Abstract
Dicycloplatin (DCP) has better solubility and stability than both cisplatin and carboplatin. Pre-clinical and phase I studies demonstrated significant antitumor activity and fewer adverse events than carboplatin. Phase II clinical trials in advanced non-small cell lung cancer found efficacy and safety of DCP-plus-paclitaxel comparable to carboplatin-plus-paclitaxel but better tolerability. This article summarizes and reviews pre-clinical and clinical data for dicycloplatin from the Chinese medical literature. We also report on new mechanistic findings in our laboratory in West Virginia, USA. Patient blood samples were collected for DCP-prototype determination by liquid chromatography mass spectrometry (LC-MS/MS). Molecular studies of ovarian cancer cells treated with DCP or cisplatin were carried out for gene-signature profiling using immunoblotting. Pharmacokinetic mass-spectrometry showed different spectrum profiles of DCP and carboplatin in plasma. Plasma concentration of DCP prototype was 17.1 μg/ml 2h after administration, with a peak concentration of 26.9 μg/ml at 0.5 h. Immunoblotting showed DCP-induced activation of DNA damage pathways, including double-phosphorylated checkpoint kinase 2 (CHK2) and breast cancer 1 (BRCA1) and triple-phosphorylated p53, compared to controls. Cisplatin produced a similar profile, with increased p53 protein. DCP and cisplatin activate DNA-damage response through similar pathways. DCP may be more soluble and stable, and better-tolerated.
Dicycloplatin is a novel platinum analog developed in China. It was approved by the State Food and Drug Administration (SFDA) of China in 2012. This compound is in the midst of pre-clinical and early-phase clinical evaluation in China. Herein we review findings that have appeared in the Chinese medical literature and report our mechanistic studies.
DCP: Chemical Structure and Properties
DCP is a platinum-based compound synthesized from platinum powder through 10 steps. The host part of DCP is carboplatin and the guest part is 1,1-cyclobutane dicarboxylic acid (CBDCA); they are linked by hydrogen bonds (Figure 1). DCP is a hydrogen-bond supramolecule; carboplatin is a covalent-bond molecule. This may explain why DCP seems to have a more stable chemical structure, good water solubility, and a better safety profile than carboplatin (1). The crystal structure of DCP has been determined. In aqueous solution, using electrospray ionization mass spectrometry (ESI-MS), a model of the structure of DCP was soluble and stable (1). No chemical decomposition of DCP in aqueous solution was observed after several years of storage at room temperature.
Pre-clinical Pharmacology
Early in vitro cytotoxicity studies of DCP in China demonstrated significant antitumor activity against a variety of human cancer cell lines, including BCG-823, BEL-7402 and EJ with an IC50 range of 25-30 μM (2-5). Studies of DCP-induced apoptosis in prostate cancer PC3 cells and lung cancer NCI/H446 cells showed that DCP can induce cell-cycle arrest, inhibit proliferation and lead to apoptosis (6-8). In a study of platinum DNA damage caused by cisplatin, carboplatin and DCP, researchers found the breaking potency on pBR322 plasmid DNA to be cisplatin>DCP>carboplatin (9).
The chemical structure of dicycloplatin and a computer-generated structural model of the dicycloplatin supramolecule in aqueous solution. Illustrative representation of the data originating from Yang et al. (1).
DCP has also shown in vivo anti-tumor effects against the experimental A549 human non-small cell lung cancer (NSCLC) tumor xenograft model, with 67-90% inhibition of tumor growth compared to 51-63% by carboplatin (Figure 2) (5, 10, unpublished data). In summary, DCP shows significant antitumor activity. These data provide a favorable pre-clinical profile for DCP.
Pre-clinical Toxicology
Acute toxicity was investigated by single intravenous and intraperitoneal administrations of different doses. The results indicate that the toxicity profile of DCP [Lethal dosage (LD50)=210 mg/kg] is higher than that of carboplatin (164 mg/kg) and cisplatin (14.27 mg/kg) (5, 11). Studies of cytotoxicity and hematological toxicity among DCP, cisplatin and carboplatin found that DCP induced renal toxicity in Sprague-Dawley (SD) rats significantly less than cisplatin, with similar episodes of bone-marrow suppression as those found with carboplatin (12, 13). In vivo evaluation of DCP toxicity suggests that DCP causes the least damage to major target organs of rats, and there exist the usual platinum-associated problems with pregnancy (14, 15). These reports provided a generally favorable safety profile for DCP, supporting the possibility for its clinical evaluation.
In vivo comparison of therapeutic efficacy of dicycloplatin (DCP) and carboplatin in A549 NSCLC xenografts. a: DCP and carboplatin in treatment of A549 NSCLC xenografts. b: Images of mice treated with DCP compared to controls. Data provided by Sopo-Xingda Pharmaceutical, Inc.
Pre-clinical Pharmacokinetics
Pharmacokinetic studies of DCP in comparison to cisplatin and carboplatin in rats suggest that DCP and carboplatin have faster renal excretion than cisplatin, and transfer proteins may be involved in excretion of DCP and its metabolites (16). Substrates of drug transporters, such as probeneside and verapamil, can reduce the renal excretion of platinum if used in combination with DCP. P-glycoprotein [P-gp] organic anion transporter (OAT) or organic anion-transporting polypeptide (OATP) may be involved in the excretion of DCP (17). A pharmacokinetic study of DCP in dogs reported that the maximum plasma concentration (Cmax), the areas under concentration (AUC) and dosage appeared linearly-correlated, and the summarized plasma pharmacokinetic parameters in rats and dogs following DCP treatment are shown in Table I. Zhao et al. reported findings in rats and dogs that plasma concentration of platinum originating from DCP decreased rapidly within the first 4h and declined slowly long-term (Figure 3a,b). This study also showed distribution of DCP with the highest concentration in plasma and some in intestine and prostate (Figure 3c) (18).
Plasma pharmacokinetic parameters in rats and dogs following dicycloplatin administration.
Early-Phase I Clinical Trial
In the early clinical phase I and phase II trials, all patients signed informed consent according to Chinese regulatory guidelines (19).
A phase I clinical and pharmacokinetic study to determine efficacy and safety was conducted at State Key Laboratory of Oncology in Southern China (20). DCP was administered to 29 patients with cancer at escalating doses from 50 mg/m2 to 650 mg/m2. Pharmacokinetic analysis was performed for 26 patients to determine the total and ultrafiltered platinum concentrations in plasma. The results demonstrated a linear pharmacokinetics characterized by distribution half-life of 1.49±0.39 h and low steady-state apparent volume of distribution (177.88±44.92 l/m2). The maximum tolerated dose of DCP was 550 mg/m2. The terminal plasma half-life of total platinum ranged from 41.86 to 77.20 h without significant dose dependency; however, the half-life of free platinum ranged from 42.34 to 61.07 h. In conclusion, DCP displayed a favorable safety profile at doses between 50 mg/m2 and 550 mg/m2 and first efficacy signals were observed. Dose-limiting toxicities were thrombocytopenia, anemia and emesis. The recommended starting dose for subsequent phase II studies is 450 mg/m2 once every three weeks (20).
Pharmacokinetics of platinum originating from dicycloplatin in rats and dogs. a: Concentration–time profiles of platinum originating from dicycloplatin in plasma following intravenous administration in male rats (dicycloplatin, 10 mg/kg) (mean±SD; n=6). b: Concentration–time profiles of platinum originating from dicycloplatin in plasma following intravenous administration in male dogs (dicycloplatin, 5 mg/kg) (mean±SD; n=6). c: Tissue distribution of DCP in rats. Mean concentrations of platinum in plasma and tissues at 6, 30 and 120 min after intravenous administration of dicycloplatin at a dose of 60 mg/kg (mean±SD, n=5). Illustrative representation of the data originating from Zhao et al. (18).
Dicycloplatin in the treatment of lung cancer with brain metastasis. Data provided by Sopo-Xingda Pharmaceutical, Inc.
Dicycloplatin may pass the blood-brain barrier demonstrated by chemosensitivity assay in brain tissue (data provided by Sopo-Xingda Pharmaceutical, Inc) and by theoretical calculation.
Phase II Clinical Trials
The first phase II clinical trial of DCP followed 39 patients with small cell lung cancer was conducted in 2003. Liu and colleagues reported an overall improved effect and tolerance for the novel platinum drug (21).
In Jan 2007, randomized multicenter phase II clinical trials among chemotherapy-naïve patients with advanced NSCLC using DCP plus paclitaxel, compared to carboplatin plus paclitaxel, was approved by Chinese FDA (19, 22, 23). These multi-center trials involved 15 hospitals nationwide and enrolled 138 patients who met the criteria (22). Patients were randomly assigned to the experimental group (DCP-plus-paclitaxel) or control group (carboplatin-plus-paclitaxel).
Unfortunately, there is no overall report; separate institutional articles were published by Cancer Center of Sun Yat-Sen University (32 patients) in 2009 (19) and Jiang-Su Tumor Hospital (20 patients) in 2011 (23). In both reports, the two arms were balanced with regard to gender, age, ECOG (Eastern Cooperative Oncology Group) performance status, disease stage and histological classification. The experimental group was given DCP at 450 mg/m2 plus paclitaxel at 175 mg/m2 q3w; the control group received carboplatin AUC=5 plus paclitaxel at 175 mg/m2 q3w. Response rate and adverse reactions were evaluated according to the RECIST criteria with survival follow-up. Results from the Sun Yat-Sen University showed a 1-year survival rate of 54.8% in the DCP arm versus 20.1% in the control arm (p=0.028) (19). At the Jiang-Su Tumor Hospital, the median progression-free survival time was 4.1 months in the control arm and 4.9 months in the DCP arm (p<0.05); there was no significant difference in the response rate, median overall survival time and 1-year survival rate between the two arms. However, the 3-year survival rate was 22.2% in the DCP group, versus 11.1% in the control group (23). Adverse reactions in these trials included myelosuppression and gastrointestinal reactions. Lympho-cytopenia and bone-marrow suppression were more frequent in the control group. No significant difference was found for other adverse reactions between the two groups. There was one report that the adverse reactions caused by DCP resulted in pancytopenia (24). Researchers concluded that efficacy and safety of the DCP-plus-paclitaxel regimen were comparable to those of carboplatin plus paclitaxel, with better tolerance of DCP-plus-paclitaxel among chemotherapy-naïve patients with advanced NSCLC (19, 22, 23).
Mass spectrometry for dicycloplatin and carboplatin. a: dicycloplatin spectrum. b: carboplatin spectrum.
Mass spectrometry demonstrates a high level of prototype of DCP in plasma 2-h after administration. Thirty minutes after DCP administration, peak plasma concentration of DCP is 26.9 μg/ml. Two hours after administration, 17.1 μg/ml of prototype DCP was measured in patient plasma, compared to 5.01 μg/ml of carboplatin.
DNA-damage response induced by cisplatin (CDDP) and dicycloplatin (DCP) in ovarian cancer A2780 cells. Cells were treated with cisplatin or dicycloplatin for 1 h. Cells were washed and re-incubated in drug-free media for the indicated times. Proteins were extracted, separated by polyacrylamide gel electrophoresis (PAGE), transferred onto PVDF membrane, and probed with the indicated antibodies.
A Case Report that Hints DCP Can Cross the Blood–Brain Barrier (BBB)
A case report was provided by Sopo-Xingda Pharmaceutical, Inc. In brief, a 42-year-old male with lung cancer (upper apical region of right lung) was diagnosed in June 2006 at Beijing Hospital based on Positron Emission Tomography/Computerized Tomography (PET/CT). The patient showed significantly increased metabolic activity of the mass and high SUV (standardized uptake value) at delayed imaging. Later, the cancer spread to lymph nodes, mediastina and meninges. An encephalic tumor was also found. A regimen of large-dose DCP (1200 mg by i.v. q3w) and oral Iressa was begun as adjuvant treatment to support cranial radiotherapy at PLA General Hospital in Beijing. This regimen continued for four months with five high-dose treatments of DCP and whole-brain irradiation to 38 Gy in 19 daily fractions, followed by a boost to the lesion area with additional 22 Gy delivered in 11 daily fractions. During treatment, the patient developed slight leucopenia and thrombocytopenia which were resolved by symptomatic treatment. The patient was discharged on 2/1/2008 with normal white blood cell count, platelet count, biochemistry, and blood coagulation tests. Magnetic resonance imaging (MRI) on 1/25/2008 showed no brain lesion (Figure 4).
That successful treatment suggested the possibility that DCP crosses the BBB. In vitro data of histoculture drug response assay [MTT: 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] in various tumor specimens demonstrated DCP chemosensitivity in brain tissue. As shown in Figure 5 (right panel), DCP sensitivity in glioma tissues is much higher compared to other chemotherapeutic agents (data provided by Sopo-Xingda Pharmaceutical, Inc.). Worth noting here, Zhao et al. reported that a visible distribution of platinum originating from DCP was seen in brains of rats (18). Theoretically, a larger polar surface area (PSA) hints at greater solubility and a higher calculated logP (ClogP) value predicts a better permeability than a smaller PSA and logP (25-27). Therefore, we calculated the PSA and logP for cisplatin, carboplatin and DCP. As shown in Figure 5 (left panel), DCP has the largest PSA and the largest ClogP. Thus, DCP seems to have better solubility and permeability based on these characteristics, with greater likelyhood of crossing the BBB. Thus, DCP has the greater likelihood of crossing the BBB.
Still, mechanistic questions remain. What is the effect-driving part of the compound? Is DCP hydrolyzed to carboplatin inside the body? What is its mechanism of action? To address these questions, we conducted molecular mechanism studies in our laboratory in light of the clinical pharmacokinetic studies of DCP in China. Our findings are presented in the following section.
Patients and Methods
Researchers at the Cancer Center of Sun Yat-Sen University, Guangzhou, China analyzed plasma level of DCP prototype in four patients. Patients gave informed consent according to Chinese regulatory guidelines (19). Patients received 450 mg/m2 of DCP by i.v. within 1-h. Blood samples were collected 30-min prior to DCP treatment and at 30-min, 1-h, 1.5-h and 2-h afterwards. The whole blood samples were mixed and immediately centrifuged for 2 min at 1,000 × g to yield plasma. Aliquot of 0.2 ml of plasma samples was added with 0.2 ml ice-cold acetonitrile; the mixtures were stored at −80°C.
Pharmacokinetic mass spectrometry. Preparation of standard drug solution: 8.1 mg of DCP (Beijing Sopo-Xingda, China) and 12.5 mg of carboplatin were added to 75% and 50% methanol, respectively, and ultracentrifuged for 3 min, each mixed for 5 min to obtain concentrations of 810 μg/ml for DCP and 1,250 μg/ml for carboplatin. Aliquots of DCP and carboplatin were stored at −20°C for instrument operation.
Prior to analysis, 0.2 ml of plasma was added to 0.2 ml ice-cold acetonitrile, then mixed and centrifuged at 14,500 × g for 10 min; 0.1 ml of supernatant was then diluted with 0.7 ml of ice-cold acetonitrile, with 10 μl used for the analysis. AP-4000™ LC-MS/MS instrumentation was used to determine the prototype level of DCP.
Molecular mechanistic studies at WVU Cancer Center are summarized below.
Cell culture and drug treatment. Human ovarian cancer A2780 cells were propagated as an adherent monolayer in RPMI-1640 medium supplemented with 10% heat-inactivated fetal bovine serum (FBS), 100 μg/ml penicillin and 100 μg/ml streptomycin (Life Technologies, Grand Island, NY, USA). Cells were grown at 37°C in a humidified atmosphere consisting of 5% CO2, 95% air and allowed to grow to 80% confluence. For drug treatment experiments, cells were plated the day before exposure to drugs.
Cisplatin (Sigma-Aldrich, St. Louis, MO, USA) was prepared fresh by first dissolving it in phosphate-buffered saline (PBS), without Ca++ or Mg++, at a concentration of 1 mg/ml, then diluting it into pre-warmed media to achieve the IC50 dose of 5-day survival (3 μM). DCP (>99.0% purity; Sopo-Xingda, Beijing, China) was prepared by first dissolving it in water and then diluting it in pre-warmed media to achieve the concentration of 500 mg/ml. Plated cells were allowed to grow for 24-h and treated with cisplatin or DCP for 1-h. At the end of 1-h exposure to drug, cells were washed twice with PBS, and then incubated with drug-free media for a required period of time.
Protein extraction and immunoblotting. Treated and untreated cells were extracted with whole cell lysis buffer [20 mM Tris-HCl (pH 7.5), 150 mM NaCl, 1 mM Na2EDTA, 1 mM EGTA, 1% Triton, 2.5 mM sodium pyrophosphate, β-glycerophosphate, 1 mM Na3VO4, 1 mM phenylmethylsulfonyl fluoride (PMSF), 10 μg /ml leupeptin, 10 μg/ml aprotinin and 5 μg/ml pepstatin] for 30 min before centrifugation (at 16,000 × g for 30 min at 4°C). Supernatant was collected as whole-cell lysate for western blot analysis. Protein concentrations of extracts were determined by Bio-Rad Protein Assay kit (Bio-Rad, Hercules, CA, USA) with bovine serum albumin as standard.
The whole-cell lysates were separated on 12% sodium dodecyl sulfate (SDS)-polyacrylamide gels and transferred to PVDF (polyvinyl difluoride) membrane (Bio-Rad) using standard electrophoresis and electroblotting procedures. Pre-stained molecular weight markers were purchased from Invitrogen (Carlsbad, CA, USA). To reduce non-specific binding, blots were pre-incubated for 1-h in a blocking buffer (5% nonfat dry milk, 1× TBS and 0.1% Tween 20). Membranes were then incubated with primary antibodies overnight at 4°C. The primary antibodies applied were against ataxia telangiectasia mutated (ATM), p-p53 phosphoserine-15, p53, CHK2 phosphothreonine-68, anti-CHK2, BRCA1 phosphoserine-1497, BRCA1, BRCA2, r-H2AX (H2A histone family member X), p48 and p21. To demonstrate equal loading of each sample, membranes were re-probed using antibody to β-actin (Sigma-Aldrich Corp. St. Louis, MO, USA). The signals of immunoreactive proteins were visualized using horseradish peroxidase-conjugated sheep anti-mouse or donkey anti-rabbit antibodies and enhanced by Chemiluminescence ECL detection system (Amersham International PLC, Buckinghamshire, UK).
Results
In China, prototype concentrations of DCP were determined using AB-4000™ LC-MS/MS instrumentation. Carboplatin prototype level was also determined. The results demonstrated that the spectrum-profiles of DCP and carboplatin are different, with a peak plasma-concentration for DCP molecule at 514.0 and a peak plasma concentration for the carboplatin molecule at 372.0 (Figure 6). DCP prototype levels in blood samples from four patients who received DCP treatment were also analyzed. In plasma, 2-h after DCP administration, 17.1 μg/ml of prototype DCP was still present following a peak concentration of 26.9 μg/ml at 0.5-h. In comparison, 5.01 μg/ml of carboplatin was seen 2-h after drug administration (Figure 7).
At WVU, for in vitro mechanistic investigations, we treated ovarian cancer cells with cisplatin or DCP, then allowed cells to grow for various periods of time before isolation for cellular proteins to be probed with antibodies selected from the DNA-damage repair signaling pathways.
Results of immunoblotting showed DCP-induced activation of the DNA damage response pathways, including phosphorylations of CHK2 at threonine 68, p53 at serine 15, and BRCA1 at serine 1497. The increases in p-Chk2, p-p53 and p-BRCA1 after 1-h DCP exposure occurred in a time-dependent manner. Quantitative analysis of phosphorylated CHK2, p53 and BRCA1 expression showed that the amount of DCP-induced phosphorylation of CHK2 and BRCA1 doubled at 48-h and tripled for phosphorylated p53 at 24-h, compared to controls (Figure 8, right panel). p53 protein expression did not seem changed after DCP treatment; however, an increase in p21 expression was seen at 48 h.
Activations induced by cisplatin in the DNA-damage response pathways were also observed. After 1-h cisplatin exposure at IC50 dose, we observed increased protein expressions of ATM and r-H2AX. The peak of CHK2 phosphorylation at Thr-68 was seen at 48 h. At 12 h after cisplatin treatment, the p53 protein level was high, accompanied by p53 phosphorylation at serine 15. In addition, cisplatin-mediated activation of p53 resulted in activation of downstream proteins p48 and p21 (Figure 8, left panel).
Discussion
As the structure of DCP is composed of carboplatin and CBDCA, it is important to consider if DCP is hydrolyzed to carboplatin inside the body. Chinese researchers measured prototype DCP concentration in patient blood samples after DCP administration, with carboplatin plasma concentration as comparison. The results showed that the majority of DCP is still present as prototype DCP in plasma 2-h after administration, with a small portion of carboplatin hydrolyzed from DCP during drug metabolism. Unfortunately, there was no quantified ratio of DCP conversion to carboplatin. Further evidence, including clinical pharmacokinetics data in more patient samples, is needed.
Li and colleagues report that DCP inhibited the proliferation of cancer cells and increased the percentage of apoptosis, including reactive oxygen species (ROS) (8). Their observations include collapse of mitochondrial membrane potential, release of cytochrome c from the mitochondria to the cytosol, and upregulation of p53, which were accompanied by activation of caspase-9, caspase-3, caspase-8, and poly (ADP-ribose) polymerase cleavage. They concluded that DCP induces apoptosis through reactive oxygen species stress-mediated death receptor pathway and mitochondrial pathway, which is similar to carboplatin. It would be interesting to confirm if this is the pathway for all platinum drug-induced apoptosis.
Our studies of DNA-damage response induced by DCP, compared to those induced by cisplatin, demonstrate that several major kinases of the DNA-damage response pathways are activated by the two platinum drugs. These observations suggest that DCP and cisplatin share similar mechanisms in platinum-induced DNA damage and cellular response, indicating DCP is a platinum analog (28-32). Meanwhile, of note, we found that DCP does not increase p53 protein level in DCP-treated cells. That is different from what we observed in the same cisplatin-treated cell model. Further investigation is needed to determine if the difference between cisplatin and DCP are relevant to clinical outcomes.
In summary, improving therapeutic efficacy and reducing adverse reactions associated with platinum chemotherapy are primary goals of cancer research (33, 34). Developing new chemotherapeutic agents with higher efficacy and lower toxicity is an important strategy. DCP seems to have unique chemotherapeutic properties and may pass the BBB and prostate lipid membrane. Further investigation of molecular mechanisms of the three platinum agents (cisplatin, carboplatin and DCP) is underway.
Acknowledgements
We acknowledge the Molecular Medicine Core Facility, Mary Babb Randolph Cancer Center, West Virginia University and Jiang-Su SOPO (Group) CO., Ltd. in China for supporting this study. We thank Dr. Chu-Biao Xue (Incyte Corp, USA) for help in calculating PSA and logP for platinum compounds; Dr. Su Li (Sun Yat-Sen University, China) for prototype-drug measurement; and Dr. Xiaobing Liang (WVU, USA) for western blot experiments.
- Received August 21, 2013.
- Revision received November 26, 2013.
- Accepted November 27, 2013.
- Copyright© 2014 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved