by radiotherapy and administration of a chemotherapeutic agent, temozolomide (TMZ). In spite of current treatment regimens, survival remains poor; median survival time ranges from 12 to 16 months 2, 3. To date, no additional treatment strategies that prolong overall survival (OS) further have been identified 4. Developments of new treatment regimens that improve patient survival in patients with GBM are needed. It is likely that a number of existing drugs used to treat other conditions have potential anticancer effects that offer significant survival benefit to GBM patients. Identification of such drugs could provide a novel treatment paradigm.

One significant limitation to the standard treatment protocol is tumour cell resistance to TMZ. O⁶‐methylguanine methyltransferase (MGMT) is a DNA repair protein which confers cellular resistance to TMZ 5. Spontaneous methylation, and subsequent silencing, of the MGMT gene promoter occurs in up to 50% of patients with GBM and has been associated with a significant increase in OS compared to patients with unmethylated MGMT promoters 6. For patients with unmethylated MGMT gene promoters, who may not respond well to standard treatment, there are currently no alternative treatment options. Furthermore, given that survival in all patients remains poor, more effective regimens are required.

In order to consider pharmacological agents that might have potential in the treatment of GBM, a thorough knowledge of GBM cancer biology is required. In summary, it is well documented that the growth and pathogenesis of GBM is driven by various genetic and epigenetic aberrations 7, 8, 9, 10. A number of recent studies have reported on the heterogeneous nature of GBM 4, 8, 10. Given this, it is probable that within a single tumour there are subpopulations of cells with different genetic alterations and potentially different susceptibilities to chemotherapeutic agents 11. It is unlikely that a ‘single‐hit’ treatment strategy will provide significant benefit to many patients diagnosed with GBM. The administration of several therapeutic agents will allow clinicians to target a range of GBM phenotypes and, as such, therapy involving multiple adjuvants might provide more benefit than standard chemotherapy with TMZ alone.

Two of the main pathological features of GBM include increased cellular proliferation and angiogenesis (Figure 1) 8. Increased cellular proliferation is a hallmark of neoplasia and in the context of GBM probably results from aberrant growth factor signalling, inactivation of tumour suppressor pathways (such as the p53 and retinoblastoma pathways), upregulation of the phosphatidyl inositol 3‐kinase (PI3K) pathway 8 and epidermal growth factor receptor (EGFR) gene mutations 10. In GBM, extensive angiogenesis might be attributed to elevated levels of vascular endothelial growth factor (VEGF) and might also involve fibroblast growth factor (FGF), nerve growth factor (NGF) and angiotensin II 12, 13.

Figure 1.

The main pathological features of glioblastoma. EGFR, epidermal growth factor receptor; FGF, fibroblast growth factor; HIF‐1, hypoxia‐inducible factor 1; PI3K, phosphatidyl‐inosotol‐3 kinase; NGF, nerve growth factor; VEGF, vascular endothelial growth factor

Additional pathological features of GBM include changes to the stromal environment such as necrosis, hypoxia and peritumoural oedema (Figure 1) 9, 14. A number of studies have reported that hypoxia might stimulate angiogenesis and increase vascularity through the upregulation of hypoxia‐inducible factor 1 (HIF‐1) 15, 16. Furthermore, hypoxic tumour cells are known to be resistant to radiotherapy, and this might also be attributed to the upregulation of HIF‐1 17, 18, 19. There is evidence in the literature that hypoxia might also confer tumour cell resistance to chemotherapy 20, 21. Several studies have shown that peritumoural oedema and necrosis in GBM have been associated with poor patient survival 9, 14.

At present, there is a pressing need to explore alternate treatment options which could improve OS and quality of life in patients with GBM. Given this, the principal aim of the present review was to identify existing drugs or therapeutic agents already registered or used in previous clinical trials that might impair one or more of the pathological processes of gliomagenesis outlined above (these will be referred to as antineoplastic drugs or antineoplastic agents), and therefore be helpful to consider in a clinical trial development portfolio.

Method

A literature search was conducted using the PubMed database. MeSH search terms and subheadings included glioma; GBM; brain neoplasms, treatment; therapeutics, combined drug therapy; therapeutics, blood–brain barrier; and pharmacology, therapeutic use. The search was restricted to primary research articles published in English. Studies were selected for review based on the inclusion of a drug that crosses the blood–brain barrier (BBB), that is primarily prescribed for noncancer indications and that might have antineoplastic actions. The ability of a drug or its active metabolite to cross the BBB and reach the tumour is paramount in the treatment of brain neoplasms. For the purpose of the present review, the ability of a drug to cross the BBB was determined using clinical pharmacology and therapeutics principles and clinical experience; drugs known to act in the central nervous system (CNS) or to elicit side effects in the CNS were included.

Exploration of the existing literature revealed a number of drug classes that might improve survival in patients with GBM. We restricted the present review to include drugs that are currently marketed, widely used in clinical practice and with favourable side‐effect profiles. This is because there is a short time to clinical use, and there is little point in developing drugs that cannot be used because of a life‐threatening side‐effect profile. In particular, we focused the present review on several agents within the following categories, with the belief that these are the most promising targets for future research: histone deacetylase inhibitors (HDACi), drugs that inhibit neurotransmitters, beta‐adrenoceptor antagonists (beta‐blockers), statins and antihypertensive medications. A total of 45 studies were included in the present review. It is important to note that this was not a comprehensive review of all drugs that might be useful in the treatment of GBM. Additional drug classes that might improve patient survival include antimalarial drugs, antiviral drugs, drugs used in the treatment of diabetes and anti‐inflammatory medications; however, these were not included in the present review. These additional drug classes are reviewed briefly in the conclusions.

Drugs that inhibit neurotransmitters

Psychotropic agents are well known to influence neurotransmission and have more recently been shown to have antiproliferative effects that might be valuable in the treatment of neoplasms 22, 23. Several subclasses of psychotropic drugs, including tricyclic antidepressants (TCAs), selective serotonin reuptake inhibitors (SSRIs), phenothiazines and antiepileptic drugs (AEDs; these will be reviewed later), have been investigated as potential adjuvants for cancer therapy 7, 22, 23, 24. The TCAs act at presynaptic adrenoceptors and serotonin receptors in the CNS to block the reuptake of noradrenaline and serotonin, respectively. SSRIs are more specific and act only on presynaptic serotonin receptors to prevent serotonin reuptake. Phenothiazines are dopamine receptor antagonists traditionally indicated for use in the treatment of psychosis and chemotherapy‐induced emesis.

Recently, TCAs (amitriptyline, imipramine, clomipramine, doxepin and citalopram) and SSRIs (paroxetine, fluoxetine and sertraline) have been reported to reduce tumour cell proliferation in a number of cell lines, including rat C6 glioma, human neuroblastoma and human astrocytoma cell lines 22, 24, 25, 26. Reduced cellular proliferation might occur through several mechanisms. Two recent studies showed an increase in caspase‐3 activity (a cysteine protease involved in the induction of the apoptotic cascade) in response to treatment with TCAs and SSRIs 22, 25. Contrary to this hypothesis, Tzadok et al. 26 demonstrated decreased cell proliferation in response to antidepressant treatment, with an absence of increased caspase‐3 activity, and hypothesized the involvement of the mitogen‐activated protein kinase pathway. A recent in vivo study by Walker et al. 24 investigated TCA use and mortality in patients with glioma or colorectal cancer. In a sample of 2592 patients with glioma and 22 524 patients with colorectal cancer, 4.1–4.2% of patients reported the use of low‐dose TCAs for the period of at least two prescriptions; however, there was no significant reduction in mortality associated with the use of TCAs in patients with glioma or colorectal cancer.

Several studies have reported on the potential antineoplastic activity of dopamine receptor antagonists, including chlorpromazine, thioridazine and perphenazine 23, 26, 27. Tzadok et al. 26 found a significant additive antiproliferative effect following treatment of U87 glioma cells with TMZ and perphenazine. Perphenazine crosses the BBB and has a strong predilection for the blockage of D2 and D3 dopamine receptors. These receptors have been implicated in gliomagenesis through stimulation of neuronal stem cells within the subventricular zone (SVZ) in the lateral ventricles 28. Blockade of D2 and D3 receptors in the SVZ might improve survival in GBM 28. Yde et al. 27 reported increased sensitivity of breast cancer cells to a first‐line therapeutic agent, tamoxifen, with adjuvant chlorpromazine therapy. Furthermore, Sachlos et al. 23 demonstrated that treatment with thioridazine reduced proliferation in acute myeloid leukaemia cell lines, potentially through selective targeting of cancer stem cells. To the best of our knowledge, there is no existing research regarding the efficacy of chlorpromazine and thioridazine as adjuvant anticancer agents in GBM.

AEDs

Epileptic seizures are a common complication of GBM, occurring in 22–60% of patients, and often treatment with an AED such as valproate (VPA), phenytoin, carbamazepine or levetiracetam (LEV) is required to provide symptomatic relief 29. A large number of studies have found a potential benefit of HDACi such as VPA in the treatment of GBM 30, 31, 32 Furthermore, other AEDs have also been hypothesized to offer a survival benefit in patients with glioma 33, 34, 35. HDACi and other AEDs are reviewed below.

HDACi

VPA has a wide spectrum of antiseizure activity and has recently been suggested as a potential adjuvant in cancer treatment 36. It is known to act on gamma‐butyric acid (GABA) levels in the brain, to block voltage‐gated sodium, potassium and calcium channels and to inhibit histone deacetylases 36. Abnormal activation of histone deacetylases can induce neoplasia; deacetylation of histones promotes the condensation of chromatin and repression of the transcription of several genes. HDACi can trigger hyperacetylation of histones, promoting growth arrest and apoptosis 31, 35, 37. Furthermore, inhibition of chromatin condensation might increase the availability of target DNA to alkylating chemotherapeutic agents, and enhance therapy 32, 38.

A number of recent studies have reported the antineoplastic effects of VPA using human glioma cell lines 31, 32, 38. Adjunctive treatment of glioma cell lines with VPA provided an additive effect on inhibition of cellular proliferation following treatment with TMZ. Furthermore, treatment with VPA has been shown to increase tumour cell sensitivity to radiotherapy 31, 32, 38. The mechanisms by which VPA might enhance the cytotoxic effects of TMZ and radiation are unknown. In addition, it has been reported that VPA might increase cellular levels of the MGMT protein 30, 38. In spite of this, VPA is still widely acknowledged to have an antiproliferative effect in glioma cells.

In addition to VPA, some of the newer AEDs, including topiramate (TPM) and the major active metabolite of LEV, 2‐pyrrolidinone‐n‐butyric acid (PBA), might act as HDACi 37. LEV is a newer AED, traditionally indicated for use as monotherapy in the treatment of partial focal seizures. Bobustuc et al. 30 found that treatment with LEV + TMZ significantly reduced cellular proliferation in several GBM cell lines. Furthermore, LEV was reported to have an inhibitory effect on MGMT protein expression in GBM cell lines but not in normal human astrocyte cells. It was hypothesized that increased levels of the p53 protein in cells treated with LEV enhanced p53 binding and inhibition of the MGMT gene promoter, consequently downregulating expression of the MGMT protein and enhancing tumour cell sensitivity to TMZ 30.

TPM is a newer AED, with several mechanisms of action that might contribute to antiseizure effects 39. TPM, in addition to VPA, is classified as a category D drug in pregnancy, owing to an increased risk of fetal malformation 40, 41. Two studies have demonstrated potential antimetastatic and antiangiogenic effects of TPM in mice injected with Lewis lung carcinoma 42, 43. In the earlier study, Ma et al. 42 demonstrated that treatment with TPM for 20 days strongly inhibited tumour metastasis in a dose‐dependent manner. More recently, these authors 43 reported that treatment with TPM for 21 days resulted in a significant reduction of VEGF expression, suggesting that TPM might effectively decrease tumour angiogenesis. It is unknown whether TPM would have similar effects in glioma cells.

Other AEDs

Enzyme‐inducing AEDs (EIAEDs), including carbamazepine and phenytoin, potently induce components of the cytochrome P450 system in the liver and consequently enhance the metabolism of chemotherapeutic agents and other drugs, including steroids that are metabolized by the same system. As such, concomitant treatment with EIAEDs might increase the dosage requirements of chemotherapeutic agents and have an effect on patient outcomes 33, 34, 35, and this is an important consideration in GBM pharmacotherapy.

Several studies have reported on the use of EIAEDs and non‐EIAEDs in patients with GBM 7, 33, 34. Jaeckle et al. 33 investigated the use of AEDs in 620 patients with GBM; 432 patients received treatment with an EIAED, 14 were prescribed a non‐EIAED and 159 did not receive any AED. The median OS in the patients who were prescribed an EIAED was 12.3 months compared to 10.7 months for patients in the other two groups. By contrast, a previous study by Oberndorfer et al. 34 found that, within a retrospective cohort of 168 patients with GBM (of whom 37 were prescribed a non‐EIAED and 43 were prescribed an EIAED), OS was 13.7 months for patients who were prescribed a non‐EIAED compared to 10.8 months in patients prescribed an EIAED. More recently, a study by Weller et al. 35 reported on the use of AEDs in a retrospective cohort of 573 patients with GBM, of whom 110 patients were treated with a non‐EIAED (predominantly VPA) and 277 patients received treatment with an EIAED. These authors found a significant survival benefit for patients treated with VPA compared to those taking other AEDs (including EIAEDs) or no AED 35. A similar retrospective study of 544 patients with primary GBM by Barker et al. 7 reported on the use of AEDs (including phenytoin, LEV, carbamazepine, phenobarbital and VPA) in 403 patients. The median OS of patients in the study was 14 months, irrespective of treatment modality. The median OS in patients treated with VPA was 16.9 months, and in those treated with radiation therapy + TMZ + VPA was 23.9 months, compared to 15.9 months without VPA. However, only a small number (n = 29) of patients were treated with VPA. Recently, there has been a large focus on VPA as a first‐line AED in the treatment of GBM.

Statins

Statins are traditionally prescribed to treat hypercholesterolaemia. More recently, statins have been indicated as potential adjunctive treatments in several cancers, owing to inhibitory effects on cellular signalling pathways involved in proliferation, migration, invasion and induction of apoptosis 44, 45, 46, 47. The primary mechanism by which statins act is inhibition of 3‐hydroxy‐3‐methylglutaryl‐coenzyme A (HMG‐CoA) reductase, preventing the conversion of HMG‐CoA to mevalonate 48. Important downstream products of mevalonate include cholesterol, Ras, Raf and Rho 48, 49. Gabrys et al. 48 discovered that growth in glioma cells might be highly dependent on the mevalonate pathway. As such, this pathway is a significant target for potential antineoplastic therapies.

Several recent in vitro studies have found statins to have proapoptotic and cytotoxic effects on glioma and other cancer cells 50, 51. Tapia‐Perez et al. 50 demonstrated statin‐mediated toxicity in human glioma cell lines U87, U138 and LN405, and in one rat glioma cell line, RGII, treated with simvastatin and lovastatin. This observed cytotoxicity was likely to have been due to stimulation of the proapoptotic protein, Bim. By contrast, Wood et al. 51 found an increase in levels of Bcl 2 (an antiapoptotic protein) in rats exposed to chronic administration of simvastatin (50 mg kg^–1 day^–1 for 21 days). In humans, the highest daily dose of simvastatin would equate to approximately 0.07 mg kg^–1 day^–1, much lower than the dosage used in the study by Wood et al. 51.

Additional in vitro studies have reported on the antiproliferative effects of statins on several cancer cell lines 46, 48, 49. Yanae et al. 46 investigated the treatment of rat C6 glioma cells and human‐derived U251MG GBM cells with mevastatin, fluvastatin and simvastatin. In this study, 5 μM of mevastatin and simvastatin, and 2.5 μM of fluvastatin inhibited cell proliferation and induced apoptosis in both cell lines. Yanae et al. 46 stated that the peak plasma concentration of fluvastatin in humans is less than 2 μM. Similarly, Wu et al. 49 found a 35% reduction in cellular proliferation in U87 and U251 human GBM cells treated with simvastatin compared to control cells. Gabrys et al. 48 demonstrated that lovastatin (0.625–50 μM) inhibited proliferation in U87MG and FaDu (hypopharyngeal squamous cell carcinoma) cell lines in a dose‐dependent manner. Mechanisms contributing to the inhibitory effects of statins on cellular proliferation might include inhibition of cell cycle progression as a result of decreased cellular signalling 48, 49. Furthermore, a study by Yongjun et al. 47 reported decreased cell migration in a U87 glioma cell line in response to treatment with atorvastatin, probably due to inhibition of matrix metalloproteinase (MMP) 2).

Antihypertensive drugs

Antihypertensive agents include angiotensin‐converting enzyme inhibitors (ACEIs), angiotensin receptor antagonists/blockers (ARBs) and beta‐blockers (these will be reviewed later) and are typically prescribed to treat hypertension, congestive heart failure and diabetic nephropathy. ACEIs reduce the production of angiotensin II, whereas ARBs selectively inhibit the activation of the angiotensin II type 1 receptor 52. Angiotensin II plays an important role in the regulation of fluid and electrolyte balance and is also involved in cell growth, differentiation, apoptosis and angiogenesis 53, 54. Existing research has hypothesized an antineoplastic role for ACEIs and ARBs, primarily due to the inhibition of angiotensin II 53, 55.

Rivera et al. 55 investigated the effects of treatment with an ARB, losartan, in 12‐week‐old rats implanted with rat C6 glioma cells and found a significant decrease in cell proliferation and a decrease in the number of capillary vessels in tumours resected from rats treated with either 40 mg kg^–1 or 80 mg kg^–1 once daily compared to controls. A more recent study from the same group reported a significant decrease in tumour volume following 30 days of treatment with either 40 mg kg^–1 or 80 mg kg^–1 losartan once daily compared to controls 53. Furthermore, Arrieta et al. 53 showed increased apoptosis and decreased levels of the proangiogenic factors VEGF, platelet‐derived growth factor and FGF in rat C6 glioma cell lines treated with losartan compared to controls. A study by Rooprai et al. 56 demonstrated downregulation of MMP‐2 and MMP‐9 in four cell lines (an ependymoma, an oligoastrocytoma, an anaplastic astrocytoma and a GBM cell line) in response to treatment with the ACEi captopril. MMPs are thought to play a role in the degradation and remodelling of the extracellular matrix and might be critically involved in GBM metastasis and invasion; as such, captopril might have potential anti‐invasive effects 57.

A number of retrospective analyses have investigated the impact of antihypertensive treatment on patient outcome in several cancers 12, 52, 58, 59, 60. Cardwell et al. 58 reported on the use of ACEIs and/or ARBs after diagnosis with either breast, colorectal or prostate cancer in over 20 000 patients in the UK. This study found a small, nonsignificant reduction in cancer‐specific mortality in colorectal and prostate cancer patients treated with an ACEI. Overall, there was no significant impact on all‐cause mortality for patients treated with either an ACEI or ARB. Holmes et al. 59 found that treatment with ACEIs, ARBs and/or beta‐blockers had no significant survival benefit in a sample of over 3000 Canadian patients with either breast, colorectal, lung or prostate cancer. Chae et al. 12 investigated the use of antihypertensive agents in 1449 patients with breast cancer. This study found no significant decrease in OS for patients prescribed antihypertensive medications, although treatment with ARBs was associated with a decreased risk of recurrence compared to controls. A smaller study by Engineer et al. 52 showed that combination treatment with a beta‐blocker and either an ACEI or ARB decreased progression and decreased hospitalizations over a 4‐year period in a sample of 262 patients with colorectal cancer compared to patients taking either a beta‐blocker or an ACEI/ARB. One study investigating the effect of treatment with ACEIs/ARBs in GBM demonstrated no significant survival benefit 60. Interestingly, Carpentier et al. 60 reported a significant reduction in preoperative oedema (as visualized on magnetic resonance imaging) in patients treated with either an ARB/ACEI compared to controls, probably due to a reduction in neovascularization that normally enables the development of extensive peritumoural oedema. Furthermore, treatment with an ACEI/ARB resulted in a significant reduction in steroid treatment requirements.

Doxazosin is an apha‐1 adrenoceptor antagonist that has been used to treat hypertension and benign prostatic hyperplasia. A recent study by Petty et al. 61 found that doxazosin inhibited the migration of A172 glioma cells in a dose‐dependent manner. In the same study, mice were implanted with prostate cancer cells and subjected to either treatment with doxazosin for 10 days or to a control treatment. The authors demonstrated a significant reduction in the number and size of lung metastases in treated mice compared to controls. They hypothesized that doxazosin might cause activation of the erythropoietin‐producing hepatocellular A2 family of receptor tyrosine kinases involved in the induction of apoptosis, and inhibition of proliferation and migration.

Beta‐blockers

Beta‐blockers are commonly prescribed in the treatment of hypertension and in prevention of cardiovascular diseases 62, 63. Studies have suggested that these agents might inhibit angiogenesis, invasion and cellular proliferation, as well as increasing apoptosis in several cancer cell lines 64, 65, 66. The most extensively studied beta‐blocker in relation to potential utility as an adjunctive treatment in neoplasia is propranolol, a nonselective beta‐blocker that competes with adrenaline and noradrenaline binding at beta‐1‐ and ‐2 adrenoceptor sites 62, 63. Evidence in the existing literature suggests that blockade of adrenergic stimulation by beta‐blockers might decrease the expression of proangiogenic factors, including VEGF, MMP‐2 and MMP‐9; however, the mechanism by which this occurs is largely unknown 62, 67, 68. Chim et al. 16 hypothesized that decreased levels of hypoxia‐inducible factor 1 associated with propranolol treatment caused a decrease in VEGF levels in haemangioma endothelial cells.

Pasquier et al. 64 investigated the use of propranolol in several cell lines, including human breast cancer, neuroblastoma and U87 GBM cell lines. In this study, treatment with propranolol resulted in decreased proliferation in all cell lines compared to controls. Results were not reported for antiangiogenic effects of propranolol in the U87 GBM cell line. A subsequent study by Pasquier et al. 65 examined the effect of a number of beta‐blockers, including metoprolol, carvedilol, nebivolol, propranolol, butoxamine and atenolol, on two neuroblastoma cell lines. The mixed alpha/beta‐receptor antagonist, carvedilol; the selective beta‐1 adrenoceptor antagonist, nebivolol; and propranolol were reported to have significant antineoplastic effects involving the induction of apoptosis and increased cell cycle arrest compared to controls. Currently, there is little evidence in the literature to support the use of selective or mixed beta‐blockers over propranolol as an adjunct in cancer treatment.

The mechanism by which beta‐adrenoceptor antagonists might promote apoptosis is not yet understood. Kozanoglu et al. 67 reported that treatment with propranolol resulted in increased expression of the proapoptotic gene Bcl‐10, decreased expression of the antiapoptotic gene Bcl‐2 and decreased activation of the nuclear factor‐kappa B (NF‐kappa B) pathway in human U266 multiple myeloma cells. NF‐kappa B is a transcription factor critically involved in many cellular processes, and excessive activation of the NF‐kappa B pathway has been linked to the promotion of cellular proliferation and inhibition of apoptosis 69.

Conclusions

GBM is a relatively rare cancer, with an incidence of 2–4 per 100 000 adults per year worldwide that is unlikely to receive significant research funding or support. Given that drug development is associated with considerable time and financial costs to the pharmaceutical industry, research teams, governments and patients, alternative strategies are required to develop more efficient and timely therapeutic protocols. The ability to use currently marketed drugs in the treatment of GBM provides a financially practicable approach to encourage the development of new therapeutic regimens to improve OS in patients with GBM.

We have demonstrated in the present review that a number of drugs currently in routine use, including antidepressants, AEDs, statins, beta‐blockers and other antihypertensive agents, exhibit promising antineoplastic effects in vitro. We believe that these drug classes offer the most promising survival benefit in GBM. Many of the drugs presented here have been shown to antagonize one or more of the pathological features of GBM. Furthermore, a number of drugs, including VPA, may potentiate the effects of standard radiotherapy and chemotherapy treatments. It is important to note that the drugs presented here do not comprise an exhaustive list of potential therapeutic agents for patients with GBM. Table 1 presents a summary of the indication for use, mechanism of action and potential antineoplastic effects of the afore‐mentioned drugs.

Table 1.

Existing drugs with potential antineoplastic effects as demonstrated through in vitro studies.

Class Drug	Primary indications for use	Primary mechanism of action	Mechanism of antineoplastic effects	References
Tricyclic antidepressants	Major depression	Inhibit reuptake of noradrenaline and serotonin at presynaptic nerve terminals	Reduce cellular proliferation and might induce apoptosis through aberrant MAPK pathway activity or inhibition of mitochondrial activity	Levkovitz et al. 22, Higgins and Pilkington 25, Tzadok et al. 26
Amitriptyline	Neuropathic pain
Imipramine
	Migraine prophylaxis
Clomipramine
Doxepin
Citalopram
Selective serotonin reuptake inhibitors	Major depression	Inhibit reuptake of serotonin at presynaptic nerve terminals	Reduce cellular proliferation and might induce apoptosis through aberrant MAPK pathway activity or inhibition of mitochondrial activity	Levkovitz et al. 22, Higgins and Pilkington 25, Tzadok et al. 26
	Bipolar disorder
	Anxiety disorders
Paroxetine
Fluoxetine
Sertraline	Bulimia nervosa
Phenothiazines	Chemotherapy‐induced emesis	Dopamine receptor antagonists	Might decrease cellular proliferation and increase cellular sensitivity to some chemotherapeutic agents	Sachlos et al. 23, Tzadok et al. 26, Yde et al. 27
Chlorpromazine
Thioridazine
Perphenazine
Valproic acid (valproate)	Seizure disorders	Acts on GABA levels in the brain to reduce voltage‐gated sodium, potassium and calcium channels	Acts as a histone deacetylase inhibitor and might contribute to chromatin condensation, growth arrest and apoptosis	Camphausen et al. 31, Chinnaiyan et al. 32, van Nifterik et al. 38
	Bipolar disorder
	Migraine prophylaxis
Levetiracetam	Partial focal seizures	Unknown	Might act as a histone deacetylase inhibitor, contributing to chromatin condensation, growth arrest and apoptosis	Bobustuc et al. 30
	Adjunct in tonic–clonic and myoclonic seizures
Statins	Hypercholesterolaemia	Inhibition of HMG‐CoA reductase, the rate‐limiting enzyme in the cholesterol synthesis pathway	Decreased activity in cell signalling pathways contributes to induction of apoptosis and decreased proliferation	Yanae et al. 46, Yongjun et al. 47, Gabrys et al. 48, Wu et al. 49, Tapia‐Perez et al. 50
Lovastatin
Pravastatin	Prevention of cardiovascular disease
Rosuvastatin
Simvastatin
	Glaucoma
Angiotensin‐converting enzyme inhibitors	Hypertension	Decrease production of angiotensin II by inhibition of angiotensin converting‐enzyme	Decreased proliferation and angiogenesis. Might also decrease invasion and migration via reduction of MMP‐2 and MMP‐9 expression	Rooprai et al. 56
	Diabetic nephropathy
Captopril
Angiotensin receptor blockers	Hypertension	Antagonists at angiotensin II type 1 receptors	Decreased angiogenesis as a result of decreased VEGF expression	Arrieta et al. 53, Rivera et al. 55
	Diabetic nephropathy
Losartan
Other antihypertensives	Hypertension	Antagonist at alpha‐1 adrenoceptors	Reduction of cell migration, proliferation and apoptosis through activation of Ephrin A2 receptors	Petty et al. 61
	Benign prostatic hyperplasia
Doxazosin
Beta‐blockers	Hypertension	Antagonists at beta‐1 and/or beta‐2 adrenoceptors	Reduction of angiogenesis through decreased expression of VEGF and MMP‐9. Also decreases cell proliferation through unknown mechanisms	Kozanoglu et al. 67, Pasquier et al. 64
Propranolol *	Prevention of cardiovascular disease
Butoxamine
Metoprolol
Nebivolol

GABA, gamma‐aminobutyric acid; HMG‐CoA, 3‐hydroxy‐3‐methylglutaryl‐coenzyme A; MAPK, mitogen‐activated protein kinase; MMP, matrix metalloproteinase; VEGF, vascular endothelial growth factor.

^*The majority of studies reporting on anti‐neoplastic effects of beta‐blockers involved only propranolol.

Recent research has identified additional classes of drugs that might have survival benefit in GBM. These drug classes have been excluded from the present review owing to poor side‐effect profiles and/or limited available data. Nonsteroidal anti‐inflammatory drugs – in particular, celecoxib – have been investigated as adjuvants in glioma therapy owing to their ability to enhance radiosensitivity in glioma cell lines; however, celecoxib was recently shown to have no survival benefit in combination with TMZ and thalidomide in a Phase II study 70, 71. Drugs used in the treatment of diabetes, including metformin and pioglitazone, might inhibit glioma cell growth 72, 73. A number of studies have reported that the antimalarial drug, chloroquine, may enhance the cytotoxic effects of TMZ in glioma cell lines 74, 75. Disulfuram, a drug used in the maintenance of alcohol abstinence, has been demonstrated to reverse tumour cell resistance to TMZ 76. Antiviral therapy with acyclovir and valganciclovir has also been hypothesized to offer a survival benefit in GBM 77, 78. Furthermore, it is likely that a number of older drugs that have been withdrawn from use might also exhibit antineoplastic effects, especially in combination with radiotherapy.

As evident in the current literature, the majority of studies reporting on the possible antineoplastic effects of existing therapeutic agents are limited to in vitro investigations, in cell lines with known differences and a lack of heterogeneity in gene expression and activity against human cancers. Further, the concentration of drug used during in vitro experiments often exceeds by several‐fold the in vivo concentration achieved with the therapeutic dosage of a given drug. It is uncertain whether antineoplastic effects reported from in vitro studies would be observed during in vivo trials. The majority of human studies to date have been limited to retrospective analyses. In these cases, researchers are unable to investigate the effects of different dosages of drugs or different phenotypic variables, and research is limited when only a small number of patients are prescribed any given drug.

There have been few studies reporting on the concurrent use of more than one class of drug for the treatment of GBM and cancer cell lines. Multidrug therapy introduces many challenges, including the management of drug interactions and the cumulative risk of additional adverse effects. In the case of GBM, where survival remains extremely poor, it is likely that the therapeutic benefits of multidrug adjunctive therapy and the known safety profile of these older drugs would greatly outweigh any therapeutic risk.

Currently, new therapeutic regimens are required to improve OS in patients with GBM. Given the evidence presented in the present review, future randomized controlled trials incorporating existing agents with potential antineoplastic effects into current therapy could demonstrate an enhancement of the treatment protocol. However, in spite of their limitations, preliminary in vitro and retrospective in vivo studies to determine effective and appropriate dosing, as well as effective drug combinations, are required. Furthermore, a retrospective in vivo study of patients with GBM, correlating OS and progression‐free survival with individual patients' prescription medication, should be carried out to provide more information on the potential survival benefit obtained from adjunctive treatment with existing drugs.

Conflict of Interest/Disclosure

All authors have completed the Unified Competing Interest form at http://www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare: no support from any organization for the submitted work; no financial relationships with any organizations that might have an interest in the submitted work in the previous 3 years; and no other relationships or activities that could appear to have influenced the submitted work.

Contributors

DR‐T wrote the manuscript. JHM was involved in the design of the research, the development of ideas and the preparation of the manuscript. LC was involved in the development of ideas and acted as a scientific adviser. RH was involved in the design of the research, the development of ideas and the preparation of the manuscript, and acted as a scientific adviser.

Rundle‐Thiele, D. , Head, R. , Cosgrove, L. , and Martin, J. H. (2016) Repurposing some older drugs that cross the blood–brain barrier and have potential anticancer activity to provide new treatment options for glioblastoma. Br J Clin Pharmacol, 81: 199–209. doi: 10.1111/bcp.12785.