Scientists at the University of Birmingham have found an anti-malarial drug was effective in treating head and neck cancer in mice.
The drug quinacrine was used extensively to prevent and treat malaria in soldiers fighting in mosquito-ridden areas during World War Two.
It is similar to the quinine that makes tonic water glow, has minimal side-effects, and is now used for treating parasite infections and other conditions.
Each year around 11,900 people are diagnosed with head and neck cancer in the UK. Current treatment relies heavily on debilitating surgery and toxic chemotherapy, but despite this, it has a poor outcome with three to seven in 10 people surviving their disease for five years or more.
The drug, quinacrine, was tested through a number of methods, including on cell cultures, in tumour biopsies from patients with head and neck cancer, and in mice.
The research results, published in Oncotarget, show that in mice quinacrine can make standard chemotherapy more effective – suggesting a lower dose may be used, reducing toxic side effects.
The results also showed the drug to be effective at reducing the growth of cancer cells grown in the lab, and in tumors.
Significantly, the research in mice showed a combination therapy of quinacrine and chemotherapy, and so allowed for the chemotherapy dose to be halved while still maintaining the same impairment of tumor growth.
Lead author Dr. Jennifer Bryant, of the University of Birmingham’s Institute of Head and Neck Studies and Education, said:
“This is important research in the laboratory and demonstrates the real potential in repurposing drugs.
“The team is now looking to translate these research findings into a clinical trial for head and neck cancer patients.”
Corresponding author Professor Hisham Mehanna, Director of the Institute of Head and Neck Studies and Education at the University of Birmingham and Consultant Head, Neck and Thyroid Surgeon at University Hospitals Birmingham NHS Foundation Trust, said drug repurposing is particularly exciting due to known safety in humans and low cost, which mean they can be rapidly translated from the lab to the clinic.
He added: “Head and neck cancer patients have limited treatment options, often associated with severe, potentially life-threatening, side effects, it is important, therefore, that we find different treatments.
“My team has developed a drug repurposing platform called “AcceleraTED’ which assesses drugs that treat other non-cancerous conditions and have been approved by the Food and Drug Administration and the European Medicines Agency to see if they have the potential to be effective anti-cancer agents against head and neck cancer.
“This research is an example of the success we are having in the laboratory through this platform in identifying promising drugs that can be candidates to be used in patients in clinic.”
Head and neck squamous cell carcinoma (HNSCC) is a debilitating disease comprising 600,000 cases per year worldwide . For advanced stage disease, treatment consists primarily of either chemoradiotherapy or surgical intervention with adjuvant treatment, but results in a poor five year survival rate – around only 50% . Due to the high morbidity of current treatments, quality of life is severely impaired, with evident unmet need necessitating more effective therapies with lower toxicity; especially as recent studies examining cetuximab as a less toxic alternative for low risk human papillomavirus positive HNSCC have shown similar toxicity, and lower efficacy compared to cisplatin [3, 4].
Repurposing existing drugs for cancer therapy can be valuable due to known safety profiles leading to higher success rates and reduced development times, and subsequently lower costs compared with the development of novel therapeutics. The time taken to gain clinical approval for repurposed drugs is usually considerably shorter (3–12 years vs 10–17 years for novel therapeutics) , and success rates for market approval approach nearly 30% compared to only around 10% for new drugs .
We established a multi-stage drug discovery and repurposing platform called AcceleraTED. On initial screening against HNSCC cell lines, quinacrine was identified as a potential hit. Quinacrine, also known as mepacrine, was initially utilized as an antimalarial agent as early as the 1930s and is considered safe, with only minimal side effects, such as headaches and gastrointestinal upset . Quinacrine has also been proposed as a treatment for numerous other indications, from colitis  to prion disease . Vassey et al. (1955)  first assessed quinacrine as an anticancer agent in mice bearing several types of tumor, such as fibrosarcoma and carcinoma. More recently, effectiveness has been demonstrated in endometrial , colon , non-small cell lung  and HNSCC .
The mechanisms through which quinacrine exerts its anticancer effects are not fully understood. Quinacrine is a potent late-stage autophagy inhibitor  and has been shown to prime cells to the effects of cisplatin via apoptosis in cervical and endometrial cancer [16, 17]. Further mechanistic insights have been demonstrated in HNSCC whereby quinacrine was able to restore the function of the tumor suppressive protein, tumor protein 53 (TP53), leading to enhanced capabilities of initiating apoptotic cell death following DNA damage with cisplatin chemotherapy ; Moreover, quinacrine treatment has been shown to suppress phosphoinositide 3-kinase (PI3K), protein kinase B (AKT), mechanistic target of rapamycin (mTOR) and nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) pathways .
Due to promising results from high throughput screening and the supportive literature, we sought to advance the repurposing potential of quinacrine as an anti-cancer therapy for HNSCC.
Quinacrine reduces cell viability of HNSCC cancer cell lines
The Prestwick library of 1280 FDA-approved drugs was initially screened at 10 μM using a high-throughput microplate platform. We identified the potential hits and completed a confirmatory screen in CAL27 and VU147 cells (Figure 1A), that revealed the ability of the anti-malarial drug, quinacrine, tested at 3.3 μM, to reduce cell viability by 96.8% and 80.7% in CAL27 and VU147 cells, respectively, compared to controls. Even higher reductions in cell viability were achieved when using quinacrine in combination with 2 μM cisplatin, causing 97.4% and 90.0% reduction in viability in CAL27 and VU147 cell lines, respectively (Figure 1A).
To expand this finding, a larger panel of HNSCC cell lines (CAL27, SCC040, FaDu, SCC47 and VU147) was exposed to a range of quinacrine concentrations. The resulting concentration-response curves illustrate that quinacrine effectively inhibits cell viability in a concentration dependent manner (Figure 1B) with IC50 values for cell lines tested ranging from 0.63 to 1.21 μM (Figure 1B and and1D),1D), which is comfortably within clinically achievable concentrations [19–21]. These data indicated that quinacrine was a viable candidate for further development.
Quinacrine increases the efficacy of cisplatin
HNSCC cell lines showed additional suppression of cell viability when quinacrine was combined with cisplatin (cell line IC50: 2, 3 or 10 μM), compared to quinacrine alone (Figure 1B). The combination of quinacrine and standard of care cisplatin was investigated further (Figure 2A). To demonstrate a concentration-dependent reduction in viability, our HNSCC cell lines were exposed to increasing concentrations of cisplatin, with and without the addition of quinacrine at 0.4, 1.5 and 3 μM. Quinacrine enhanced the ability of cisplatin to suppress cell viability in all cell lines. This reduction was more evident at lower concentrations of cisplatin, since cisplatin concentrations of 0.1 mM (10–4 M) or above resulted in dramatic suppression of viability of all cell lines, such that additional suppression by the addition of quinacrine was not possible. For example, when treated with a cisplatin concentration of 0.3 μM (3 × 10–7 M) alone, SCC040 showed cell viability suppression of 12%, compared to suppression of 17%, 48% and 79% following the addition of 0.4, 1.5 and 3 μM quinacrine to 0.3 μM cisplatin, respectively. In comparison, at a cisplatin concentration of 0.1 mM (10–4M), cell viability was reduced by cisplatin alone by 90%, with only marginal additional suppression by increasing doses of quinacrine.
Quinacrine displays synergy with cisplatin
To confirm the above findings and to assess potential synergy of quinacrine when combined with cisplatin, Chou-Talalay analysis was undertaken . Synergy was observed at lower concentrations of quinacrine and cisplatin, as demonstrated by a combination index (CI) number less than 1 (Figure 2B) when using a fixed ratio of quinacrine to cisplatin concentrations of 1:4 based on their IC50 values (IC50 values for quinacrine given in Figure 1D; cisplatin IC50 ~ 2 μM for CAL27 and SCC040, and 10 μM for SCC47). Dose reduction indexes (DRI) refer to the amount that one drug can be reduced by to maintain the same cell viability reduction, termed the fraction affected (Fa) (Figure 2C). DRI values higher than 1 indicate that one drug can be reduced by adding a second whilst achieving the same reduction in cell viability. For example, to achieve a 50% reduction in viability (Fa = 0.5) in SCC040 cells, the concentration of cisplatin can be reduced by 1.27 times when used in combination with quinacrine. Synergy appears to be reduced at higher cisplatin concentrations, likely due to the dramatic reduction in cell viability caused by high concentration cisplatin treatment alone.
Quinacrine reduces cell viability of primary tumor cultures and potentiates the effects of cisplatin
Whilst cell lines are a good model, demonstration of anti-tumor activity against primary tumor cells direct from patients is much more valuable. Therefore, cells established from six HNSCC patients were cultured and treated at a range of quinacrine concentrations. All six primary patient HNSCC samples tested in vitro exhibited similar responses to quinacrine to those observed for the cell lines, with a mean IC50 of ~2 μM (Figure 1B and and1D1D).
When comparing the sensitivity of primary HNSCC cultures to the different cell lines at 1 μM quinacrine (10–6 M) (Figure 1C), primary cells exhibited cell viability suppression of 21.1%, similar to the SCC040 (14.3%) and VU147 (35.8%) cells lines, and lower than the CAL27 (72.3%), FaDu (56.7%) and SCC47 (47.3%) cells.
The primary cell cultures showed additional suppression of cell viability when 2 μM cisplatin was added to quinacrine (Figure 1B). As described in the cell lines above, increasing doses of quinacrine (0.4, 1.5, 3 and 6 μM quinacrine) enhanced the ability of cisplatin to suppress cell viability, more prominently at the lower doses of cisplatin (Figure 2A).
Clonal survival is reduced by quinacrine
Clonogenic assays were used to assess whether differences in viability were due to reduced clonogenic survival of HNSCC cells. All four cell lines (CAL27, SCC040, FaDu and SCC47) tested demonstrated a concentration-dependent reduction in survival fractions when treated with increasing concentrations (0.3, 0.6, 1.2, 2.4 and 3 μM) of quinacrine (Figure 3A and and3B).3B). The relative sensitivities of the different cell lines to 1.2 μM quinacrine alone or in combination with cisplatin (0.25 μM) and irradiation (0.5 Gy) are displayed in Figure 3C. CAL27 and FaDu cells had the largest decrease in survival when treated with 1.2 μM quinacrine alone, reduced by 88.4% and 88.9% compared with untreated controls, respectively. SCC040 cells demonstrated 52.9% and SCC47 cells 66.6% reduced survival compared with controls.
Further reductions were also evident upon the addition of 0.25 μM cisplatin and 0.5 Gy irradiation (Cis-IR) to 1.2 μM quinacrine (Figure 3C). However, reductions in cell viability due to adding Cis-IR to quinacrine (adjacent bars on Figure 3A) only reached statistical significance (p < 0.05) at quinacrine concentrations equal to or lower than 1.2 μM. This is due to the large reductions in clonogenic survival caused by higher doses of quinacrine, such that radiotherapy and cisplatin could not cause significant additional reductions in survival.
Quinacrine induces apoptosis
To understand the mechanisms of quinacrine-induced cell death in HNSCC cell lines, we first investigated apoptosis using the Annexin V apoptosis assay.
The table in Figure 4A demonstrates how various groups of live, dead and dying cells were identified by flow cytometry. Cells that are alive (within the lower left quadrant) became slightly more dispersed following the addition of quinacrine (32) compared to those not receiving quinacrine (green).
This is caused by the DNA-intercollation and subsequent auto-fluorescence of quinacrine within the FITC channel. These qualities do not interfere with assessment of apoptotic and non-apoptotic cell death, as quinacrine emits light at 488 nm, which is spectrally distinct from PI fluorescence (610 nm) and Annexin V-Cy5 fluorescence (680 nm).
CAL27, SCC040, SCC47 and FaDu cells were exposed to cisplatin alone (2 μM and 10 μM), increasing concentrations of quinacrine (0.6, 1.2, 1.8 and 2.4 μM) alone or a combination of increasing quinacrine concentrations and 2 μM cisplatin for 48 hours prior to Annexin V/PI quantification. CAL27, FaDu and, to a lesser extent, SCC040 cells showed a dose-dependent increase in Annexin V positive and PI positive cells with increasing quinacrine concentrations (Figure 4B). FaDu cells showed the largest increase in apoptosing cells and apoptotic cell death when exposed to quinacrine, with 39.4% of cells displaying Annexin V positivity when treated with 2.4 μM quinacrine alone and 49.1% positive when treated with 2.4 μM quinacrine in combination with 2 μM cisplatin. CAL27 cells also showed considerable apoptosis following 2.4 μM quinacrine treatment alone, with 30.1% cells positive for Annexin-V. Interestingly, SCC47 cells did not show sensitivity towards quinacrine or 2 μM cisplatin alone, reflecting the diversity in cell lines utilized.
Autophagic flux is altered by quinacrine treatment
The ability of quinacrine to mediate autophagy has been shown in other cancer types, but not in HNSCC to date. We therefore analyzed LC3 levels to determine whether quinacrine affects autophagic flux within CAL27 and SCC040 cells (Figure 5). LC3-I is constitutively expressed within the cytosol and converted to LC3-II upon recruitment to autophagosome membranes; therefore LC3-I/LC3-II is often utilized as a marker of autophagy initiation. LC3-II is later degraded within autolysosomes upon completion of autophagy . Quinacrine has been shown to sequester hydrogen ions within autolysosomes, which raises pH and impairs the breakdown of cellular components and therefore inhibits completion of autophagy, leading to a build-up of LC3-II  (Figure 5C).
Cells were exposed to cisplatin alone, quinacrine at 1.2, 1.8 and 2.4 μM alone or 1.8 μM quinacrine plus 2 μM cisplatin for 48 hours. Expression of LC3-I and LC3-II was then determined by Western blotting. Both cell lines displayed a dramatic increase in LC3-II at all concentrations of quinacrine treatment tested (Figure 5A–5B), whereas LC3-I remained unchanged in CAL27 cells and showed a non-significant increase in SCC040 cells. The accumulation of LC3-II in this setting is therefore indicative of quinacrine inhibiting the completion of autophagy. In contrast, cisplatin addition did not significantly alter levels of LC3-I or LC3-II in either cell line tested.
We assessed p53 expression in both SCC040 and CAL27 cell lines following 48 hours exposure to quinacrine (1.2, 1.8 and 2.4 μM) alone and with 1.8 μM quinacrine + 2 μM cisplatin to assess the potential influence of p53 on apoptosis in these cell lines. CAL27 cells possess a mutation in p53, rendering it non-functional . They demonstrated no change in expression of p53 or the downstream protein p21 following quinacrine or cisplatin treatment (Supplementary Figure 1). SCC040 cells express wild-type p53 . Again, no change in p53 or p21 expression was seen as a result of quinacrine or cisplatin treatment after 48 hours (Supplementary Figure 1) This would indicate that the mechanisms of action of quinacrine, within our experimental setup, are likely to be independent of TP53 status and p53 function.
Tumor xenograft growth is impaired by quinacrine and cisplatin
To substantiate the findings in the HNSCC cell lines and in patient-derived tumor cells, NSG mice bearing FaDu xenograft tumors were treated orally with 100 mg/kg quinacrine every two days, with and without 1 mg/kg or cisplatin (2 mg/kg or 1 mg/kg) via intraperitoneal injection on days 4, 8 and 12. Tumor volumes at day 19, when the first control animal was culled, show reductions in the size of treated tumors, mainly in those treated with quinacrine and 2 mg/kg cisplatin (p = 0.0001) (Figure 6A and and6B6B).
Quinacrine alone showed early inhibition of tumor growth, and modestly extended the mean time needed to reach maximum volume by 2 days (from 20 to 22 days), compared to untreated control (Figure 6A and and6C).6C). In comparison, higher dose cisplatin (2 mg/kg) alone extended the mean time to reach maximum tumor volume by 8 days, from 20 to 28 days compared to untreated control. Quinacrine was particularly effective when combined with high dose cisplatin, whereby the time taken for tumors to reach their maximum volume could be extended by 12 days, taking a median of 32 days compared to 20 days for untreated controls to reach a tumor volume of 1000 mm3 (p < 0.0001), and was also significantly higher than the time taken to reach maximum tumor volume in the group treated by quinacrine alone, from 22 days to 32 days (p = 0.0002). There was a difference in the time needed to reach maximum tumor volume between the combined quinacrine and 2 mg/kg cisplatin group and the 2 mg/kg cisplatin alone group of 4 days, however this did not reach statistical significance (p = 0.6), possibly due to the small sample size and corrections for multiple comparisons.
The Kaplan-Meier plot (Figure 6D) further emphasizes the extended time for treated tumors to reach their maximum tumor volume (MTV) when mice were treated with cisplatin (2 mg/kg) and quinacrine, even though one animal was culled early due to a deterioration of their overall condition and not due to tumor volume. The Hazards ratio for time taken to reach MTV in the 2 mg/kg cisplatin and quinacrine group compared to the 2 mg/kg cisplatin alone group was 0.4 (95% CI, Mantel-Haenszel test).
Importantly, the in vitro experiments demonstrated that the synergy between quinacrine and cisplatin was highest at lower doses of cisplatin, where quinacrine potentiated the effect of cisplatin most. At higher concentrations of cisplatin tumor kill was so high that there was little possibility for quinacrine to add an effect. We therefore also tested the combination of quinacrine with a lower dose of cisplatin (1 mg/kg – half the higher dose used). At that concentration, quinacrine and 1 mg/kg cisplatin resulted in the same growth rate as 2 mg/kg cisplatin alone, indicating that by adding quinacrine, the dose of cisplatin can be halved whilst maintaining the same anti-cancer efficacy, which could have significant benefits to patients by reducing toxicity related side effects.
Quinacrine and cisplatin were well tolerated throughout the in vivo experiment. The weights of mice remained well within the 20% weight loss cut-off (Figure 6E). A slight yellowing of the skin of animals treated with quinacrine was apparent due to the auto fluorescent properties of the drug (Figure 6F).
Quinacrine is present in plasma and tumor tissue
Quinacrine was administered at 100 mg/kg to mice every 48 hours via oral gavage. To establish whether a steady, therapeutic concentration was maintained within the circulation of mice receiving oral quinacrine treatment, blood samples were taken 48 hours after the 9th dose of quinacrine (on day 19 of the toxicity experiment). Analysis carried out by LC-MS (Supplementary Figure 2) revealed that even 48 hours after the last dose, plasma concentrations of 0.5 mg/mL were still present (Figure 6G), equating to 1.25 μM, a dose that had resulted in a significant response using the alamarBlue® viability assay in all cell lines tested (Figure 1B–1D) and initiated apoptosis in CAL27 and FaDu cells (Figure 4B).
To establish whether quinacrine successfully entered tumors from the blood, sections were taken and quinacrine fluorescence visualized at 488nm. Tumor samples from mice receiving quinacrine treatment showed clear fluorescence throughout their tumors, whereas those receiving only control treatment did not (Figure 6H); thus confirming that quinacrine is able to reach tumor tissue following oral administration.
More information: Repurposed quinacrine synergizes with cisplatin, reducing the effective dose required for treatment of head and neck squamous cell carcinoma. Oncotarget. DOI: 10.18632/oncotarget.27156
Journal information: Oncotarget
Provided by University of Birmingham