Repurposing Drugs in Oncology (ReDO)-diclofenac as an anti-cancer agent.

Introduction

Diclofenac (DCF) is a commonly used non-steroidal anti-inflammatory drug (NSAID) used in the treatment of pain in rheumatoid arthritis and other musculoskeletal conditions, migraine, fever, acute gout and post-operative pain. First developed by Ciba-Geigy (later merging with Sandoz to become Novartis), the drug is now available globally as a generic medication. DCF is also commonly available as a gel for topical application for localised pain or for the treatment of actinic keratosis. In some countries low-dose formulations of oral DCF (typically 25 mg tablets in small pack sizes of 12 – 18 tablets) are available over-the-counter (OTC) as a general purpose analgesic or anti-pyretic. It is also generally available OTC in the gel format. Common trade names include Voltaren, Voltarol, Cataflam, Cambia, Zipsor and Zorvolex.

DCF has an established role in oncological practice in the treatment of cancer-related pain and, as a topical treatment for actinic keratosis, which is commonly viewed as a pre-cancerous lesion. As these are licensed and common uses of DCF they are outside of the scope of this paper, except in the case where topical DCF is being investigated for other cancer indications.

Current usage

Dosage

DCF, which is available as a sodium or potassium salt, is used in tablet, gel/emulsion, injection and suppository forms. Dosages vary by format and indication. Typical doses for rheumatic disease and musculoskeletal disorders are in the range 75–150 mg in 2–3 divided doses, orally or rectally. Post-operative pain may be treated with diclofenac injections, either deep intramuscularly or intravenously, at a dose of 75–150 mg, with a maximum of 150 mg in 24 hours. The gel formulation utilises diclofenac sodium 3% in a sodium hyaluronate base and is applied twice daily for 60–90 days in the treatment of actinic keratosis [ 1 ].

Toxicity

While it is a non-selective inhibitor of both isoforms of the cyclooxygenase enzyme (COX-1 and COX-2), DCF has a preferential binding to COX-2 [ 2 ], which may explain its intermediate risk profile for gastro-intestinal (GI) events in comparison with some other NSAIDs. Common side effects include abdominal pain, constipation, diarrhoea, dyspepsia, flatulence, heartburn, nausea and headache. Less common side effects include rash, itching, bloating, GI ulceration, oedema and dizziness. Rare but serious adverse events include GI bleeding, anaemia, liver failure, pancreatitis and pneumonia. As with all NSAIDs, long-term use DCF is also associated with a small increase in the risk of cardiovascular events, particularly myocardial infarction and stroke. A recent meta-analysis reported that the vascular risks of long-term DCF use were similar to those of selective COX-2 inhibitors, with a rate ratio of 1.41 for vascular events for DCF, compared to 1.37 for selective COX-2 inhibitors and 1.44 for ibuprofen [ 3 ].

DCF is contra-indicated in patients with previous history of hypersensitivity to aspirin or any other NSAID, suffering from congestive heart failure, ulcerative colitis or other inflammatory bowel condition, active GI ulcer or bleeding. It is also recommended that DCF be avoided in the final trimester of pregnancy and caution be exercised during lactation.

Pharmacokinetics

Oral DCF is rapidly absorbed and almost completely distributed to plasma and tissues with little evidence of drug accumulation after repeated dosing within the normal therapeutic range [ 4 – 6 ]. Peak plasma concentration following a single 50 mg enteric coated diclofenac sodium tablet is 5.0 μM, attained in around 2 hours. The potassium salt of DCF is absorbed more rapidly, and a 50 mg tablet reaches a peak plasma concentration of 3.8 μM in 20–60 minutes. Terminal half-life is 1.8 hours after oral dosing. About 60% of the drug and its metabolites are eliminated in the urine and the balance through bile in the faeces. More than 90% of an oral dose is accounted for in elimination products within 72 hours, with only 1% of an oral dose excreted as unchanged parent compound in urine.

Pre-clinical evidence in cancer - In Vitro and In Vivo

This paper focuses on the evidence of an anticancer effect of DCF treatment, including data that is specific to DCF and other data that is in line with DCF’s effects as an NSAID. In particular the emphasis is on DCF treatment post-diagnosis rather than on evidence of efficacy in a cancer chemo-prevention context (see [ 7 ] and [ 8 ] for reviews on NSAIDs in chemo-prevention).

Fibrosarcoma

The first evidence for a possible anti-tumour effect of DCF was shown in experimental studies in implanted tumours (fibrosarcoma and hepatoma) in a rat model in 1983 [ 9 ]. Chemically induced rat tumours treated with a range of prostaglandin synthase inhibitors (indomethacin, diclofenac and aspirin) showed reduced growth rate and levels of vascularisation. Subsequent work indicated that DCF increased tumour blood flow in vivo , possibly via a role for prostaglandins in vascular permeability [ 10 ].

The in vivo effect of DCF on implanted fibrosarcoma tumours in mice was confirmed in 2002 by Hofer and colleagues [ 11 , 12 ]. G:5:113 murine fibrosarcoma cells were implanted in male C3H/DiSn mice and tumours allowed to develop for five days before drug treatment with DCF, ibuprofen or flurbiprofen at an i.p. dose of 0.15 mg/mouse commenced. Two regimens were used, drug for five days (regimen A) and for fourteen days (regimen B) continuously and animals were monitored for a period of 15 weeks. The dose was selected by authors as it approximated the typical human NSAID dose of 1–2.5 mg/kg/day. Regimen B was superior to regimen A for all three drugs tested. Tumour growth inhibition in DCF treated animals was observed three weeks after application in regimen B (46% of control values, P = 0.019). In vitro analysis showed a decrease in cell numbers in response to DCF concentrations of 5 μM, 10 μM and 20 μM.

Colorectal cancer

The anti-proliferative effects of a range of NSAIDs, including DCF, were assessed in three human colon cancer cell lines (HT-29, SW480, and DLD-1) in vitro in 1994 [ 13 ]. DCF was found to exert an anti-proliferative effect and had an IC50 of 55 μM, 37 μM and 170 μM respectively, making DCF one of the most potent of the panel of drugs tested. Later investigators studied the in vivo effect of topical application of DCF with hyaluronan on implanted colon-26 adenocarcinoma tumours in a BALB/c murine model [ 14 ]. Topical application at a dose of 6 mg/kg retarded and then stopped tumour growth compared to controls.

Additional evidence for an effect in colon cancer cell lines came from a study in 2003 by Falkowski and colleagues [ 15 ]. In addition to in vitro results which showed that DCF had a dose dependent effect on the C-26 murine colon adenocarcinoma cell lines, the authors also treated Balb/c x C57 BL/6 mice bearing syngeneic colon tumours with DCF at the dose of 250 mg/L in drinking water. Treatment commenced four days after tumour cell implantation and proceeded for 12 days. Tumour growth was reduced compared to untreated controls by day four of treatment and continued until treatment end. Prostaglandin E2 (PGE 2 ) and thromboxane B2 (TBX 2 ), both metabolites involved in the arachidonic acid cascade, were also significantly reduced by treatment end.

DCF is a component of the anti-angiogenic combinational drug combination TL-118, the other components being cimetidine, low dose cyclophosphamide and sulfasalazine. The efficacy of TL-118 was investigated in a mouse model of liver metastases from colorectal cancer and compared to treatment with rapamycin and the B20 anti-VEGF antibody [ 16 ]. CT-26-murine colorectal adenocarcinoma cells were injected into the spleen of male CB6F1 mice, leading to the formation of hepatic tumour nodules within 17 days of inoculation. Mice were treated with TL-118 intraperitoneally. The DCF dose is reported as 30 mg/kg, and is included in the TL-118 protocol on days one and four of each six day cycle of treatment. Only mice treated with TL-118 showed significant tumour growth delay, with both partial and complete remissions recorded. Overall survival in both partial and complete remission groups was significantly longer than untreated controls and animals treated rapamycin and B20 ( P < 0.0005).

Neuroblastoma

DCF was also tested for activity in neuroblastoma cell lines and xenograft models [ 17 , 18 ]. Johnsen et al showed that COX-2 was over-expressed in 27 of 28 (96%) tissue samples from paediatric neuroblastoma patients, with no staining in surrounding tissues. In vitro DCF and the selective COX-2 inhibitor celecoxib inhibited cell growth in a panel of neuroblastoma cell lines. The IC50 ranged from 12.5 to 50 μM for celecoxib and 100 to 600 μM for DCF, with evidence of increased apoptosis in response to DCF. In vivo nude rats carrying SH-SY5Y neuroblastoma xenografts were treated with DCF in drinking water at a dose of 200 mg/L or 250 mg/L. Tumour growth was significantly inhibited after 2 days of DCF treatment (200 mg/L, P = 0.042; 250 mg/L, P = 0.024) compared with untreated controls. At the higher dose tumour growth was inhibited throughout the treatment period (11 days after the appearance of palpable tumours), and tumour weight at autopsy was lower than untreated controls for both doses, (median tumour weight 1.52 g, 0.22 g and 0.21 g for control, 200 mg/L and 250 mg/L groups respectively, P = 0.009).

In their most recent work, this same group of investigators have identified a high-risk, inflammatory subset of neuroblastomas associated with deletion of chromosome 11q [ 19 ]. In addition to analysis of patient samples, an in vivo model was used (nude mice inoculated with SK-N-AS cells harbouring an 11q-deletion) to test the effect of DCF on tumour growth. Mice were treated with DCF at a dose of 250 mg/L in drinking water. The difference in tumour volume between DCF-treated mice and controls was significant by day 8 (approximately 33% lower, P = 0.01) and day 9 (approximately 40% lower, P = 0.008).

The TL-118 drug combination was also tested against neuroblastoma [ 20 ]. An aggressive orthotopic model was generated by implanting human SK-N-BE (2) cells into NOD-SCID mice. TL-118, was tested alone and in combination with gemcitabine or 13-cis-retinoic acid. Initial testing with TL-118 showed a high level of toxicity and therefore a reduced dose was used. TL-118 alone reduced tumour growth rate and extended survival 1.5-fold ( P < 0.0001). Treatment with gemcitabine alone also extended survival, but in combination with TL-118 the effect was significantly more pronounced, increasing survival 2.5-fold ( P < 0.001), suggesting a synergistic effect.

Ovarian cancer

Zerbini and colleagues assessed the combinatorial effect of combination treatment with NSAIDs in ovarian cancer cell lines and in vivo [ 21 , 22 ]. The effect of a panel of NSAIDs was tested against four ovarian cancer cell lines (SKOV-3, CAOV-3, SW626 and 36M2) and assessed singly and, for the most potent, in combination. DCF (in the range 20–200 μM) and sulindac sulphide showed the strongest activity, inducing apoptosis and inhibiting cell growth in all four cell lines. The combinations of DCF plus sulindac sulphide, DCF plus naproxen, sulindac sulphide plus naproxen and sulindac plus ebselen were more effective than single drug treatments. In vivo SCID mice were injected with SKOV-3 cancer cells and were fed a control diet or diet supplemented with DCF (dose 100 ppm) or sulindac sulphide (dose 200 ppm). All mice developed tumours, but DCF or sulindac-fed mice bore tumours with volume 20% or 30% lower than controls ( P < 0.05).

Later work by Valle and colleagues also investigated the use of NSAIDs, specifically DCF and indomethacin, in ovarian cancer cell lines and an in vivo model [ 23 ]. Serous ovarian adenocarcinoma cell lines HEY, OVCAR5 and UCI-101 were treated with varying concentrations of the two NSAIDs in the range 0–500 μM for 24 hours and assessed for cell viability. DCF treatment significantly ( P < 0.05) reduced cell viability at concentrations of 50 μM in the HEY and OVCAR5 cells, and at 250 μM in the UCl-101 line. The HEY cell line was used in the in vivo experiments in athymic nude mice. For the DCF group treatment commenced 3 days after inoculation, DCF was administered intraperitoneally twice a week for four weeks at a dose of 18 mg/kg. DCF treatment reduced tumour growth compared to controls by 33% ( P = 0.016), whereas treatment with indomethacin showed reduced tumour growth of 22% compared to their control group ( P = 0.031).

Other cancers

Results in an orthotopic syngeneic murine pancreatic cancer model showed that DCF treatment inhibited tumour growth compared to untreated controls [ 24 ]. Mice inoculated with PANC02 cells developed pancreatic tumours which readily metastasised to the peritoneal area around the incision site for tumour cell inoculation. Treatment with DCF was at a dose of 30 mg/kg (animal weight) given orally in drinking water and commenced three days after inoculation and continued for 11 days. Treated mice developed primary tumours with a mean weight 60% lower than untreated controls ( P ≤ 0.01), while mean weight of metastatic tumours was also lower in DCF animals but the difference did not reach statistical significance. Analysis of tumour samples showed evidence of increased apoptosis and decreased angiogenesis compared to controls. However, in vitro experiments did not show evidence of apoptotic effect in PANC02 cells cultured with DCF at a concentration of 10 μM and 50 μM for four days.

DCF was also used in a murine glioma model [ 25 ]. The effect of increasing DCF concentrations between 100 μM–600 μM was assessed in cultured GL261 glioma cells. DCF below 200 μM impaired cell growth and concentrations above 300 μM caused cell death.

Lactate production by cells was significantly reduced at a concentration of 100 μM. In vivo female C57BL/6 mice were orthotopically inoculated with GL261 glioma cells and then treated with DCF (at a dose of 25 mg/kg). As with the in vitro analysis, DCF decreased lactate production compared to controls, though not to a statistically significant level. However, DCF-treated mice had a statistically significantly higher median overall survival than control mice (30.5 days versus 24 days, P = 0.0156). Concurrent treatment of DCF and R848 (the TLR7/8 agonist resiquimod) did not show improvement in survival compared to either treatment alone. Subsequent work has shown that the IC50 for DCF treatment against a panel of human glioblastoma cell lines (HTZ-349, U87MG, and A172) is in the range 50–200 μM, which are physiologically relevant [ 26 ].

The combination of DCF and sorafenib was the subject of an in vitro study using a panel of nine melanoma cell lines [ 27 ]. The combination had been selected after a functional screen to identify promising synergistic combinations of drugs with activity against melanoma cell lines displaying the major genetic drivers of the disease (BRAF, NRAS, CDKN2A etc). Combination treatment of sorafenib and DCF was effective against all cell lines, regardless of genotypic status. A different research group also investigated the in vitro activity of sorafenib and a number of COX inhibitors, including DCF, in the HepG2 hepatocellular carcinoma (HCC) cell line [ 28 ]. The results showed that DCF treatment at a concentration of 50 μg/mL significantly reduced proliferation ( P < 0.01).

In vivo evidence for an effect of DCF in melanoma has also been published. Gottfried et al investigated the effect of DCF on the Myc transcription factor and glucose metabolism in leukaemia, prostate cancer and melanoma cell lines [ 29 ]. Additionally the group studied the effect of DCF in a syngeneic murine model, (C57/BL6 mice inoculated with B16 melanoma cells). Fourteen days post-inoculation DCF intraperitoneal treatment commenced at a dose of 15 mg/kg. A significant growth inhibitory effect was apparent within three days compared to controls ( P < 0.05), and tumour weight and volume were significantly reduced ( P < 0.001) at the end of the experiment (23 days after tumour inoculation).

Inoue and colleagues investigated the in vivo and in vitro effect of topical DCF application in prostate cancer [ 30 ]. Using two prostate cancer cell lines, one of which was transfected to over-express COX-2, the investigators treated each line with varying concentrations of DCF in the range 0 to 1000 μM for 72-hours. There was a dose dependent reduction in cell viability, with the COX-2 cells more sensitive to DCF. Cell viability in the COX-2 versus non-COX-2 cell lines was 74.0% and 95.7% ( P = 0.0094), 51.6% and 73.8% at 50 μM ( P < 0.0001) respectively. The IC50 was calculated as 42.2 μM and 91.6 μM, respectively. Further in vitro experiments showed that the cells over-expressing COX-2 were more resistant to radiotherapy than non-COX-2 over-expressing cells. The addition of DCF to cultures increased the effect of radiotherapy in the COX-2 cell line, significantly decreasing the survival fraction at a 2 Gy dose from 78.6% to 35.5% ( P = 0.0225). This effect was confirmed in a xenograft model, with male BALB/c nu/nu mice inoculated with the COX-2 over-expressing cell line and treated when tumours reached 0.5 cm in diameter. In addition to untreated controls, three treatment groups were used, topical DCF alone, radiotherapy alone and topical DCF + radiotherapy (at a dose of 3 Gy). On day 36 following treatment, the mean tumour volume for the DCF group was 32% of the control group, 44% for radiotherapy group and 15% for the combination group (all P < 0.05).

Human data

In contrast to the wide range of in vitro and in vivo results, there is a relative paucity of clinical data with respect to the use of DCF as an anticancer agent rather than as an analgesic. While there has been much clinical interest in the use of NSAIDs in cancer therapy, much of this has been focused on selective COX-2 inhibitors such as celecoxib. However, data is not completely lacking and is outlined below.

Forget and colleagues reported on a retrospective analysis of breast cancer patients treated with conservative surgery, with and without intraoperative NSAIDs (DCF or ketorolac) [ 31 ]. Patients treated pre-incisionally with ketorolac (20 mg -30 mg) or DCF (75 mg) showed improved DFS (HR = 0.57, 95% confidence interval CI: 0.37–0.89, P = 0.01) and an improved OS (HR = 0.35, CI: 0.17–0.70, P = 0.03), compared to patients not treated with NSAIDs. The proportion of NSAID-treated patients who received DCF was 29% (147/510), which was insufficient for an adequately powered analysis of DCF, therefore the data for DCF and ketorolac was pooled in the study. Subsequently a Phase III prospective randomised trial using ketorolac has been instituted ( NCT01806259 ) to investigate the impact on distant relapse and overall survival of patients undergoing breast cancer surgery [ 32 ]. Accrual for this trial completed in August 2015 and primary end-point analysis (recurrence-free survival) is due in September 2017 (personal communication, Patrice Forget).

A similar retrospective study, by the same authors, looked at cohorts of patients for cancers of the breast, lung and kidney who had undergone surgical resection [ 33 ]. Of note results in non-small cell lung cancer (NSCLC) showed a statistically significant impact of pre-operative DCF on the risk of distant metastases (HR = 0.14 CI = 0.02–0.99, P = 0.05) and a tendency to improved mortality risk (HR = 0.61, CI: 0.35–1.06, P = 0.08) compared to no NSAID use.

As mentioned previously, TL-118 is a four-drug combination treatment that includes DCF. It is produced by Tiltan Pharma Ltd, Israel. A report was published outlining the use of standard of care chemotherapy (gemcitabine) and long-term use of TL-118 in a case of inoperable pancreatic adenocarcinoma, (not biopsy-confirmed) [ 34 ]. Initial treatment was associated with a sustained reduction in the CA 19 -9 tumour marker and a radiologically confirmed near-complete remission. TL-118 treatment was suspended due to toxicity, (weakness and vomiting), followed by a later suspension of gemcitabine treatment, during which time serum CA 19-9 increased. Gemcitabine was re-introduced but CA 19-9 continued to rise. Re-introduction of TL-118 caused a sharp reduction of CA 19-9 again. The patient was still under combined treatment of gemcitabine and TL-118 and showing progression-free response 16 months post-diagnosis. By way of comparison the authors quote figures of 6 and 9 months as averages for progression-free survival and overall survival for patients treated with first-line chemotherapy for pancreatic cancer.

Desmoid tumours, also known as aggressive fibromatoses, are rare non-metastasising tumours that arise from fibroblastic cells. While they do not metastasise they are locally invasive and are treated with surgical resection where possible, however, recurrence is a frequent event. Lackner and colleagues reported on two cases of unresectable disease in paediatric patients who were treated with tamoxifen (at a dose of 1 mg/kg, orally, BID) and DCF (at a dose of 2 mg/kg, rectally, twice a day) [ 35 ]. Both patients exhibited long term (four years and two years) disease control. No toxicity was reported. The same authors subsequently reported positively on four additional patients in an 11-year observational study [ 36 ]. An additional case report from Teshima et al outlined a case of aggressive recurrent disease treated with DCF at an oral dose of 50 mg BID for a period of two years [ 37 ]. Long term treatment was without apparent toxicity and caused considerable reduction in tumour size and symptoms. Furthermore the positive effects were sustained even after cessation of treatment.

Inflammatory myofibroblastic tumour (IMT) is another rare soft tissue tumour that can be locally invasive and which has features of both benign and malignant disease. Standard treatment is surgical resection, but recurrent or non-resectable disease is common. Tao and colleagues reported on a rare case of retroperitoneal IMT which was not amenable to complete resection and was therefore successfully treated with a combination of methotrexate, cisplatin and DCF (dose unspecified) following tumour debulking [ 38 ].

Clinical trials

As of 21st September 2015 there are four clinical trials on-going. Note that these trials are specifically looking for an anticancer effect from DCF or drug combinations including DCF. Trials for non-cancer indications or trials in cancer in which DCF is used for analgesia are not included.

NCT01935531 – This is a single-arm open label trial of topical DCF gel (3% DCF in 2.5% hyaluronic acid) in patients with actinic keratosis with the aim of assessing the impact of DCF on lactate production and tumour metabolism. Biopsy samples will be compared before and after three months of treatment. The primary outcome will be the level of lactate before and after treatment. Secondary outcomes include lactate levels in healthy skin in subset of patients, measures of metabolic change (e.g. glycolytic proteins, glucose levels etc).

There are also a number of clinical trials involving DCF as a component of TL-118, mentioned previously. The oral treatment is designed to be taken on six days out of seven: sulfasalazine is included on each treatment day, DCF and cyclophosphamide on days one and four and cimetidine on days two, three, five and six [ 16 ]. NCT00684970 is a multi-centre Phase IIB trial for metastatic castration resistant prostate cancer. The primary end point is progression free survival from 24 weeks after commencement of treatment up to 3 years. Secondary end points include overall survival, time to PSA progression, PSA response and pain response in evaluable patients. NCT01509911 is an international multi-centre trial in metastatic pancreatic cancer for patients starting gemcitabine treatment. The primary outcome is the disease control rate after 16 weeks of treatment. NCT01659502 is a single centre study in pancreatic cancer. The primary outcome is clinical benefit measurement (a composite score based on pain, performance status and weight) in a two-year time frame.

Mechanism of action

There are multiple mechanisms of action posited to explain the diverse anticancer effects of DCF. Many of these are common to other NSAIDs, particularly for COX-2 inhibitors such as celecoxib and similar drugs. Of particular significance in this respect is the role of the prostaglandins, especially PGE 2 . PGE 2 is formed from the breakdown of arachidonic acid to prostaglandin H 2 by COX-1 and COX-2 followed by further processing by microsomal prostaglandin synthase 1 (mPGES-1). Elevated levels of mPGES-1 and PGE 2 are found in a range of different cancer types and are associated with the chronic inflammation that is associated with a pro-tumour microenvironment [ 39 , 40 ]. DCF, in common with other inhibitors of the COX enzymes also acts to reduce PGE 2 synthesis and therefore many of the anticancer effects of DCF are associated, directly or indirectly, with reductions in PGE 2 levels. However, there is considerable variation in COX-1/COX-2 selectivity between different NSAIDs [ 2 , 41 ], and some evidence that DCF binds to COX-2 via a different mechanism to other commonly used drugs [ 42 ], therefore in the discussion that follows DCF-specific evidence is referenced where available but in some cases reference is made to generic COX-2/PGE 2 mechanisms.

Relevant mechanisms of action include: Anti-angiogenic Immunomodulation Pro-apoptotic Platelet function Actions on Myc and glucose metabolism Treatment Sensitivity

Angiogenesis

Inhibition of tumour neo-angiogenesis was one of the earliest anticancer mechanisms identified for DCF. Early work by Peterson and colleagues, in 1983, established that administration of DCF to animal models of cancer (fibrosarcoma and hepatoma) diminished the growth rate and degree of vascularisation of tumours, as did indomethacin or aspirin [ 9 ]. Earlier work had already shown that indomethacin and aspirin had growth inhibitory effects in cancer, although initially it was hypothesised that this was primarily due to reversal of the immunosuppressive effects of PGE 2 [ 43 ]. A purely immune-related explanation for the growth inhibition of indomethacin and aspirin was discounted experimentally and a number of alternative mechanisms investigated [ 44 ].

Both topical and oral administration of DCF was found to retard implanted colon-26 tumour growth in BALB/c mice with a corresponding decrease in tumour angiogenesis which was associated with a reduction of PGE 2 synthesis [ 14 ]. One mechanistic explanation is that PGE 2 upregulates the production of VEGF via the prostanoid E receptors (EP1 – EP4) [ 45 , 46 ]. Indeed genetic deletion of mPGES-1 in MMTV/NDL mice, which are genetically predisposed to the development of HER2/neu breast cancer, showed reduced incidence of mammary tumours, reduced levels of PGE 2 and VEGF-A expression, and lower levels of angiogenesis (measured as microvessel density) [ 47 ]. Similarly, analysis of high-risk neuroblastoma subsets by Larsson et al , showed that high mPGES-1 expression correlated with poor patient survival and that treatment with DCF down-regulated PGE 2 and that this correlated with reduced tumour growth volumes in an in vivo murine model [ 19 ].

The effect of DCF on VEGF expression has also been directly assessed in a number of tumour types. For example, after analysing COX-1, COX-2, VEGF-A and VEGF-C expression in a large panel of oesophageal carcinoma tumour samples ( n = 123), von Rahden et al assessed the effect of three COX inhibitors, including DCF, in three oesophageal carcinoma cell lines (OSC-1, OSC-2 and OE-33) [ 48 ]. Treatment with DCF, at a concentration of 10 μM, significantly reduced expression of VEGF-A after 6 hours exposure in the OSC-1 and OSC-2 cell lines, and reduced expression of VEGF-C after 6 and 12 hours exposure respectively in the OSC-1 and OE-33 cell lines.

Similarly, Mayorek and colleagues compared VEGF levels in mice bearing orthotopic syngeneic pancreatic tumours treated with DCF and untreated controls [ 24 ]. Mice treated with DCF at a dose of 30 mg/kg/day, administered in drinking water starting 3 days after inoculation with PANC02 cells, developed tumours 60% lower in weight than in untreated controls. Tumours from DCF-treated animals also showed significantly lower levels of VEGF expression and lower levels of VEGF in peritoneal fluid, however plasma VEGF levels showed no difference. Ex-vivo analysis using rat aortic rings treated with DCF at a concentration of 10 μM showed that sprouting area was inhibited 2.5 fold compared to untreated controls.

In addition to VEGF, DCF may affect other angiogenic pathways. Kaur and Sanyal investigated the role of DCF in a chemically-induced murine colorectal cancer model [ 49 ]. In addition to down-regulation of VEGF, they investigated the role of two chemokines, monocyte chemoattractant protein (MCP-1) and macrophage inflammatory protein (MIP-1α), in angiogenesis. Treatment of animals with a known carcinogenic agent (1, 2-Dimethylhydrazine) with and without co-administration of DCF, at an oral dose of 8 mg/kg/day, showed that DCF reduced VEGF expression. DCF also decreased MCP-1 expression, another known marker of angiogenesis. However, DCF increased the expression of MIP-1α (aka CCL3) which is involved in both inflammatory and angiogenic processes.

The pro-angiogenic factors VEGF and basic fibroblast growth factor (bFGF) enhance the expression of the CXCR4 chemokine receptor on endothelial cells rendering them more responsive to CXCL12 (also known as stromal-derived factor 1α) signalling, associated with both increased angiogenesis and the metastatic cascade [ 50 ]. This process can be upregulated by PGE 2 , and it has been shown that reducing its expression using the COX-2 inhibitors piroxicam and NS398, it is possible to reduce CXCR4 expression and subsequent level of angiogenesis by 50–60% in an in vivo matrigel plug experiment [ 51 ].

Furthermore, Colleselli et al showed that COX-2 had an effect on endothelial progenitor cells (EPC), which are mobilised from the bone marrow and are involved in tumour angiogenesis [ 52 ]. Of the two co-enzymes COX-2 inhibition was associated with a greater reduction in EPC proliferation and an increase in the rate of apoptosis. DCF, at a concentration of 10 μM, produced a statistically significant decrease in EPC numbers, a result in line with that for celecoxib but not for the COX-1 inhibitor acetylsalicylic acid. In terms of apoptosis, DCF at a concentration of 50 μM and celecoxib at a concentration of 25 μM produced statistically significant increases in apoptosis compared to controls.

Immunomodulation

In addition to having pro-angiogenic effects, it is known that tumour-associated PGE 2 has negative effects on anti-tumour immunity [ 39 , 53 ]. In addition to in vitro and in vivo evidence of immunosuppressive effects, there is also evidence from patient samples which correlate COX-2/PGE 2 expression with immunosuppression in a number of different cancers [ 54 – 56 ]. The effects of prostaglandin inhibition was also investigated in vitro in lymphocyte subsets derived from breast cancer patients [ 57 , 58 ]. Analysis had shown that irradiation was associated with immunosuppression via reduction in mitogen response in lymphocyte populations in response to increased prostaglandin synthesis. In vitro treatment with a range of prostaglandin inhibitors reversed this effect and could enhance mitogen responses, with DCF having the most potent effect.

While tumour-associated immunosuppression is a complex and multi-factorial process, a number of immune cell sub-populations are particularly implicated, including myeloid derived suppressor cells (MDSCs) and regulatory T cells (T-reg).

PGE 2 has been shown to induce the differentiation of bone marrow stem cells into MDSCs in a number of animal models of cancer. For example Sinha et al showed that BALB/c mice carrying 4T1 mammary carcinomas had delayed tumour growth and reduced MDSC populations when the EP2 PGE 2 receptor was knocked out, compared with wild-type mice [ 59 , 60 ]. Treatment of wild-type mice carrying 4T1 tumours with a COX-2 inhibitor (SC58236) also reduced tumour growth rates and reduced the accumulation of MDSC cells. Similarly, Fujita and colleagues showed that in a mouse model of glioma COX-2 blockade using aspirin or celecoxib inhibited PGE 2 production and delayed tumour progression [ 61 ]. This was associated with reduced accumulation of granulocytic MDSCs and an increased presence of cytotoxic T lymphocytes (CTLs). Of note it was also reported that treatment with aspirin was only effective if administered prior to tumour development, whereas the selective COX-2 inhibitor celecoxib was effective when administered starting 21-days post-tumour implantation. The celecoxib dose used was 30 mg/kg/d, which the authors reported as equivalent to a human dose of 400 mg/day, which is a clinically relevant dose. Finally, Veltman et al used a murine model of mesothelioma and showed that treatment with dietary celecoxib reduced the local and systemic expansion of MDSC sub-populations, and that this correlated with a reduction in immune suppression [ 62 ].

In addition to animal models of cancer-related MDSCs there is some data from ex-vivo patient samples presented by Mao et al [ 63 ]. Different populations of mononuclear cells were isolated from melanoma patients and tested for immunosuppressive activity. CD14 + HLADR low/- cells significantly inhibited the cytolytic activity and IFNγ production of autologous non-activated natural killer (NK) cells, with the effect due to release of TGFβ. PGE 2 enhanced the production of TGFβ by monocytic cells. Using a mouse model the authors showed that abrogation of tumour COX-2 expression reversed immunosuppression and increased the lytic activity of NK cells.

While there is no direct evidence for a DCF-specific action on MDSCs, there is little doubt that it is a potent inhibitor of PGE 2 [ 64 , 65 ]. Similarly, as we have seen above there is evidence that a range of selective and non-selective COX-2/PGE 2 inhibitors can reduce MDSC populations, and that therefore we would expect similar activity from DCF.

Certain regulatory T cell populations are also known to be associated with tumour-associated immunosuppression, particularly CD4 + CD25 + FOXP3 + cells [ 66 – 68 ]. Of particular interest is the role of COX-2/PGE 2 in the increase in T-reg cell numbers and immunosuppressive phenotype, which has been established in a number of tumour models [ 53 , 56 , 69 , 70 ]. Reduction of tumour-induced PGE 2 using both selective and non-selective COX inhibitors has been shown to reduce T-reg populations and activity [ 71 – 74 ]. In terms of DCF-specific evidence, Chirasani and colleagues showed both in vitro and in vivo that DCF was able to reduce the intra-tumoural accumulation and activation of T-regs in a murine glioblastoma model [ 25 ]. DCF, at a concentration of 1.5 μM, was also used in vitro to reduce the suppressive activity of T-reg of head and neck squamous cell carcinoma cell lines [ 70 ].

PGE 2 is also implicated in tumour-associated immunosuppression via inhibition of antigen presenting cells (APC)/dendritic cell (DC) induction and maturation. In addition to evidence from primary tumour cultures [ 75 ], there is also evidence that some stromal cell populations also secrete PGE 2 and are involved in inhibition of DC maturation [ 76 ]. Eruslanov et al showed in an in vitro study that PGE 2 skewed the differentiation of Th1 APCs towards MDSCs or tolerogenic M2-polarised macrophages [ 77 ]. It was further shown that co-culture with the COX-2 inhibitor LM-1685 partially restored expression of CD11c, a DC marker.

However, there is also some evidence that PGE 2 is required for activated DC migration to lymphoid tissues [ 78 , 79 ]. Yen and colleagues used an in vivo model to show that DCs matured within inflammatory sites require both CCR7 and PGE 2 -induced MMP-9 for their directional migration to draining lymph nodes [ 79 ]. There are, therefore, both immunostimulatory and immunosuppressive roles for PGE 2 in the full life-cycle of DCs [ 80 ].

Analysis by Trabanelli et al suggests that the disappointing clinical responses of DC vaccines in oncology are due to the induction of tolerogenic responses mediated by PGE 2 , and that the positive effects of PGE 2 on DCs are mitigated by the upregulation of the immunosuppressive enzyme indoleamine 2, 3-dioxygenase-1 (IDO1) [ 81 ]. The clinical implications of these diverse results are discussed in the Our Take section of this paper.

Apoptosis

In addition to modulation of angiogenesis and immune suppression, there is some evidence for a pro-apoptotic mechanism of action for DCF in cancer. One of the earlier reports outlining an anticancer effect of topical DCF with hyaluronan showed that incubation of colon-26 cells with DCF at concentrations between 30 – 300 μM induced a significant increase in apoptosis [ 14 ]. Some indication that this effect was independent of COX-2/PGE 2 inhibition was provided by Kusuhara and colleagues who showed that apoptosis induction by DCF in cultured rat gastric mucosa cells was associated with caspase-dependent DNA fragmentation [ 82 , 83 ]. This finding was in line with similar contemporary reports using other NSAIDs, including sulindac [ 84 ], aspirin [ 85 ] and indomethacin [ 86 ]. In contrast, Ashton showed no such increase in apoptosis in guinea-pig gastric mucosal cells exposed to 250 μM for 24 hours [ 87 ].

Gardner and colleagues further elucidated the mechanism of action with respect to human colorectal cancer in a number of NSAIDs, including DCF [ 88 ]. Using the SW480 human colorectal cancer cell line, which does not express COX-2, they investigated the effects of indomethacin, sulindac sulphide, sulindac sulphone, rofecoxib and DCF on proliferation, apoptosis, β-catenin and cyclin-D1 in vitro . DCF, at a concentration of 200 μM, induced a statistically significant anti-proliferative effect but this was associated with a decrease in apoptosis. DCF was also associated with a decrease in β-catenin protein levels and cyclin D. Overall, DCF was shown to have moderate anti-proliferative and weak pro-apoptotic activity in this cell line. Additional in vitro confirmation of an inhibitory effect on β-catenin came from the work of Lu et al who confirmed that a panel of NSAIDs, including DCF, repressed β-catenin via high-level expression of peroxisome proliferator-activated receptor-γ (PPAR-γ) [ 89 ].

Inoue et al investigated the apoptotic activity of DCF in the HL-60 human promyelocytic leukaemia cell line [ 90 ]. DCF, at concentrations above 100 μM, induced DNA fragmentation and apoptosis, trigged the caspase cascade and release of cytochrome c. This was associated with an increase in intracellular reactive oxygen species (ROS), with a downstream inhibition in Akt phosphorylation via a PI3 kinase (PI3K) pathway. Johnsen et al also reported a pro-apoptotic effect of DCF in neuroblastoma, with in vitro analysis revealing evidence of DNA fragmentation and a caspase-dependent pathway [ 17 ]. Inhibition of Akt/PI3K signalling was also apparent in colorectal cancer in work by Rana et al [ 91 ]. Similarly, Albano et al reported that apoptosis in the human melanoma cell lines A2058 and SAN was associated with an increase in intracellular ROS and increase of caspase-9 and -3, reduction of Bcl-2/Bax ratio and mitochondrial release of cytochrome c [ 92 ].

Singh et al also reported on apoptosis in the leukemic cell lines HL-60 and THP-1, and in 43 samples from acute myeloid leukaemia patients [ 93 ]. Induction of apoptosis was via the activation of several AP-1 family transcription factors, (such as c-Jun, JunB and Fra-2), and subsequent induction of GADD45α which in turn activates JNK to trigger apoptosis.

Braun et al investigated the pro-apoptotic effects of DCF, acetylsalicylic acid (ASA) and sodium salicylate (NaS) on cutaneous T-cell lymphoma cell lines (CTCL) [ 94 ]. CTCL is a heterogeneous group of non-Hodgkin lymphomas that includes mycosis fungoides and Sézary syndrome, and CTCL cells are known to be resistant to apoptosis triggered via death receptors [ 95 ]. In vivo treatment with NSAIDs, including DCF at a concentration 200 μM, restored sensitivity to tumour necrosis factor-related apoptosis-inducing ligand (TRAIL)-induced apoptosis. In addition to in vitro analysis based on established CTCL cell lines, ex vivo analysis of T-cells from four Sézary syndrome patients showed enhanced apoptotic response in three of them compared to healthy controls. In addition to the work of Braun et al , other workers have also investigated the use of COX-2 inhibition in CTCL, for example in vivo work using celecoxib in a mouse model of mycosis fungoides [ 96 ].

Another COX-2/PGE 2 -independent pro-apoptotic pathway is via increased expression of non-steroidal anti-inflammatory drug-activated gene 1 (NAG-1), also known as macrophage inhibitory cytokine-1. NAG-1 is a member of the transforming growth factor-beta (TGF-β) superfamily with evidence of both pro- and anti-cancer activity, possibly related to stage of disease [ 97 ]. Over-expression of NAG-1 has been related to induction of apoptosis in a range of cancer types, possibly as a down-stream target of p53 signalling [ 97 – 100 ]. Kim et al tested the relative effect of different NSAIDs on the induction of NAG-1 in an oral squamous cell carcinoma line (SCC 1483) [ 101 ]. DCF was the most potent of the panel of NSAIDs used, increasing NAG-1 expression five-fold at a concentration of 100 μM. NAG-1 expression increased prior to the induction of apoptosis, with which it was highly correlated.

In a PANCO2 pancreatic cancer model, Mayorek and colleagues showed that DCF treatment at a dose of 30 mg/kg caused a 60% reduction in tumour weight compared to untreated controls, and that the reduction in tumour weight was caused by an increased rate of apoptosis [ 24 ]. The effect was not evident in vitro and further analysis showed both a reduced rate of angiogenesis and an increased level of arginase activity in tumour stroma and peritoneal macrophages. The increased level of arginase activity was associated with a reduction in nitric oxide (NO) and arginine depletion in the peritoneal cavity and serum, although it was not shown how this could enhance the apoptotic effect of DCF treatment.

Apoptosis due to NSAID treatment, including DCF, was also shown in a panel of ovarian cancer cell lines by Zerbini and colleagues [ 22 ]. The induction of apoptosis was mediated by the pro-apoptotic cytokine melanoma differentiation associated gene-7/Interleukin-24 (mda-7/IL-24), which mediates the induction of GADD45α expression and activation of the JNK pathway.

Platelet function

There is increasing interest in the role of platelets in cancer, with emerging evidence of a role in tumour progression and metastasis. A number of mechanisms are known to be active in the pro-cancer role of platelets including the release of pro-angiogenic factors, ‘cloaking’ of tumour cells from NK cells and a role in preparing metastatic niches [ 102 ]. Cancer-associated thrombocytosis is a common clinical occurrence and is associated with poorer outcomes in a number of cancers [ 103 – 105 ]. Of note there is also some evidence that the putative anti-cancer effects of aspirin may be related to its anti-platelet effects via irreversible inhibition of COX-1 [ 106 , 107 ]. Similarly there is some evidence that other anti-thrombotic therapies, for example low molecular weight heparins may also have anti-cancer or anti-metastatic activity [ 108 , 109 ].

As a non-selective COX inhibitor, albeit with a preference for COX-2, DCF also has clinically relevant actions on platelet function via COX-1 inhibition. Van Hecken et al studied the effects of a panel of NSAIDS in healthy volunteers and found that at steady state dosing of DCF at 50 mg three times a day for six days ex vivo COX-1 level (expressed as thromboxane B2 generation in clotting whole blood) was reduced by 53% compared to base-line, and platelet aggregation was also significantly reduced ( P < 0.001) [ 64 ]. This effect on platelet function has also been confirmed in a clinical setting, for example Bajaj et al reported a 64% reduction in platelet aggregation in patients treated with a single 75 mg dose of DCF administered pre-operatively [ 110 ].

We may speculate, therefore, that DCF may also exert anti-angiogenic, immunomodulatory and other anti-cancer effects via inhibition of platelet function in addition to COX-2/PGE 2 mediated actions.

Myc and glucose metabolism

There is also some evidence that DCF has an impact on tumour metabolism that is independent of its action as a COX-inhibitor. Gottfried and colleagues showed that DCF down-regulated Myc gene expression and glucose metabolism in a number of leukaemia, prostate cancer and melanoma cell lines in vitro and in an in vivo melanoma model [ 29 ]. Of note neither aspirin nor the COX-2 inhibitor NS-398 had an effect on Myc expression or glucose metabolism. Furthermore, DCF inhibited lactate efflux, causing an increase in cellular lactate levels which was independent of the effect on Myc gene expression. The increase in cellular lactate, leading to a decreasing proliferation rate, was also matched by a decrease in extra-cellular lactate. Similarly, COX-independent effects on lactate were reported in glioblastoma, both in vivo [ 25 ] and in human glioblastoma cell lines [ 26 ].

The effects on glucose metabolism may be related to impacts on glycolytic pathways via STAT3 inhibition [ 26 ], although there is some evidence that the effect may also be mediated via hypoxia-related down-regulation of glucose-transporter 1 (GLUT1) [ 111 ].

Further evidence for an effect on Myc expression was also provided by Sareddy and colleagues, who showed that DCF and celecoxib caused a reduced expression of Wnt/β-catenin/Tcf signalling in a two glioblastoma cell lines (U87 and U251) [ 112 ]. In vitro treatment showed that both drugs, (DCF concentration in the range 50–200 μM, celecoxib in the range 20–80 μM) significantly reduced expression of down-stream targets of β-catenin signalling, including c-Myc and cyclin D1. These effects were associated with reduced glioblastoma cell proliferation, colony formation and invasion.

Treatment sensitivity

There is some evidence that COX-2 expression may correlate with sensitivity to chemotherapy or radiotherapy in different cancer types. For example, analysis of 104 cases of primary invasive breast cancer indicated that increased expression of COX-2 correlated, ( P < 0.0001), with increased expression of multi-drug resistance gene (MDR1) and P-glycoprotein (P-gp), both mechanistically implicated in resistance to chemotherapy [ 113 ]. Furthermore expression of MDR1/P-gp had prognostic significance in terms of both PFS and OS in this patient population, (both P < 0.0001). In an analysis of advanced ovarian cancer cases, it was also found that COX-2 correlated to treatment resistance, ( P = 0.0072), although this varied by chemotherapy drug, with resistance to platinum-based drugs more susceptible to COX-2 expression than treatment with paclitaxel.

Based on these and other results there has been an interest in the use of COX-2 inhibitors to potentiate sensitivity to chemotherapy [ 114 , 115 ]. There has been limited clinical investigation of this strategy to date. A Phase II trial in heavily pre-treated ovarian cancer patients with recurrent disease, ( n = 45, of whom 23 were platinum-resistant), showed that the combination of carboplatin and low-dose celecoxib (400 mg/day) had a response rate of 28.9%, including three complete regressions. Median PFS was 5 months overall, but among responders the PFS was 8 months and OS 17 months (statistical significance not shown) [ 116 ].

In chronic myeloid leukaemia (CML), the standard treatment for the chronic phase of the disease is with the targeted agent imatinib mesylate. Transport of the drug into CML cells is via the human organic cation transporter-1 (OCT-1) and low OCT-1 functional activity is associated with treatment resistance and poor patient outcomes [ 117 ]. In an investigation of drug-drug interactions between imatinib and a panel of common NSAIDs, DCF and ibuprofen were shown to have significant interactions [ 118 ]. Specifically, DCF at a clinically relevant concentration of 10 μM was shown to increase OCT-1 activity and to statistically significantly reduce the IC50 of imatinib in two CML cell lines (K562 and KU812). In contrast ibuprofen, at a clinically relevant concentration of 130 μM decreased OCT-1 activity and increased the IC50 value of imatinib. The use of DCF was also tested ex vivo using mononuclear cells (MNC) from newly diagnosed CML patients. DCF treatment in MNC samples with low baseline OCT-1 activity increased OCT-1 activity and reduced the IC50 of imatinib, clearly suggesting that DCF may be useful to sensitise patients at highest risk of treatment failure due to drug resistance.

Another mechanism of chemoresistance involves the repopulation of tumour masses by the accelerated proliferative response of cancer stem cells after chemotherapy, as has been shown by Kurtova et al in bladder cancer [ 119 ]. In vitro work showed that this process was driven by PGE 2 and that it could be reversed by PGE 2 inhibition. In vivo xenograft models of bladder carcinoma (T24 and a patient derived chemoresistant line), showed that with combined treatment of gemcitabine and celecoxib resistance to treatment did not occur.

Wasserman and colleagues investigated the long-term outcomes for women treated with adjuvant radiotherapy in the treatment of primary breast cancer [ 57 ]. They showed that local radiotherapy for breast cancer (45 Gy) caused a severe lymphopenia with reductions of both T- and non-T-lymphocyte counts and reactivity and that recovery of T-cells was still impacted 10–11 years after treatment. Furthermore mortality was greater in women with impaired T-cell function up to eight years after treatment. However, in vitro treatment with DCF showed improved reactivity of lymphocytes from radiation-treated women three months after the completion of their treatment.

Crokart et al investigated the effect of NSAID administration on oxygen pressure in two syngeneic murine cancer models (TLT liver tumours and FSaII fibrosarcomas) [ 120 ]. DCF, at a dose of 20 mg/kg by weight, showed an increase in tumour oxygen pressure around 30 minutes after administration, results which were similar to those for piroxicam and indomethacin. Further analysis using the COX-2 inhibitor NS-398 showed that this increase in oxygen pressure was associated with tumour regrowth delay in mice treated with radiotherapy.

Our take

The pre-clinical and clinical data, summarised in Table 1 , indicate that DCF has a number of distinct anti-cancer effects, summarised in Figure 1 , which warrant further investigation in a clinical setting. These effects are mediated by both COX-dependent and independent mechanisms of action, suggesting that at least some of the effect is specific to DCF rather than being associated generically with other NSAIDs such as celecoxib, indomethacin and aspirin. DCF benefits from potent down-regulation of both COX-1 and COX-2, addressing both platelet- and PGE 2 -mediated pro-cancer pathways and also displays a range of COX-independent pathways. These multiple mechanisms of action make DCF one of the more interesting NSAIDs in the context of cancer treatment.

The combination of anti-angiogenic activity with positive effects on immunity is especially interesting in the context of surgical intervention in cancer. There is increasing evidence that the ‘wound healing’ response initiated by surgical intervention against tumours is implicated in distant metastatic relapse. Evidence for this effect comes both from retrospective analyses of patient outcomes and from in vivo models [ 121 – 123 ]. There are multiple mechanisms posited to be at work, many of them focused on the post-surgical inflammatory cascade leading to an up-regulation of angiogenic signalling and a sustained immune suppression [ 124 – 126 ]. In response there has been a new focus on those peri-operative interventions which may have an impact on the post-surgical relapse rate by selective targeting of aspects of this wound healing response, particularly with respect to the choice of anaesthesia [ 127 – 130 ]. A number of drugs have been identified which may have a positive effect when used in the peri-operative or post-operative setting, including ketorolac [ 32 , 131 , 132 ], cimetidine [ 133 , 134 ], and DCF [ 31 , 33 ]. Therefore, further investigation of the peri-operative use of DCF is warranted in a number of cancers in which post-surgical distant metastases are a frequent occurrence, including osteosarcoma, oesophageal carcinoma, NSCLC, ovarian and breast cancer.

The complex role that PGE 2 plays in the complete life-cycle of DCs may mean that long-term use of DCF and COX-2 inhibitors may have some negative effects on anti-tumour immunity, although the picture remains unclear as to whether the effects are overall positive or negative. This suggests that caution may need to be exercised in the treatment schedule such that the negative effects are minimised and the positive maximised. Certainly the short-term use in peri-operative interventions may be positive precisely because of the timing of the treatment. Alternatively it is suggested by Trabanelli et al that blockade of IDO1 may also be a viable strategy to ameliorate the negative effects on DCs [ 81 ].

While there is some evidence that DCF has pro-apoptotic activity, much of this evidence is in vitro and uses relatively high drug concentrations. The current level of evidence does not support the use of DCF as an inducer of apoptosis and it is likely that the anticancer effects are primarily due to the other mechanisms of action.

We note also the strong pre-clinical evidence that DCF has an effect in neuroblastoma, a disease with a dismal prognosis for patients with refractory or metastatic disease. While there are a number of new targeted agents being investigated for this high-need population of patients, there are few clinical trials that have progressed to Phase III. The addition of DCF to either existing standard of care or with new targeted agents has the potential for clinical benefit and therefore warrants further investigation.

Additionally, given the potent effects that DCF has on PGE 2 expression, there is merit in investigating the addition of DCF to existing standard of care therapy in those cancers in which PGE 2 upregulation is associated, including breast, head and neck and colorectal cancers. For example, it is known that PGE 2 promotes colorectal cancer growth via an upregulation of β-catenin signalling [ 135 , 136 ], and that DCF can inhibit this in vivo [ 137 ].

Finally, while there have been a number of interesting and positive case reports of DCF activity against both desmoid and inflammatory myofibroblastic tumours, there have been no randomised clinical trials to confirm these results. Given the apparent low toxicity of DCF and the positive results that have been reported, investigation of DCF is clearly required, particularly as many of the agents currently being trialled, (examples include sorafenib, imatinib and crizotinib), have greater toxicity and costs associated with them.

Next steps

The evidence is strongest for clinical trials of DCF, in combination with other agents, in the following cancer types: Desmoid tumours (metronomic treatment) Inflammatory myofibroblastic tumours High-risk refractory or metastatic neuroblastoma

The peri-operative use of DCF is also of interest in the following cancers: Osteosarcoma Head and neck cancers Oesophageal cancer Breast cancer Ovarian cancer Non-small Cell Lung Cancer

Conclusion

Drawing on in vitro , in vivo and human data we have summarised the evidence for an anti-cancer effect of DCF treatment. The established pharmacokinetics and known toxicity profile make this generic drug a strong candidate for repurposing as an oncological treatment, both in combination with existing standard of care treatments or in a cocktail with other repurposed drugs. A number of possible multi-drug combinations are outlined in the supplementary materials for specific cancer indications.

Author contributions

Primary author: Pan Pantziarka. Contributing authors: Vidula Sukhatme, Gauthier Bouche, Lydie Meheus, Vikas P. Sukhatme. All authors read and approved the final manuscript.

Competing interests

The authors declare that they have no competing interests. All the authors are associated with not for profit organisations that aim to repurpose drugs for oncology treatments.

References

1. Joint_Formulary_Committee British national formulary 2015 69 BMJ Group and Pharmaceutical Press

2. Cryer B Feldman M Cyclooxygenase-1 and cyclooxygenase-2 selectivity of widely used nonsteroidal anti-inflammatory drugs Am J Med 1998 104 98 413 421 10.1016/S0002-9343(98)00091-6 9626023

3. Baigent C Bhala N Emberson J Vascular and upper gastrointestinal effects of non-steroidal anti-inflammatory drugs: meta-analyses of individual participant data from randomised trials The Lancet 2013 382 9894 769 779 10.1016/S0140-6736(13)60900-9 PMC3778977 23726390

4. Davies NM Anderson KE Clinical pharmacokinetics of diclofenac Therapeutic insights and pitfalls Clin Pharmacokinet 1997 33 3 184 213 10.2165/00003088-199733030-00003 9314611

5. Gan TJ Diclofenac: an update on its mechanism of action and safety profile Curr Med Res Opin 2010 26 7 1715 1731 10.1185/03007995.2010.486301 20470236

6. Su SF Chou CH Kung CF In vitro and in vivo comparison of two diclofenac sodium sustained release oral formulations Int J Pharm 2003 260 1 39 46 10.1016/S0378-5173(03)00237-0 12818808

7. Fischer SM Hawk ET Lubet RA Coxibs and other nonsteroidal anti-inflammatory drugs in animal models of cancer chemoprevention Cancer Prev Res 2011 4 11 1728 1735 10.1158/1940-6207.CAPR-11-0166 PMC3208782 21778329

8. Gurpinar E Grizzle WE Piazza GA COX-independent mechanisms of cancer chemoprevention by anti-inflammatory drugs Front Oncol 2013 3 181 10.3389/fonc.2013.00181 23875171 PMC3708159

9. Peterson HI Effects of prostaglandin synthesis inhibitors on tumor growth and vascularization Experimental studies in the rat Invas Metastas 1983 3 3 151 159 6203870

10. Peterson HI Alpsten M Skolnik G Karlsson L Influence of a prostaglandin synthesis inhibitor and of thrombocytopenia on tumor blood flow and tumor vascular permeability Experimental studies in the rat Anticancer Res 1985 5 3 253 257 4015039

11. Hofer M Hoferová Z Fedoročko P Hematopoiesis-stimulating and anti-tumor effects of repeated administration of diclofenac in mice with transplanted fibrosarcoma cells Physiol Res 2002 51 629 632 12511188

12. Hoferová Z Fedorocko P Hofmanová J The effect of nonsteroidal antiinflammatory drugs ibuprofen, flurbiprofen, and diclofenac on in vitro and in vivo growth of mouse fibrosarcoma Cancer Invest 2002 20 4 490 498 10.1081/CNV-120002149 12094544

13. Hixson LJ Alberts DS Krutzsch M Antiproliferative effect of nonsteroidal antiinflammatory drugs against human colon cancer cells Cancer Epidemiol Biomarkers Prev 1994 3 5 433 438 7920212

14. Seed MP Brown JR Freemantle CN The inhibition of colon-26 adenocarcinoma development and angiogenesis by topical diclofenac in 2.5% hyaluronan Cancer Res 1997 57 9 1625 1629 9134996

15. Falkowski M Skogstad S Shahzidi S The effect of cyclooxygenase inhibitor diclofenac on experimental murine colon carcinoma Anticancer Res 2003 23 3B 2303 2308 12894507

16. Edrei Y Gross E Corchia N Improved efficacy of a novel anti-angiogenic drug combination (TL-118) against colorectal-cancer liver metastases; MRI monitoring in mice Br J Cancer 2012 107 4 658 666 10.1038/bjc.2012.322 22805330 PMC3419965

17. Johnsen JI Lindskog M Ponthan F Cyclooxygenase-2 is expressed in neuroblastoma, and nonsteroidal anti-inflammatory drugs induce apoptosis and inhibit tumor growth in vivo Cancer Res 2004 64 20 7210 7215 10.1158/0008-5472.CAN-04-1795 15492235

18. Johnsen JI Lindskog M Ponthan F NSAIDs in neuroblastoma therapy Cancer Lett 2005 228 1-2 195 201 10.1016/j.canlet.2005.01.058 15975708

19. Larsson K Kock A Idborg H COX/mPGES-1/PGE 2 pathway depicts an inflammatory-dependent high-risk neuroblastoma subset Proc Natl Acad Sci USA 2015 112 26 8070 8075 10.1073/pnas.1424355112 26080408 PMC4491767

20. Komar-Stossel C Gross E Dery E TL-118 and gemcitabine drug combination display therapeutic efficacy in a MYCN amplified orthotopic neuroblastoma murine model – evaluation by MRI PLoS One 2014 9 3 e90224 10.1371/journal.pone.0090224 24603724 PMC3946152

21. Zerbini LF Czibere A Wang Y A novel pathway involving melanoma differentiation associated gene-7/interleukin-24 mediates nonsteroidal anti-inflammatory drug-induced apoptosis and growth arrest of cancer cells Cancer Res 2006 66 24 11922 11931 10.1158/0008-5472.CAN-06-2068 17178890

22. Zerbini LF Tamura RE Correa RG Combinatorial effect of non-steroidal anti-inflammatory drugs and NF-κB inhibitors in ovarian cancer therapy PLoS One 2011 6 9 e24285 10.1371/journal.pone.0024285 21931671 PMC3171406

23. Valle BL D’Souza T Becker KG Non-steroidal anti-inflammatory drugs decrease E2F1 expression and inhibit cell growth in ovarian cancer cells PLoS One 2013 8 4 e61836 10.1371/journal.pone.0061836 23637916 PMC3634839

24. Mayorek N Naftali-Shani N Grunewald M Diclofenac inhibits tumor growth in a murine model of pancreatic cancer by modulation of VEGF levels and arginase activity PLoS One 2010 5 9 1 10 10.1371/journal.pone.0012715 PMC2939880 20856806

25. Chirasani SR Leukel P Gottfried E Diclofenac inhibits lactate formation and efficiently counteracts local immune suppression in a murine glioma model Int J Cancer 2013 132 4 843 853 10.1002/ijc.27712 22752934

26. Leidgens V Seliger C Jachnik B Ibuprofen and diclofenac restrict migration and proliferation of human glioma cells by distinct molecular mechanisms PLoS One 2015 10 10 e0140613 10.1371/journal.pone.0140613 26485029 PMC4617646

27. Roller D Axelrod M Capaldo B Synthetic lethal screening with small molecule inhibitors provides a pathway to rational combination therapies for melanoma Mol Cancer Ther 2012 9 2505 2515 10.1158/1535-7163.MCT-12-0461 22962324 PMC3496043

28. Yagi K Kawasaki Y Nakamura H Differential combined effect of COX inhibitors on cell survival suppressed by sorafenib in the HepG2 cell line Biol Pharm Bull 2014 37 7 1234 1240 10.1248/bpb.b13-00963 24989014

29. Gottfried E Lang SA Renner K New aspects of an old drug—diclofenac targets MYC and glucose metabolism in tumor cells PLoS One 2013 8 7 e66987 10.1371/journal.pone.0066987 23874405 PMC3706586

30. Inoue T Anai S Onishi S Inhibition of COX-2 expression by topical diclofenac enhanced radiation sensitivity via enhancement of TRAIL in human prostate adenocarcinoma xenograft model BMC Urol 2013 13 1 10.1186/1471-2490-13-1 23289871 PMC3561196

31. Forget P Bentin C Machiels JP Intraoperative use of ketorolac or diclofenac is associated with improved disease-free survival and overall survival in conservative breast cancer surgery Brit J Anaesth 2014 113 Suppl 1 1 6 10.1093/bja/aet464 24464611

32. Forget P Berlière M van Maanen A Perioperative ketorolac in high risk breast cancer patients Rationale, feasibility and methodology of a prospective randomized placebo-controlled trial Med Hypotheses 2013 81 4 707 712 10.1016/j.mehy.2013.07.033 23937996

33. Forget P Machiels J-P Coulie PG Neutrophil:lymphocyte ratio and intraoperative use of ketorolac or diclofenac are prognostic factors in different cohorts of patients undergoing breast, lung, and kidney cancer surgery Ann Surg Oncol 2013 20 Suppl 3 S650 S660 10.1245/s10434-013-3136-x 23884751

34. Breuer S Maimon O Appelbaum L TL-118-anti-angiogenic treatment in pancreatic cancer: a case report Med Oncol 2013 30 2 585 10.1007/s12032-013-0585-9 23609193

35. Lackner H Urban C Kerbl R Noncytotoxic drug therapy in children with unresectable desmoid tumors Cancer 1997 80 2 334 340 10.1002/(SICI)1097-0142(19970715)80:2&lt;334::AID-CNCR22&gt;3.0.CO;2-U 9217047

36. Lackner H Urban C Benesch M Multimodal treatment of children with unresectable or recurrent desmoid tumors: an 11-year longitudinal observational study J Pediatr Hematol Oncol 2004 26 8 518 522 10.1097/01.mph.0000130219.26284.b3 15284591

37. Teshima M Iwae S Hirayama Y Nonsteroidal anti-inflammatory drug treatment for desmoid tumor recurrence after surgery Otolaryngol Head Neck Surg 2012 147 5 978 979 10.1177/0194599812453561 22760417

38. Tao YL Wang ZJ Han JG Inflammatory myofibroblastic tumor successfully treated with chemotherapy and nonsteroidals: a case report World J Gastroenterol 2012 18 47 7100 7103 10.3748/wjg.v18.i47.7100 23323014 PMC3531700

39. Nakanishi M Rosenberg DW Multifaceted roles of PGE 2 in inflammation and cancer Semin Immunopathol 2013 35 2 123 137 10.1007/s00281-012-0342-8 22996682 PMC3568185

40. Dubois RN Role of inflammation and inflammatory mediators in colorectal cancer Trans Am Clin Climatol Assoc 2014 125 358 373 25125751 PMC4112706

41. Giuliano F Warner TD Ex vivo assay to determine the cyclooxygenase selectivity of non-steroidal anti-inflammatory drugs Brit J Pharmacol 1999 126 1824 1830 10.1038/sj.bjp.0702518 10372826 PMC1565969

42. Rowlinson SW Kiefer JR Prusakiewicz JJ A novel mechanism of cyclooxygenase-2 inhibition involving interactions with Ser-530 and Tyr-385 J Biol Chem 2003 278 46 45763 45769 10.1074/jbc.M305481200 12925531

43. Plescia OJ Smith AH Grinwich K Subversion of immune system by tumor cells and role of prostaglandins Proc Natl Acad Sci USA 1975 72 5 1848 1851 10.1073/pnas.72.5.1848 1057174 PMC432644

44. Lynch NR Castes M Astoin M Mechanism of inhibition of tumour growth by aspirin and indomethacin Br J Cancer 1978 38 4 503 512 10.1038/bjc.1978.237 365211 PMC2009769

45. Ben-Av P Crofford LJ Wilder RL Induction of vascular endothelial growth factor expression in synovial fibroblasts by prostaglandin E and interleukin-1: a potential mechanism for inflammatory angiogenesis FEBS Lett 1995 372 1 83 87 10.1016/0014-5793(95)00956-A 7556649

46. Amano H Hayashi I Endo H Host prostaglandin E(2)-EP3 signaling regulates tumor-associated angiogenesis and tumor growth J Exp Med 2003 197 2 221 232 10.1084/jem.20021408 12538661 PMC2193807

47. Howe LR Subbaramaiah K Kent CV Genetic deletion of microsomal prostaglandin e synthase-1 suppresses mouse mammary tumor growth and angiogenesis Prostaglandins Other Lipid Mediat 2013 106 6 99 105 10.1016/j.prostaglandins.2013.04.002 23624019 PMC3830707

48. Von Rahden BHA Stein HJ Pühringer F Coexpression of cyclooxygenases (COX-1, COX-2) and vascular endothelial growth factors (VEGF-A, VEGF-C) in esophageal adenocarcinoma Cancer Res 2005 65 12 5038 5044 10.1158/0008-5472.CAN-04-1107 15958546

49. Kaur J Sanyal SN Diclofenac, a selective COX-2 inhibitor, inhibits DMH-induced colon tumorigenesis through suppression of MCP-1, MIP-1α and VEGF Mol Carcinog 2011 50 9 707 718 10.1002/mc.20736 21268133

50. Guo F Wang Y Liu J CXCL12/CXCR4: a symbiotic bridge linking cancer cells and their stromal neighbors in oncogenic communication networks Oncogene 2015 154 1 11 10.1038/onc.2015.139 25961926

51. Salcedo R Zhang X Young HA Angiogenic effects of prostaglandin E2 are mediated by up-regulation of CXCR4 on human microvascular endothelial cells Blood 2003 102 6 1966 1977 10.1182/blood-2002-11-3400 12791666

52. Colleselli D Bijuklic K Mosheimer BA Inhibition of cyclooxygenase (COX)-2 affects endothelial progenitor cell proliferation Exp Cell Res 2006 312 15 2933 2941 10.1016/j.yexcr.2006.05.021 16893539

53. Kalinski P Regulation of immune responses by prostaglandin E2 J immunol 2012 188 1 21 8 10.4049/jimmunol.1101029 22187483 PMC3249979

54. Obermajer N Muthuswamy R Odunsi K PGE 2-induced CXCL 12 production and CXCR4 expression controls the accumulation of human MDSCs in ovarian cancer environment Cancer Res 2011 71 24 7463 7470 10.1158/0008-5472.CAN-11-2449 22025564 PMC4993027

55. Johnson SD De Costa AMA Young MRI Effect of the premalignant and tumor microenvironment on immune cell cytokine production in head and neck cancer Cancer 2014 6 2 756 770 10.3390/cancers6020756 PMC4074802 24698959

56. Baratelli F Lee JM Hazra S PGE (2) contributes to TGF-beta induced T regulatory cell function in human non-small cell lung cancer Am J Transl Res 2010 2 4 356 367 20733946 PMC2923860

57. Wasserman J Blomgren H Rotstein S Immunosuppression in irradiated breast cancer patients: in vitro effect of cyclooxygenase inhibitors B New York Acad Med 1989 65 1 36 44 PMC1807786 2513994

58. Blomgren H Rotstein S Wasserman J In vitro capacity of various cyclooxygenase inhibitors to revert immune suppression caused by radiation-therapy for breast-cancer Radiother Oncol 1990 19 329 335 10.1016/0167-8140(90)90033-S 2126634

59. Sinha P Clements VK Fulton AM Prostaglandin E2 promotes tumor progression by inducing myeloid-derived suppressor cells Cancer Res 2007 67 9 4507 4513 10.1158/0008-5472.CAN-06-4174 17483367

60. Ostrand-Rosenberg S Sinha P Myeloid-derived suppressor cells: linking inflammation and cancer J Immunol 2009 182 8 4499 4506 10.4049/jimmunol.0802740 19342621 PMC2810498

61. Fujita M Kohanbash G Fellows-Mayle W COX-2 blockade suppresses gliomagenesis by inhibiting myeloid-derived suppressor cells Cancer Res 2011 71 7 2664 2674 10.1158/0008-5472.CAN-10-3055 21324923 PMC3075086

62. Veltman JD Lambers MEH van Nimwegen M COX-2 inhibition improves immunotherapy and is associated with decreased numbers of myeloid-derived suppressor cells in mesothelioma Celecoxib influences MDSC function BMC Cancer 2010 10 464 10.1186/1471-2407-10-464 20804550 PMC2939552

63. Mao Y Sarhan D Steven A Inhibition of tumor-derived prostaglandin-E2 blocks the induction of myeloid-derived suppressor cells and recovers natural killer cell activity Clin Cancer Res 2014 20 15 4096 4106 10.1158/1078-0432.CCR-14-0635 24907113

64. Van Hecken A Schwartz JI Depré M Comparative inhibitory activity of rofecoxib, meloxicam, diclofenac, ibuprofen, and naproxen on COX-2 versus COX-1 in healthy volunteers J Clin Pharmacol 2000 40 10 1109 20 11028250

65. Schwartz JI Dallob AL Larson PJ Comparative inhibitory activity of etoricoxib, celecoxib, and diclofenac on COX-2 versus COX-1 in healthy subjects J Clin Pharmacol 2008 48 6 745 754 10.1177/0091270008317590 18434566

66. Ha TY The role of regulatory T cells in cancer Immune Netw 2009 9 6 209 235 10.4110/in.2009.9.6.209 20157609 PMC2816955

67. Byrne WL Mills KHG Lederer JA Targeting regulatory T cells in cancer Cancer Res 2011 71 22 6915 6920 10.1158/0008-5472.CAN-11-1156 22068034 PMC4287207

68. Darrasse-Jèze G Podsypanina K How numbers, nature, and immune status of Foxp3+ regulatory T-cells shape the early immunological events in tumor development Front Immunol 2013 4 SEP 1 10 10.3389/fimmu.2013.00292 24133490 PMC3784046

69. Baratelli F Lin Y Zhu L Prostaglandin E2 induces FOXP3 gene expression and T regulatory cell function in human CD4+ T cells J Immunol 2005 175 3 1483 1490 10.4049/jimmunol.175.3.1483 16034085

70. Bergmann C Strauss L Zeidler R Expansion of human T regulatory type 1 cells in the microenvironment of cyclooxygenase 2 overexpressing head and neck squamous cell carcinoma Cancer Res 2007 67 18 8865 8873 10.1158/0008-5472.CAN-07-0767 17875728

71. Yaqub S Henjum K Mahic M Regulatory T cells in colorectal cancer patients suppress anti-tumor immune activity in a COX-2 dependent manner Cancer Immunol Immunother 2008 57 6 813 821 10.1007/s00262-007-0417-x 17962941 PMC11030670

72. Lee SY Choi HK Lee KJ The immune tolerance of cancer is mediated by IDO that is inhibited by COX-2 inhibitors through regulatory T cells J immunother 2009 32 1 22 28 10.1097/CJI.0b013e31818ac2f7 19307990

73. Lönnroth C Andersson M Arvidsson A Preoperative treatment with a non-steroidal anti-inflammatory drug (NSAID) increases tumor tissue infiltration of seemingly activated immune cells in colorectal cancer Cancer Immun 2008 8 5 18307280 PMC2935782

74. Ogawa F Amano H Eshima K Prostanoid induces premetastatic niche in regional lymph nodes J Clin invest 2014 124 11 4882 4894 10.1172/JCI73530 25271626 PMC4347225

75. Sombroek CC Stam AGM Masterson AJ Prostanoids play a major role in the primary tumor-induced inhibition of dendritic cell differentiation J Immunol 2002 168 9 4333 4343 10.4049/jimmunol.168.9.4333 11970975

76. Shiraishi H Yoshida H Saeki K Prostaglandin E2 is a major soluble factor produced by stromal cells for preventing inflammatory cytokine production from dendritic cells Int Immunol 2008 20 9 1219 1229 10.1093/intimm/dxn078 18640970

77. Eruslanov E Daurkin I Ortiz J Pivotal advance: tumor-mediated induction of myeloid-derived suppressor cells and M2-polarized macrophages by altering intracellular PGE 2 catabolism in myeloid cells J Leukoc Biol 2010 88 5 839 848 10.1189/jlb.1209821 20587738 PMC8454925

78. Krause P Bruckner M Uermösi C Prostaglandin e2 enhances T-cell proliferation by inducing the costimulatory molecules OX40L, CD70, and 4-1BBL on dendritic cells Blood 2009 113 11 2451 2460 10.1182/blood-2008-05-157123 19029446

79. Yen JH Khayrullina T Ganea D PGE 2 -induced metalloproteinase-9 is essential for dendritic cell migration Blood 2008 111 1 260 270 10.1182/blood-2007-05-090613 17925490 PMC2200811

80. De Keijzer S Meddens MBM Torensma R The multiple faces of prostaglandin E2 G-protein coupled receptor signaling during the dendritic cell life cycle Int J Mol Sci 2013 14 4 6542 6555 10.3390/ijms14046542 23528886 PMC3645653

81. Trabanelli S Lecciso M Salvestrini V PGE 2 -induced IDO1 inhibits the capacity of fully mature DCs to elicit an in vitro antileukemic immune response J Immunol Res 2015 2015 253191 10.1155/2015/253191 25815345 PMC4357138

82. Kusuhara H Matsuyuki H Matsuura M Induction of apoptotic DNA fragmentation by nonsteroidal anti-inflammatory drugs in cultured rat gastric mucosal cells Eur J Pharmacol 1998 360 2-3 273 280 10.1016/S0014-2999(98)00679-7 9851595

83. Kusuhara H Komatsu H Sumichika H Reactive oxygen species are involved in the apoptosis induced by nonsteroidal anti-inflammatory drugs in cultured gastric cells Eur J Pharmacol 1999 383 3 331 337 10.1016/S0014-2999(99)00599-3 10594327

84. Shiff SJ Qiao L Tsai LL Sulindac sulfide, an aspirin-like compound, inhibits proliferation, causes cell cycle quiescence, and induces apoptosis in HT-29 colon adenocarcinoma cells J Clin Invest 1995 96 1 491 503 10.1172/JCI118060 7615821 PMC185223

85. Elder DJE Hague A Hicks DJ Differential growth inhibition by the aspirin metabolite salicylate in human colorectal tumor cell lines: enhanced apoptosis in carcinoma and in vitro-transformed adenoma relative to adenoma cell lines Cancer Res 1996 56 10 2273 2276 8625297

86. Sawaoka H Kawano S Tsuji S Cyclooxygenase-2 inhibitors suppress the growth of gastric cancer xenografts via induction of apoptosis in nude mice Am J Physiol 1998 274 6 Pt 1 G1061 G1067 9696706 10.1152/ajpgi.1998.274.6.G1061

87. Ashton M Hanson PJ Disparate effects of non-steroidal anti-inflammatory drugs on apoptosis in guinea-pig gastric mucous cells: inhibition of basal apoptosis by diclofenac Brit J Pharmacol 2002 135 2 407 416 10.1038/sj.bjp.0704497 11815376 PMC1573157

88. Gardner SH Hawcroft G Hull MA Effect of nonsteroidal anti-inflammatory drugs on beta-catenin protein levels and catenin-related transcription in human colorectal cancer cells Brit J Cancer 2004 91 1 153 163 10.1038/sj.bjc.6601901 15188006 PMC2364748

89. Lu D Cottam HB Corr M Repression of beta-catenin function in malignant cells by nonsteroidal antiinflammatory drugs P Natl Acad Sci USA 2005 102 51 18567 18571 10.1073/pnas.0509316102 PMC1317972 16352713

90. Inoue A Muranaka S Fujita H Molecular mechanism of diclofenac-induced apoptosis of promyelocytic leukemia: dependency on reactive oxygen species, Akt, Bid, cytochrome c, and caspase pathway Free Radical Bio Med 2004 37 8 1290 1299 10.1016/j.freeradbiomed.2004.07.003 15451068

91. Rana C Piplani H Vaish V Downregulation of PI3-K/Akt/PTEN pathway and activation of mitochondrial intrinsic apoptosis by diclofenac and curcumin in colon cancer Mol Cell Biochem 2015 402 1–2 225 241 10.1007/s11010-015-2330-5 25644785

92. Albano F Arcucci A Granato G Markers of mitochondrial dysfunction during the diclofenac-induced apoptosis in melanoma cell lines Biochimie 2013 95 4 934 945 10.1016/j.biochi.2012.12.012 23274131

93. Singh R Cadeddu RP Fröbel J The non-steroidal anti-inflammatory drugs sulindac sulfide and diclofenac induce apoptosis and differentiation in human acute myeloid leukemia cells through an AP-1 dependent pathway Apoptosis 2011 16 9 889 901 10.1007/s10495-011-0624-y 21739277

94. Braun FK Al-Yacoub N Plötz M Nonsteroidal anti-inflammatory drugs induce apoptosis in cutaneous T-cell lymphoma cells and enhance their sensitivity for TNF-related apoptosis-inducing ligand J Investig Dermatol 2012 132 2 429 439 10.1038/jid.2011.316 22011910

95. Braun FK Fecker LF Schwarz C Blockade of death receptor-mediated pathways early in the signaling cascade coincides with distinct apoptosis resistance in cutaneous T-cell lymphoma cells J Investig Dermatol 2007 127 10 2425 2437 10.1038/sj.jid.5700868 17495957

96. Kopp KLM Dabelsteen S Krejsgaard T COX-2 is a novel target in therapy of mycosis fungoides Leukemia 2010 24 12 2127 2129 10.1038/leu.2010.221 20882047

97. Wang X Baek SJ Eling TE The diverse roles of nonsteroidal anti-inflammatory drug activated gene (NAG-1/GDF15) in cancer Biochem Pharmacol 2013 85 5 597 606 10.1016/j.bcp.2012.11.025 23220538 PMC3566326

98. Okazaki R Moon Y Norimura T Ionizing radiation enhances the expression of the nonsteroidal anti-inflammatory drug-activated gene (NAG1) by increasing the expression of TP53 in human colon cancer cells Radiat Res 2006 165 2 125 130 10.1667/RR3492.1 16435911

99. Kim CH Kim MY Moon JY Implication of NAG-1 in synergistic induction of apoptosis by combined treatment of sodium salicylate and PI3K/MEK1/2 inhibitors in A549 human lung adenocarcinoma cells Biochem Pharmacol 2008 75 9 1751 1760 10.1016/j.bcp.2008.02.002 18358453

100. Wang C Wang J Bai P Troglitazone induces apoptosis in gastric cancer cells through the NAG-1 pathway Mol Med Rep 2011 4 1 93 97 21461570 10.3892/mmr.2010.381

101. Kim KS Yoon JH Kim JK Cyclooxygenase inhibitors induce apoptosis in oral cavity cancer cells by increased expression of nonsteroidal anti-inflammatory drug-activated gene Biochem Biophys Res Commun 2004 325 4 1298 1303 10.1016/j.bbrc.2004.10.176 15555568

102. Gay LJ Felding-Habermann B Contribution of platelets to tumour metastasis Nature Rev Cancer 2011 11 2 123 134 10.1038/nrc3004 21258396 PMC6894505

103. Digklia A Voutsadakis IA Thrombocytosis as a prognostic marker in stage III and IV serous ovarian cancer Obstet Gynecol Sci 2014 57 6 457 463 10.5468/ogs.2014.57.6.457 25469333 PMC4245338

104. Rachidi S Wallace K Day TA Lower circulating platelet counts and antiplatelet therapy independently predict better outcomes in patients with head and neck squamous cell carcinoma J Hematol Oncol 2014 7 1 65 10.1186/s13045-014-0065-5 25260646 PMC4189675

105. Kim KH Park TY Lee JY Prognostic significance of initial platelet counts and fibrinogen level in advanced non-small cell lung cancer J Korean Med Sci 2014 29 4 507 511 10.3346/jkms.2014.29.4.507 24753697 PMC3991793

106. Su BB Chen JH Shi H Aspirin may modify tumor microenvironment via antiplatelet effect Med Hypotheses 2014 1 7 9 10.1016/j.mehy.2014.05.007 24908358

107. Mikami J Kurokawa Y Takahashi T Antitumor effect of antiplatelet agents in gastric cancer cells: an in vivo and in vitro study Gastric Cancer 2015 1 10 10.1007/s10120-015-0556-2 26486505

108. Mousa SA Petersen LJ Anti-cancer properties of low-molecular-weight heparin: preclinical evidence Thromb Haemost 2009 102 2 258 267 19652876 10.1160/TH08-12-0832

109. Pfankuchen DB Stölting DP Schlesinger M Low molecular weight heparin tinzaparin antagonizes cisplatin resistance of ovarian cancer cells Biochem Pharmacol 2015 97 2 147 157 10.1016/j.bcp.2015.07.013 26239805

110. Bajaj P Ballary CC Dongre NA Comparison of the effects of parecoxib and diclofenac in preemptive analgesia: a prospective, randomized, assessor-blind, single-dose, parallel-group study in patients undergoing elective general surgery Curr Ther Res Clin Exp 2004 65 5 383 397 10.1016/j.curtheres.2004.10.004 24672093 PMC3964533

111. Palayoor ST Tofilon PJ Coleman CN Ibuprofen-mediated reduction of hypoxia-inducible factors HIF-1alpha and HIF-2alpha in prostate cancer cells Clin Cancer Res 2003 9 8 3150 3157 12912967

112. Sareddy GR Kesanakurti D Kirti PB Nonsteroidal anti-inflammatory drugs diclofenac and celecoxib attenuates Wnt/β-catenin/Tcf signaling pathway in human glioblastoma cells Neurochem Res 2013 38 11 2313 2322 10.1007/s11064-013-1142-9 24013885

113. Surowiak P Materna V Matkowski R Relationship between the expression of cyclooxygenase 2 and MDR1/P-glycoprotein in invasive breast cancers and their prognostic significance Breast Cancer Res 2005 7 5 R862 R870 10.1186/bcr1313 16168133 PMC1242165

114. de Groot DJA de Vries EGE Groen HJM Non-steroidal anti-inflammatory drugs to potentiate chemotherapy effects: from lab to clinic Crit Rev Oncol/Hematol 2007 61 1 52 69 10.1016/j.critrevonc.2006.07.001 16945549

115. Hiľovská L Jendželovský R Fedoročko P Potency of non-steroidal anti-inflammatory drugs in chemotherapy Mol Clin Oncol 2015 3 1 3 12 25469262 10.3892/mco.2014.446 PMC4251142

116. Legge F Paglia A D’Asta M Phase II study of the combination carboplatin plus celecoxib in heavily pre-treated recurrent ovarian cancer patients BMC Cancer 2011 11 214 10.1186/1471-2407-11-214 21627839 PMC3123659

117. White DL Dang P Engler J Functional activity of the OCT-1 protein is predictive of long-term outcome in patients with chronic-phase chronic myeloid leukemia treated with imatinib J Clin Oncol 2010 28 16 2761 2767 10.1200/JCO.2009.26.5819 20421539

118. Wang J Hughes TP Kok CH Contrasting effects of diclofenac and ibuprofen on active imatinib uptake into leukaemic cells Brit J Cancer 2012 106 11 1772 1778 10.1038/bjc.2012.173 22531634 PMC3364120

119. Kurtova AV Xiao J Mo Q Blocking PGE-induced tumour repopulation abrogates bladder cancer chemoresistance Nature 2014 10.1038/nature14034 PMC4465385 25470039

120. Crokart N Radermacher K Jordan BF Tumor radiosensitization by antiinflammatory drugs: evidence for a new mechanism involving the oxygen effect Cancer Res 2005 65 17 7911 7916 16140962 10.1158/0008-5472.CAN-05-1288

121. Forget P Vandenhende J Berliere M Do intraoperative analgesics influence breast cancer recurrence after mastectomy? A retrospective analysis Anesth Analg 2010 110 6 1630 1635 10.1213/ANE.0b013e3181d2ad07 20435950

122. Retsky M Demicheli R Hrushesky WJM Reduction of breast cancer relapses with perioperative non-steroidal anti-inflammatory drugs: new findings and a review Curr Med Chem 2013 20 33 4163 4176 10.2174/09298673113209990250 23992307 PMC3831877

123. Antonio N Bønnelykke-Behrndtz ML Ward LC The wound inflammatory response exacerbates growth of pre-neoplastic cells and progression to cancer EMBO J 2015 1 18 10.15252/embj.201490147 PMC4585460 26136213

124. Neeman E Ben-Eliyahu S Surgery and stress promote cancer metastasis: new outlooks on perioperative mediating mechanisms and immune involvement Brain Behav Immun 2013 30 Suppl S32 S40 10.1016/j.bbi.2012.03.006 22504092 PMC3423506

125. Gottschalk A Sharma S Ford J The role of the perioperative period in recurrence after cancer surgery Anesth Analg 2010 110 6 1636 1643 10.1213/ANE.0b013e3181de0ab6 20435944

126. Opitz I Arni S Oberreiter B Perioperative diclofenac application during video-assisted thoracic surgery pleurodesis modulates early inflammatory and fibrinolytic processes in an experimental model Eur Surg Res 2013 50 1 14 23 10.1159/000341670 23429206

127. Ash SA Buggy DJ Does regional anaesthesia and analgesia or opioid analgesia influence recurrence after primary cancer surgery? An update of available evidence Best Pract Res Clin Anaesthesiol 2013 27 4 441 456 10.1016/j.bpa.2013.10.005 24267550

128. Shakhar G Ben-Eliyahu S Potential prophylactic measures against postoperative immunosuppression: could they reduce recurrence rates in oncological patients? Ann Surg Oncol 2003 10 8 972 992 10.1245/ASO.2003.02.007 14527919

129. Heaney Á Buggy DJ Can anaesthetic and analgesic techniques affect cancer recurrence or metastasis? Brit J Anaesth 2012 109 Suppl 1 i17 i28 10.1093/bja/aes421 23242747

130. Horowitz M Neeman E Sharon E Exploiting the critical perioperative period to improve long-term cancer outcomes Nature Rev Clin Oncol 2015 1 14 10.1038/nrclinonc.2014.224 PMC5497123 25601442

131. Retsky M Rogers R Demicheli R NSAID analgesic ketorolac used perioperatively may suppress early breast cancer relapse: particular relevance to triple negative subgroup Breast Cancer Res Treat 2012 134 2 881 888 10.1007/s10549-012-2094-5 22622810

132. Guo Y Kenney SR Cook LS A novel pharmacologic activity of ketorolac for therapeutic benefit in ovarian cancer patients Clin Cancer Res 2015 21 22 5064 72 10.1158/1078-0432.CCR-15-0461 26071482 PMC4644688

133. Pantziarka P Bouche G Meheus L Repurposing drugs in oncology (ReDO)-cimetidine as an anti-cancer agent Ecancermedicalscience 2014 8 485 10.3332/ecancer.2014.485 25525463 PMC4268104

134. Deva S Jameson M Histamine type 2 receptor antagonists as adjuvant treatment for resected colorectal cancer Cochrane Database Syst Rev 2012 8 8 CD007814 22895966 10.1002/14651858.CD007814.pub2 PMC11627143

135. Castellone MD Teramoto H Williams BO Prostaglandin E2 promotes colon cancer cell growth through a Gs-axinbeta-catenin signaling axis Science 2005 310 5753 1504 1510 10.1126/science.1116221 16293724

136. Greenhough A Smartt HJM Moore AE The COX-2/PGE 2 pathway: key roles in the hallmarks of cancer and adaptation to the tumour microenvironment Carcinogenesis 2009 30 3 377 386 10.1093/carcin/bgp014 19136477

137. Kaur J Sanyal SN PI3-kinase/Wnt association mediates COX-2/PGE 2 pathway to inhibit apoptosis in early stages of colon carcinogenesis: chemoprevention by diclofenac Tumor Biol 2010 31 6 623 631 10.1007/s13277-010-0078-9 20617408

Supplementary Material

Introduction

The following drugs warrant further investigation in combination with diclofenac (DCF) and existing standard of care cancer treatments in a range of cancers. These combinations, listed in Table 1 , have been selected on the basis of existing pre-clinical and clinical experience in each of the indications. In some cases these combinations replicate existing protocols currently being tested in clinical trials, but substitute known and repurposed drugs for the newer and/or more toxic agents currently being investigated. All of these proposed combinations are expected to display relatively low toxicity and use low cost and generally available agents. The following drugs are not listed in order of priority .

Pharmacological interventions

The following proposed drug combination are primarily based on postulated synergies between DCF and other agents, including chemotherapeutics and a number of repurposed non-cancer drugs. Combination anti-PD1/anti-CTLA4 therapies – A major clinical focus in cancer immunotherapy is the use of immune checkpoint inhibitors, particularly anti-cytotoxic T-lymphocyte antigen 4 (anti-CTLA-4) and anti-programmed death-1 (anti-PD-1), to reverse immune suppression. Initial positive results using ipilimumab and nivolumab in melanoma have provoked interest in other cancer types and in the use of combination treatments [ 1 – 3 ]. However, despite some startling successes, some patients do not show a response, or acquire resistance to these therapies. There is some evidence that the degree of benefit may be associated with a pre-existing anti-tumour T-cell response [ 4 , 5 ]. Spranger et al showed that in melanoma tumour-associated β-catenin signalling was associated with a lack of T-cell response and subsequent treatment failure in a murine model [ 6 ]. DCF has been shown to inhibit β-catenin signalling [ 7 – 9 ]. Additionally Zelenay et al showed that both aspirin and celecoxib synergised with anti-PD-1 blockade in murine models of melanoma and colorectal cancer [ 10 ]. Therefore the use of DCF, possibly in combination with other agents, in order to prime the T-cell response and improve the clinical benefit of checkpoint inhibitors is clearly warranted. Cimetidine – The H2 receptor antagonist cimetidine remains a widely used antacid for both short-term and long-term administration. There is a range of evidence, both pre-clinical and clinical, for a number of anticancer effects, particularly with respect to colorectal cancer [ 11 , 12 ]. The anticancer mechanisms of action may be related to effects on cell adhesion, angiogenesis and a number of different effects on immunity. In respect to colorectal cancer the clinical evidence shows a positive effect on survival when used perioperatively for early stage colorectal cancer [ 11 ]. In one trial long-term use of cimetidine was associated with a significantly improved long-term survival in patients with colorectal cancer with high levels of sialyl Lewis-X and sialyl Lewis-A epitope expression on tumour cells [ 13 ]. The use of DCF pre-operatively and long-term cimetidine post-operatively warrants clinical investigation. Itraconazole – Itraconazole, a generic broad-spectrum anti-fungal drug, has been shown to have potent anticancer effects in a range of cancers [ 14 ]. The main mechanisms of action which have been posited include Hedgehog pathway inhibition, anti-angiogenic activity and reversal of multi-drug resistance. There is some evidence that itraconazole also targets the cancer stem cell (CSC) fraction in multiple myeloma [ 15 ]. Recent evidence suggests that elevated expression of COX-2 may be related to resistance via CSC populations in colorectal cancer [ 16 ]. DCF has also shown interesting activity in blocking CSC repopulation of tumour masses following debulking in a bladder cancer model [ 17 ]. Combination therapies which target the stem cell populations with agents such as DCF and itraconazole in addition to therapies targeting the non-CSC populations warrant additional pre-clinical and clinical study. Metronomic chemotherapy – Metronomic chemotherapy is an alternative to maximum tolerated dose (MTD) chemotherapy and is characterised by frequent, low-dose administration of many standard chemotherapy drugs. Where MTD chemotherapy aims to maximise tumour kill rates, metronomic chemotherapy appears to deliver its therapeutic effects via anti-angiogenic and immunomodulatory activities – with an attendant reduction in adverse effects and an improved quality of life [ 18 – 20 ]. To date a broad range of chemotherapy agents have been used metronomically, mostly via oral formulations, including cyclophosphamide, capecitabine, methotrexate, etoposide, vinorelbine and temozolomide [ 21 ]. There is much interest in the addition of further anti-angiogenic agents in combination with metronomics, with a number of trials using celecoxib in this respect [ 22 – 24 ]. As there is evidence that DCF has both anti-angiogenic and immunomodulatory effects the combination with metronomic chemotherapy is worthy of further clinical investigation, particularly in those cancers with high unmet needs, such as soft tissue sarcoma, metastatic breast cancer and ovarian cancer.

Non-pharmacological interventions

DCF co-treatment may be a valid addition to a number of non-pharmaceutical cancer treatment modalities. Photodynamic Therapy (PDT) – PDT is a locally ablative cancer treatment that involves the administration of a photosensitive agent which preferentially accumulates in tumour tissue, followed by the application of a light source of specific frequency to the tumour such that the photosensitive agent reacts and causes tumour cell death via both apoptosis and necrosis. A number of studies have suggested that PGE 2 is associated with increased resistance and cell survival in PDT [ 25 , 26 ]. Therefore inhibition of COX-2/PGE 2 has been investigated as a mechanism to improve response to treatment, primarily focusing on COX-2 inhibitors such as celecoxib and NS-398 [ 27 – 29 ]. While PDT is a local treatment, there is interest in the systemic immune effects which treatment is known to elicit and in particular with the development of protocols which increase anti-tumour immunity [ 30 , 31 ]. The combination of PDT with short-term DCF treatment may serve to both reduce treatment resistance by inhibition of PGE 2 and improve the systemic immune response. Perioperative – There is evidence that the perioperative period following primary tumour resection may be associated with an increased risk of distant metastases in a range of tumour types [ 32 – 34 ]. A number of mechanisms of action have been posited, with evidence in particular that the post-surgical immunosuppression associated with the inflammatory wound-healing response is involved [ 35 , 36 ]. There is evidence from retrospective studies that the recurrence rate may be modulated by the use of NSAIDs, specifically ketorolac and DCF [ 37 – 40 ]. Two Phase 3 prospective clinical trials of pre-incisional ketorolac are currently underway in breast cancer ( NCT01806259 [ 41 ] and NCT02141139 ). Additional trials are warranted in other cancers, including osteosarcoma, non-small cell lung cancer and oesophageal cancer. In addition to ketorolac there is sufficient data to suggest that other agents may be of some value in the perioperative setting, including DCF, cimetidine, Polysaccharide-K (PSK) and other agents, alone or in combinations. Radiotherapy – Radiotherapy remains a key treatment modality in many forms of cancer, with much interest in the development of new agents to improve therapeutic response [ 42 ]. COX-2/PGE 2 inhibition has been explored as a potential radiosensitisation strategy in a number of tumour models, with positive results [ 43 – 46 ]. There have been a number of small clinical trials exploring this strategy in a range of cancers, including NSCLC [ 47 ], nasopharyngeal carcinoma [ 48 ], head and neck cancer [ 49 ] and other malignancies. Given the relative COX-2 selectivity of DCF, clinical exploration may be warranted, particularly in the case of intraoperative radiotherapy regimens where DCF may provide positive effects both in radiosensitisation and in reversing the post-surgical inflammatory response. In the case of glioblastoma multiforme (GBM), there is evidence that post-irradiation recurrence is associated with hypoxia-induced regrowth of damaged vasculature, a process mediated by SDF-1/CXCR4 interaction and the recruitment of bone-marrow derived cells [ 50 ]. Inhibition of SDF-1/CXCR4 interactions using plerixafor has been shown to inhibit the vasculogenic cascade. A combination of plerixafor with DCF, noting that PGE 2 inhibition can also down-regulate CXCR4-related angiogenesis [ 51 ], would therefore be of some interest in the case of GBM radiotherapy.

References

1. Johnson DB Peng C Sosman JA Nivolumab in melanoma: latest evidence and clinical potential Ther Adv Med Oncol 2015 7 2 97 106 10.1177/1758834014567469 25755682 PMC4346215

2. Wolchok JD Kluger H Callahan MK Nivolumab plus ipilimumab in advanced melanoma N Engl J Med 2013 369 2 122 133 10.1056/NEJMoa1302369 23724867 PMC5698004

3. Larkin J Chiarion-Sileni V Gonzalez R Combined nivolumab and ipilimumab or monotherapy in untreated melanoma N Engl J Med 2015 373 1 23 34 10.1056/NEJMoa1504030 26027431 PMC5698905

4. Winograd R Byrne K Evans RA Induction of T cell immunity overcomes complete resistance to PD-1 and CTLA-4 blockade and improves survival in pancreatic carcinoma Cancer Immunol Res 2015 22 399 412 10.1158/2326-6066.CIR-14-0215 25678581 PMC4390506

5. Spranger S Koblish HK Horton B Mechanism of tumor rejection with doublets of CTLA-4, PD-1/PD-L1, or IDO blockade involves restored IL-2 production and proliferation of CD8(+) T cells directly within the tumor microenvironment J Immunother Cancer 2014 2 1 3 10.1186/2051-1426-2-3 24829760 PMC4019906

6. Spranger S Bao R Gajewski TF Melanoma-intrinsic β-catenin signalling prevents anti-tumour immunity Nature 2015 523 7559 231 235 10.1038/nature14404 25970248

7. Sareddy GR Kesanakurti D Kirti PB Nonsteroidal anti-inflammatory drugs diclofenac and celecoxib attenuates Wnt/β-catenin/Tcf signaling pathway in human glioblastoma cells Neurochem Res 2013 38 11 2313 2322 10.1007/s11064-013-1142-9 24013885

8. Cho M Gwak J Park S Diclofenac attenuates Wnt/β-catenin signaling in colon cancer cells by activation of NF-κB FEBS Lett 2005 579 20 4213 4218 10.1016/j.febslet.2005.06.049 16051228

9. Gardner SH Hawcroft G Hull MA Effect of nonsteroidal anti-inflammatory drugs on beta-catenin protein levels and catenin-related transcription in human colorectal cancer cells Br J Cancer 2004 91 1 153 163 10.1038/sj.bjc.6601901 15188006 PMC2364748

10. Zelenay S van der Veen AG Bötcher JP Cyclooxygenase-dependent tumor growth through evasion of immunity Cell 2015 1 14 10.1016/j.cell.2015.08.015 PMC4597191 26343581

11. Pantziarka P Bouche G Meheus L Repurposing drugs in oncology (ReDO)-cimetidine as an anti-cancer agent Ecancermedicalscience 2014 8 485 10.3332/ecancer.2014.485 25525463 PMC4268104

12. Kubecova M Kolostova K Pinterova D Cimetidine: an anticancer drug? Eur J Pharm Sci 2011 42 5 439 444 10.1016/j.ejps.2011.02.004 21329756

13. Matsumoto S Imaeda Y Umemoto S Cimetidine increases survival of colorectal cancer patients with high levels of sialyl Lewis-X and sialyl Lewis-A epitope expression on tumour cells Br J Cancer 2002 86 2 161 167 10.1038/sj.bjc.6600048 11870500 PMC2375187

14. Pantziarka P Sukhatme V Bouche G Repurposing drugs in oncology (ReDO)-itraconazole as an anti-cancer agent Ecancermedicalscience 2015 9 521 10.3332/ecancer.2015.521 25932045 PMC4406527

15. Yang Y Shi J Tolomelli G RARα2 expression confers myeloma stem cell features Blood 2013 122 8 1437 1447 10.1182/blood-2013-02-482919 23847194 PMC3750340

16. Valverde A Peñarando J Cañas A Simultaneous inhibition of EGFR/VEGFR and cyclooxygenase-2 targets stemness-related pathways in colorectal cancer cells PLoS One 2015 10 6 e0131363 10.1371/journal.pone.0131363 26107817 PMC4479446

17. Kurtova AV Xiao J Mo Q Blocking PGE-induced tumour repopulation abrogates bladder cancer chemoresistance Nature 2014 517 7533 209 13 10.1038/nature14034 25470039 PMC4465385

18. Lien K Georgsdottir S Sivanathan L Low-dose metronomic chemotherapy: a systematic literature analysis Eur J Cancer 2013 49 16 3387 3395 10.1016/j.ejca.2013.06.038 23880474

19. Drevs J Fakler J Eisele S Antiangiogenic potency of various chemotherapeutic drugs for metronomic chemotherapy Anticancer Res 2004 24 3a 1759 1763 15274352

20. Scharovsky OG Mainetti LE Rozados VR Metronomic chemotherapy: changing the paradigm that more is better Curr Oncol 2009 16 2 7 15 10.3747/co.v16i2.420 19370174 PMC2669231

21. André N Carré Pasquier E Metronomics: towards personalized chemotherapy? Nat Rev Clin Oncol 2014 11 7 413 31 10.1038/nrclinonc.2014.89 24913374

22. Perroud HA Rico MJ Alasino CM Safety and therapeutic effect of metronomic chemotherapy with cyclophosphamide and celecoxib in advanced breast cancer patients Future Oncol 2013 9 3 451 462 10.2217/fon.12.196 23469980

23. Bhatt RS Merchan J Parker R A phase 2 pilot trial of low-dose, continuous infusion, or ‘metronomic’ paclitaxel and oral celecoxib in patients with metastatic melanoma Cancer 2010 116 7 1751 1756 10.1002/cncr.24902 20120033 PMC2847062

24. Khan OA Blann AD Payne MJ Continuous low-dose cyclophosphamide and methotrexate combined with celecoxib for patients with advanced cancer 2011 49 24 Pt 1 Nature Publishing Group 6896 900 10.1038/bjc.2011.154 PMC3111194 21587257

25. Henderson BW Donovan JM Release of prostaglandin E2 from cells by photodynamic treatment in vitro Cancer Res 1989 49 24 I 6896 6900 2531034

26. Penning LC Keirse MJ VanSteveninck J Ca(2+)-mediated prostaglandin E2 induction reduces haematoporphyrin-derivative-induced cytotoxicity of T24 human bladder transitional carcinoma cells in vitro Biochem J 1993 292 Pt 1 237 240 10.1042/bj2920237 8503851 PMC1134294

27. Ferrario A Fisher AM Rucker N Celecoxib and NS-398 enhance photodynamic therapy by increasing in vitro apoptosis and decreasing in vivo inflammatory and angiogenic factors Cancer Res 2005 65 20 9473 9478 10.1158/0008-5472.CAN-05-1659 16230411

28. Ferrario A Gomer CJ Targeting the tumor microenvironment using photodynamic therapy combined with inhibitors of cyclooxygenase-2 or vascular endothelial growth factor Methods Mol Biol 2010 635 121 132 10.1007/978-1-60761-697-9_9 20552344

29. Kim CH Chung CW Lee HM Synergistic effects of 5-aminolevulinic acid based photodynamic therapy and celecoxib via oxidative stress in human cholangiocarcinoma cells Int J Nanomedicine 2013 8 2173 2186 23807846 10.2147/IJN.S44394 PMC3685402

30. Shams M Owczarczak B Manderscheid-Kern P Development of photodynamic therapy regimens that control primary tumor growth and inhibit secondary disease Cancer Immunol Immunother 2015 64 3 287 297 10.1007/s00262-014-1633-9 25384911 PMC4341021

31. Reginato E Wolf P Hamblin MR Immune response after photodynamic therapy increases anti-cancer and anti-bacterial effects World J Immunol 2014 4 1 1 11 10.5411/wji.v4.i1.1 25364655 PMC4214901

32. Horowitz M Neeman E Sharon E Exploiting the critical perioperative period to improve long-term cancer outcomes Nat Rev Clin Oncol 2015 1 14 10.1038/nrclinonc.2014.224 PMC5497123 25601442

33. Gottschalk A Sharma S Ford J The role of the perioperative period in recurrence after cancer surgery Anesth Analg 2010 110 6 1636 1643 10.1213/ANE.0b013e3181de0ab6 20435944

34. Forget P De Kock M Perspectives in anaesthesia for cancer surgery J Cancer Res Clin Oncol 2014 140 3 353 359 10.1007/s00432-013-1522-1 24077864 PMC11824043

35. Roxburgh CS Horgan PG McMillan DC The perioperative immune/inflammatory insult in cancer surgery: time for intervention? Oncoimmunol 2013 2 12 e27324 10.4161/onci.27324 PMC3909539 24498571

36. Forget P Simonet O De Kock M Cancer surgery induces inflammation, immunosuppression and neo-angiogenesis, but is it influenced by analgesics? F1000Res 2013 2 102 24358839 10.12688/f1000research.2-102.v1 PMC3752648

37. Forget P Vandenhende J Berliere M Do intraoperative analgesics influence breast cancer recurrence after mastectomy? A retrospective analysis AnesthAnalg 2010 110 6 1630 1635 10.1213/ANE.0b013e3181d2ad07 20435950

38. Forget P Machiels JP Coulie PG Neutrophil:lymphocyte ratio and intraoperative use of ketorolac or diclofenac are prognostic factors in different cohorts of patients undergoing breast, lung, and kidney cancer surgery Ann Surg Oncol 2013 20 Suppl 3 S650 S660 10.1245/s10434-013-3136-x 23884751

39. Forget P Bentin C Machiels JP Intraoperative use of ketorolac or diclofenac is associated with improved disease-free survival and overall survival in conservative breast cancer surgery Brit J Anaesth 2014 113 Suppl 1 1 6 10.1093/bja/aet464 24464611

40. Retsky M Demicheli R Hrushesky WJM Reduction of breast cancer relapses with perioperative non-steroidal anti-inflammatory drugs: new findings and a review Curr Med Chem 2013 20 33 4163 4176 10.2174/09298673113209990250 23992307 PMC3831877

41. Forget P Berlière M van Maanen A Perioperative ketorolac in high risk breast cancer patients. Rationale, feasibility and methodology of a prospective randomized placebo-controlled trial Med Hypotheses 2013 81 4 707 712 10.1016/j.mehy.2013.07.033 23937996

42. Deorukhkar A Krishnan S Targeting inflammatory pathways for tumor radiosensitization Biochem Pharmacol 2010 80 12 1904 1914 10.1016/j.bcp.2010.06.039 20599771 PMC3090731

43. Milas L Kishi K Hunter N Enhancement of tumor response to gamma radiation by an inhibitor of cyclooxygenase-2 enzyme J Natl Cancer Inst 1999 91 17 1501 1504 10.1093/jnci/91.17.1501 10469752

44. Kishi K Petersen S Petersen C Preferential enhancement of tumor radioresponse by a cyclooxygenase-2 inhibitor Cancer Res 2000 60 5 1326 1331 10728694

45. Davis TW O’Neal JM Pagel MD Synergy between celecoxib and radiotherapy results from inhibition of cyclooxygenase-2-derived prostaglandin E2, a survival factor for tumor and associated vasculature Cancer Res 2004 64 1 279 285 10.1158/0008-5472.CAN-03-1168 14729635

46. Crokart N Radermacher K Jordan BF Tumor radiosensitization by antiinflammatory drugs: evidence for a new mechanism involving the oxygen effect Cancer Res 2005 65 17 7911 7916 16140962 10.1158/0008-5472.CAN-05-1288

47. Liao Z Komaki R Milas L A phase I clinical trial of thoracic radiotherapy and concurrent celecoxib for patients with unfavorable performance status inoperable/unresectable non-small cell lung cancer Clin Cancer Res 2005 11 9 3342 3348 10.1158/1078-0432.CCR-04-1741 15867233

48. Mohamma-Dianpanah M Shafizad A Khademi B Efficacy and safety of concurrent chemoradiation with weekly cisplatin ± low-dose celecoxib in locally advanced undifferentiated nasopharyngeal carcinoma: a phase II-III clinical trial J Cancer Res Ther 2011 7 4 442 10.4103/0973-1482.92013 22269407

49. Halamka M Cvek J Kubes J Plasma levels of vascular endothelial growth factor during and after radiotherapy in combination with celecoxib in patients with advanced head and neck cancer Oral Oncol 2011 47 8 763 767 10.1016/j.oraloncology.2011.05.009 21696999

50. Kioi M Vogel H Schultz G Inhibition of vasculogenesis, but not angiogenesis, prevents the recurrence of glioblastoma after irradiation in mice J Clin Invest 2010 120 3 694 705 10.1172/JCI40283 20179352 PMC2827954

51. Salcedo R Zhang X Young HA Angiogenic effects of prostaglandin E2 are mediated by up-regulation of CXCR4 on human microvascular endothelial cells Blood 2003 102 6 1966 1977 10.1182/blood-2002-11-3400 12791666

52. Deva S Jameson M Histamine type 2 receptor antagonists as adjuvant treatment for resected colorectal cancer Cochrane Database Syst Rev 2012 8 8 CD007814 22895966 10.1002/14651858.CD007814.pub2 PMC11627143

53. Bastiaannet E Sampieri K Dekkers OM Use of aspirin postdiagnosis improves survival for colon cancer patients Br J Cancer 2012 106 1532–1827 (Electronic) 1564 1570 10.1038/bjc.2012.101 22454078 PMC3341868

54. Pantziarka P Bouche G Meheus L Repurposing drugs in oncology (ReDO)-mebendazole as an anti-cancer agent Ecancermedicalscience 2014 8 443 10.3332/ecancer.2014.485 25075217 PMC4096024

55. Kelleher FC Cain JE Healy JM Prevailing importance of the hedgehog signaling pathway and the potential for treatment advancement in sarcoma Pharmacol Ther 2012 136 153 168 10.1016/j.pharmthera.2012.08.004 22906929

56. André N Abed S Orbach D Pilot study of a pediatric metronomic 4-drug regimen Oncotarget 2011 2 12 960 965 10.18632/oncotarget.358 22156656 PMC3282100

57. Doudican NA Byron SA Pollock PM XIAP downregulation accompanies mebendazole growth inhibition in melanoma xenografts Anticancer Drugs 2013 24 2 181 188 10.1097/CAD.0b013e32835a43f1 23059386

58. Lackner H Urban C Kerbl R Noncytotoxic drug therapy in children with unresectable desmoid tumors Cancer 1997 80 2 334 340 10.1002/(SICI)1097-0142(19970715)80:2&lt;334::AID-CNCR22&gt;3.0.CO;2-U 9217047