Near-infrared spectroscopic evaluation of lyophilized viral vaccine formulations

Use of machine learning tools and NIR spectra to estimate residual moisture in freeze-dried products

Residual Moisture (RM) in freeze-dried products is one of the most important critical quality attributes (CQAs) to monitor, since it affects the stability of the active pharmaceutical ingredient (API). The standard experimental method adopted for the measurements of RM is the Karl-Fischer (KF) titration, that is a destructive and time-consuming technique. Therefore, Near-Infrared (NIR) spectroscopy was widely investigated in the last decades as an alternative tool to quantify the RM. In the present paper, a novel method was developed based on NIR spectroscopy combined with machine learning tools for the prediction of RM in freeze-dried products. Two different types of models were used: a linear regression model and a neural network based one. The architecture of the neural network was chosen so as to optimize the prediction of the residual moisture, by minimizing the root mean square error with the dataset used in the learning step. Moreover, the parity plots and the absolute error plots were reported, allowing a visual evaluation of the results. Different factors were considered when developing the model, namely the range of wavelengths considered, the shape of the spectra and the type of model. The possibility of developing the model using a smaller dataset, obtained with just one product, that could be then applied to a wider range of products was investigated, as well as the performance of a model developed for a dataset encompassing several products. Different formulations were analyzed: the main part of the dataset was characterized by a different percentage of sucrose in solution (3%, 6% and 9% specifically); a smaller part was made up of sucrose-arginine mixtures at different percentages and only one formulation was characterized by another excipient, the trehalose. The product-specific model for the 6% sucrose mixture was found consistent for the prediction of RM in other sucrose containing mixtures and in the one containing trehalose, while failed for the dataset with higher percentage of arginine. Therefore, a global model was developed by including a certain percentage of all the available dataset in the calibration phase. Results presented and discussed in this paper demonstrate the higher accuracy and robustness of the machine learning based model with respect to the linear models.

Application of Near-Infrared Spectroscopy to statistical control in freeze-drying processes

Batch freeze-drying of pharmaceutical products in vials may result in a high degree of intra-batch variability due to several reasons, e.g. non uniform heating rate in the drying chamber. Therefore, product quality in the final product has to be checked in a statistically significant number of samples, in particular in the stage of process development. Here, Fourier-Transform Near-Infrared Spectroscopy is proposed as a fast, non-destructive technique for an off-line Statistical Quality Control application. At first, results obtained in a batch where product features are satisfactory are used to identify a target quality threshold. Then, a statistical controller is developed in such a way that in a production run it is possible to quickly check if product quality exceeds the desired threshold or not. Two approaches based on multivariate analysis are presented: one employs the Hotelling T² and Mahalanobis statistics to calculate control charts, the other is an application of Partial Least Squares for discriminant analysis (PLS-DA). Control charts and PLS-DA were trained with samples obtained in a run where sucrose solution was processed and validated in other runs where the final product was known to have the desired qualitative characteristics or not. Overall, out-of-specification samples can be predicted by control charts and PLS-DA with 99% and 98% accuracy respectively. PLS-DA was shown to be able to better identify samples correctly processed, while the control charts where more accurate to identify vials where something went wrong. Focusing on residual moisture of the final product, all samples where it was higher than the target value were always correctly identified.

Lyophilization and nebulization of pulmonary surfactant-coated nanogels for siRNA inhalation therapy

RNA interference (RNAi) enables highly specific silencing of potential target genes for treatment of pulmonary pathologies. The intracellular RNAi pathway can be activated by cytosolic delivery of small interfering RNA (siRNA), inducing sequence-specific gene knockdown on the post-transcriptional level. Although siRNA drugs hold many advantages over currently applied therapies, their clinical translation is hampered by inefficient delivery across cellular membranes. We previously developed hybrid nanoparticles consisting of an siRNA-loaded nanosized hydrogel core (nanogel) coated with Curosurf®, a clinically used pulmonary surfactant (PS). The latter enhances both particle stability as well as intracellular siRNA delivery, which was shown to be governed by the PS-associated surfactant protein B (SP-B). Despite having a proven in vitro and in vivo siRNA delivery potential when prepared ex novo, clinical translation of this liquid nanoparticle suspension requires the identification of a long-term preservation strategy that maintains nanoparticle stability and potency. In addition, to achieve optimal pulmonary deposition of the nanocomposite, its compatibility with state-of-the-art pulmonary administration techniques should be evaluated. Here, we demonstrate that PS-coated nanogels can be lyophilized, reconstituted and subsequently nebulized via a vibrating mesh nebulizer. The particles retain their physicochemical integrity and their ability to deliver siRNA in a human lung epithelial cell line. The latter result suggests that the functional integrity of SP-B in the PS coat towards siRNA delivery might be preserved as well. Of note, successful lyophilization was achieved without the need for stabilizing lyo- or cryoprotectants. Our results demonstrate that PS-coated siRNA-loaded nanogels can be lyophilized, which offers the prospect of long-term storage. In addition, the formulation was demonstrated to be suitable for local administration with a state-of-the-art nebulizer for human use upon reconstitution. Hence, the data presented in this study represent an important step towards clinical application of such nanocomposites for treatment of pulmonary disease.

Analysis of protein glycation in human fingernail clippings with near-infrared (NIR) spectroscopy as an alternative technique for the diagnosis of diabetes mellitus

Background:

Glycated keratin allows the monitoring of average tissue glucose exposure over previous weeks. In the present study, we wanted to explore if near-infrared (NIR) spectroscopy could be used as a non-invasive diagnostic tool for assessing glycation in diabetes mellitus.

Methods:

A total of 52 patients with diabetes mellitus and 107 healthy subjects were enrolled in this study. A limited number (n=21) of nails of healthy subjects were glycated in vitro with 0.278 mol/L, 0.556 mol/L and 0.833 mol/L glucose solution to study the effect of glucose on the nail spectrum. Consequently, the nail clippings of the patients were analyzed using a Thermo Fisher Antaris II Near-IR Analyzer Spectrometer and near infrared (NIR) chemical imaging. Spectral classification (patients with diabetes mellitus vs. healthy subjects) was performed using partial least square discriminant analysis (PLS-DA).

Results:

In vitro glycation resulted in peak sharpening between 4300 and 4400 cm−1 and spectral variations at 5270 cm−1 and between 6600 and 7500 cm−1. Similar regions encountered spectral deviations during analysis of the patients’ nails. Optimization of the spectral collection parameters was necessary in order to distinguish a large dataset. Spectra had to be collected at 16 cm−1, 128 scans, region 4000–7500 cm−1. Using standard normal variate, Savitsky-Golay smoothing (7 points) and first derivative preprocessing allowed for the prediction of the test set with 100% correct assignments utilizing a PLS-DA model.

Conclusions:

Analysis of protein glycation in human fingernail clippings with NIR spectroscopy could be an alternative affordable technique for the diagnosis of diabetes mellitus.