Search

The aim of this study was to assess the ability of Raman spectroscopy to discriminate and quantify five antineoplastic drugs in an aqueous matrix at low concentrations before patient administration. Five antineoplastic drugs were studied at therapeutic concentrations in aqueous 0.9% sodium chloride: 5-fluorouracil (5FU), gemcitabine (GEM), cyclophophamide (CYCLO), ifosfamide (IFOS) and doxorubicin (DOXO). All samples were packaged in glass vials and analyzed using Raman spectrometry from 400 to 4000cm-1. Discriminant analyses were performed using Partial Least Squares Discriminant Analysis (PLS-DA) and quantitative analyses using PLS regression. The best discrimination model was obtained using hierarchical PLS-DA models including three successive models for concentrations higher than the lower limit of quantification (0% of fitting and cross-validation error rate with an excellent accuracy of 100%). According to these hierarchical discriminative models, 90.8% (n=433) of external validation samples were correctly predicted, 2.5% (n=12) were misclassified and 6.7% (n=32) of the external validation set were not assigned. The quantitative analysis was characterized by the RMSEP that ranged from 0.23mg/mL for DOXO to 3.05mg/mL for 5FU. The determination coefficient (R2) was higher than 0.9994 for all drugs evaluated except for 5FU (R2=0.9986). This study provides additional information about the potential value of Raman spectroscopy for real-time quality control of cytotoxic drugs in hospitals. In some situations, this technique therefore constitutes a powerful alternative to usual methods with ultraviolet (UV) detection to ensure the correct drug and the correct dose in solutions before administration to patients and to limit exposure of healthcare workers during the analytical control process.

Polycyclic aromatic hydrocarbons (PAHs) are environmental contaminants with carcinogenic effect raising worldwide concerns. Hydroxylated PAHs (OH-PAHs) could be formed in the body as metabolites of PAHs in human urine samples and thus considered as biomarkers of PAH exposure. In this study, five OH-PAHs including 3-phenanthrol, 1-naphthol, 2-hydroxy fluorene, 1-hydroxprene and 6-hydroxy chrysene in human urine samples were selectively enriched by C18 solid-phase microextraction (SPME), then SPME fiber was connected high voltage and then was inserted into a glass-capillary to elute and ionize the analytes for mass spectrometric (MS) detection. The coupling of SPME-MS showed excellent analytical performance for detection of urinary OH-PAHs under optimal conditions, providing an easy operation for rapid detection of a single sample within minutes. By use of internal standard (i.e., 2-hydroxy fluorene-d9), the limit of detection (LOD) and limit of quantitation (LOQ) of OH-PAHs were found to be less than 0.05 ng L−1 level (S/N > 3) and less than 0.1 ng L−1 level (S/N > 10), respectively. The dynamic ranges of five OH-PAHs were found to be a range at 0.1–5.0 ng L−1 with excellent coefficient (R2 > 0.99). This method also showed good precisions (intra-day: 3.4–5.5%, inter-day: 7.0–9.8%, n = 5) and good accuracy (85.3–95.3%, n = 5). Moreover, ion suppression and matrix effect in detection of OH-PAHs in urine were also investigated. Human urine samples collected from 12 volunteers including 6 non-smokers and 6 smokers have been successfully analyzed, it was found that individual OH-PAHs in smokers were higher than in non-smokers. This study demonstrated that SPME coupled with glass-capillary nanoESI-MS is a sensitive method for rapid detection of urinary OH-PAHs for health risk assessment of PAHs exposure.

Implementation of ICH guideline for validation of analytical methods was tested in the case of two enzymatic assays of determination of superoxide dismutase (SOD) activity. Analytical figures of merits of two tested methods (NBT reference method and the so called WST-1 method) demonstrate the feasibility of such approach. Nevertheless specification usually admitted for physicochemical method (e.g., HPLC), needs to be clearly enlarged, for instance, up to 10% for repeatability and 20% for reproducibility. On the other hand, the cross-validation performed between the two enzymatic activity determination techniques, though based on the same principle, clearly shows that the quality of results depends on small variation in the experimental conditions. This shows that an enzymatic activity determination should be strictly related to the technique used, especially in the pharmaceutical control quality field, and confirms that analytical figures of merits are strongly function of the technique used. Finally, it was demonstrated that both NBT reference method and WST-1 method give strictly similar results.

A general approach, still few exploited so far and never associated with microbore-LC, consisting of detection of various lipid classes (i.e. phospholipids, triglycerides, ceramides and glycosphingolipids) by non-covalent association with 1,6-diphenyl-1,3,5-hexatriene (DPH) fluorescence probe is developed. This mode of detection was coupled with non-aqueous reversed-phase microbore-LC (C18) by using classical post-column fluorescence detection. The classical LC system was first adapted to microbore-chromatography (internal diameter 1 mm) without apparatus miniaturization of the solvent delivery system and the detection cell. For this purpose, the detection parameters (probe concentration, post-column flow rate, post-column reactor length and post-column system temperature) were optimized by a central composite design (CCD) using a mixture of phosphatidylcholine (PC) species as a lipid model and DPH (lambda(ex) = 350 nm, lambda(em) = 430 nm) as a fluorescence probe. The optimal conditions of detection for the various molecular species of PC were determined for a DPH concentration of 3.35 micromol/L, a post-column flow rate of 0.5 mL/min, a reactor length of 1.4 m and a temperature of 35 degrees C. The fluorescence response was linear over a wide range of PC species from 5 microg/mL to 100 microg/mL and the lower limit of detection (signal/noise = 3) was about 1 microg/mL, that is equivalent to evaporative light scattering detection (ELSD). Others molecular species of various classes of lipids, i.e. triglycerides, ceramides and glycosphingolipids were also easily detected. Thus, this study demonstrated the versatility of the proposed system of detection which was shown to be sensitive, easy to perform, non-destructive and allowed, in contrast to ELSD, for a linear response with various polarity lipid classes.

Ceramide analysis was developed with gradient elution in non-aqueous reversed-phase liquid chromatography with evaporative light scattering detection (ELSD) or postcolumn fluorescence detection. Fluorescence detection (excitation, 360 nm; emission, 425 nm) after postcolumn formation of mixed assemblies between eluted ceramides and 1,6-diphenyl-1,3,5-hexatriene was developed. In comparison with ELSD, fluorescence detection allows a better detection of the minor species ceramide from ceramide type III (commercial mixture of non-hydroxy fatty acid-sphingosine) and appears to be more sensitive for quantitation of ceramides at low concentrations. The fluorescence response is linear over a wide range of injected amount of ceramide III (expressed as stearoyl-phytosphingosine): 10 ng to 1000 ng. The response of ELSD is non linear but can be linearized in double logarithmic coordinates for calculations over a narrow range, e.g. between 10 to 350 ng ceramide III injected. The lower quantitation limits of these two detectors are similar: 5 ng ceramide III was injected.

A precise and sensitive method is described for the determination of N-nitrosodiethanolamine (NDELA) at the 1 μl l -1 level. NDELA was found to be extracted from aqueous solution by ethyl acetate at pH 1.0. The technique is based on the alkaline denitrosation of NDELA and reaction of the nitrite liberated with Coumarin 120 (4-methyl-7-aminocoumarin) in sulphuric acid medium to give the corresponding 7-diazo compound. After hydrolysis at 100°C, the corresponding 7-hydroxycoumarin (4-methylumbelliferone) is detected fluorimetrically at 380 nm after excitation at 325 nm. HPLC is essential to separate excess Coumarin 120 from 4-methylumbelliferone. Linearity was obtained from 1 to 30 μl l -1 (r = 0.997). The limit of detection was 0.8 μg l -1 for a signal-to-noise ratio of 3. The repeatability expressed as R.S.D. was 6.8% at 1 μl l -1 and 5.5% at 30 μg l -1 . The reproducibility, established by two operators on three consecutive days at 5 μl l -1 (n = 6), was 9.2% (R.S.D.). Under the conditions adopted, detection of trace levels of NDELA in ethanolamine derivatives is possible.