Stability of voriconazole 10 mg/mL ophthalmic solution during 90 days

Introduction

Keratitis is an inflammation of the cornea, the transparent membrane that covers the iris and pupil of the eye. Migration of inflammatory cells results in corneal opacification and can lead to a complete blindness. Its incidence is constant, due to steroids or immunosuppressive treatments and contact lenses [1]. Infectious keratitis in contact lens wearers are more and more frequent and constitute a real diagnostic and therapeutic emergency [2, 3]. The risk factors are essentially: poor hygiene, permanent wearing, contamination of rinse solution [4]. In developing countries it is more commonly caused by ocular trauma sustained during agricultural work [1, 5, 6]. The minimal annual incidence is estimated of 23.6 for 100,000 people, with the highest rate in Asia (33.9) and Africa (13.5), and the lowest in Europe (0.02) [7]. Keratitis represent a therapeutic emergency where even a delay of a few hours can affect the ultimate visual result [8]. If untreated, keratitis can lead to corneal scarring and blindness [9]. It’s important to treat keratitis before the corneal tissue is destroyed and scare is formed. If the corneal ulcer is severe, corneal sample is realised by scarping the corneal epithelium and an emergency treatment by association of broad spectrum antibiotic (ceftazidime, gentamycine, vancomycine) called “fortified antibiotics” eye drops is used empirically before microorganism identification. Once the microorganism is identified the treatment is adapted, and administered over a long period.

In the case of fungal keratitis, the antifungal agents that can be used are: polyenes (amphotericin B, natamycin), azoles (fluconazole, itraconazole, voriconazole, and posaconazole), pyrimidine (5 flucytosine) and echinocandins (caspofungin) [1]. The only commercially available antifungal eye drop is natamycine 5% Natacyn^®, however it has a poor penetration, and is only effective against superficial infections [9, 10]. Besides, 10 mg/mL voriconazole solution is a broad spectrum antifungal that has been shown to be effective in treating keratitis caused by many saprophytic fungi, most commonly, Aspergillus, Fusarium, and Candida [11]. Except for natamycin, no ophthalmic solutions are available on the market justifying their preparation in hospital pharmacies. To face increasing needs of these ophthalmic solutions stability studies are required to implement hospital preparations with an extended shelf life.

Several stability datas on voriconazole ophthalmic solutions are available [12], [13], [14], [15], nevertheless they did not address long storage refrigerated conditions without the use of benzalkonium chloride preservative. Several shortfalls from previous papers are addresses by the present work e.g. long refrigerated storage instead of frozen, microbiological assessment study during storage and in use. Moreover, to the best of our knowledge no data are available related on eye drop solutions microbiological contamination issue during real-life simulation use. Considering all these data this study aims to assess the physico-chemical and microbiological stability of 10 mg/mL voriconazole ophtalmic solution for 90 days of storage at +2 to 8 °C in two different High-Density-Polyethylene (HDPE) vials, one with a sterility preserving cap Novelia^® (Nemera) and the other without classical sterility preserving cap. In addition, an evaluation of the microbiological stability in real-life was carried out for 15 days by simulation of use and storage between at +2° and +8 °C away from light.

Material and methods

Results

Physicochemical stability

Validation of the HPLC-UV assay method

The method was linear (r²>0.9999). The regression was different from zero y=25.282x + 1.1118. The interday repeatability of the five quality control voriconazole solution was fully satisfactory, with accuracies ranging from 100.26 to 100.84% and the precision (RSD %) ranging from 0.237 to 0.523%. The RSD % values for precision were in compliance with ICH standards, which require a RSD of ≤2% and an accuracy of 98–102%. The detection (LOD) and quantification (LOD) limits were 0.07 and 0.23 mg/mL respectively (Table 1).

Table 1:

Interday repeatability of voriconazole stock solutions.

	QC1	QC2	QC3	QC4	QC5
Theorical concentration, mg/mL	3	7	10.2	13	17
Mean, mg/mL	3.02	7.03	10.28	13.10	17.04
SD	0.02	0.03	0.03	0.04	0.04
RSD %	0.51	0.40	0.27	0.30	0.23
Accuracy %	100.64	100.45	100.83	100.74	100.26

The method was stability indicating showing degradation products in stressed conditions (acid, basic and oxidative). Their retention times were: 4.3 and 4.5 min (Table 2). No degradation product was observed for light exposure and dry heat (Figure 1).

Table 2:

Table of the degradation study.

	Dry heat 70 °C 1H	HCl 1N to 80 °C for 1H	NaOH 1N to 80 °C for 1H	H2O2 30% 3H at 100 °C	UV 48H 30W
AUC 1 (Tr: 4.3 min)	5.89	15.23	204.53	118.27	0.78
AUC 2 (Tr: 4.5 min)	0.71	1.90	27.93	7.03	0.08
AUC voriconazole	237.76	220.56	NA	219.74	234.17
AUC total	244.36	237.70	232.46	345.04	235.04
% of degradation	3%	7%	100%	36%	0%

Figure 1:

Chromatograms of voriconazole reference solution (A), and degradation products after stressed conditions UV (B), 70 °C dry heat (C), acid (HCl 1N at 80 °C) (D), basic (NaOH 1N at 80 °C) (E), H₂O₂ 30% at 100 °C (F).

In both containers no significant variation of visual aspect (colour, limpidity), pH (6.17 ± 0.1), osmolality (549 ± 5.06 mOsm/kg) and concentration for 90 days (Table 3) was observed. Value of pH and osmolarity remained stable during the study. The results of sub visible particle analyses are presented in Table 3 (Table 3), the particulate contamination remained stable during the study and there was no difference between the two contents.

Table 3:

Physicochemical and sub visible particles count results of eye drop solutions at D0 and D90 days in HDPE vials with classical cap and vials with Novelia^® cap.

		HDPE vials with classical cap		HDPE vials with NOVELIA® cap
		D0	D90	D0	D90
pH		6.11 ± 0.06	6.21 ± 0.18	6.16 ± 0.10	6.46 ± 0.15
Osmolality, mOsm/kg		549.0 ± 4.4	541.0 ± 3.8	543.0 ± 8.2	534.0 ± 6.0
Concentration, mg/mL		10.65 ± 0.18	10.07 ± 0.03	10.18 ± 0.35	9.97 ± 0.12
% of the initial concentration		100%	94.55%	100%	97.93%
Sub visible particles/mL	≥10 µm	7.3 (±3.1)	38 (±25.2)	3.7 (±1.5)	20.3 (±9.0)
	≥25 µm	1.0 (±1.0)	6.7 (±7.4)	1.0 (±1.7)	4.7 (±3.8)

1. Results are mean ± SD of three determinations.

Mean concentrations for each sample time were determined and converted to percentages of initial concentration. The initial concentration of voriconazole at D0 in each container was designated as 100%, and subsequent sample concentrations were expressed as percentages of the initial concentration (Table 3).

Concentrations remained stable for 90 days at 4 °C, the mean concentration for the three batches was 10.06 ± 0.34 mg/mL. At time zero, impurity have already been present over all chromatogram and there retention time were 4.3 and 4.5 min. No additional peak was observed during the study period for vials without sterility preserving classical cap and those with Novelia^® cap (Figure 2). Voriconazole 10 mg/mL ophthalmic solutions were considered stable at +2 to 8 °C during 3 months, according to their concentrations, always within the 90–110% stability range.

Figure 2:

Chromatograms of voriconazole eye drop solutions 5 mg/mL in HDPE vials with classical eye drop insert at D0 (A) et D90 (B) and HDPE vials with Novelia^® insert at D0 (C) and after 90 days (D) of storage at +2 to 8 °C. (A′), (B′), (C′) and (D′) are respectively the close-up of the chromatogram baseline.

Discussion

Several data on voriconazole ophtalmic solution stability were published and their conditions were different from ours in term of storage conditions, formulation, physicochemical analytical method and sterility study method. Two of them studied stability at −20 °C for 90 days [14, 15] and two between +2° and +8 °C [13, 23]. To implement a batch production and for easier management of the conservation, study was performed between +2° and +8 °C for 90 days.

For physicochemical stability, our results are comparable to Al-Badriyeh et al. [13] for the same study conditions (+2 to 8 °C, in HDPE lightproof vials, HPLC-UV analytical method). Most of the stability studies where performed with non chiral method, except Roche et al. which demonstrate the lack of the racemic form and the results were comparable to ours. Dupuis et al. [12] assess too short conservation condition to perform sterility test. Al-Badriyeh et al. [13] added preservative in their formulation. Frozen storage and the stability of the frozen solutions studied by Amoros-Reboredo et al. and Roche et al. [14] requires additional equipment and accurate validation of the thawing process. According to previous published works, physico-chemical stability was expected in our storage conditions at +2 to 8 °C in HDPE lightproof vials equipped with a classical cap.

Battour et al. [24] showed the influence of sterilization method on pH, their study found a decrease of the eye drop solutions pH stored in gamma radiation sterilized LDPE eye drop vials. This study highlighted the impact of the container nature, sterilisation method and the drug solution in contact with container on the drug stability. HDPE are also known to be affected by gamma irradiation, the thickness of the vials wall and the direct interaction with the drug solution [25].

However in our study, the pH remained stable for 90 days irrespectively of the sterilisation method of the HDPE vials sterilized by gas (ethylene oxide) or gamma irradiation methods for the vials without and with sterility preservative cap respectively.

For sub visible particles, European Pharmacopoeia recommends that the average number of particles present in the containers tested does not exceed 1,000 per millilitre for particles equal to or greater than 10 µm in size and does not exceed 100 per millilitre for particles equal to or greater than 25 µm in size [26]. Sub visible particle contamination is in the acceptable range defined in the latest monographie of EP for ophthalmic solutions [26] and are similar to Roche et al. [15].

To study the activity of voriconazole, Roche et al. [15] used chiral chromatographic to identify racemization of voriconazole which is not an active form, whereas Amorós-Reboredo et al. [14] performed an in vitro sensitivity test to determine minimum inhibitory concentration (MIC) over the study period. Both found that voriconazole activity was preserved after a long term storage of 90 days, considering the previous published datas voriconazole activity was maintained throughout the 90 days, however MIC determination could complete our study.

Lack of microbiological contamination of ophtalmic solutions is required by European Pharmacopeia [26], thus a complete sterility assay as described in the PE [22] was performed. This method is the preferential method to study product with an antimicrobial activity. It give additional information on the microbiological quality of the eye drop solution free from microbiological agent, aseptically prepared by sterile filtration through 0.22 µm filter, after long storage and simulated administration use for 15 days which was not studied before.

Voriconazole eye drops solution for both contents remain sterile at D90.

For in use microbiological stability, solutions of multidose eye drops were found to be microbiologically contaminated, as well as the eyedropper and the vial cap [27] and even with ophthalmic solutions containing preservatives [28]. In our study, simulating the patient’s use during 15 days which is the maximum duration of use [26], we found one microbiological contamination among the three samples in vials with a classical polyethylene dropper. Conversely, no contamination was found for eyedroppers with Novelia^® cap specially designed to avoid air ingress during administration thanks to patented valve system [29]. As a result, the system allows to avoid the use of toxic preservative such as benzalkonium chloride still used as the main preservative in eye drops formulation despite its known adverse events [30].

According to “GERPAC Consensus Conference – Guidance on the Assignment of Microbiological Shelf life for Hospital Pharmacy Aseptic Preparations” [31] our process was previously validated using media broth simulation tests of our maximum batch size of 20 units. Physical integrity testing of the eye drop vials would advantageously complete our work.

However, we validated the routine physico-chemical and sterility methods for batch voriconazole eye drop production. Taking into consideration previous works showing the preservation of antimicrobial activity of voriconazole after 90 days long term storage [14, 15], we can safely implement batch production of voriconazole eye drops with a 90-day storage period.

With regards to microbiological risk during administration, Novelia^® cap system brings an advantage in comparison to classical cap due to the eyedropper valve giving an airtight system preventing microbiological contamination during patient use.

Conclusions

Fungal keratitis is rare but severe, stock hospital preparation of antifungal eye drops permanently available allows to implement an emergency treatment for patients and optimize the production organisation. Our study showed that voriconazole 10 mg/mL ophthalmic solution was stable during 90 days at +2 to 8 °C in different HDPE content with different dropper caps, with or without sterility preservative cap. Additionally, the Novelia^® cap system thanks to its patented valve, was found to be safer than a classical dropper cap considering the microbiological risk of contamination during eye drop administration.