In Vivo Intraocular Distribution and Safety of Periocular Nanoparticle Carboplatin for Treatment of Advanced Retinoblastoma in Humans


To study the intraocular distribution and safety of polymethylmethacrylate nanoparticles loaded with carboplatin after posterior subtenon injection in humans.


Prospective, interventional, comparative case series.


Six patients (mean age: 26.83 ± 7.5 years), scheduled to undergo planned uniocular enucleation in an institutional setting, were randomly divided into 3 groups. Each group received a 10 mg/mL posterior subtenon injection of nanoparticle carboplatin in the eye to be enucleated. Two eyes were enucleated 6, 24 and 72 hours post injection. Intravenous blood was collected during enucleation. The concentration of carboplatin reaching various intraocular tissues was determined by inductively coupled plasma atomic emission spectroscopy. The drug toxicity in the ocular tissues was assessed by histopathology and high-resolution transmission electron microscopy.


The highest level of carboplatin was detected in retinas (8.33 ± 1.69 mg/g), up to 24 hours post treatment. The intravitreal concentration continued to increase gradually until 72 hours (3.46 ± 0.26 mg/g). The choroids and lenses showed very low levels of carboplatin after 6 hours, with negligible amounts at 72 hours. No signs of tissue damage were observed on histopathology or electron microscopy. Intravenous concentration of carboplatin was undetectable in all patients.


Results may indicate an increased facilitated trans-scleral transport of nanoparticle carboplatin, with a sustained-release behavior but without any associated short-term ocular or systemic side effects in humans. The very high concentrations achieved in vitreous and retina after a single posterior subtenon injection may be clinically useful for adjunctive treatment of advanced intraocular retinoblastoma with vitreous seeds. However, further studies are needed to assess long-term toxicity and clinical efficacy.

Retinoblastoma is a common tumor of infancy and childhood. In the past decade, because of ongoing concerns regarding external-beam radiotherapy, systemic intravenous chemotherapy has become the mainstay in the management of intraocular retinoblastoma. Systemic chemotherapy has been shown to decrease the size of intraocular retinoblastoma effectively, allowing for more conservative, globe-preserving therapeutic treatment options. However, systemic chemotherapy is not without its associated life-threatening systemic toxicities in these infants.

Ideally, chemotherapy administered for intraocular retinoblastoma would be distributed exclusively to the intraocular space of the affected eye, thus increasing intraocular concentration, but without any significant systemic exposure and therefore no systemic toxicity. This has prompted recent investigations to evaluate high-dose, focal chemotherapy protocols. Periocularly injected conventional carboplatin by itself has been reported to have a low efficacy in the treatment of advanced intraocular retinoblastoma. Studies have, however, shown better efficacy when adjunctive periocular injection of carboplatin is used along with systemic chemotherapy than when patients have been treated with systemic chemotherapy alone for advanced retinoblastoma. It is believed that a high intracellular carboplatin level obtained in the tumor tissue is associated with increased tumor control. This has led to postulates that if we could further increase the intraocular concentration of carboplatin by periocular injections, it would not only be a good adjunct to intravenous systemic chemotherapy for intraocular retinoblastoma, but it may also be a potential option for ameliorating or decreasing the amounts of intravenous chemotherapy the patients are currently subjected to.

Periocular carboplatin is being used globally at many centers, and reports of local toxicity, such as atrophy of orbital fat, ocular motility restriction, optic nerve damage, and preseptal cellulitis have been reported. However, these undesired effects are thought to be due to injection technique, use of multiple injections, or rapid dispersion of carboplatin to the orbital tissue and vasculature.

Thus, to prevent periocular toxicity and to improve efficacy, a sustained-release system of carboplatin, which can improve the trans-scleral migration of carboplatin, along with maintenance of the drug therapeutic window with a single injection, may be required. With the advent of nanotechnology, polymeric biocompatible nano-carriers have emerged as a suitable vehicle for passive targeting of the drug to the site of action. These multifunctional biocompatible nano-structures can prolong the half-life of drug circulation because of their unique characteristics of relatively large surface (functional) area, improved drug stability and sustained-release behavior in vivo, which may also reduce side effects and without increasing immunogenicity (as may occur in other drug carriers like albumin or human protein).

In this preliminary study, our group has attempted to examine the safety of these nanoparticles loaded with carboplatin in vivo in human eyes and to assay the concentrations of carboplatin attained within the various intraocular tissues after posterior subtenon injection in eyes prescheduled to undergo enucleation for other causes.

Materials and methods

Patient characteristics

The study was an institutional, prospective, double-blinded, interventional comparative-case series approved by the Apollo Hospitals Institutional Review Board, Hyderabad, India, with informed consent by patients to participate in research. Written informed consent was obtained from the patients or their guardians (if minor) for the injection of periocular nanoparticle carboplatin prior to their scheduled enucleation. The described research protocol adhered to the tenets of the Declaration of Helsinki. Six patients who were scheduled to undergo enucleation of 1 eye because of intraocular diseases were identified. Detailed patient characteristics are provided in Table 1 .

Table 1

Characteristics of the 6 Patients Selected to Study the Intraocular Distribution and Safety of Periocularly Injected Nanoparticle Carboplatin for Treatment of Advanced Retinoblastoma a

Patient Age (y) Sex Eye Involved Type of Disease Reason for Enucleation
1 18 Male Right Anterior staphyloma with disorganized anterior chamber Painful blind eye for cosmetic enucleation
2 21 Female Left Glaucomatous eye Painful blind eye for cosmetic enucleation
3 33 Male Right Glaucomatous eye Painful blind eye for cosmetic enucleation
4 27 Female Right Anterior staphyloma with disorganized anterior chamber Painful blind eye for cosmetic enucleation
5 38 Male Left Glaucomatous eye Painful blind eye for cosmetic enucleation
6 24 Male Right Anterior staphyloma with disorganized anterior chamber Painful blind eye for cosmetic enucleation

a These patients were scheduled to undergo planned uniocular enucleation due to other intraocular diseases and were randomly divided into 3 separate groups of 2 patients each. Each group received 10 mg/mL posterior subtenon injection of nanoparticle carboplatin in the eye to be enucleated. Two eyes each were enucleated 6, 24 and 72 hours post injection.

Inclusion criteria were patients with painful blind eyes, corneal opacities or anterior staphyloma, glaucomatous blind eyes and any sort of retinal degenerative disease.

Exclusion criteria were (1) patients with retinoblastoma; (2) patients with any traumatic damage to the anterior or posterior segment (to ensure that the sclera is intact and not damaged); (3) patients who had undergone prior ocular surgery; and (4) patients with any previous treatment by platinum-based chemotherapy.

Preparation and physicochemical characterization of carboplatin-loaded polymethylmethacrylate nanoparticles

Polymethylmethacrylate nanoparticles were prepared by free-radical emulsion polymerization of methyl methacrylate from an aqueous solution containing carboplatin in the presence of sodium dodecyl sulfate and thermal initiator ammonium persulphate, with modifications (D. Kalita. Carboplatin-loaded polymeric nanoparticles as potential therapeutic candidates for ocular and other carcinomas. Indian Institute of Technology, Mumbai, PhD dissertation, 2013). Freshly distilled methylmethacrylate was added drop-wise, in an inert atmosphere, to a solution of sodium dodecyl sulfate and excipient free carboplatin, in Millipore water (50 mL) at 78°C–80°C in a 100 mL Schlenk flask fitted with a thermometer and a water condenser. The polymerization was initiated by slow addition of initiator ammonium persulphate at 78°C–80°C on a hot well plate for 24 hours in an inert atmosphere, with high-speed magnetic stirring, followed by cooling to room temperature by stirring for 10 hours at 25°C. The white dispersion of polymethylmethacrylate nanoparticles was purified by dialysis against deionized water with dialysis membrane for another 24 hours and stored at 8°C for further use. A part of the nanoparticles was lyophilized (at −54°C, 0.12 mbar) to get free-flowing powder and was stored at −20°C. The particle size was analyzed by dynamic light scattering, and the bonding and structure of the molecule was analyzed by 1 H, 13 C nuclear magnetic resonance spectroscopy. Zeta-potential measurements were used to measure the surface electric characteristics of the particle and its variation by using various preparation parameters. X-ray diffraction was used to measure surface crystallinity and gel-permeation chromatography was used for estimation of the molecular weight of polymethylmethacrylate. Dynamic light scattering was used over a period of 1 month to assess the stability of nanoparticle in the fluid state. Fourier transform infrared spectroscopy was used to examine the nature of bonding or chemical interaction between carboplatin and the polymethylmethacrylate in the lyophilized sample and for elucidating the polymer backbone. Transmission electron microscopy and scanning electron microscopy measurements were carried out to estimate the stability of particle size in the formulation, surface morphology, poly-dispersity, and particle shape. The in vitro release pattern of carboplatin from the polymethylmethacrylate matrix was evaluated by dialysis membrane in phosphate-buffered saline (pH 7.4).

Protocol design for posterior subtenon injection and surgery

Six patients were randomly divided into 3 treatment groups of 2 patients each. Each group received a fixed dose of 1 mL of 10 mg/mL posterior subtenon injection of freshly prepared nanoparticle carboplatin suspension at 6 hours, 24 hours or 72 hours prior to enucleation, respectively. Drug injection was performed carefully by an ophthalmologist trained in ocular oncology (DS). Care was taken to avoid penetration of the globe, and the globe was evaluated in detail after the injection. An elective cosmetic myoconjunctival enucleation technique with silicone ball insertion was performed within the scheduled treatment time, depending on the group to which the patient had been randomized. Intravenous blood samples were taken from each patient immediately after the enucleation procedure and stored at 4°C for drug estimation.

Tissue sampling

The external area of each eye was cleaned with cold saline, and the eyes were wrapped in gauze with cold saline. Vitreous was removed with an 18 gauge needle and stored at −80°C in pre-weighed vials for analysis of carboplatin. Each eye was fixed in 10% formalin and then sectioned. Post dissection, the individual tissue parts were transferred to pre-weighed vials, and the weight of each tissue was noted. The lens, retina, choroid, and sclera were dissected apart and stored in 10% formalin for spectroscopy analysis. Separate aliquots were preserved in formalin for histopathology. Individual tissues were labeled in a blinded manner (the code known to only 1 person) by a laboratory technician and stored at 4°C before analysis.

Estimation of carboplatin in vitreous and tissues by inductively coupled plasma atomic emission spectroscopy

The determination of the level of carboplatin in ocular tissues was achieved by the measurement of platinum present in it by Inductively Coupled Plasma Atomic Emission Spectroscopy (Ultima 2; JobinYvon Technologies, Horiba Scientific, Kyoto, Japan), limit of detection, 10 parts per billion, with a plasma gas (argon) flow rate of 12 L/min, an auxiliary gas flow rate of 0.2 L/min, a sample uptake of 2.5 mL/min, and an integration time of 5 seconds. To remove the organic content in tissues, the tissue samples were digested with concentrated nitric acid (60%) at 120°C for 6 hours and then evaporated to dryness for 3 repetitive cycles. This was diluted with Millipore water prior to analysis. The investigator was blinded to the patient sample. The estimated amount of carboplatin was based on the total platinum content in the tissue, irrespective of the binding nature or uncertain degradation mechanism of the carboplatin molecule in vivo. All data were represented as mean weight of carboplatin per dry tissue weight in milligram per gram (mg/g).

Ocular histopathology and transmission electron microscopy

The intraocular tissue parts were fixed in 4% paraformaldehyde overnight at 4°C, processed routinely, sectioned and stained with hematoxylin and eosin. Light-microscopic examination was performed on all histopathologic sections, which were interpreted by an experienced ocular pathologist (KC). Eyes were evaluated for evidence of choroidal and retinal toxic effects. Any morphologic disruptions in tissues were noted.

Tissue retrieved from paraffin wax with xylene at 4°C was put in sodium cacodylate buffer at 4°C overnight and then washed with sodium cacodylate buffer twice at room temperature and stained with 1% osmium tetroxide for 2 hours. After washing and staining with 5% uranyl acetate, the ocular tissues were then dehydrated with gradually increasing concentrations of ethanol. The anhydrous tissues were washed with propylene oxide and then embedded with epoxy 332–732 resin (Electron Microscopy Sciences, Hatfield, Pennsylvania, USA). The blocks were sectioned at 70 nm using a diamond knife. Ultrathin sections were collected on 200-mesh copper grids. After staining with 2% uranyl acetate, the grids were viewed in a JEOL 2100 High Resolution Transmission Electron Microscopy (JEOL, Tokyo, Japan). Multiple images were grabbed at differing locations because of the huge number of tissues under investigation.

Statistical analysis

Mean ± SD were reported for each group, and significance of the data was analyzed by the Student t test. P < 0.05 was considered statistically significant.


The average diameter of polymethylmethacrylate nanoparticles in situ loaded with carboplatin was determined to be 110 ± 10 nm based on transmission electron microscopy and scanning electron microscopy analyses. The particles were nearly spherical, with smooth surface morphology, as observed on high resolution transmission electron microscopy. Hydrodynamic diameters of the nanoparticles were comparable to those obtained from electron microscopy, and they were nearly monodisperse (polydispersity index ∼0.2). The surface charge on the nanoparticles was highly negative, as obtained from zeta potential measurements. The lyophilized nanoparticles were readily dispersible in water as well as in phosphate-buffered saline 7.4, with mild vortexing. The encapsulation efficacy of carboplatin in the nanoparticle formulation was 56%, with a quantitative carboplatin content of 8597.05 μg/g of lyophilized powder. The in vitro release profile of carboplatin from polymethylmethacrylate nanoparticles followed a biphasic pattern, which indicated the sustained-release behavior of these nanoparticles.

The mean age of the patients was 26.83 ± 7.5 years. The maximum concentration of carboplatin was found in retinas until 24 hours post injection (8.33 ± 1.69 mg/g), which decreased at 72 hours (0.39 ± 0.11 mg/g). The second highest concentration was found in the vitreous, and it increased gradually to 3.46 ± 0.26 mg/g at 72 hours. The choroid and lens showed very low levels of carboplatin after 6 hours, with negligible amounts at the end of 72 hours ( Table 2 ; Figure ). The histopathologic evaluation of choroid, lens and retina showed no signs of necrosis or inflammation in these tissues. The high-resolution transmission electron microscopy images of various tissues also revealed the minute details of preservation of these ocular tissues. Retinal pigment epithelium consisted of monolayer of cells containing intracytoplasmic melanosomes, which were clearly visible at all treatment durations. The inner and outer nuclear layers were also intact. In choroid also, except for procedural artifacts, no signs of tissue damage or architectural distortion were noted; red blood cells and melanocytes were intact. Serum plasma concentrations of carboplatin in each patient were undetectable by inductively coupled plasma atomic emission spectroscopy analysis.

Jan 8, 2017 | Posted by in OPHTHALMOLOGY | Comments Off on In Vivo Intraocular Distribution and Safety of Periocular Nanoparticle Carboplatin for Treatment of Advanced Retinoblastoma in Humans
Premium Wordpress Themes by UFO Themes