Intravitreal injection of polysorbate 80 : a functional and morphological study

Objective: to determine the functional and morphological effects at rabbits retina of PS80 concentration used in the preparation of intravitreal drugs. Methods: eleven New Zealand rabbits received a intravitreal injection of 0.1ml of PS80. As control, the contralateral eye of each rabbit received the same volume of saline. Electroretinography was performed according to a modified protocol, as well as biomicroscopy and retina mapping before injection and seven and ten days after. Animals were euthanized in the 30th day and the retinas were analyzed by light microscopy. Results: eyes injected with PS80 did not present clinical signs of intraocular inflammation. Electroretinography did not show any alteration of extent and implicit time of a and b waves at scotopic and photopic conditions. There were no morphological alterations of retinas at light microscopy. Conclusion: intravitreal injection of PS80 in the used concentration for intravitreal drug preparations do not cause any functional or morphological alterations of rabbit retinas. These results suggest that PS80 is not toxic to rabbit retinas and may be safely used in the preparation of new lipophilic drugs for intravitreal injection.


INTRODUCTION
D rug access to the retina and choroid has always been a challenge to ophthalmologists due to the existence of two anatomic barriers (internal and external blood-retinal barriers) that impairs penetration of drugs in the posterior segment of ocular bulbus.Treatment of blindness secondary to most prevalent retina and choroid diseases (macular degeneration related to age and diabetic retinopathy) has changed dramatically with the use of intravitreal injection of therapeutic agents in the posterior segment of ocular bulbus 1 .Intravitreal injection of drugs overcomes external blood-retinal barrier and assures that retina and choroid receive therapeutic level of drugs, lowering significantly systemic absorption and consequent toxicity.According to Brazilian and World legislation, intravitreal injection of drugs is a surgical procedure and must be performed under rigorous aseptic technique.
Most commonly injected drugs in the vitreous are monoclonal antibodies (particularly inhibitors of the vascular endothelium growth factor), corticosteroids and antibiotics, but, in theory, any drug can be injected in the vitreous.However, some pharmacological aspects must be considered, such as the aqueous solubility, pharmacokinetics and biochemical proprieties of the compounds, as well as their interaction with the vitreous 2 .
Polysorbates, a class of non-ionic surfactants, are very useful excipients in several pharmaceutic formulations for intravenous use with different objectives.
Polysorbates increase drug solubility in suspensions with low or no-solubility, to obtain aqueous dispersions.In those cases, surfactant concentration varies from 0.05% to 0.5%, depending on the solid content of formulation.
Polysorbates also are used in the formula of injectable solutions to increase absorption of soluble drugs due to micelle formation.Also, polysorbates are useful to sta- PS80 has no neurotoxicity in newborn rats following administration of high oral doses during pregnancy, and do not cause development disturbances, functional alterations of central nervous system, and alterations of locomotion or of reflexes 3 .In adult animals, oral intake of high doses of PS80 is safe in mice, rats, dogs and apes 4 .
However, intraperitoneal injection of PS80 in newborn female rats cause morphological and functional alterations of uterus and ovaries 5 .Also, PS80 may be associated to non-immune anaphylactic reaction, following intravenous administration during pregnancy 6 .PS80 effects on the eye surface were studied in several experimental models.Sub-tenon injection of PS80 in rabbits caused less toxicity in eye surface than other commonly used excipients commonly used in topic formulations for ocular use, such as carboxymethylcellulose, polyethylene glycol, benzylic alcohol, benzalkonium chloride and methylcellulose 7 .PS80 also seems to have a protective mechanism in the corneal epithelium of cells maintained in culture, reducing the toxicity induced by benzalkonium chloride, a commonly used excipient used in eye drops 8,9 .
Formula most used commercially of triamcinolone acetonide (TA) contains PS80.TA is a synthetic glucocorticoid with long-lasting effect that has been widely used in the treatment of retinal diseases by intravitreal injection, but safety studies show controversial results.Some in vivo experimental studies suggest that TA intravitreal injection, which formula contains PS80, is safe 10- 12 .However, other experimental studies suggest that TA formulation without preservative is less toxic to retina after intravitreal injection than most common formulas [13][14][15] .Since TA vehicle formulation has many compounds, such as benzylic alcohol, carboxy-methylcellulose, PS80, sodium hydroxide and hydrochloric acid, the role of each compound in retinal toxicity is still uncertain [16][17][18] .
Although PS80 is frequently used in the preparation of formulations for ocular use, including drugs for intravitreal use, its effect on retina after intravitreal injection has never been studied.The objective of the present study is to determine functional and morphological alterations of rabbit retina caused by PS80, at the same concentration used for the preparation of new drugs for intravitreal use.

METHODS
Eleven New Zealand non-pigmented rabbits (weighting from 2 to 3 kg) were used.

Intravitreal Injection
Right before intravitreal injection, it was performed paracentesis of anterior camera (30G needle), removing 0.1ml of aqueous humor to avoid significant increase of ocular pressure.Under direct visualization, right eye of each animal were submitted to an intravitreal injection of 0.1ml of PS80 (0.4% w/v, pH 6.6-6.

Ophthalmologic Exam
Animals were submitted to biomicroscopy and indirect binocular ophthalmoscopy before and right after intravitreal injections, repeated in the 7th and 14th days after injections.

Electroretinography
Both eyes were submitted to full-field electroretinography (ERG) before and after seven and 14 days of injection.For ERG, contact lens were applied attached to bipolar corneal electrodes in both eyes and a ground electrode was fixed at the animal ear.Animals were positioned in a Faraday cage (60x60cm) and the luminous stimulation was generated by a Ganzfeld stimulator controlled by a computer system.ERG signs

Morphological Analysis
Animals were sacrificed 30 days after intravitreal injections and their eyes were processed for light microscopy, after euthanasia, posterior eye segments were fixed in ALFAC solution.After inclusion in paraffin, they were submitted to 7µm slices that were dyed with hematoxylin and eosin and analyzed under light microscopy.Thickness and retinal organization were analyzed at retinal inferior medium periphery of all eyes.

Statistical Analysis
Amplitude and implicit times were described as medium ± standard deviation.Results were analyzed by ANOVA variance analysis test using repeated measures.Fisher test was used as post hoc test to determine significant difference among medias identified by ANO-VA.Naka-Rushton equation parameters (b wave amplitude versus intensity of luminous stimulus) were initially evaluated by ANOVA variance analysis test of one and two factors, with adequate Bonferroni correction to the number of comparisons between groups and intervals.
Differences were considered significant when p was lower than 0.05.

Clinical Aspects
No alterations were observed at biomicroscopy and indirect binocular ophthalmoscopy during the follow-up period (cataract, cells at anterior and posterior cameras, retinal lesion and endophthalmitis).(amiodarone, ciclosporin and decetaxel) 20 and it is used as excipient in vaccines 21 .Although PS80 is usually considered a safe component for systemic use and of several drugs for intravitreal use that include it in their formula, clinical and experimental studies of its safety are controversial, regarding intravitreal injections 9,[13][14][15]22 . Sine formula of drugs injected at vitreous contain many other agents (preservatives, surfactants, solvent and agents that stabilize pH and tonicity), the role of each agent regarding retinal toxicity is still uncertain [16][17][18]23 .One of the agents present in TA preparation injected in vitreous is benzylic alcohol, that has preservative and antibacterial proprieties.It has already been shown that benzylic alcohol causes early non-immunologic contact reaction in humans.Also, experimental data on teratogenesis and toxicity to reproductive processes are still controversial 24 .

Electroretinography
Maia et al. 15 evaluated clinical and morphological alterations of rabbits retina secondary to sub-retinal injection of supernatants of TA solutions containing benzylic alcohol or not.Both tested solutions contained PS80 in their formula.Authors showed that eyes injected with TA supernatant that did not contain benzylic alcohol had lower grade of retinal lesion, suggesting that the presence of benzylic alcohol may, at least in part, be related to retinal toxicity.
Biochemical parameters also have a very important role in drug retinal toxicity.Osmolarity and pH may be responsible for alterations detected at ERG, indirect binocular ophthalmoscopy, angiography with fluorescein and histology [25][26][27][28] .Eyes that received intravitreal injections of compounds with non-physiologic pH and osmolarity may present retinal detachment 27 , alterations of a and b waves at ERG (lowering of amplitude and increase of implicit time) [25][26][27] and extra-and intracellular edema 25 .PS80 used in this study is the commercially This study suggests that PS80, at the used concentration in the preparation of drugs for intravitreal use, is not toxic to rabbits retina and may be used safely as a component of the preparation of suspension of lipophilic drugs.However, pharmacological and additional retinal toxicity studies are needed to determine the safety of PS80 in multiple intravitreal injections in the same eye, since this is a very common treatment in daily practice.
Animals were treated according to the recommendations of the Association for Research in Vision and Ophthalmology Statement for the Use of Animals in Ophthalmic and Vision Research.Experiments were approved by the Ethic Commission of Animal Experimentation of Biomedical Science Institute of the University of São Paulo #029, sheet 43, book 2, and by the Ethic Commission of Research in Animals of the Psychological Institute of University of São Paulo (#07.2010).Animals were kept in individual cages in a cleardark cycle of 12 hours, and free access to water and food.Pupils were dilated with tropicamide 0.5% eye drops and eyes were anesthetized with proxymetacaine eye drops.Before intravitreal injection, electroretinography and euthanasia, animals were anesthetized with intramuscular injection of ketamine hydrochloride (35mg/ kg) and xylazine hydrochloride (5mg/kg).Animals were sacrificed by intravenous injection of sodium pentobarbital (40mg/kg).
were amplified and digitalized.Data were analyzed by LabVIEW® computer software.Luminous stimuli band was calibrated to vary from 0.3 to 1000 Hz.The protocol used for ERG acquisition was the one suggested by the International Society for Clinical Electrophysiology of Vision (ISCEV)19 modified for acquirement of some additional information in experimental studies.For obtaining scotopic answers, animals were adapted in the dark for 30 minutes and were submitted to stimuli with five different luminous intensity (0.001, 0.01, 0.1, 1 and 10 cd.s/m 2 ).After adaptation for ten minutes to light, they were submitted to luminous stimuli with 1cd.s/m 2 with background illumination of 25cd/m 2 .A and b waves were recorded and their amplitude and implicit time were analyzed.A wave amplitude was measured from baseline to minimum amplitude registered after presentation of stimuli.Implicit time was measured from the beginning of luminous stimulus until the a wave peak.B wave amplitude was measured from a wave peak to b wave peak, and the implicit time of b wave corresponded to the necessary time for that peak.ERG dynamic interval at scotopic condition was evaluated by a graphic of median amplitude versus luminous stimulus intensity.Curves were obtained by the equation of Naka-Rushton: V=Vmax.In/Kn + In; Vmax is the saturation amplitude of b wave, I is the intensity of luminous stimulus, K is necessary luminous intensity for obtaining 50% of Vmax and n is the curve inclination, representing the dynamic interval of the measured wave.

Figure 3 .
Figure 3. Ratio of scotopic V max x photopic b wave amplitude.

Figure 5 .HistologyFigure 6
Figure 5. B wave amplitude x stocopic and photopic a wave.
available formula that is universally used in the preparation of drugs for intravitreal use (Tween® 80).Twe-en® 80 has a pH very close to normal (6.6-6.8), and is iso-osmolar (288 a 318 mOsm/kg H 2 O).Therefore, it is very unlikely that biochemical factors associated to PS80 used in this study (such as pH and osmolarity) may cause retinal toxicity.Since this is the first publication about the retinal effects of intravitreal injection of PS80, it is not possible to compare it directly with other results.However, PS80 is present is several drugs that are injected in vitreous of animals and studies don't show any retinal toxicity, such as Triesence® (a new TA formulation without preservative, specifically produced for intravitreal injection), Remicade® (infliximabe)[29][30][31] and Humira® (adalimumabe)[32][33][34] .These last two are monoclonal antibodies that block tumor necrosis factor approved for the treatment of gastrointestinal, rheumatic and dermatologic diseases, that have been used for the treatment of auto-immune uveitis.This study has some limitations.Only one concentration of PS80 was tested.It did not allow us to determine the maximal safe dose for intravitreal injection, but the concentration tested is used in all formulations of drugs for intravitreal use.Also, no immune-histochemical analysis or ultramicroscopic studies were performed to detect subtle or subclinical alterations of retinal toxicity.This is an experimental study and the results may not represent integrally the findings of human inflamed eyes.Limitations of the use of rabbit eyes in the studies of drug retinal toxicity include retinal vascularization differences in relation to human eye, and differences of the eye volume of rabbits and humans.In spite of the cited limitations, this study results have low variability, in special of ERG results, even considering that exists several variability factors that are very difficult to control in studies with ERG in animals, that could influence the results35 .
Left eye received an intravitreal injection of sterile saline and used as control.