Received date: October 30, 2012; Accepted date: November 23, 2012; Published date: November 26, 2012
Citation: Kuehl PJ, De S, Eppler B, Marsters J, Matthews L, et al. (2012) Development and Validation of an HPLC Assay For Dual Detection of Gentamicin Sulfate and Leucine From a Novel Dry Powder For Inhalation. J Anal Bioanal Tech 3:152. doi: 10.4172/2155-9872.1000152
Copyright: © 2012 Kuehl PJ, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Visit for more related articles at Journal of Analytical & Bioanalytical Techniques
A stability indicating HPLC assay with UV detection was developed for the simultaneous quantification of Gentamicin Sulfate and L-leucine from NanoGENT™ dry powder for inhalation. In order to support the development of the NanoGENT™ (dry powder gentamicin sulfate for inhalation) the assay was created to allow simultaneous detection of the active pharmaceutical ingredient (gentamicin sulfte) and the primary excipient (L-leucine). In order to quantify Gentamicin Sulfate by UV detection derivatization was required. The assay resolved L-leucine from all four Gentamicin Sulfate peaks and all four Gentamicin Sulfate peaks from each other with a reversed phase isocratic assay. Once developed the assay was validated in accordance with regulatory guidance in order to support regulatory approval of the NanoGENT™ dry powder inhalation formulation. The validation data indicated that the dual detection assay meets or exceeded all criteria for use as a stability indicating assay.
Gentamicin sulfate; L-leucine; NanoGENTTM; Dual detection; Dry powder; Inhalation
Orally inhaled therapeutics is a mainstay in the treatment of asthma and COPD and is growing rapidly in the treatment of other pulmonary and systemic indications. The growth of orally inhaled therapeutics to treat atypical indications has broadened beyond vaccines and insulin to include biological threats. This growth has exacerbated the fact that there are currently a minimal number of GRAS excipients for inhalation delivery . Further, the excipients that are considered GRAS do not span the range of physical chemical properties that are often required for formulations of a material for delivery to the lungs.
The path for regulatory approval for oral inhalation delivery of an excipient is the same as an active pharmaceutical ingredient (API). Both must undergo preclinical safety and stability assessment to support submission of an IND. As such, a validated analysis assay for the quantification and stability of the API and excipient(s) is needed [2,3]. The current status of analytical technologies allows the potential for dual detection of both an API and excipient in one assay.
NanoGENTTM Gentamicin Sulfate inhalation powder is being developed as an inhaled, antibiotic treatment to be administered to the lungs. NanoGENTTM is approximately 80% Gentamicin Sulfate and each dose of 25 mg of free-flowing powder delivering approximately 10-15 mg Gentamicin Sulfate to the respiratory tract per inhalation [4,5]. While HPLC assays have been published for Gentamicin Sulfate NanoGENTTM also includes L-leucine in this dry-powder jet-milled formulation, therefore an assay was required to resolve and quantify both components . Therefore in order to support the preclinical safety studies and the formulation stability studies a stability indicating assay was developed to quantify L-leucine and all four peaks of Gentamicin Sulfate.
Gentamicin Sulfate and L-leucine reference standards were purchased from the USP. NanoGENTTM was used as received from Nanotherapeutics. LC-MS Grade water was received from EMD. Potassium hydroxide, boric acid, phthaldialdehyde, thiogylycolic acid, sodium 1-heptanesulfonate monohydrate, glacial acetic acid were procured from Sigma Aldrich. Isopropanol (IPA) was used from Burdick & Jackson and LC-MS grade Methanol (MeOH) was used from J.T. Baker.
All samples were derivatized prior to analysis using the following method: 440 μL of isopropanol and 160 μL of Reagent 2 (1.0 g Phthaldialdehyde in 5 mL Methanol, 95 mL Reagent 1 [0.4 M Boric acid adjusted to pH 10.4 using 8 N Potassium hydroxide] and 2 mL Thioglycolic acid, the resulting solution was adjusted to pH 10.4 using 8 N Potassium hydroxide) were added to each sample (0.4 mL sample in 2 mL autosampler vial). Each sample was then vortexed for 10 seconds ± 1 second and heated in an oven at 60°C ± 3°C for 15 minutes ± 1 minute. The samples were allowed to return to room temperature prior to analysis.
Analysis was conducted using an HPLC and UV detection. Separation was achieved with a C18, 5 μ, 4.6×150 mm) column and a with a matching C18 guard column (4×3 mm). The column temperature was set at 20°C with detection at 330 nm. The isocratic method used a flow rate of 2.0 ml/min with the mobile phase details in (Table 1).Under these conditions Gentamicin Sulfate produced four peaks with retention times of ~ 3, 9, 13, and 14 minutes and the L-leucine retention time was ~ 2 minutes.
|Mobile phase||5.0 g sodium 1-heptanesulfonate monohydrate in 250 ml HPLC grade water, 50 ml glacial acetic acid and 650 ml HPLC grade MeOH.|
|Gradient||100% mobile phase A|
|Column||C18, 5µ, 4.6 x 150 mm|
|Guard Column||C18, 4 × 3 mm|
|Flow Rate||2.0 ml/min|
|Run time||20 min|
|Injection Volume||10 µL|
|Detection Wavelength||330 nm|
Table 1: Instrument set-up and parameters.
HPLC assay validation criteria
|Test||Performance Criteria / Testing Method|
|System Suitability||Column efficiency of >1200
Five replicate injections RSDa ≤ 2.0%
Asymmetry < 1.5
Capacity Factor (C1) between 2 and 7
|Robustness||±2.5% change in methanol composition in mobile phase, ±5°C change in derivatization temperature, ±10% change of mobile phase flow rate, different injection volume (10, 15, and 20 µl)|
|Specificity||No interfering peaks of the same retention time as either Gentamicin Sulfate or L-leucine with peak area above 1% of the corresponding lowest standard for both non stressed and acid stressed samples|
|Accuracy/Recovery||85 – 115% of Theoretical (80 – 120% at LOQb)|
|Repeatability||≤ 15% RSDa(≤ 20% RSDa at LOQb)|
|Intermediate Precision||≤ 15% RSDa(≤ 20% RSDa at LOQb)|
|LOQb||Tested at: Gentamicin Sulfate = 50 µg/mL
L-leucine = 0.5 µg/mL
|LODc||Tested at: Gentamicin Sulfate = 5 µg/mL
L-leucine = 0.2 µg/mL
|aRelative Standard Deviation
bLimits of Quantification
c Limits of Detection
Table 2: Analytical qualification performance criteria for the determination of Gentamicin Sulfate and L-leucine in formulations.
In order to characterize the system suitability of the instrument, 6 consecutive injections of an L-leucine and Gentamicin Sulfate standard were made from one standard solution. For each injection the L-leucine peak was evaluated for its resolution from the solvent front, the % RSD from the response and the asymmetry of the peak.
The Gentamicin Sulfate peaks were evaluated for their resolution (with the previous peak), the capacity factor (k’) for the C1 (Gent Peak 1) was to be between 2 to 7, the number of theoretical plates (N) for the C2 peak was to not be less than 1200, the asymmetry (10% height) of any peak of interest should not to exceed 1.5 for each peak, and the RSD for the area of each peak was not to exceed 2%.
The robustness of the HPLC assay was evaluated by changing the methanol composition in the mobile phase ± 2.5% (target of 650 mL in 1 L), changing the derivatization temperature ± 5°C (target of 60°C), altering the flow rate ± 10% (target of 2.0 mL/min), evaluating three different injection volumes (10, 15 and 20 μL). The robustness of the assay was evaluated in terms of the stability of the mobile phase by preparing ~ 9500 mL of mobile phase, stored at room temperature, and using it to assay standard solutions over 6 weeks. In a similar manner working solutions of USP reference standards and NanoGENTTM formulation were prepared, stored at 2-8°C, and assayed over 6 weeks.
The robustness of the assay to each of the changes was evaluated by comparing the retention times, asymmetry and the percentage of target amounts in the samples.
Specificity was determined in two manners. First by comparing the chromatograms from the blank injections described below with the low standard for each of Gentamicin Sulfate and L-leucine.
• Injection of purified water
• Injection of mobile phase
• Injection of blank aerosol filter manipulated in purified water
For the first specificity testing no peaks were to have the same retention time as either Gentamicin Sulfate or L-leucine above a limit of 1% relative to the peak area of the lowest standard.The second specificity testing compared two solutions of NanoGENTTM following derivatization. One sample was unaltered following derivatization the other was mixed with 1 N HCl. These samples were then assayed to evaluate the change in retention time.
The linearity was determined with five calibration solutions prepared with both Gentamicin Sulfate and L-leucine. The Gentamicin Sulfate concentrations spanned 50 μg/ml to 2000 μg/ml, and the L-leucine concentrations spanned 0.5 μg/ml to 50 μg/ml. The linearity curves were evaluated by triplicate measurements of each of the standard concentrations from which the mean peak area was used to perform a back calculation of each of the standard concentrations. The correlation coefficient of the linearity curve was required to be more than 0.98.
The intended purpose for the validated assay was to quantify both Gentamicin Sulfate and L-leucine from extracts of aerosol and cascade impactor samples. Therefore the initial accuracy/recovery work was conducted to determine the recovery of the Gentamicin Sulfate and L-leucine in the presence of the aerosol excipients. These samples were prepared at three difference concentrations (Gentamicin Sulfate at 40, 80 and 120 μg/mL and L-leucine at 1, 2 and 3 μg/mL). Subsequently, accuracy/recovery was evaluated by spiking blank aerosol filter samples and simulated extraction solutions with both Gentamicin Sulfate and L-leucine. Ratios of Gentamicin Sulfate and L-leucine were maintained in the same range as the formulated NanoGENTTM dry powder for inhalation.
The two test articles were spiked and recovered from the same filters at the three different levels (low, mid and high). The amounts on filter for Gentamicin Sulfate were 250 μg, 5 mg and 15 mg and the amounts on filter for L-leucine were 2.5 μg, 50 μg and 150 μg. The final accuracy/ recovery experiments were conducted to specifically mimic a sample for analysis. In this test aerosol filter samples were preloaded with the placebo inhalation formulation (containing the dry powder excipients) and then spiked with both Gentamicin Sulfate and L-leucine. The samples were extracted and assayed.
The precision of the assay was evaluated in two different manners. Initial assay precision was conducted to evaluate the effect of different analysts and different instruments on the assay performance. This was conducted by having three analysts each prepare 6 independent samples of NanoGENTTM . Each analyst assayed their independently prepared samples on a different HPLC system.
In addition to the above precision the intermediate precision was evaluated by preparing the accuracy/recovery filters (described above) on three different days and with two different analysts. This resulted in a total of 9 independently prepared samples at each of the three amounts for Gentamicin Sulfate and another 9 independently prepared samples at each of three amounts for L-leucine. The RSD was determined both intraday and interday for these samples.
Aerosol filter extraction
The extraction of Gentamicin Sulfate and L-leucine from filters was performed as follows. The filter was folded and transferred into a 7-ml glass vial. 4 ml of purified water was added to the vial and the vial was capped using an Aluminum lined lid. The vial was rotated on a rotary shaker at a setting of 40 ± 5 for 30 minutes ± 5 minutes, sonicated for 10 minutes ± 1 minutes and then vortexed for 60 ± 5 seconds. 400 μL of the extract or the diluted extract was transferred into a 2-ml auto sampler vial with Teflon-lined septum for derivatization and analysis on HPLC/UV.
Limit of quantitation (LOQ)
The determination of the assay LOQ was conducted by preparing six independent samples at or near the hypothesized LOQ (50 μg/ mL for Gentamicin Sulfate and 0.5 μg/mL for L-leucine) and assaying them. The final LOQ was determined to be the concentration for each compound that could be calculated to be within ± 20% of theoretical concentration and the RSD between the calculated concentration for all six independent samples be less than 20%.
Limit of detection (LOD)
The LOD was determined by prepared six independent samples at several concentrations at or near the hypothesized LOD (0.5 μg/mL for Gentamicin Sulfate and 0.2 μg/mL for L-leucine). Samples were then assayed and the retention time determined between the six independent samples at each concentration. The lowest concentration with a retention shift of less than 5% was determined to be the LOD.
The results for the L-leucine and Gentamicin Sulfate system suitability are shown in (Table 3) below. Not shown is the RSD of the sum of the four Gentamicin Sulfate peaks, which was 0.27%.
|Test||L-leucine||Gent Peak 1||Gent Peak 2||Gent Peak 3||Gent Peak 4|
|Column Efficiency (theoretical plates)||3461||1489||2407||2467||2597|
|Replicate Injections RSD||0.3||0.4||0.4||0.3||0.3|
Table 3: System suitability results (quantification assay / stability indicating assay).
The results for the change in the methanol composition of the mobile phase are shown in (Table 4). The retention time of the L-leucine did not change as a function of methanol composition but the retention time of each Gentamicin Sulfate peak decreased with increased methanol composition. The percent of target for both Gentamicin Sulfate (GS) and L-leucine were all within the targeted range of 80%-120% for all conditions.
|Methanol (%)||L-leucineRT (min)||C1 RT (min)||C1a RT (min)||C2a RT (min)||C2 RT (min)||% Target (GS)||% Target(L-leucine)|
Table 4: Robustness of the assay following changes in methanol composition in the mobile phase.
The robustness of the assay to withstand changes in the derivatization temperature showed that there no change was seen for ± 5°C change (55, 60 and 65°C) in derivatization temperature. In terms of robustness of the assay to withstand changes in mobile phase flow rate (± 10% of target) no change in retention time and/or percent of target was seen. The robustness of the assay to withstand change as the mobile phase aged showed no change in retention time of any of the peaks out to six weeks.
The robustness of the assay in terms of varied injection volume is shown in (Table 5). For all three injection volumes the retention time and percent recovery did not change. Further the peak asymmetry for the C1 peak decreased from 1.0 at 10 μL to 0.8 at 20 μL. While this does begin to suggest column overloading even at 20 μL the peak asymmetry is still acceptable. The robustness of the assay to withstand stability induced changes in the standard solutions is shown in (Table 6). The data show that the percent of target for both the USP standard solution and the NanoGENTTM solutions did not change over the 6 weeks evaluated.
|Injection Volume (μL)||L-leucine RT(min)||C1 RT (min)||C1a RT (min)||C2a RT (min)||C2 RT (min)||% Target (GS)||% Target(L-leucine)||C1 Asy. (10%)|
Table 5: Robustness Studies for increase in injection volume.
|Week||USP Standard Gentamicin Sulfate (%TC)||NanoGENTTM Sample Gentamicin Sulfate (%TC)||USP Standard L-leucine (%TC)||NanoGENTTM Sample L-leucine (%TC)|
Table 6: Robustness studies for stability of USP standards and NanoGENT samples.
The specificity of the assay was conducted to evaluate the assay for potential sources of interfering peaks from the matrices used in sample preparation. No interfering peaks were seen from water used to prepare mobile phase, the mobile phase or from a blank filter media. The specificity of the assay in the presence of acid degraded samples is shown in (Table 7). The non-stressed NanoGENTTM samples displayed even smaller shifts in retention time and therefore are not presented. The data indicate that the assay is specific for L-leucine and all Gentamicin Sulfate peaks.
|Sample ID||L-leucineRT (min)||C1 RT (min)||C1a RT (min)||C2a RT (min)||C2 RT (min)|
|USP Standard (avg. of injs. #1 - 6)||2.0||3||9||12||14|
|% RPD (NanoGENTTM Sample inj #1 vs. # 2 of sample)||0.0||0.1||0.2||0.3||0.0|
|% RPD (Placebo inj #1 vs. # 2 of sample)||0.0||NA|
|% RPD (NanoGENTTM Sample vs. USP Standard||0.1||0.1||0.1||0.2||0.2|
|%RPD (Placebo vs. USP Standard)||0.0||NA|
|% RPD: percent relative difference|
Table 7: Specificity studies for acid.
The linearity for both Gentamicin Sulfate and L-leucine was determined to have a correlation coefficient (R2) for Gentamicin Sulfate of 1.0000 and for L-leucine of 1.0000. Back calculation of the standard concentration from the linear curves resulted in no standard deviating from theoretical concentration by more than 2.5%.
Average recovery for Gentamicin Sulfate and L-leucine from the spiked extraction solutions showed recovery of 94% and 98%, respectively, over all concentration ranges. In a similar manner the recovery of Gentamicin Sulfate from spiked aerosol filter samples ranged from 96 to 101% recovery and the L-leucine recovery ranged from 91% to 101%. These data indicated that the extraction procedure and assay were accurate for the recovery of both Gentamicin Sulfate and L-leucine from aerosol samples. In order to confirm the assay and extraction procedure in the presence of the dry powder excipients the recovery from the excipient loaded filters was characterized. These results showed an average recovery of 97% for Leucine and 99% for Gentamicin sulfate. Both establish the assay as viable for recovery of Leucine and Gentamicin sulfate in the presence of the dry powder excipients.
The assay precision studies determined that between three different analysts and three different HPLC systems the RSD for each analyst ranged from 0.5 to 1.9% for Gentamicin Sulfate and ranged from 0.7 to 2.5% for L-leucine. The overall RSD for all Gentamicin Sulfate was 2.1% and for L-leucine it was 2.2%.
The precision studies for the spiked aerosol filter samples for Gentamicin Sulfate showed that the assay precision (measured by RSD between three consecutive days of analysis) was less than 2.0%. For L-leucine the assay precision was less than 6.5%.
Limit of Quantification (LOQ)
The LOQ of the quantification assay was determined to be 50 μg/mL Gentamicin Sulfate and 0.5 μg/ml for L-leucine. The average calculated concentration for the six samples of Gentamicin Sulfate at 50 μg/mL ranged from 49.94 to 51.03 μg/mL with an RSD of 0.75%. The average calculated concentration for the six samples of L-leucine prepared at 0.5 μg/mL ranged from 0.50 to 0.52 μg/mL with an RSD of 1.6%.
Limit of detection (LOD)
The LOD was determined by the variability in the retention times of samples prepared at the LOD (5 μg/ml Gentamicin Sulfate and 0.2 μg/ml L-leucine). This solution was injected six times. The difference in retention time of the six injections for all of the peaks was less than 5%, specifically it ranged from 0.0 to 0.5%. From the results of this test, the LOD was accepted as 5 μg/ml Gentamicin Sulfate and 0.2 μg/ml L-leucine.
A HPLC assay with UV detection has been developed and validated for the dual detection of L-leucine and Gentamicin Sulfate from the NanoGENTTM dry powder inhalation formulation. The assay resolves L-leucine from all four Gentamicin Sulfate peaks in a 20 minute run time. The system suitability, robustness, specificity, linearity, accuracy/ recovery, repeatability, intermediate precision, LOQ and LOD have been established and shown to meet or exceed the requirements for validation of the assay as a dual detection assay.
Make the best use of Scientific Research and information from our 700 + peer reviewed, Open Access Journals