In Vivo Toxicity Study of Engineered Lipid Microbubbles in Rodents.

Real-time intraoperative imaging for brain tumor surgery is crucial for achieving complete resection. We are developing novel lipid-based microbubbles (MBs), engineered with specific ligands, which are able to interact with the integrins overexpressed in the endothelium of the brain tumor vasculature. These MBs are designed to visualize the tumor and to carry therapeutic molecules into the tumor tissue, preserving the ultrasound acoustic properties of the starting plain lipid MBs. The potential toxicity of this novel technology was assessed in rats by intravenous injections of two doses of plain MBs and MBs engineered for targeting and near-infrared fluorescence visualization at two time-points, 10 min and 7 days, for potential acute and chronic responses in rats [(1) MB, (2) MB-ICG, (3) MB-RGD, and (4) MB-ICG-RGD]. No mortality occurred during the 7-day study period in any of the dosing groups. All animals demonstrated a body weight gain during the study period. Minor, mostly reversible changes in hematological and biochemical analysis were observed in some of the treated animals. All changes were reversible by the 7-day time-point. Histopathology examination in the high-dose animals showed development of foreign body granulomatous inflammation. We concluded that the low-dose tested items appear to be safe. The results allow for proceeding to clinical testing of the product.

Introduction

Gliomas are neoplasms that arise within the cerebral parenchyma and are characterized by an infiltrative growth pattern. Glioblastoma, the most malignant form of cerebral gliomas, is associated with median survival of only about 14 months.[1] Maximal surgical resection, however, has been shown to have a positive effect on progression-free and overall survival.[2,3] Complete surgical resection of these infiltrative tumors is often hampered by the absence of intraoperative real-time imaging to enable the demonstration of the tumor’s boundaries during the operation. Certain surgical modalities, such as intraoperative neuronavigation, 5-aminolevulinic acid (5-ALA) fluorescence-guided tumor resection, and intraoperative magnetic resonance imaging (MRI) enable the surgeons to evaluate the extent of resection intraoperatively, but each of these modalities has its own limitations.[4,5] For example, 5-ALA leads to the accumulation of fluorescent porphyrins in the high-grade but not low-grade glioma tissue. Intraoperative MRI supports the surgeons with detailed image information, but it requires a high constructive, logistic, and pecuniary effort associated with high costs and prolongation of the total operative time. Intraoperative ultrasound (US) is a more practical and cost-effective tool, allowing real-time imaging with less effort compared with the other modalities. It has the potential to avoid brain shift due to cerebrospinal fluid (CSF) loss and tissue edema, which renders neuronavigation less reliable. One of its major limitations is the difficult image interpretation.[6] Therefore, improving intraoperative US image quality combined with direct microscopy visualization is highly desirable to better delineate the tumor margins for maximizing its resection.

Microbubbles (MBs) have become a well-established contrast agent for US imaging, improving both the US image quality and visualization of the tumor vasculature.[7] Conjugation of specific ligands to a MB’s shell against overexpressed receptors in a glioblastoma may allow for specific binding to the tumor and also may serve as a novel theranostic platform to carry drugs specifically into the tumor site.[8] For example, integrins, a family of heterodimeric cell adhesion molecules, play a crucial role in cell–cell adhesion and cell–extracellular matrix interactions.[8,9] Conjugation of the MB shell with RGD motif, tripeptide Arg-Gly-Asp, a ligand for integrins αvβ3 and αvβ5, which is overexpressed in the glioblastoma vasculature,[10] may lead to selective adherence of MBs to the tumor microvessel endothelium and specific accumulation in the tumor site.[8] The MBs’s lipid shell may also be labeled with fluorophores, such as indocyanine green (ICG), for direct fluorescent visualization through the operative microscope.[11,12]

In this work, a cyclic derivative of RGD with a cysteine moiety included in the molecular scaffold, cRGDfC (Scheme ), is used as this form of RGD favors the interaction with the target integrins[13] and offers a quantitative coupling to the MB surface bearing a maleimide moiety.

Scheme 1

Structure of cRGDfC Peptide

The thiol, −SH, highlighted in the circle is the reactive group used for coupling to MBs.

Such modified lipid MBs have potential as novel multimodal tools for tumor visualization, allowing for real-time intraoperative imaging during tumor resection via both US and in parallel fusion to the operative microscope.

The objective of this study is to assess the potential toxic effects of newly designed US contrast agents (UCAs), that is naked MB, MB labeled with ICG (MB-ICG), MB coupled to RGD (MB-RGD), and MB labeled with ICG and coupled to RGD (MB-ICG-RGD), following a single intravenous (IV) injection to male Sprague-Dawley rats for better tumor vasculature visualization in humans. The results reported in this work contribute to the understanding of the not yet complete toxicology scenario involving MBs.

Materials and Methods

1,2-Distearoyl-sn-glycero-3-phosphocholine (DSPC), 1,2-dipalmitoyl-sn-glycero-3-phospho-(1′-rac-glycerol) sodium salt (DPPG-Na), and 1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-[maleimide(polyethylene glycol)-2000] ammonium salt (DSPE-PEG2000-Mal) were purchased from Avanti Polar Lipids, Alabaster, Alabama (USA). tert-Butanol (t-BuOH), palmitic acid (PA), poly(ethylene glycol) with molecular weight 3500–4500 (PEG 4000), ethylenediaminetetraacetic acid (EDTA), phosphate-buffered saline (PBS, pH 7.4), and ethanol were Sigma-Aldrich, Milan, (Italy) products. BODIPY FL l-cystine and gel-immobilized (tris(2-carboxyethyl)phosphine) disulfide (TCEP) reducing agent were purchased from Thermo Fisher Scientific, Monza, MB (Italy). cyclo-(Arg-Gly-Asp-D-Phe-Cys) (c-RGDfC) purchased from Peptides International, Louisville, Kentucky (USA), sulfur hexafluoride (SF6) from Rivoira Gas, Rome, RM (Italy), and ICG from Pulsion Medical System, Munich (Germany) were used without further purification. Water of Milli-Q purity grade (18.2 MΩ·cm) was produced with an Elga PureLab Classic, High Wycombe, (UK) deionization apparatus.

Synthesis of Engineered MBs

MBs-RGD

Thirty mg of DSPC, 30 mg of DPPG-Na, 4.8 mg of DSPE-PEG2000-Mal, and 6 mg of PA were dissolved together with 2.94 g of PEG 4000 in 31 mL of tert-butanol in a round bottom glass vessel under reflux (82 °C). After complete dissolution, the clear solution was rapidly frozen with liquid nitrogen and lyophilized by the freeze dryer Lio 5P, CinquePascal, Milan (Italy). An aliquot (300 mg) of the obtained cake was introduced in a 20 mL glass vial. The vial was closed with a rubber stopper and evacuated, and the headspace was replaced by SF6. MBs were formed by injecting PBS (7.2 mL, 10 mM) through the stopper into the vial and by 20 s of vigorous shaking of the suspension to mix all components until a milky homogeneous liquid had been obtained.

The MB suspension was centrifuged twice: the infranatant was discharged to remove excess components of nonforming MB shells, whereas the MBs in the supernatant were resuspended in 7.2 mL of 10 mM PBS containing 5 mM EDTA. A total of 900 μg of cysteine-tagged RGD, c-RGDfC (Scheme ) was kept in reducing conditions in the presence of 3.6 mM TCEP and 5 mM EDTA to avoid oxidation and disulfide bonds formation between the cysteine moiety of the c-RGDfC molecules in solution. After 1 h at room temperature, the TCEP reducing gel was discarded by centrifugation and c-RGDfC, in large excess as compared to the molar content of DSPE-PEG2000-Mal moiety contained on the MB shells, was incubated with MBs for 1 h at 4 °C (Scheme ). To remove unreacted c-RGDfC, targeted MBs-RGD were centrifuged twice and the infranatant was replaced by PBS (7.2 mL, 10 mM) containing maltose at 10% w/v after each centrifugation step. The MBs-RGD suspension was then poured in a 10 mL glass vial, frozen with liquid nitrogen, and lyophilized.

Scheme 2

RGD-MBs Synthesis

Ligand Quantitation on MBs

A direct quantitative estimation of the c-RGDfC ligand is difficult to accomplish. To overcome such a hurdle, without changing the coupling reaction to MBs and its stoichiometry, c-RGDfC is replaced with fluorescent, thiol-containing reduced BODIPY-cystine (see Scheme ). BODIPY-cystine was incubated with MBs after reduction with an equivalent volume of TCEP disulfide-reducing gel. The amount of maleimide groups using a tenfold excess of reduced BODIPY-cystine reproduced exactly the reaction conditions used for the coupling of c-RGDfC to the MBs. The number of fluorophore-conjugated MBs (BODIPY-MBs) was determined by optical microscopy using the counting chamber of a hemocytometer. The MBs were then disrupted by sonication, and the fluorescence of the resulting solution was assessed by RF-5301 PC Shimadzu, Milan, (Italy) with excitation and emission slit widths of 1.5 nm and sensitivity set to “high”. The determination of the concentration of bound BODIPY was based on a “fluorescence intensity versus free BODIPY concentration” calibration curve covering the BODIPY concentration ranging from 0.02 to 0.45 μM (see Figure S1 of Supporting Information). The fluorescence intensity measured of the BODIPY-MBs dispersion was related to the MB concentration to determine the c-RGDfC density on the MB surface. This method provided an extent of conjugation of 1.9 × 106 molecules of ligand per MB.

Naked MBs

Nontargeted lipid MBs (naked MBs) as the control test item were also prepared according to the procedure described above, with the exclusion of DSPE-PEG2000-Mal in the MB formulation and by skipping the reaction step with c-RGDfC.

MB Reconstitution and ICG Labeling

MB samples were freshly prepared before dosing the animals. The chemical and physical integrity of the freeze-dried cake containing DSPC, DPPG-Na, and PEG 4000 from which the MBs are prepared, was tested by differential scanning calorimetry (DSC). As shown in Figure S2, the thermogram displays an endothermic peak at 60 °C, typical of the melting of PEG 4000, which is the major component of the cake. MBs were stored as lyophilized powders of naked MBs and MBs-RGD in glass vials. For their reconstitution, the powder was purged with SF6 gas through the glass vial stopper septum with inlet and outlet needles to remove air and to fill the headspace of the container. Before use, 5 mL physiological saline was added to 25 mg of the freeze-dried powder per vial and vigorously shacked, Figure S3. To obtain fluorescent-labeled MBs, for example, MBs-ICG and MBs-ICG-RGD, lyophilized MBs (either naked or RGD engineered) were reconstituted by adding 1 mL of ICG solution (5 mg/mL in water) and up to 5 mL of physiological saline to reach a final ICG concentration of 1 mg/mL. Samples were shaken vigorously after reconstitution to obtain a milky MBs suspension. One vial of the freeze-dried MBs yielded a suspension of approximately 108 MBs/mL.

A characterization of naked MBs suspension after reconstitution is reported in the Supporting Information section together with the instrumental setup used for the determinations. The size and size distribution of naked MBs is shown in Figure S4. An average diameter of 2.8 μm with a standard deviation of 1 μm was determined. The US acoustic properties of the MBs depend on their size distribution. We tested the US behavior of the synthetized naked MBs from 0.5 to 10 MHz with the setup described in Figure . Attenuation spectra (Figure S5) were recorded at different MBs floating times.

Figure 4

Naked MB and MB-ICG, 2520 mg/kg, 7 days sacrifice time-point. Animal 3M16 (upper-left), 5M29 (upper-right), high magnification sections of the lungs (×20). Presence of homogenous greyish foreign body associated with granulomatous reaction within the blood vessel is considered to be the test compound (yellow arrow). Animal 7M57 (lower-left), 7M40 (lower-right), sections of the lungs (×20). Presence of homogenous greyish foreign body with (7M40) and without (7M57) granulomatous reaction within the blood vessel is considered to be the test compound (yellow arrow).

Animals

Fifty-four eight-week-old male Sprague-Dawley rats weighing 300 g ± 20% for 18 test groups were obtained from Envigo RMS, Israel. All dosing solutions were injected by a single IV injection into one of the tail veins of the animals. Test items were injected in two volume doses of 2 and 20 mL/kg, corresponding to 252 and 2520 mg/kg, respectively, of formulations containing naked MBs, MB-ICGs, MB-RGDs, and MB-ICG-RGDs for either 10 min or 7 days postdosing. These time-points were chosen to assess the acute and chronic response to the IV MBs administration.

As control, a saline was injected at the highest volume dosage of 20 mL/kg (see Supporting Information Table S1). The study was approved by the Animal Care and Use Committee (ACUC) of the National Council for Animal Experimentation, and all procedures were conducted within compliance of their regulations.

Clinical Signs and Body Weight

Careful clinical examinations were carried out, and the results were recorded immediately following the dosing. For animals scheduled for sacrifice at 7 days postdosing, additional once-daily observations were performed for the detection of changes in skin, fur, eyes, mucous membranes, occurrence of secretions and excretions (e.g., diarrhea), and autonomic activity (e.g., lacrimation, salivation, piloerection, pupil size, and unusual respiratory pattern). Changes in gait, posture, and response to handling, as well as the presence of irregular behavior, tremors, convulsions, or coma were also inspected. The eight-week-old male Sprague-Dawley rats were expected to increase their body weight during the study period of 7 days, accordingly determinations of individual body weights of the animals were carried out shortly prior to dosing (day 0), on day 2, and on day 7 (at study termination).

Clinical Pathology

Hematology and biochemistry parameters listed in Tables S2 and S3 in Supporting Information were determined in all animals before the scheduled necropsy. For animals scheduled for sacrifice 10 min postdosing, blood was drawn at 10 min postdosing and immediately after the animals were sacrificed. Individual blood samples were obtained by retro-orbital sinus bleeding under CO2 anesthesia. Following completion of blood collection, all samples (whole blood and serum) were kept at 2–8 °C until transported for analysis to the American Medical Laboratories (Israel) Ltd.

Necropsy Procedure and Macroscopic Examination

Following the respective termination time-points, the animals were euthanized by CO2 asphyxiation and subjected to a fully detailed necropsy and gross pathological examination. At necropsy, the animals underwent thorough examinations, including the external surface of the body, all orifices, and cranial, thoracic, and abdominal cavities and their contents. Any abnormality or gross pathological change observed in tissues and/or organs was recorded.

Organ/Tissue Collection, Weighing, and Fixation

The brain, kidneys, liver, lungs, and spleen were collected from each animal during the respective scheduled necropsy session and fixed in 10% neutral buffered formalin (approximately 4% formaldehyde solution). In addition, any other organ/tissue with gross macroscopic change/s was collected, recorded, and preserved in 10% neutral buffered formalin. Slide preparations and histopathological examinations for abnormalities detected during necropsy were performed. The wet brains, kidneys, livers, lungs, and spleens were weighed as soon as possible following their dissection.

Histology and Histopathology

The harvested tissues from each animal were transferred to Patho-Lab Diagnostics Ltd. for slide preparation. The tissues were trimmed, embedded in paraffin, sectioned at an approximately 5 μm thickness, and stained with hematoxylin and eosin (H&E). The histopathological changes were described and scored by the study pathologist using semiquantitative scoring of five grades (0–4), taking into consideration the severity of the changes (0 = no lesion, 1 = minimal change, 2 = mild change, 3 = moderate change, 4 = marked change). After histopathological evaluation, the finding in each test item-treated animal was concurrently compared to the respective control animal to verify any potential treatment-related effect.

Results

There was no mortality among any of the test items-treated or saline-treated control animals throughout the entire 7-day study period. Clinical signs were confined to all high-dose-treated groups immediately following dosing, and they consisted of either decreased motor activity or slight dyspnea. All clinical signs faded within a few hours and no other signs were observed throughout the remaining 7-day study period. All animals who reached the end of the 7-day study period made comparable body weight gains expected for eight-week-old male Sprague-Dawley rats.

Hematology

There were increases in the red blood cell (RBC) count and hemoglobin (HGB) levels in all high-dose test groups compared to the control. An increase in hematocrit (HCT) was evident in all high-dose test groups compared to the control, and it reached above the normal range for this species. There was a reduction in platelet counts in all high-dose test groups at the 10 min sacrifice time-point compared to the control, and it dropped below the normal range for this species. The reduction was more pronounced in the RGD-containing MBs (with or without ICG)-treated animals. One animal from the low-dose MBs-ICG-treated group displayed an absolute lymphocytosis, with a total white blood cell (WBC) count of 7.8 × 103 cells/μL (∼4 time higher than the upper limit of the normal value for this species), 100% lymphocytes, and a two-fold reduction in RBC count compared to normal values. An increase in total WBC count above the normal range was observed in the group of low-dose MBs-RGD. An increased total WBC was also observed in all other groups, including the vehicle-treated rats compared to their respective 10 min counterparts, although the values were within the normal range for this species. The increase did not change the neutrophils-to-lymphocytes ratio in the blood (Table S2, in Supporting Information).

Blood Biochemistry

There was an increase in glucose levels in the high-dose-treated groups, an increase in total serum bilirubin in the MBs-ICG with or without RGD-treated group, a decrease in total protein and serum albumin levels in the MBs-ICG with or without RGD-treated groups, and a decrease in potassium levels in the high-dose-treated groups (below the normal range for these species) after 10 min. The 7-day test and vehicle control treatment groups exhibited comparable values (Table S3, in Supporting Information).

Organ Weight

Aside from the changes found in spleen weight, which may indicate erythrocyte release and contribute to the rise in circulating RBCs, there was no apparent toxicity resulting from the noted changes in organ weight. At the 10 min sacrifice time-point, a reduction in the spleen weight ranging between ∼10 and 20% was noted in high-dose naked MBs, high-dose MBs-ICG, high-dose MBs-ICG-RGD, low-dose MBs-RGD, and low-dose MBs-ICG-RGD compared to the vehicle control-treated animals. A similar-sized reduction in lung weight was measured in all test groups at the 10 min sacrifice time-point with the exception of the low-dose MBs-RGD-treated group. At the 7 day sacrifice time-point, there was an increase in the lung weight ranging between ∼20 and 30% in the low-dose naked MBs, low-dose MBs-ICG-, and low-dose MBs-RGD-treated groups (Table ).

Table 1

Organ Weights (g) and Weight Variations Compared to Control-Treated Animalsa

		brain		lung		spleen		kidney		liver
		weight [g]	variation [g]	weight [g]	variation [g]	weight [g]	variation [g]	weight [g]	variation [g]	weight [g]	variation [g]
before treatment		5.2	0.3	5.7	0.6	3.1	0.1	3.6	0.1	43	1.0
naked MBs 2 mL/kg	10 min	5.5	0.2	b5.5	0.1	3.3	0.4	3.8	0.1	b48	1.7
naked MBs 2 mL/kg	7 days	4.8	0.3	6.7	0.6	2.7	0.2	3.3	0.3	44.7	1.0
naked MBs 20 mL/kg	10 min	5.5	0.0	5.4	0.2	2.6	0.2	3.4	0.1	43.4	2.8
naked MBs 20 mL/kg	7 days	5.1	0.2	6.0	0.8	2.9	0.2	3.5	0.1	45.8	1.8
MBs-ICG 2 mL/kg	10 min	5.5	0.1	6.0	0.7	3.3	0.4	3.5	0.4	47.9	4.7
MBs-ICG 2 mL/kg	7 days	5.0	0.4	7.3	2.2	2.8	0.3	3.5	0.2	47.7	1.2
MBs-ICG 20 mL/kg	10 min	5.2	0.3	5.9	0.2	2.7	0.1	3.3	0.2	43.5	2.7
MBs-ICG 20 mL/kg	7 days	5.2	0.2	6.3	0.8	2.7	0.4	3.3	0.3	43.2	5.0
MBs-RGD 2 mL/kg	10 min	b6.0	0.3	6.7	0.7	2.8	0.5	3.7	0.2	b4.6	1.2
MBs-RGD 2 mL/kg	7 days	5.1	0.2	6.7	+0.6	3.1	0.4	3.4	0.3	44.5	3.4
MBs-RGD 20 mL/kg	10 min	5.8	0.3	6.0	0.0	3.1	0.4	4.0	0.4	b46.4	0.9
MBs-RGD 20 mL/kg	7 days	5.4	0.3	5.5	0.2	2.9	0.4	3.4	0.2	45.6	5.4
MBs-ICG-RGD 2 mL/kg	10 min	5.5	0.2	5.9	1.1	2.6	0.3	3.6	0.4	45.2	0.2
MBs-ICG-RGD 2 mL/kg	7 days	5.0	0.6	5.8	0.4	2.8	0.9	3.6	0.4	46.2	5.1
MBs-ICG-RGD 20 mL/kg	10 min	5.5	0.1	5.6	0.4	2.6	0.1	3.6	0.2	46.8	4.3
MBs-ICG-RGD 20 mL/kg	7 days	5.0	0.5	6.1	0.9	2.5	0.3	3.4	0.2	44.1	2.9

Columns are differently colored for reader’s convenience.

Statistical significance: p > 0.05.

Gross Necropsy Observations

A single animal from the MBs-RGD high-dose-treated group displayed hemorrhage-like lesions in a 1–2 mm interval striped appearance throughout the small intestine up to the cecum. No other treatment-related abnormalities were noted. A green discoloration in some organs was observed in animals treated with the green dye ICG at the 10 min sacrifice time-point. The groups treated with low doses displayed discoloration in the duodenum, and the groups treated with high doses displayed discoloration in the duodenum, liver, pancreas, and jejunum (Table S4 in Supporting Information).

Histopathological Evaluation

Histopathology examination revealed treatment-related changes in the kidneys and lungs. In the kidneys, mild tubular dilation was noted in all animals injected with the dose level of 2520 mg/kg and sacrificed 10 min postdosing. In particular, the tubular dilation was seen in the high-dose naked MB (Figure ), high-dose MB-ICG (Figure ), and high-dose MB-RGD and high-dose MB-ICG-RGD animals (Figure ).

Figure 1

Naked MB, 2520 mg/kg, 10 min sacrifice time-point, kidney. Animal 3M13. Upper row: Low (left ×4) and high (right ×10) magnification view of the kidney. Mild tubular dilatation in the cortex and outer medulla (yellow arrow). Lower row: Higher magnification (×10). Mild tubular dilatation in the outer medulla (left) and in the inner medulla (right) (yellow arrow).

Figure 2

MB-ICG, 2520 mg/kg, 10 min sacrifice time-point, kidney. Animal 5M26. Upper row: Low (left ×4) and high (right ×10) magnification view of the kidney. Mild tubular dilatation in the cortex and outer medulla (yellow arrow). Lower row: Higher magnification (×10). Mild tubular dilatation in the outer medulla (left) and in the inner medulla (right) (yellow arrow).

Figure 3

MB-ICG-RGD, 2520 mg/kg, 10 min sacrifice time-point, kidney. Animal 9M50. Upper row: Low (×4) and high (×10) magnification view of the kidney. Mild tubular dilatation in the cortex and outer medulla (yellow arrow). Lower row: Higher magnification (×10). Mild tubular dilatation in the outer medulla (left) and in the inner medulla (right) (yellow arrow).

The tubular dilation involved the entire surface of the kidney, extending in the cortex, outer and inner medulla, and papilla. The lumen of the dilated tubules was empty, and the lining epithelium was either normal or minimally flattened. No degeneration, necrosis, inflammation, or abnormal contents was noted in the affected tubules. As no obstruction of the tubules was noted, it was assumed that the dilation was not related to distal absorption of the tubules. The terminology of tubule dilation was selected according to the criteria set in Frazier et al. in 2012.[14] No treatment-related renal changes were noted in any of the animals from the same dose group that were sacrificed 7 days postdosing. It may therefore be considered that the primary tubular dilation observed after 10 min from dosing had recovered after 7 days postdosing. In addition, it can be concluded that the tubular dilation did not progress to cause damage of the lining epithelium. It is suggested that the induction of tubular dilation was apparently related to the large-dose volume (i.e., 20 mL/kg) containing the MBs. In the lungs, minimal-to-mild intravascular accumulation of grayish amorphous (fluid-like) material was noted only in the animals injected with the dose level of 2520 mg/kg (naked MB, MB-ICG, MB-RGD, and MB-ICG-RGD) (Figures and 5).

Figure 5

MB-ICG-RGD, 2520 mg/kg, 10 min sacrifice time-point, high magnification of the lung (×20). Animal 9M49 (left) and 9M52 (right). Presence of homogenous greyish foreign body within the blood vessel is considered to be the test compound (yellow arrow).

The changes observed in the animals sacrificed 10 min postdosing were characterized by the presence of intravascular accumulation of grayish amorphous (fluid-like) material, whereas in the animals sacrificed 7 days postdosing, the intravascular accumulation of grayish amorphous (fluid-like) material was associated with foreign body granulomatous reaction, indicating that there was a clearing process of the intravascular accumulated material. Given that the intravascular accumulation of the injected material was associated with the development of foreign body granulomatous inflammation (i.e., a change observed only in the 2520 mg/kg-dosed groups), it is considered as an adverse effect, in accordance with the recently published Society of Toxicologic Pathology (STP) “Recommended (“Best”) Practices for Determining, Communicating, and Using Adverse Effect Data from Nonclinical Studies” (11). All other observed changes were considered as being spontaneous incidental findings and as having no relation to the treatment.

Discussion

Image-guided intraoperative navigation has become an integral component of modern cranial neurosurgery. Although preoperative MRI and CT scans can be reliably used for preoperative planning and intraoperative navigation, they are unable to account for intraoperative brain shifts due to patient positioning, CSF loss, tumor debulking, gravity, and cerebral edema.[15,16] Intraoperative US provides an improved ability to account for such anatomical changes.[17] Intraoperative contrast-enhanced US offers an evolving tool for neuronavigation and intraoperative imaging, with the unique advantages of combining dynamic imaging and real-time perfusion data. Contrast agents used for this purpose are composed of MBs that remain within the vasculature, thereby allowing characterization of the arterial and venous network. This offers the potential for differentiating normal brain from neoplastic tissue, delineating feeding arteries and draining veins within the surgical bed, and providing information with regard to hemodynamic alterations and real-time extent of the resection.[17,18] Notably, the safety profile of MBs allows multiple injections throughout the operation, highlighting changes in enhancement patterns at different stages of the surgery.[19,20] SonoVue MBs, a known contrast agent for US imaging, is made of MBs stabilized by a shell of phospholipids, encapsulating sulphur hexafluoride (SF6), which is an innocuous gas. Recent advancements in US and MBs technology have shown that MBs can be molecularly targeted by adding binding ligands onto the shell that allow them to specifically accumulate at certain target area sites.[21] We aim to develop a combined imaging technology for the diagnostic and tailored therapeutic intervention for patients bearing gliomas. As part of creating a new multimodal imaging system, we will employ a new generation of MBs that simultaneously act as contrast agents for intraoperative US imaging and intraoperative fluorescence-guided microscopic resection of gliomas. For this purpose, we have designed novel MBs attached to specific glioma ligands and engineered to act as platforms for delivering drugs to the tumor site, for which a toxicology investigation was necessary.

Our novel engineered MBs are targeted to molecules that are differentially expressed on the vasculature named αvβ3 integrins.[10] In this study, we assessed the potential toxic effects of (1) the naked MBs, (2) the MBs having the shells conjugated with RGD motif, (3) the MBs labeled with the fluorophore ICG for direct fluorescent visualization through the operative microscope, and (4) the MBs carrying both RGD and ICG molecules on the shells following a single IV injection to male Sprague-Dawley rats. No toxic effect was noted in our study following a single IV injection of any of the test items at a dose level of 252 mg/kg related to a single molecule or to a combination of molecules attached to the MBs surface.

The increased level of RBCs at the 10 min time-point should be interpreted with caution because the size of the injected MBs is comparable to that of the RBCs in the circulation (∼7 μm in diameter), and hence, the count may be skewed and the reported result may be an artifact. Rapid release of RBCs either from the spleen and/or bone marrow may also explain the increase in the RBC count. The spleen can serve as an RBC reservoir, and reversible release of RBCs in response to stress was reported in the rats.[22] In the current study, with the exception of the high-dose MBs-RGD-treated group, there was a corresponding decrease in the spleen size in the high-dose treatment groups (∼12–18% compared to vehicle control), although it may be still be attributed to weight variation among individual rats. However, the histopathological examination revealed no treatment-related changes in the spleen. The increase in RBC count and HGB levels and the decrease in platelets number in the high-dose treatment groups were reversible by the 7-day time-point. Absolute lymphocytosis was noted in a single case. As the group size is small (n = 3/time-point/group), it is difficult to determine whether this is treatment-related, especially because there were no comparable occurrences in the animals treated with high doses of the same test item.

All changes in blood biochemistry noted at the 10 min time-points were reversible by the 7-day time-point. The increase in the blood glucose level may be attributed to the high level of maltose (disaccharide of glucose) in the MBs formulation (>97%). The decrease in total protein and serum albumin levels and the increase in bilirubin levels in the high-dose ICG-treated groups are expected changes because ICG is known to bind serum albumin and to increase serum bilirubin, probably by inhibition of its hepatic uptake.[23]

Increase in organ weight (Table ), that is 10–20%, was noted in the spleen and lungs at the 10 min sacrifice time-point for any type of dose and MBs conjugation. At 7 day postdosing, the lung weight increased by 20–30% for the low-dose-treated animals. Changes in the organ weights were noted in the spleen and lungs in both time-points. However, aside from the changes in the spleen weight which may indicate erythrocyte release and contribute to the rise in circulating RBCs, there was no apparent toxicity from the changes observed in organ weight.

Histopathological evaluation revealed that the treatment-related changes in the kidneys consisted of mild tubular dilation, and it was seen in all animals injected with the dose level of 2520 mg/kg and sacrificed 10 min postdosing. The lumen of the dilated tubules was empty, and the lining epithelium was normal or minimally flattened. The treatment-related changes in the lungs consisted of minimal-to-mild intravascular accumulation of a grayish amorphous (fluid-like) material, possibly comprising of the injected test compound, which was observed only in the animals injected with the dose level of 2520 mg/kg. This accumulation was associated with foreign body granulomatous reaction at 7 days postdosing, indicating that a clearing process had started. In fact, it was reported that debris originating from MBs deflation and fragmentation during body circulation and US exposure remains on the surface of endothelial cells for a significant period of time during which they can be recognized by the immune system.[24]

For conclusions, we compared potential toxic effects of lipid naked MBs, ICG adsorbed onto MBs surface (MB-ICG), tripeptide motif Arg-Gly-Asp (MB-RGD) tethered to MBs and combinations of all of the above (MB-ICG-RGD) following a single IV injection in male Sprague-Dawley rats for better tumor vasculature visualization in the human brain. No definite toxic effect was noted in any of the test items at a dose level of 252 mg/kg, related to a single molecule or to a combination of molecules attached to MBs surface. This finding leads us to conclude that these agents are safe. The 10-fold higher dose level of 2520 mg/kg, however, of each of the tested items was associated with adverse effects. These results allow us to go ahead with planning an efficacy study before going to the first attempt to use this technology in clinical practice.