Antitumor Effect of the Atypical Retinoid ST1926 in Acute Myeloid Leukemia and Nanoparticle Formulation Prolongs Lifespan and Reduces Tumor Burden of Xenograft Mice

Acute myeloid leukemia (AML) is a genetically heterogeneous disorder characterized by clonal expansion of myeloid blast progenitor cells in the bone marrow and peripheral blood of patients (1). AML is associated with a highly variable prognosis dictating a high mortality rate with an overall survival exceeding 2 years only in 20% of elderly patients and 5 years in less than 50% of adult patients (2). Furthermore, standard chemotherapy with or without hematopoietic stem cell transplantation leads to 40% cure rates in approximately 40% of adult patients but only in 10% of elderly patients (3). Thus, developing effective and safer therapies remains urgently needed.

Retinoids are natural vitamin A derivatives or synthetic molecules with vitamin A activities that regulate a wide range of biological processes, including development, differentiation, proliferation, and cell death, particularly in hematopoietic cells (4). Differentiation therapy using the natural retinoid all-trans retinoic acid (ATRA) became the paradigm in management of an AML subtype, the acute promyelocytic leukemia (APL; ref. 5). However, in non-APL AML patients, ATRA is only effective on those presenting with nucleophosmin-1 (NPM1) mutations without fms-like tyrosine kinase 3 Internal tandem duplication (FLT-3 ITD) mutations, and is often linked with acquired resistance and disease relapse (4, 6). Arsenic trioxide combined with ATRA displayed a potent and selective efficacy in this category of AML patients both ex vivo and in vivo (7, 8). Although clearance of AML blasts was observed in few treated patients, no cure was achieved. In a randomized clinical trial, the combination of decitabine and ATRA revealed beneficial response rates (9–11). In many trials, retinoid related molecules were developed to overcome ATRA limitations by increasing their specificity and decreasing their toxicity (12, 13). So far, one of the most potent and less toxic retinoid-related molecules is the synthetic adamantyl retinoid ST1926; a derivative of CD437 (14, 15). ST1926 has shown a potent antitumor activity in several in vitro and in vivo cancer models of ovarian carcinoma (16, 17), lung carcinoma (18), neuroblastoma (19, 20), rhabdomyosarcoma (21), teratocarcinoma (22), in vitro two-dimensional and three-dimensional breast cancer models (23), and several leukemia animal models (13, 22, 24). This antitumor activity is independent of retinoic acid receptors (RAR) and p53 signaling pathways. ST1926 displayed a favorable pharmacokinetic profile when compared to CD437 (14) offering promise in cancer therapeutics. ST1926 exhibits its growth inhibitory effects by inducing early DNA damage in various types of tumor cells (18, 21–25). ST1926-induced resistance resulted in delayed and reduced DNA damage, highlighting the critical role of DNA damage pathways in ST1926-mediated cell death (18, 26). ST1926 is orally bioavailable (13), while rapidly achieving effective micromolar (μmol/L) concentrations in mouse plasma (21). Consequently, ST1926 reached phase I clinical trials for patients with advanced ovarian carcinoma (27, 28). However, its development was halted due to rapid glucuroconjugation resulting in sharp decrease of plasma concentrations to sub-μmol/L levels (27).

Nanomedicine has recently gained widespread attention, as it enables more efficient drug delivery, increased stability, and bioavailability, and reduced drug toxicity (29). The promise of nanomedicine in therapeutics is expanding rapidly. It has been applied to several cancer applications where the FDA (USA) has approved some formulations that are currently undergoing clinical trials (30). For instance, Doxil (doxorubicin hydrochloride liposome injection, Orthobiotech) was the first FDA approved nanodrug used to treat metastatic ovarian cancer and Kaposi's sarcoma, where it improved the balance between treatment, efficacy, and toxicity (31). The albumin bound nanoparticle (NP) formulation of paclitaxel, Abraxane (Bioscience, Astra), is currently approved in the clinic against metastatic breast, pancreatic, and non–small cell lung cancer (32, 33). Interestingly, ATRA loaded in polymer poly(lactide-coglycolide) NPs reversed AML cell growth and induced cell differentiation and apoptosis (34). This latter formulation was also shown to increase ATRA's anticancer activity and bioavailability relative to its free form in liver carcinoma models (35).

In this study, we investigated the preclinical efficacy of ST1926 and polymer-stabilized ST1926 NPs (ST1926-NP) in both AML in vitro and in vivo models. We showed that ST1926 at low sub-μmol/L concentrations potently inhibited the growth of several tested human ATRA-resistant AML cell lines, and primary AML patients-derived blasts, while sparing resting and activated normal leukocytes as well as hematopoietic and mesenchymal stem cells (MSCs). ST1926 induced early DNA damage and massive apoptosis in all tested AML cell lines. In a murine AML xenograft mouse model, ST1926 and ST1926-NP significantly reduced tumor burden and prolonged survival. Strikingly, ST1926-NP prolonged survival in AML xenografted animals at four-fold lower concentrations than the naked drug with no detectable toxicity on normal animals. More importantly, ST1926-NP significantly reduced bone marrow leukemia burden in THP-1 xenograft mice and this effect was similar to that observed at four-fold higher concentrations of the naked ST1926. These results highlight the promise of ST1926 in AML therapy and warrant further clinical development of this drug.

THP-1 (DSM ATCC 16), KG1-α (ATCC CCL246.1), MOLM-13 (DSM ACC 554), ML-2 (DSM ACC15), and HEL (DSM ACC 11) human AML cell lines (kindly provided by Dr. Fred Mazurier in 2014) were cultured in RPMI1640 (Lonza) medium supplemented with 10% FBS (Sigma) and 50 U/mL penicillin–streptomycin antibiotics (Lonza). Cell lines were authenticated in December 2016 by Laragen Cell Line Authentication and tested in 2016 for mycoplasma contamination by Hoechst staining. Cells were passaged on average 10× between thawing and use. Primary AML cells from the bone marrow of two patients were collected as previously described (10). Peripheral blood mononuclear cells (PBMC) were obtained from four healthy donors following Ficoll-Hypaque (Lymphoprep) separation. Activated PBMCs were cultured as previously described (22). Human samples were collected after approval by the Institutional Review Board at the American University of Beirut and after consented agreement of patients according to Helsinki's Declaration.

ST1926 (E-3-(4′-hydroxy-3′-adamantylbiphenyl-4-yl) acrylic acid) was kindly provided by Biogem, dissolved in DMSO at a concentration of 1 × 10⁻² mol/L, and stored in amber tubes at −80°C. The final DMSO concentration never exceeded 0.1% which showed no effect on the growth of all tested cells. Caffeine (Sigma Chemical Co.) was dissolved in water to a final concentration of 100 μmol/L, and diluted to a 2 μmol/L final concentration in cell culture media.

Cell growth was assessed using thiazolyl blue tetrazolium bromide (MTT) dye (Sigma). Optical density (OD) was measured by the microplate ELISA reader (Multiscan EX) at 595 nm. Cell viability was determined by trypan blue dye exclusion assay. Cell-cycle analysis was performed using propidium iodide (50 μg/mL; Sigma) staining and a FACSAria SORP flow cytometer (Becton Dickinson), and cell-cycle distribution was analyzed using FACSDiva software (Becton Dickinson), as previously described (36).

Mitochondrial membrane potential was quantified using Rhodamine (R123) retention (Sigma) as previously described (24).

Cellular protein lysates [0.25 mmol/L Tris-HCL (pH 7.4), 20% β-mercaptoethanol, and 5% SDS] were prepared, quantified, and separated by SDS-PAGE, and were then transferred onto nitrocellulose membranes. Membranes were blocked with 5% skimmed milk in TBS (50 mmol/L Tris-HCL and 150 mmol/L NaCl), and were incubated overnight with specific primary antibodies at 4°C. Secondary antibodies were added after mild washing for 2 hours at room temperature while shaking. Proteins were visualized by enhanced chemiluminescence (ECL) using the ECL system. The following antibodies were used: p53 (sc-126) and PARP (sc-7150; Santa Cruz Biotechnology), GAPDH (MAB5476; Abnova), and γ-H2AX (2577; Cell Signaling).

ST1926-NP formulations were prepared by Flash NanoPrecipitation (FNP; ref. 37). ST1926 and polystyrene-b-poly(ethylene oxide) diblock copolymer (PS-PEO; Polymer Source, Dorval, Canada) at a mass ratio of ST1926/PS-PEO of 1/5 were dissolved in 3 mL of tetrahydrofuran (THF). The PS-PEO used has a polystyrene block size of 1,500 g/mole, and a poly(ethylene oxide) block size of 2,400 g/mole. To form NPs, the THF solution of ST1926 and PS-PEO was mixed in a confined volume with water using a multi-inlet vortex mixer (MIVM) at a volumetric ratio of 1/9 THF/H₂O. Flow rates of the THF and water streams were set to 12 and 108 mL/min, respectively, using Harvard apparatus PHD2000 syringe pumps. The resulting NP suspension was collected at the mixer outlet for further processing and analysis.

NP formulations used in mice experiments were concentrated using 50-kDa MicroKros hollow fiber filter (Spectrum, D04-E050-05-S) with 235 cm² surface area and maximum pressure of 30 psig (2 bar) as previously described (37). The concentration of ST1926 in the NP formulation was determined by UV analysis. A sample of ST1926 NP was dissolved in THF to obtain a molecularly dissolved solution. The A360 UV absorbance was determined using a DeNovix DS-11 spectrophotometer (DeNovix), and correlated to the ST1926 concentration using a calibration curve obtained for ST1926 in solution.

The NP size was determined using dynamic light scattering (DLS; Brookhaven Instruments, BI-200SM) following NP formation (38). The hydrodynamic particle size distribution was determined by measuring light scattering at 90°. NP formulations were freshly prepared each week weekly for in vivo treatments, and the particle size monitored by DLS.

NOD SCID and NOD-scid IL2rγ^null (NSG) mice were obtained from Jackson Laboratories. Mice protocols were approved by the Institutional Animal Care and Use Committee of the American University of Beirut. The comparative toxicity of the naked drug ST1926 to its NP formulation was first assessed in healthy NOD SCID mice. ST1926 at 7.5 and 30 mg/kg in its naked form or encapsulated in NP at the same corresponding concentrations, as well as the solvent and NP controls were tested in groups of five healthy adult mice. Animals were treated every other day for a period of 4 weeks. Mice were monitored on a daily basis for any change (fur, movement, …) and weighed on a weekly basis to check for any signs of toxicity.

For survival experiments, NOD SCID male and female mice (6 weeks old; average weight of 20 g) were intraperitoneally injected with 3 × 10⁶ THP-1 AML cells. Animals were housed in a specific pathogen-free facility. Mice were divided randomly into five groups: control only injected with THP-1 cells (n = 2), ST1926 control (n = 5), ST1926 treatment (n = 7), NP control (n = 7), and ST1926-NP treatment (n = 7). Health status of the mice and nodule formation around the abdominal region were monitored every other day, over the period of the treatment. Mice were humanely euthanized when moribund as defined by: weight loss >15% to 20%, lethargy, ruffled fur, slow motion, after deep anesthesia by isoflurane followed by cervical dislocation.

For ST1926 treatment, 0.6 mg of ST1926 was dissolved in 10 μL DMSO and subsequently diluted in 90 μL of 1:10 ethanol/cremophor: 1×PBS solution. ST1926-treated animals received 100 μL of ST1926 intraperitoneally (equivalent to 30 mg/kg body weight) as this dosage was reported to cause survival advantage in AML mice with no signs of toxicity (13). ST1926-NP-treated animals were intraperitoneally injected with 200 μL of the suspension, which provides an ST1926 dose of 7.5 mg/kg body weight. This concentration was selected based on optimal NP encapsulation efficacy and formulation stability and was determined by UV analysis. Treatment protocols started 3 days after malignant cell injection, and were delivered three times a week for up to 5 weeks. Mice survival was recorded. Peritoneal volumes were measured and recorded once a week by a digital caliper (Model DC150-S) applying the general formula: V = (4/3) × π × r³.

To assess bone marrow tumor burden, 3 × 10⁶ THP-1 cells were injected into the tail vein of 8-week-old male and female NSG mice (n = 3 per group). On day 3 post THP-1 injection, mice were treated intraperitoneally with 7.5 or 30 mg/kg of naked ST1926, NP charged with 7.5 mg ST1926, and corresponding control for 3 days a week, every other day over a period of 3 weeks.

Bone marrow from the femurs and tibias of euthanized animals were flushed at the end of week 3 post-THP-1 cells inoculation. Cell surface staining was performed on 100 μL of sample using 20 μL of the anti-human CD45 PerC-P antibody (Becton Dickinson, cat. no. 345809) as described (39). After incubation for 15 minutes in the dark, erythrocytes were lysed using 1 mL FACS Lyse (Becton Dickinson, cat. no. 345809). Labeled samples were washed twice and analyzed by Guava flow cytometer.

Statistical comparisons were done using Microsoft Excel 2010. Nonpaired and paired t test was used for comparison of two groups, whereas one-way ANOVA was used for three or more groups of treatments. *, **, and *** indicate P values ≤ 0.05, 0.01, and 0.001, respectively. Kaplan–Meier method with statistical significance was used for survival analysis. Log-rank (Mantel–Cox) and Gehan–Breslow–Wilcoxon tests were used for statistical analysis between the different groups; differences were considered significant only when P values were less than 0.05.

To test the effect of ST1926 on cell growth and viability, we used five human non-APL AML cell lines (THP-1, KG-1α, MOLM-13, ML-2, and HEL) harboring different genetic mutations. These AML cells were shown to be resistant to ATRA (7). Using the MTT cell proliferation assay, we observed that pharmacologically achievable sub-μmol/L concentrations of ST1926 as low as 0.1 μmol/L, with long plasma retention times (21, 27) significantly inhibited the proliferation of all tested AML cells, irrespective of their mutational signatures, in a time-dependent manner (Fig. 1A). ST1926 at 0.1 μmol/L concentrations resulted in approximately 70% growth inhibition at 72 hours in tested cells. Similar trends were observed when cell viability was assessed using trypan blue exclusion assay (Supplementary Fig. S1). Despite the growth inhibitory effect of ST1926 in AML cells, no effect was noticed on their differentiation status, as assessed by the similar CD11b expression in untreated or 0.5 μmol/L ST1926-treated cells up to 72 hours.

Furthermore, we have tested ST1926 effect on primary leukemic blasts derived from the bone marrow of two AML patients [patient 1: AML-M1/NPM1/FLT3-ITD positive; patient 2: APL/t(15;17) PML-RARα positive], and have demonstrated that 0.5 μmol/L ST1926 treatment for 48 hours reduced their viability by approximately 80% (Fig. 1B). Importantly, high concentrations of ST1926 up to 10 μmol/L had no effect on resting (Fig. 1C) and phytohemagglutinin-activated normal PBMCs from four healthy donors (Fig. 1D) as previously reported (22). Furthermore, minimal effect was observed on the growth of normal human CD34⁺ fraction of mononuclear cells and MSCs at the working concentrations of 0.5 and 1 μmol/L (Supplementary Fig. S2). In summary, 0.5 μmol/L ST1926 concentrations were used in subsequent experiments as they drastically reduced cell growth and viability of all tested AML cells.

To assess whether ST1926-growth inhibition was sustained after drug removal, THP-1 and MOLM-13 cells, harboring p53 and FLT-3 mutations (40), respectively, were pretreated with 0.5 μmol/L ST192 for 24 hours, then cells were resuspended in drug-free media for up to 3 days. Using MTT assay, our results indicate that ST1926 growth suppression was irreversible where THP-1 and MOLM-13 growth was significantly reduced by 70% and 60%, respectively, 2 days postdrug removal (Fig. 2A). To study the mechanisms involved in ST1926-induced growth inhibition, cell-cycle analysis was performed in THP-1, KG1-α, MOLM-13, and ML-2 cells treated with 0.5 μmol/L ST1926 for up to 48 hours. No major variation in cell-cycle distribution was observed between control AML cell lines (Fig. 2B and Supplementary Fig. S3). ST1926-treated THP-1, MOLM-13, and ML-2 cells induced a massive accumulation of cells in the presumably apoptotic sub-G₁ region of the cell cycle as early as 24 hours except for KG1-α, where this accumulation was pronounced after 48 hours of treatment (Fig. 2B). In fact, approximately 80%, 40%, 70%, and 80% of THP-1, KG1-α, MOLM-13, and ML-2 treated cells for 48 hours, respectively, accumulated in the sub-G₁ region (Fig. 2B). We observed a minor G₀–G₁ cell-cycle arrest in KG1-α, MOLM-13, and ML-2 treated cells for 24 hours, whereas S-phase arrest was detected in THP-1-treated cells (Fig. 2B).

Apoptosis induction upon ST1926 treatment was confirmed in AML cells. THP-1, KG1-α, MOLM-13, and ML-2 cells treated with 0.5 μmol/L ST1926 revealed PARP (113 kDa) cleavage into its death associated fragments (89 and 24 kDa) as early as 24 hours (Fig. 3A). ST1926-induced apoptosis in AML cells was associated with dissipation of the mitochondrial membrane potential as early as 24 hours in 0.5 μmol/L ST1926-treated THP-1 and MOLM-13 cells (Fig. 3B). The intracellular fluorescence of Rhodamine-123 dye decreased significantly by approximately 60% in THP-1 and MOLM-13 treated cells for two days (Fig. 3B). To further understand ST1926-mediated growth inhibition and cell death, p53 protein levels were monitored in THP-1 and MOLM-13 cell lines harboring a mutated and wild type p53 profile (40), respectively. MOLM-13-treated cells with 0.5 μmol/L ST1926 displayed a remarkable increase in total p53 protein levels whereas as expected no changes were detected in THP-1-treated cells with mutated p53 (Fig. 3C).

ST1926 has been previously characterized as a genotoxic drug that causes early DNA damage in various types of tumor cells (15, 22–26). Here, we show a substantial increase in the phosphorylation of the DNA damage marker, H2AX to γ-H2AX, as early as 30 minutes in THP-1 and MOLM-13 cells treated with 0.5 μmol/L ST1926 (Fig. 3D). To further investigate the role of DNA damage in ST1926 effects, Caffeine was used in combination with ST1926 in AML cells. Caffeine is known to inhibit the DNA damage response pathway (41) and to mitigate ST1926-induced S-phase arrest in rhabdomyosarcoma cells (21). In our studies, the effect of ST1926 on THP-1 and MOLM-13 cell growth inhibition was significantly reversed upon cotreatment with Caffeine (Fig. 3E). Altogether, these results show that ST1926 is a potent inducer of apoptosis and DNA damage in AML cells at sub-μmol/L concentrations independently of p53 status.

Solubility and bioavailability problems are major reasons for which many drug candidates fail in clinical development. ST1926 is a highly hydrophobic drug (Fig. 4A) that was found to undergo major glucuroconjucation leading to its poor bioavailability and rapid excretion by the liver (27). To enhance ST1926 bioavailability, we generated polymer-coated ST1926-NP using Flash NanoPrecipitation technique (37, 42). ST1926 was formulated into NP with a drug to polymer mass ratio of 1:5 using 5 mg of ST1926 and 25 mg of polystyrene-b-poly(ethylene oxide) block copolymer (PS1.5-PEO2.4) with a polystyrene hydrophobic block. The resulting solution was mixed at an optimized rate using the MIVM mixer (Fig. 4B). ST1926-NP's size was determined using the DLS apparatus and was found to be on average 190 nm and to remain stable at room temperature even after 24 hours (Fig. 4C).

To test whether formulating ST1926 into NPs attenuates its efficacy, THP-1 and MOLM-13 cells were treated with the same concentrations of naked drug and formulated NPs, and their cell growth was determined using the MTT proliferation assay. Interestingly, both ST1926 and ST1926-NP exhibited similar growth-inhibitory effects with an approximate IC₅₀ values of 0.5 μmol/L at 48 hours in THP-1 and MOLM-13 cells (Fig. 4D). This suggests that drug–carrier interactions are not rate-limiting for drug diffusion out of the NP, which could otherwise reduce its efficacy.

To assess the in vivo efficacy of ST1926 and its formulated NPs, we injected 3 million THP-1 cells intraperitoneally into NOD SCID mice. Three days post-AML cell injection, THP-1 xenografted mice were intraperitoneally treated with either 30 mg/kg ST1926, 7.5 mg/kg ST1926-NP, or their respective vehicles once daily, 5 days a week, for up to 5 weeks. ST1926-NP concentration was selected at optimal formulation stability to reduce aggregation and other instabilities (43).

We observed a marked survival advantage in both ST1926 and ST1926-NP-treated animals with a median survival of 63 and 62 days, respectively, when compared to their respective controls with a median survival of 42 and 46 days (Fig. 5A). Strikingly, even at four-fold lower concentrations, ST1926-NP exhibited similar survival profile as ST1926 (Fig. 5A). Importantly, after 24 days of treatment, we noted a sharp increase in the peritoneal volume of control mice reaching more than 100 cm³ versus less than 50 cm³ in ST1926 and ST1926-NP treated animals (Fig. 5B and Supplementary Table S1). One out of seven mice survived for 242 days due to ST1926 treatment at 30 mg/kg and for 280 days, in ST1926-NP treated animals at four-fold lower concentrations (Fig. 5A). These results indicate that ST1926 treatment prolongs the survival and reduces the peritoneal volume of AML xenografted animals and its NP formulation results in similar effects at significantly lower concentrations than the naked drug.

We have used a well-established AML orthotopic mouse model where THP-1 cells were injected intravenously in NSG mice (39, 44). Three days postleukemic cells injection, animals were treated with 7.5 or 30 mg/kg of naked ST1926, or 7.5 mg ST1926-NP, and compared to untreated controls. All mice were sacrificed at the end of the third week of treatment. Since THP-1 cells express the human CD45 (hCD45⁺) marker (44, 45), bone marrow cells were harvested from untreated or treated animals and stained for hCD45⁺ to evaluate leukemia burden. ST1926-NP containing 7.5 mg of drug significantly reduced bone marrow leukemia burden in AML mice (P < 0.01; Fig. 5C). Importantly, the effect of 7.5 mg/kg ST1926-NP was similar to that of the naked drug at four-fold higher concentrations. In contrast, naked ST1926 at the same concentrations of 7.5 mg/kg did not show any reduction in tumor burden underscoring the superiority of the new ST1926-NP formulation (Fig. 5C).

These results highlight the promising impact of this new formulation in reducing AML tumor burden at lower drug concentrations.

In this study, we showed that ST1926 potently inhibited the growth and induced massive apoptosis of ATRA-resistant AML cells at low sub-μmol/L concentrations through the activation of DNA damage and dissipation of the mitochondrial membrane potential. This ST1926 mechanism of action is similar to previously reported results in other tested hematological malignancies (13, 22, 24–26) and solid tumors (16, 19, 21, 23, 46). Importantly, sub-μmol/L ST1926 concentrations exerted the same inhibitory effect in primary AML patients' derived blasts, while sparing normal leukocytes and normal human hematopoietic and MSCs. We also report that ST1926 cell death effect is independent of p53 status as previously published (14, 21–23) and of several commonly encountered mutations in AML.

ST1926 has been previously characterized as a genotoxic drug that causes early DNA damage in various types of tumor cells (18, 21–26). Here we show that ST1926-treated AML cells at sub-μmol/L concentrations resulted in early DNA damage. Co-treatment with DNA damage inhibitors reversed the drug's effects highlighting the crucial role of ST1926 in DNA damage. In fact, ST1926 resistance was demonstrated to be due to a defective DNA damage response in the NB4 APL (26), H460 lung carcinoma (18), and colon carcinoma HCT116 cells (unpublished results). Previous work has established that ST1926 inhibits the growth of the p53 mutated AML NB4 cells, and of the p53 null HL-60 AML cells, underlying a p53-independent mechanism of action (13). Similarly to its CD437 parental compound, ST1926 was shown to work independently of the RAR signaling pathway (15, 23, and our unpublished data). Recently, CD437 was demonstrated to directly inhibit DNA polymerase α, the enzyme responsible for initiating DNA synthesis during the S-phase of cell cycle, which is encoded by POLA1 (47). It would be interesting to check on the status of POLA1 mutations and/or expression in AML cells that develop resistance to ST1926 and to investigate whether this mutation confers resistance to CD437 as well.

ST1926 clinical testing was halted in phase I for ovarian cancer, as it was found to undergo glucuroconjugation on its phenolic hydroxyl group leading to its poor bioavailability and rapid excretion by the liver (27). In fact, short-lived ST1926 μmol/L plasma concentrations were observed in humans (27) and mice (21), which abruptly dropped to more sustained sub-μmol/L levels. Given ST1926 potent activities and lack of toxicity in humans and in various tumor models, multiple efforts were taken to synthesize analogs with more favorable properties (48). Another approach was to enhance ST1926 bioavailability through the use of NP. Furthermore, NPs have great potential as cellular drug delivery vehicles. They are shown to promote targeted efficacy by enhancing the drugs' stability, protecting them from degradation, reducing side effects as well as bioavailability, and retention at the target site of action (49, 50). Moreover, most approved NP-based therapies are nowadays administered to patients by intravenous injection underscoring their promising clinical use (51). Here, we report the generation of polymer-stabilized ST1926-NP formulations using Flash NanoPrecipitation technique (37, 42).

Both ST1926 and NP formulations significantly prolonged survival, and reduced peritoneal volume and bone marrow tumor burden in AML xenografted mice. Furthermore, both treatments were well-tolerated in animals and no signs of behavioral abnormalities, undesirable side effects, or toxicities were noted. Interestingly, ST1926 antitumor properties in AML xenografted mice were observed at four-fold lower concentrations when formulated in NPs versus the naked drug. Indeed, the higher potency of ST1926-NP obtained solely in vivo when compared to the naked drug, may be due to the complexity of the tumor microenvironment (52) that might be providing enhanced accessibility of ST1926-NP than the naked drug after circulation in the plasma. Further in vivo studies are required to study the mechanism of action of ST1926 and ST1926-NP in AML models. PEG used as hydrophilic block has been demonstrated to prolong the NPs circulation time in vivo and to provide a steric barrier to the particle and a reduction in its opsonization (53). We have preliminary data showing that the NP formulation did not enhance the plasma ST1926 concentrations than the naked drug. It remains to be determined whether ST1926-NP results in lower blood absorption but greater extravascular tissue and bone marrow distribution than the free drug. Future assays with high sensitivity should be optimized to detect ST1926 and/or its glucuroconjugated form in plasma, bone marrow, and urine. These pharmacokinetic studies may explain our observed beneficial effect of using ST1926-NP in AML xenograft mouse models. Interestingly, we have previously observed a significant extravascular distribution of the naked ST1926 in mice which may explain the observed antitumor effect of ST1926 in animal models (21).

The observed potent antitumor properties of ST1926 in several types of cancers that are resistant to conventional therapies and independently of p53 signaling as well as its low drug toxicity underline the need for its further clinical development. In summary, we highlight the promise of ST1926 NP formulation in AML therapy and call for its drug development through formulation strategies or synthesis of analogs with more favorable pharmacological properties (48).

No potential conflicts of interest were disclosed.

Conception and design: L. El-Houjeiri, W. Saad, N. Tawil, H. El Hajj, N. Darwiche

Development of methodology: L. El-Houjeiri, W. Saad, B. Hayar, N. Tawil, A. Mancinelli, H. El Hajj

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): L. El-Houjeiri, W. Saad, B. Hayar, P. Aouad, R. Abdel-Samad, R. Hleihel, M. Hamie, H. El Hajj

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): L. El-Houjeiri, W. Saad, B. Hayar, R. Hleihel, H. El Hajj, N. Darwiche

Writing, review, and/or revision of the manuscript: L. El-Houjeiri, W. Saad, B. Hayar, P. Aouad, N. Tawil, R. Abdel-Samad, H. El Hajj, N. Darwiche

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): L. El-Houjeiri, B. Hayar, P. Aouad, R. Hleihel, H. El Hajj

Study supervision: W. Saad, C. Pisano, H. El Hajj, N. Darwiche

The authors thank Ms. Zaynab Jaber for her help with the nanoparticle preparation, Dr. Marwan El Sabban and Ms. Jamal Al Saghir for their technical expertise with the CD45⁺ mononuclear and MSCs experiments, and Mr. Abdel Rahman Itani for his assistance with intravenous injections in mice. We thank Dr. Samira Kaissi for her thorough editing of the manuscript.

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.