Physicochemical Stability of 5 mg/mL Pediatric Prednisone Oral Suspension in Syrspend^® SF PH4

Reagents

Micronized prednisone was bought to Inresa (Bartenheim, France) and Syrspend^® SF PH4 to Fagron (Thiais, France). Methanol was of HPLC grade and was purchased from VWR Chemicals (Fontenay-sous-Bois, France). Prednisone and prednisolone standards were procured from Merck (Darmstadt, Germany).

Water for HPLC was distilled and passed through a reverse osmosis system. Hydrochloric acid (HCl) solutions 0.1 M and 1 M were purchased from VWR Chemicals (Fontenay-sous-Bois, France), sodium hydroxide (NaOH) 0.1 M and 1 M were bought to Merck (Darmstadt, Germany), oxygen peroxide (H₂O₂) 10 volumes was acquired from Gifrer (Decines-Charpieu, France).

Instrumentation and chromatographic conditions

The HPLC system included a 717 plus autosampler, 2487 UV detector, and a HPLC 515 pump (Waters, Milford, USA). Chromatographic separation of the analytes was carried out on Waters Xterra RP C18 5 µm column (4.6*150 mm). The volume injection of the sample was 10 µL. The flow rate was kept at 1 mL/min and prednisone was detected at 254 nm [9, 10]. The oven temperature was kept at 45 degrees Celsius.

Multiple mobile phases were tested to achieve an optimal elution. Fourteen different combinations of organic solvent and water were tried: acetonitrile/water, acetonitrile/methanol/water and methanol/water, each at various proportions.

Data were acquired and processed with EMPOWER Software (Warers, Milford, USA). All calculations were performed using Microsoft Excel 2010 (Microsoft Corporation, USA).

Preparation of oral suspensions

Three batches of 5 mg/mL oral suspension were prepared, using 350 mg of micronized prednisone and 70 mL of Syrspend^® SF PH4. These were packaged in amber glass vials (60 mL) to protect from light and stored at room temperature throughout the study. According to the European Pharmacopoeia [11], prednisone could be protected from light. That why, the study was realized with amber glass.

Prednisone concentrations were determined on days 0, 1, 4, 10, 30, 60, 90 after the first day of production.

Linearity

The response function was performed on three different days with the six calibrations curves. In order to check whether an excipient effect existed, two types of ranges were produced: a range only containing the API and a reconstituted range, with Syrspend^® SF PH4. The method was considered as linear if the coefficient of determination was over 0.99 for the mean standard curve.

Accuracy

The accuracy was determined using the data obtained during study of the response function. For each QC, the accuracy was measured by calculating the report between theoretical and calculated concentrations. This report allows the access to a recovery rate, whereby, it had to be less than 5 % in order to be accepted.

Precision

To determine the repeatability of this method, six points of the middle of the range was performed (125 µg/mL), and each point was prepared by independent weighing. Then, the intermediate precision was determined on three different days, using the relative standard variation (RSD), which had to be less than 5 %.

Stability indicating method

A stability indicating method is a process that is able to distinguish the API from its degradation products. The method has to be sufficiently sensitive to detect these degradation products in low quantity and sufficiently resolute to distinguish products with potentially close structures [13, 14].

Forced degradation experiments were carried out on prednisone under various conditions explained in the ICH guideline: heat, acid, alkaline and oxidative conditions.

A thermal degradation study was performed at two different temperatures: room temperature (24 °C±1 °C) and 80 °C in a water bath. The exposure time was one hour in both conditions.

Acid hydrolysis was studied by adding HCl solutions at 0.1 M, 0.5 M and 1 M in prednisone solution. At one hour of exposure time, neutralization was performed with respectively 0.1 M, 0.5 M, and 1 M NaOH solutions. The alkaline hydrolysis was performed by doing the opposite procedure of acid hydrolysis (exposition with NaOH, then neutralization with HCl). Acid and alkaline degradation was performed in a water bath heated to 80 °C, for one hour.

Oxidative degradation was assessed by exposed prednisone solution in 10 volumes of H₂O₂ for three hours in a water bath heated to 100 °C.

The degradation was confirmed if a reduction of more 5 % of prednisone concentration was observed.

Related substances

A solution of prednisolone was prepared in methanol to obtain a concentration of 5 mg/mL to differentiate prednisone to prednisolone. A work concentration of 125 µg/mL of prednisolone was compared to the same concentration of prednisone solution by HPLC. The same analysis method was used for the two molecules.

The resolution between these two peaks had to be more than 1.5 to validate the separation.

Visual and odour inspection

A visual inspection was done in order to detect precipitation or variation of colour with time. A control of odour suspension was also performed.

Osmolality

The osmolality of the suspension was measured with Advanced Instruments Model 3250 Osmometer (Radiometer, Neuilly-Plaisance, France). A 250 µL aliquot of each suspension were taken and diluted (1:2 v/v) in water for injection before measurement.

pH

The pH was determined with a pH-meter: pHenomenal VWR^® (VWR Chemicals, Fontenay-sous-Bois, France).

Measure of concentration

Prednisone concentration was quantified in triplicates immediately after preparation and after 1, 4, 10, 30, 60 and 90 days. Before removing samples, the containers were handshaken manually to ensure a homogeneous suspension, according to the 9th edition of the European Pharmacopoeia [11]. Then, 400 µL sample of each preparation was diluted (1:10 v/v) in methanol and centrifuged at 4000 rpm for five minutes to sediment insoluble excipients. Another dilution (1:4 v/v) of supernatant was performed in methanol for HPLC analysis. The chemical stability was determined by calculating the percentage of the initial concentration remaining at each time interval: prednisone concentration in subsequent samples greater than 95 % were considered stable.

Statistical and data analysis Statistical tests were performed by Excel^® software (Microsoft Office, USA, 2007) with a risk α set at 5 %. Statistical significance was specified as p<0.05.

Before using the Student t-test, the Shapiro-Wilk test was performed to show a normality test data (α=0.05).

Linearity

The coefficients of determination were 0.9999, 0.9994 and 0.9998, for days 1, 2 and 3, respectively, in the suspension without Syrspend^® SF PH4.

For the suspension with Syrspend^® SF PH4, the coefficients of determination were 0.9999, 0.9999 and 0.9998, for days 1, 2 and 3, respectively.

The average of regression equations was without and with Syrspend^® SF PH4: y=22096x-1129 (r²=0.9997) and y=22791x-35,573 (r²=0.9998), respectively. Student statistical test did not show significant differences between the slopes and the y-intercepts (α=5 %). No excipient effect was observed between curves with and without Syrspend^® SF PH4.

Accuracy

The accuracy rates for the 75 µg/mL, 125 µg/mL and 175 µg/mL QC were estimated to be 0.50 %, 0.59 % and 1.12 %, respectively. These results related to ranges with excipient.

Precision

This method is repeatable because of a RSD of 1.64 %, 2.16 % and 1.37 %, for days 1, 2 and 3, respectively.

The results of the intermediate precision for three days showed a RSD of 1.72 %.

Stability indicating method

No degradation was observed at room temperature (20 °C) (Figure 3) and heating at 80 °C (Figure 4).

Figure 3:

chromatogram of prednisone at room temperature.

Figure 4:

chromatogram of prednisone heating at 80 °C.

The acid hydrolysis showed a reduction of prednisone concentration of 8.7 % with HCl 1 M (Figure 5). No degradation was observed with 0.1 and 0.5 M of HCl (0.8 % and 4 % respectively).

Figure 5:

chromatogram of prednisone and products of degradation with HCl 1 M.

The alkaline hydrolysis was observed with low concentration of NaOH. In fact, a reduction of 80 % of prednisone concentration was visible with NaOH 0.1 M (Figure 6). The same result was noted with 0.5 and 1 M of NaOH.

Figure 6:

chromatogram of prednisone and products of degradation with NaOH 0.1 M.

No variation was observed with H₂O₂ (reduction of 4.5 %).

The prednisone degradation is poor in acid environment but major in basic environment.

Related substances

The resolution between peaks of prednisone and prednisolone was calculated at 2.9 (Figure 7).

Figure 7:

chromatogram of prednisone (on the left) and prednisolone (on the right).

Visual and odour inspection

No precipitation, no variation of colour or odour on the suspension was observed in the three batches.

Osmolality

During the study period, all osmolality measurements were included between 42 and 46 mOsm/Kg (Table 1). No significant variation of osmolality was observed.

pH

During the study, all pH measurements were included between 4.18 and 4.22 (Table 1). No significant variation of pH was observed.

Measure of concentration

After validation of the analytical method, prednisone concentrations remained stable within more or less 5 % of nominal value over 60 days. Two of the three batches showed a concentration more than 5 % of nominal value on the day 90 (Table 2). Results of the Shapiro-test showed that samples followed a normality test data (p-value=0.9638). Student test did not show a significant difference from day 0 to day 60 (t calc<t table). However, a significant difference from the day 90 was observed (t calc=5.3 and t table=2.92).