NIR Spectroscopy for Quality Control of an Ebola Virus Drug

December 11, 2017

Near-infrared (NIR) spectroscopy is an important technique in the pharmaceutical industry because of its ability to provide information about bulk material without sample preparation. Multivariate calibration techniques are frequently used to analyze the NIR data. Robert Lodder, who is a professor in the Department of Pharmaceutical Sciences at the University of Kentucky in Lexington, Kentucky, uses NIR spectroscopy along with an interesting alternative calibration technique, molecular factor computing, in his work with an experimental drug for combating the Ebola virus. We recently spoke with him about his research.

Near-infrared (NIR) spectroscopy is an important technique in the pharmaceutical industry because of its ability to provide information about bulk material without sample preparation. Multivariate calibration techniques are frequently used to analyze the NIR data. Robert Lodder, who is a professor in the Department of Pharmaceutical Sciences at the University of Kentucky in Lexington, Kentucky, uses NIR spectroscopy along with an interesting alternative calibration technique, molecular factor computing, in his work with an experimental drug for combating the Ebola virus. We recently spoke with him about his research.

In a presentation at SciX 2017 (1), you described the use of an NIR-based approach for identifying the proper formulation of BSN389, an experimental drug for treating active Ebola virus infections. What are the advantages of NIR spectroscopy for this application?

NIR is a rapid spectrometric method that is rugged and potentially deployable in the field in Africa. The cost of instrumentation for molecular factor computing (MFC) is low because filter-type instruments can be used (instead of monochromators or Fourier transform devices), and computing requirements are reduced (because the chemometric procedures like principal component analysis are implemented by the optical filters themselves). BSN389 is provided as a lyophilized formulation, and NIR works through the walls of intact vials without potentially exposing the sterile contents to contamination. NIR can be used in identification and quantification of the drug, in stability studies, and to monitor potency of drug in the field.

What is the mode of action of BSN389, and how does it compare with those of other drugs being developed for post-infection treatment of Ebola patients?

BSN389 targets the mechanism the body uses to replicate the virus, not the virus itself. For this reason, BSN389 is not specific to a particular strain of Ebola virus. Other drugs (such as antibody therapies) and supportive techniques are then also employed along with BSN389. Most groups are working on vaccines, which can be specific to certain strains of virus. Unlike vaccines, BSN389 can be used after a person has contracted the disease. BSN389’s mechanism of action permits it to be potentially used against all strains of the virus, a characteristic that vaccines might not share. Ebola outbreaks are usually self-limited to a few hundred patients. Africa has over 1 billion inhabitants. If all of these people were immunized against Ebola, there is a good chance there would be more adverse events in people from the vaccine than there would be from Ebola virus infections.

Can you briefly describe the instrumental setup used in the study?

A 12-V, 100-W tungsten-halogen broadband source with a 1400-nm long-pass filter was used as the source of broadband NIR light. The tungsten-halogen light source has more intense radiation in the shorter NIR wavelength region. To avoid saturating the detector with short-wavelength NIR radiation that contains little chemical information about the samples, the 1400-nm long-pass filter was used to block the short wavelength radiation. The source beam was modulated with an optical chopper at a frequency of 280 Hz. The light beam was focused onto an InGaAs photodiode through a convex lens after it passed through the molecular filter cuvette and sample cuvette.

Your NIR system is based on a molecular factor computing approach. How does molecular factor computing differ from ordinary spectroscopy with principal component analysis chemometrics? What are its advantages for your analysis?

Multivariate calibration is a well-established method in chemometrics for the analysis of NIR, UV–visible, and Raman spectra. Conventional measurement of physical or chemical properties from spectra is conducted by constructing a predictive model. Two of the most commonly used methods to build a predictive model are partial least squares (PLS) and principal component regression (PCR). In a conventional spectrometer with typical chemometrics, data collection and processing of raw data can be computationally expensive and time consuming, especially when spatial relationships (image data) are essential. Methods for selecting small but highly relevant variables to represent the original data in a reduced coordinate space and methods for integrated sensing and processing (ISP) are therefore being considered.

ISP aims to design and optimize sensing systems that integrate the traditionally independent units of sensing, signal processing, communication and targeting. By employing ISP, mechanical and computational complexity within the traditional sensing system has been significantly reduced through defining efficient low-dimensional representations of those sensing problems that were first posed in high-dimensional settings by traditional sensing architecture.

Molecular absorption filters can be used as mathematical factors in spectral encoding to produce a factor-analytic optical calibration in a high-throughput spectrometer, which we call molecular factor computing. The molecules in the filter effectively compute the calibration function by weighting the signals collected at each wavelength over a broad range of wavelengths. Given a set of training spectra gathered at all available wavelengths, it is possible to rationally select molecular filter materials to complete a factor analysis procedure like principal component analysis (PCA). PCA is designed to maximize the signals from the spectral regions with the most variability by most heavily weighting them. The same idea is used in the absorbance filters.

What other applications might benefit from the use of an MFC-based approach?

Any situation that requires a low-noise, stable, small, and rugged spectrometer is an ideal situation for an MFC instrument. Process control and portable instruments that must be used in the field are good application targets.

What are the next steps in your research?

We are working on getting FDA buy-in to the clinical trial plan for BSN389.

Reference

1.    R. Lodder and M. Hernandez-Murtillo, “Near-IR Quality Control of a Drug to Treat a Rare Tropical Disease,” presentation at SciX 2017 in Reno, Nevada.