Purpose: The clinical success of arsenic trioxide (As2O3) in hematologic malignancies has not been replicated in solid tumors due to poor pharmacokinetics and dose-limiting toxicity. We have developed a novel nanoparticulate formulation of As2O3 encapsulated in liposomal vesicles or “nanobins” [(NB(Ni,As)] to overcome these hurdles. We postulated that nanobin encapsulation of As2O3 would improve its therapeutic index against clinically aggressive solid tumors, such as triple-negative breast carcinomas.
Experimental Design: The cytotoxicity of NB(Ni,As), the empty nanobin, and free As2O3 was evaluated against a panel of human breast cancer cell lines. The plasma pharmacokinetics of NB(Ni,As) and free As2O3 were compared in rats to measure drug exposure. In addition, the antitumor activity of these agents was evaluated in an orthotopic model of human triple-negative breast cancer.
Results: The NB(Ni,As) agent was much less cytotoxic in vitro than free As2O3 against a panel of human breast cancer cell lines. In contrast, NB(Ni,As) dramatically potentiated the therapeutic efficacy of As2O3 in vivo in an orthotopic model of triple-negative breast cancer. Reduced plasma clearance, enhanced tumor uptake, and induction of tumor cell apoptosis were observed for NB(Ni,As).
Conclusions: Nanobin encapsulation of As2O3 improves the pharmacokinetics and antitumor efficacy of this cytotoxic agent in vivo. Our findings demonstrate the therapeutic potential of this nanoscale agent and provide a foundation for future clinical studies in breast cancer and other solid tumors. Clin Cancer Res; 16(14); 3607–17. ©2010 AACR.
Arsenic trioxide (As2O3) is a highly effective therapy for acute promyelocytic leukemia and has antitumor activity in preclinical models of solid tumors; however, clinical trials of As2O3 in several solid tumors reveal a narrow therapeutic window that limits wider application. Nanoscale drug carriers have increased the therapeutic index of several cytotoxic agents by increasing tumor drug delivery, enhancing antitumor efficacy, and attenuating systemic toxicity. We have developed a novel high-density nanoparticulate formulation of As2O3 that is encapsulated in 100-nm liposomal vesicles or nanobins. Nanobin encapsulation of As2O3 [NB(Ni,As)] dramatically improves pharmacokinetic properties of the active agent and leads to greater therapeutic efficacy compared with free As2O3 in an orthotopic model of triple-negative breast cancer. Moreover, we show that NB(Ni,As) is well tolerated in this model, suggesting that this nanoscale platform may have the potential to expand the clinical utility of As2O3 as a cancer therapeutic agent.
Breast cancer is the second leading cause of cancer mortality for women in the United States (1). Although preventive agents and targeted therapies directed at the estrogen receptor, progesterone receptor, and human epidermal growth factor 2 receptor (HER2/neu) have resulted in improved clinical outcomes for many women with breast cancer, formidable challenges remain in treating tumors that do not express these molecular targets. These “triple-negative” breast carcinomas represent 15% of newly diagnosed breast cancer cases and often exhibit a basal epithelial or basal-like gene expression profile that is associated with poor survival (2–4). Consistent with its aggressive nature, triple-negative breast cancer is characterized by high rates of distant recurrence, particularly in the lung and brain, within the first five years after diagnosis despite adjuvant chemotherapy (5, 6). Hence, development of new therapeutic agents for these clinically intractable tumors is highly desirable.
Arsenic trioxide (As2O3) is a Food and Drug Administration–approved treatment for refractory acute promyelocytic leukemia (APL) and has shown preliminary activity in patients with relapsed/refractory multiple myeloma (7–10). Several mechanisms of action have been proposed for As2O3 activity, including induction of apoptosis mediated by reactive oxygen species, promotion of cellular differentiation, and inhibition of angiogenesis (9, 11–13). As2O3 has also been shown to reduce migration and invasion of cervical and ovarian cancer cells in vitro (14, 15). Preclinical studies of As2O3 have shown antitumor activity in murine solid tumor models, including breast, brain, liver, gastric, prostate, renal, and bladder cancer (16, 17). Unfortunately, little or no efficacy has been observed in clinical trials of As2O3 when evaluated in solid tumors (7). Two factors seem to have limited the utility of As2O3 in the clinic: rapid renal clearance, which limits tumor uptake, and dose-limiting toxicity (18–20).
Nanoscale drug delivery vehicles have been reported to increase the therapeutic index of cytotoxic drugs by prolonging their serum half-lives, increasing tumor accumulation, and reducing systemic toxicity (21–23). Liposomes are one such nanoscale delivery vehicle that can be used to deliver cytotoxic payloads (24, 25). Two technical advances have greatly improved the clinical utility of nanoliposomes: (a) passivation of liposomes with polyethelene glycol (pegylation) to reduce opsonization and prolong serum half-life, and (b) nanoscale (100 nm diameter) synthesis of liposomes to enhance tumor accumulation via extravasation of nanoliposomes through fenestrated tumor vasculature. The latter property of nanoscale drug delivery vehicles is known as the enhanced permeability and retention effect (EPR) and enables passive targeting of tumors (22, 26).
Previous attempts at encapsulating As2O3 in liposomes have been reported; however, these formulations were not stable, resulting in rapid leakage of the active agent (27). The predominant form of As2O3 in aqueous solution at physiologic pH is arsenous acid [As(OH)3]; it rapidly crosses lipid bilayers, so passive encapsulation of the drug in these previous formulations resulted in rapid loss of As(OH)3 from the liposomal vesicles (27–29). We have developed a novel nanoparticulate formulation of As2O, in which the As2O3 is stabilized as a nanoscale precipitate inside a pegylated 100-nm liposome (Fig. 1A). These arsenic nanobins [NB(Ni,As)] are characterized by a nanoparticulate core containing extremely high densities of arsenic (>270 mmol/L) and Ni2+ cations that both stabilize and potentiate drug activity (30, 31). NB(Ni,As) is stable for at least 12 months at 4°C with <10% leakage of free As2O3. The release of arsenic from NB(Ni,As) is triggered by exposure to low pH environments (5.0-6.5), such as the acidic tumor milieu and inside endocytotic vesicles in tumor cells and tumor macrophages. We had previously shown that sequestration of As2O3 inside the nanobins attenuates in vitro cytotoxicity compared with free As2O3 because the As2O3 inside the vesicle is not bioactive until released (30–32). These results led us to postulate that the nanobin formulation may also reduce the systemic toxicity of As2O3 by shielding normal tissues from the cytotoxic agent, which may overcome some of the limitations observed with free As2O3 in clinical studies.
In this study, we evaluated the in vitro and in vivo activity of our novel NB(Ni,As) using breast cancer cells and a mouse model of triple-negative breast cancer. We hypothesized that NB(Ni,As) would enhance the antitumor activity of As2O3 by improving its pharmacokinetics in vivo, promoting tumor uptake of As2O3 via the EPR effect, and reducing systemic toxicity by shielding nontumor tissues from drug exposure. We show here that NB(Ni,As) exhibits improved pharmacokinetic characteristics and enhanced therapeutic efficacy in an orthotopic model of triple-negative breast cancer compared with the free drug. Our findings provide the first proof-of-principle evidence for NB(Ni,As) antitumor efficacy in vivo and suggest that NB(Ni,As) may represent a novel therapeutic agent for triple-negative breast cancer as well as other solid tumors.
Materials and Methods
1,2-distearoyl-glycero-3-phosphocholine (DSPC) and 1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N[methoxy(polyetheylene glycol)-2000] (ammonium salt) were purchased from Avanti Polar Lipids. Cholesterol (Chol), arsenic trioxide, nickel(II) acetate, sodium chloride, HEPES, Blasticiden-S, arsenic trioxide, and Sephadex G-50 were obtained from Sigma, and were the highest grade available. Nickel(II) acetate was also purchased from Strem Chemical. The As2O3 used in the pharmacokinetic study was Trisenox (Cephalon). Trace metal grade nitric acid (69%) for inductively coupled plasma mass spectrometry (ICP-MS) was purchased from Fisher Scientific.
Preparation of arsenic trioxide–loaded nanobins
Arsenic trioxide–loaded nanobins were prepared from a modified procedure developed in our laboratory (31). Briefly, a thin lipid film consisting of DSPC/Chol/DSPE-PEG2000 = 51/4/45 mol % was prepared by dissolving the lipids in chloroform followed by rotary evaporation to dryness in a round bottom flask. The resulting lipid film was placed under high vacuum overnight to remove any residual solvent. The dry lipid films were hydrated with either (a) 300 mmol/L Ni(OAc)2 for arsenic trioxide loaded nanobins [NB(Ni,As)] or (b) 20 mmol/L HEPES, 150 mmol/L NaCl, pH 7.4 buffer for vehicle control nanobins [NB(NaCl)] at 60°C for 1 hour. The hydrated lipid suspensions were subjected to 10 freeze-thaw cycles (alternating between an ethanol/dry-ice bath and 60°C water bath). The hydrated lipids were downsized to 100 nm using a Lipex Extruder (Northern Lipids) operated at 60°C sequentially through 200- and 100-nm polycarbonate filters (Whatman International). Buffer exchange to remove unencapsulated Ni(OAc)2 was done using a Sephadex G-50 column equilibrated with 20 mmol/L HEPES, 150 mmol/L NaCl, pH 6.8 or with continuous diafiltration with tangential flow filtration (TFF) using Minimate TFF cartridges (100 kD MWCO) or Spectrum Labs Hollow Fiber Cartridges (500 kD MWCO) with 20 mmol/L HEPES, 150 mmol/L NaCl (pH 6.8). The resulting 100-nm vesicles were either (a) incubated with a As2O3 solution for 2 hours at 60°C to generate NB(Ni,As), or (b) pH adjusted to 7.4 in the case of NB(NaCl). Extraliposomal arsenic trioxide was removed from NB(Ni,As) using a Sephadex G-50 column preequilibrated with 20 mmol/L HEPES, 150 mmol/L NaCl, pH 7.4 or with TFF as described above into 20 mmol/L HEPES, 150 mmol/L NaCl (pH 7.4). The molar ratios of As/P (phospholipid) were measured using an inductively coupled plasma-optical emission spectrometer (ICP-OES; Vista MPX). The size and stability (pH, thermal) of the nanobins was determined by transmission electron microscopy on a Hitachi HF2000 electron microscope and by dynamic light scattering on a Zetasizer Nano ZS (Malvern Instruments).
In vitro drug release assay
NB(Ni,As) was mixed with fetal bovine serum (FBS) in a volume/volume ratio of 2:8 (80% serum), resulting in a final lipid concentration of 1 mmol/L (pH 7.4 or pH 5.5), 37°C. At various time points, aliquots were applied to a Sephadex G-50 column to remove the arsenic and nickel species that had leaked out of the nanobins. The excluded nanobin fractions were evaporated prior to digestion with concentrated trace-metal free grade HNO3 (69%) before analysis with ICP-OES to determine the drug to lipid molar ratios. The drug release percentage (%) was calculated as [(ro-ri)/ro] × 100%, where ro is the initial As(Ni) to lipid molar ratio and ri the As(Ni) to lipid molar ratio at a specific time point (30).
Unless indicated otherwise, human cell lines were purchased from the American Type Culture Collection (ATCC) and were grown in media containing 4 mmol/L L-glutamine and 100 units/mL of penicillin/streptomycin. All sera, media, and additives were purchased from Invitrogen unless otherwise noted. MDA-MB-231 and BT-20 breast cancer cells were grown in MEM supplemented with 10% FBS (ATCC). MDA-MB-231-mCherry cells were kindly provided by Dr. Jennifer Koblinski (Northwestern University) and were grown in DMEM and Ham's F12 (DMEM/F12) supplemented with 1 ug/mL Blasticidin S (Sigma) and 5 % HI (heat-inactivated)-FBS. SK-BR-3 breast cancer cells were grown in McCoy's 5A media supplemented with 10% FBS (ATCC). MCF-10A cells were grown in phenol red-free DMEM/F12 with 5% heat-inactivated horse serum, 10 μg/mL insulin, 20 ng/mL epidermal growth factor (EGF), 100 ng/mL cholera toxin, and 0.5 μg/mL hydrocortisone.
The cytotoxicity of As2O3 and NB(Ni,As) was assessed by MTS assay using the CellTiter 96 AQueous MTS (Promega). Cells (2.4 ×104 cells/well) were plated in half-area 96-well plates (Greiner Bio-One) overnight in phenol red-free DMEM supplemented with 10% FBS. Serial dilutions of As2O3-loaded nanobins [NB(Ni,As)] and empty nanobins [NB(NaCl)] in phenol red-free media were transferred to the cells. After 96 hours, MTS solution was added to the 96-well plates and the absorbance was measured at 495 nm 2 hours later. The normalized viability of cells was plotted versus the log concentration of elemental arsenic and a sigmoidal dose response was fit (Prism, Graphpad). The IC50 values are based on two to three independent experiments.
Apoptosis was measured by fluorescence-activated cell sorting (FACS) using the Annexin V-PE Apoptosis Detection Kit I (BD Bioscience) following the manufacturer's protocol. Cells were treated with 50 μmol/L As2O3, NB(Ni,As) or media with or without 50 μmol/L Z-VAD-FMK (Promega) for 48 hours and analyzed by flow cytometry. Z-VAD-FMK, a pan-caspase inhibitor, was used to test whether cell death was caspase dependent.
Migration and invasion assays
MDA-MB-231-mCherry cells were grown in serum-free media for 24 hours. Cell migration and invasion were measured as described (33) with modifications. Briefly, cell suspensions (500 μL, 2.5 ×104 cells) were seeded on the top of uncoated (migration assay) and Matrigel-coated (invasion assay) transwells (8-μm pore diameter; BD Biosciences). Serum-free cell suspensions (500 μL) containing NB(NaCl), As2O3 (0.1, 1.0, 10 μmol/L) or NB(Ni,As) (0.1, 1.0, 10 μmol/L) were added to the top chamber of the transwell. The lower chambers contained drug-free, complete media (750 μL) supplemented with 5% serum. Cells invading the lower chamber were stained with 0.5% crystal violet (60% PBS, 40% EtOH), photographed at ×5 magnification with an inverted microscope, and counted. The same cell suspensions (100 μL, 5.0 ×103 cells) were plated in uncoated (migration) and Matrigel-coated (invasion assay) 96-well plates. After addition, 100 μL of complete media were added to the wells, and the viability after 24 hours (migration) or 48 hours (invasion) was determined by MTS assay. The results from at least two independent experiments in triplicate are presented.
Double jugular catheterized 10-week-old female Sprague Dawley (SD) rats (approximate weight of 220 grams) were purchased from Charles River Laboratories. Rats (n = 5) were treated i.v. by the left jugular catheter with 4 mg As equivalents/kg of either the NB(Ni,As) or Trisenox (As2O3). Blood samples were collected in lithium-heparin-coated tubes from the right jugular catheter at 15, 30, 60, 120, 240, 480, and 1,440 min. The blood samples were centrifuged (2,000 g × 5 minutes), and plasma was collected and stored at 4°C. Plasma samples were analyzed for As and Ni concentration by ICP-MS (methods are detailed in Supporting Information).
Noncompartmental pharmacokinetic parameters were determined by the following methods, using WinNonlin Version 4.1 software (Pharsight Corp.). The area under the time concentration curve (AUC∞) was calculated using the linear trapezoidal rule with extrapolation to time infinity. Clearance (CL) was calculated from dose/AUC∞. Apparent volume of distribution (Vd) was calculated from dose/Co (concentration at time zero calculated from extrapolation of the plasma time curve). Plasma half-life (t1/2) was calculated from 0.693/slope of the terminal elimination phase (λ). NCI-Frederick is accredited by AAALAC International and follows the Public Health Service Policy for the Care and Use of Laboratory Animals. Animal care was provided in accordance with the procedures outlined in the Guide for Care and Use of Laboratory Animals (34).
Orthotopic breast cancer model
Human triple-negative MDA-MB-231-mCherry breast cancer cells (1 × 106) suspended in 100 μL of chilled Matrigel (BD Biosciences) were injected bilaterally into the lactiferous ducts of the 4th mammary gland of 5- to 6-week-old female athymic nu/nu mice (Harlan Sprague-Dawley). Twelve days postinoculation, the mice were randomized into four treatment groups (8 mice per group): PBS alone, empty vehicle nanobins [NB(NaCl)], As2O3 (4 mg/kg), and NB(Ni,As) (4 mg/kg). Stock solutions of As2O3 were prepared by dissolving solid As2O3 in 5 mol/L NaOH, then the concentrated stock solution was diluted into PBS and the pH was readjusted to 7.4. Stock solutions of NB(NaCl) and NB(Ni,As) were diluted with PBS. Each group was treated twice weekly for 3 weeks (7 total treatments) by i.p. injection. Tumors were measured with digital calipers and tumor volumes were calculated using the equation VTumor = (w2 × l × π)/6. Mice in which tumors did not take, comprising three from PBS, one from NB(NaCl), and one from As2O3, were excluded from analysis. Mice were weighed twice weekly.
In a second study tissue samples for arsenic biodistribution analysis and histology were obtained from mice (n = 3) treated with PBS, NB(NaCl), As2O3 (4 mg/kg), and NB(Ni,As) (4 mg/kg) for 3 doses or 5 doses in the same schedule described above and sacrificed 48 hours after the last treatment. Tumors (n = 6) were also harvested, with one half fixed in 10% formalin and embedded in paraffin for sectioning and the other half frozen for arsenic determination by ICP-MS. Immunohistochemical staining for active cleaved caspase-3 was done by the R. H. Lurie Cancer Center Pathology Core Facility using an active caspase-3 antibody (Biocare Medical, CP 229) and standard immunohistochemical methods. Three randomly selected fields from each tumor were scored by an individual who was blinded to the treatment status. Cleaved caspase-3 values were normalized to the vehicle NB(NaCl) control group. In a third study the effect of arsenic therapy on organ function was measured in mice treated with PBS, NB(NaCl), As2O3 (4 mg/kg), and NB(Ni,As) (4 mg/kg) on the schedule of the efficacy study. At the termination of the study, blood was collected from euthanized mice and a peripheral smear and metabolic profile was analyzed by RADIL. All animal experiments were conducted under protocols approved by the Animal Care and Use Committee of Northwestern University.
Inductively coupled plasma-mass spectrometry
Tumor and tissue arsenic levels were determined by ICP-MS using a Thermo XSeries II ICP-MS (Thermo-Fisher). Samples were prepared by digesting tissues in 500 μL of concentrated trace metal free grade nitric acid (65-70%) in capped, metal-free falcon tubes for 2 hours at 60°C. At 20-minute intervals during the digestion, the sample tubes were vortexed and vented in a fume hood. After 2 hours, the digests were filtered through a 0.45-μm polytetrafluoroethylene filter into a fresh metal-free falcon tube. For ICP-MS analysis, a portion of the filtered digest was diluted with ultra-pure laboratory grade water (18 MΩ-cm) and an internal standard mixture of Sc, Tb, Y, In, Bi (CPI International) was added. Standards between 0 and 90 ppb were made using a custom mixed element solution (CPI International). The final ICP-MS samples and elemental standards were prepared in a matrix of 2% nitric acid.
Statistical analysis described in experimental sections was done with GraphPad Prism. Statistical significance was determined by one-way ANOVA, followed by Tukey's posthoc analysis for tumor volumes, animal weights, and arsenic levels, or Bonferroni posttest for the invasion and migration assays. Statistical significance was determined by pairwise t-tests for cleaved caspase-3 staining. P < 0.05 was considered significant.
Preparation of arsenic trioxide–loaded nanobins
Previous protocols for producing NB(Ni,As) and NB(NaCl) were scaled up and modified to produce sufficient material for xenograft studies (30, 31). Encapsulation of arsenic trioxide in nanobins is represented schematically in Fig. 1A. To measure loading efficiency, we determined the arsenic to phospholipid ratio by ICP-OES (As:P) to be 0.45:0.50 which is consistent with our previous work (30, 31). Characterization of NB(Ni,As) by transmission electron microscopy revealed intact 100-nm particles (Fig. 1B). Dynamic light scattering established the average diameter of the NB(Ni,As) particles to be 95 nm for both preparations. A low polydispersity index (PDI = 0.06-0.09) revealed highly homogenous particle size. The hydrodynamic radius of the NB(Ni,As) was also measured in water, 10 mmol/L NaCl, and in PBS at multiple dilutions, and the size was not dependent on concentration or the dispersing media (Supplementary Fig. S1 and Supplementary Table S1). The thermal stability of the NB(Ni,As) was evaluated from 20°C to 60°C in 5°C increments, revealing no significant change in diameter (Supplementary Fig. S2 and Supplementary Table S2). The effect of pH (2.3-11.3) on the hydrodynamic size of the nanobins was evaluated by titration, and the pH did not affect the diameter in the pH range tested (Supplementary Fig. S3 and Supplementary Table S3). Thus, the arsenic-loaded nanobins exhibited remarkable size stability under these conditions. Control empty nanobins NB(NaCl) were similar in size (102 nm diameter) and polydispersity (PDI = 0.06).
The stability of the NB(Ni,As) was assessed by measuring the release of As2O3 in 80% FBS at 37°C, pH 7.4 and at pH 5.5 (Fig. 1C). Over 48 hours, 22% of the total As2O3 was released from the nanobins at physiologic pH, confirming that the majority of the As2O3 is sequestered within the nanobins. At the acidic pH encountered in the acidic tumor interstitium and in the endocytic pathway of tumor macrophages, the NB(Ni,As) undergoes a triggered-release in which >50% of the arsenic is released over 48 hours. Long-term stability measurements show that the NB(Ni,As) does not increase in size over time or release arsenic (Supplementary Fig. S4 and Supplementary Tables S4 and S5).
Nanobin encapsulation of arsenic trioxide attenuates its cytotoxicity in vitro
The cytotoxicity of As2O3, NB(Ni,As), and NB(NaCl) was evaluated using a panel of human breast cancer cell lines and an immortalized mammary epithelial cell line (MCF-10A) by MTS cell viability assay. These human cancer cell lines represent the major subgroups of breast cancer, including estrogen receptor positive (MCF-7), HER2/neu positive (SK-BR-3), and triple negative/basal-like (MDA-MB-231 and BT-20; ref. 35). Dose response curves for As2O3, NB(Ni,As), and NB(NaCl) revealed that nanobin encapsulation of As2O3 resulted in a 3 -to 4-fold increase in the IC50 compared with free As2O3 in all cell lines tested (Fig. 2 and Table 1A). These results corroborate findings in other cancer cell lines (30, 31) and underscore the attenuated cytotoxicity of nanobin encapsulated As2O3. We next examined the cytotoxic mechanism of NB(Ni,As) and free As2O3 in vitro using quantitative flow cytometric analysis of apoptosis by Annexin V labeling (Fig. 3A). MDA-MB-231-mCherry triple-negative breast cancer cells were treated with these agents (50 μmol/L) for 48 hours with or without a 1-hour preincubation with the pan-caspase inhibitor Z-VAD-FMK (50 μmol/L). Consistent with the MTS assay results, NB(Ni,As) was less potent than free As2O3 at inducing Annexin V–positive cells. Moreover, Z-VAD-FMK partially rescued cell death induced by As2O3 or NB(Ni,As). These findings indicate that caspase-mediated apoptosis contributes to the cell death induced by NB(Ni,As) and As2O3.
The effect of NB(Ni,As) and free As2O3 on the migration and invasion of MDA-MB-231-mCherry cells was determined in transwell migration and Matrigel invasion assays. Both NB(Ni,As) and free As2O3 inhibited migration (Fig. 3B) and invasion (Fig. 3C) in a dose-dependent manner. Under the conditions tested, cell viability was not significantly affected during migration (24 hours) or invasion (48 hours; Supplementary Fig. S6), which suggests that these agents have antimigratory and anti-invasive activity.
Intravenous administration of NB(Ni,As) leads to increased plasma arsenic and decreased tissue biodistribution compared with free drug
The plasma pharmacokinetics of As2O3 and NB(Ni,As) were measured in double jugular catheterized, 10-week-old, female SD rats. Intravenous administration of NB(Ni,As) resulted in a dramatically altered plasma concentration profile compared with that of free As2O3 measured by ICP-MS (Fig. 4). NB(Ni,As) exhibited a monophasic decay of arsenic and nickel components, whereas free As2O3 showed a biphasic decay. Selected pharmacokinetic parameters from noncompartmental are shown (Table 1B). The volume of distribution (Vd) of NB(Ni,As) was approximately 50 mL/kg, which is roughly equal to the plasma volume of the rat, indicating minimal distribution of NB(Ni,As) outside of the plasma compartment. The peak (C0) arsenic concentration of NB(Ni,As) was approximately 100 times greater than that of the free drug, whereas clearance (Cl) was decreased ∼300-fold and total exposure, AUC∞, was increased ∼300-fold. Simultaneous measurement of plasma nickel revealed that the nickel levels paralleled plasma arsenic, and had similar C0, Vd, AUC∞ and Cl. Because the nickel is released very slowly from NB(Ni,As) (Fig. 1), the parallel serum levels of As and Ni suggest that NB(Ni,As) is a robust delivery platform that is stable in vivo. The stability and low reticuloendothelial system sequestration of the NB(Ni,As) platform seem to contribute to the favorable pharmacokinetic profile in comparison with free As2O3.
Arsenic trioxide–loaded nanobins inhibit tumor growth in an orthotopic model of triple-negative breast cancer
The antitumor efficacy of NB(Ni,As), empty nanobins, and free As2O3 was evaluated in female athymic nude mice bearing orthotopic human MDA-MB-231 triple-negative mammary tumors. We determined that a well-tolerated dose of NB(Ni,As) was 4 mg/kg (twice weekly; data not shown). Mice were treated with vehicle, empty nanobins [NB(NaCl)], NB(Ni,As) (4 mg As2O3/kg), or free As2O3 (4 mg As2O3/kg) twice weekly by i.p. injection. Although free As2O3 had no effect in the study, the corresponding arsenic concentration-equivalent of NB(Ni,As) robustly inhibited MDA-MB-231 tumor growth (Fig. 5A). Moreover, a higher dose of free As2O3 (8 mg/kg i.p. given on the same dosing schedule) also did not inhibit mammary tumor growth in this model (data not shown). Consistent with the enhanced antitumor efficacy of NB(Ni,As), the concentration of elemental As was 3- to 5-fold higher in MDA-MB-231 tumors treated with NB(Ni,As) than in tumors treated with free As2O3 (Fig. 5B). Mammary tumors from mice treated with NB(Ni,As) had increased caspase-3 activation as determined by cleaved caspase-3 immunostaining compared with vehicle- or empty nanobin–treated mice (Fig. 5C). The observation that free As2O3 induces apoptosis in tumor cells, at least transiently, but does not suppress mammary tumor growth in mice suggests that the improved pharmacokinetics and tumor delivery of nanobin encapsulated As2O3 are responsible for its antitumor efficacy in vivo.
Importantly, treatment of mice with NB(Ni,As) at 4 mg/kg, twice weekly for three weeks, did not cause significant weight loss during the study period (Fig. 5D). To further assess toxicity, we obtained completed blood counts, liver function, kidney function, and serum chemistries (Supplementary Table S8). An isolated elevation of blood urea nitrogen was observed in mice treated with NB(Ni,As); however, electrolytes (sodium, potassium, calcium, phosphorus, and chloride) and creatinine were normal. Moreover, histologic analysis of H&E-stained tissues (skin, kidneys, liver, heart, and lungs) from NB(Ni,As)-treated mice did not reveal any histopathology (Supplementary Fig. S10). Measurement of total arsenic in the liver, kidney, and the heart and lungs after five treatments revealed modest accumulation of arsenic in the liver and kidney (Supplementary Fig. S11). Taken together, our findings indicate that NB(Ni,As) is an effective and well-tolerated therapeutic agent in this murine model of breast cancer.
We observed that encapsulation of As2O3 in a novel nanobin, NB(Ni,As), led to increased in vivo antitumor efficacy of As2O3 in the MDA-MB-231 orthotopic model of human triple-negative breast cancer. NB(Ni,As) inhibited mammary tumor growth at doses (4 mg/kg twice weekly) that are significantly lower than the anticipated efficacious dose of the parent drug based on published reports in other preclinical models of solid tumors (36, 37). Indeed, twice the equivalent dose of free As2O3 had no effect on tumor growth in our study (data not shown). We attribute the enhanced antitumor efficacy in vivo of NB(Ni,As) to the reduced plasma clearance and the increased tumor accumulation of arsenic via the EPR effect of the nanobin platform. Specifically, we observed that the AUC∞, an estimate of drug exposure, was increased approximately 300-fold for NB(Ni,As) compared with free As2O3. Moreover, the tumor arsenic concentrations were 3-5 fold higher in NB(Ni,As)-treated mice. The As2O3 that was delivered to tumors by NB(Ni,As) seemed to be released over an extended time period, leading to metronomic-like dosing compared with free As2O3 and resulting in prolonged antitumor activity even after treatment was stopped (Fig. 5A). In contrast, free As2O3 had no antitumor activity at the dose and schedule used and may have caused a “rebound” tumor growth. The latter phenomenon has been observed in maximal tolerated dose (MTD)-based regimens where neovascularization and tumor growth resume during the necessary recovery periods from drug-induced toxicity (38). NB(Ni,As) may block this tumor rebound effect by loading a tumor with drug that is continuously released and does not follow the peak-trough kinetics typically associated with MTD-based chemotherapy. To that end, NB(Ni,As) may behave like a metronomically dosed or depot agent, although that aspect of NB(Ni,As) pharmacology remains to be explored.
The inhibition of NB(Ni,As)- and free As2O3-induced cell death by Z-VAD-FMK, and the enhanced caspase-3 activation in mammary tumors observed in mice treated with these cytotoxic agents, indicate that the antitumor effects of NB(Ni,As) are at least partly mediated by a caspase-dependent apoptotic mechanism. Although we observed similar rates of mammary tumors apoptosis in the NB(Ni,As) and free As2O3 groups 48 hours after treatment, only NB(Ni,As) reduced tumor burden in vivo. This seeming paradox likely reflects the rapid clearance of free As2O3 from the circulation, leading to transient induction of apoptosis in the mammary tumors and potential rebound growth (36). In contrast, NB(Ni,As) prolongs the pharmacokinetics of As2O3 and increases tumor uptake, resulting in sustained antitumor effects. Intriguingly, free As2O3 and NB(Ni,As) inhibit migration and invasion at concentrations well below their IC50 values, suggesting that these agents may have antimetastatic activity in addition to their cytotoxicity.
Although As2O3 is currently approved for use in APL, several clinical trials in patients with solid tumors have failed to show a clinical benefit of As2O3 at doses in the 0.25 to 0.35 mg/kg/d range (39, 40). APL patients receive 0.16 mg/kg/d of As2O3, and this dose is associated with grade 3/4 toxicities such as peripheral neuropathy, and hepatic and cardiac toxicity (10, 20). These dose-limiting toxicities have limited further dose escalation of As2O3 in other malignancies. Conversion between mouse and human dose levels by body weight (mg/kg) can be estimated by dividing the mouse dose by 12.3 (41). Thus, a 4 mg/kg dose in the mouse is estimated to be equivalent to a human dose of 0.32 mg/kg, which results in a comparable weekly dose of free As2O3 used in APL patients because the nanobin is given twice weekly rather than daily. Hence, clinically efficacious concentrations of NB(Ni,As) are associated with minimal systemic toxicity in this preclinical model.
As noted, we observed that NB(Ni,As) is more effective at suppressing tumor growth in vivo than the parent drug As2O3; however, free As2O3 is much more cytotoxic in vitro. This disparity suggests that standard in vitro cytotoxicity assays of nanoparticle-encapsulated drugs may be a poor predictor of in vivo antitumor activity because they fail to capture the effects of drug encapsulation on pharmacokinetics and biodistribution in vivo. Hence, early in vivo testing of nanoparticle-encapsulated cytotoxic agents in animal models of cancer is of paramount importance and an important component of the development of this class of drugs.
Many potentially effective cancer drugs have been abandoned due to unacceptable systemic toxicity, and poor pharmacokinetics and/or biodistribution (22, 42). We have devised a nanoparticulate platform (nanobin) in which the nanoscale size leads to the concentration and retention of drug in the tumor, while the nanobin encapsulation improves the pharmacokinetic characteristics and potentially the toxicologic properties of the encapsulated drug. This platform has been applied to the encapsulation and reformulation of a highly cytotoxic drug, As2O3, which is currently limited in its use clinically to APL and other hematologic malignancies. The reformulated As2O3 nanobin, NB(Ni,As), differs substantially from the parent drug in both its pharmacologic and efficacy profile. These results provide a foundation for additional preclinical development and future clinical interventions in breast cancer and other solid tumors.
Disclosure of Potential Conflicts of Interest
T.V. O'Halloran and H. Chen are inventors on a patent covering the NB(Ni,As) filed by Northwestern University.
We thank Dr. Andrew Mazar for helpful comments and discussion and Dr. Jennifer Koblinski for providing MDA-MB-231-mCherry breast cancer cells. We acknowledge the Northwestern University Analytical Services Laboratory, High Throughput Analysis Laboratory, the Quantitative Bioelemental Imaging Center in the Chemistry of Life Processes Institute, the Northwestern University Atomic and Nanoscale Characterization and Experimental Center, and the Robert H. Lurie Comprehensive Cancer Center Pathology Core Facility. We also acknowledge Sarah Skoczen and Matthew Hansen at the Nanotechnology Characterization Lab for assistance with nanobin physicochemical characterization and ICP-MS of tissue samples from the pharmacokinetic study.
Grant Support: NIH grants R01GM054111, The Center of Cancer Nanotechnology Excellence U54CA119341, and the Robert H. Lurie Comprehensive Cancer Center Core Grant P30CA060553; the CDMRP Breast Cancer Research Program BC073413 and BC076723, the Breast Cancer Research Foundation, and the Malkin Fellowship. NCL work was funded in whole or in part with federal funds from the National Cancer Institute, NIH, under contract HHSN261200800001E. The content of this publication does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the U.S. Government.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Received January 11, 2010.
- Revision received May 6, 2010.
- Accepted May 11, 2010.