A-TEEM™ combines simultaneous absorbance, transmittance, and fluorescence excitation-emission matrix measurements into a rapid, non-destructive 3D-plotted molecular analysis.
HORIBA’S New Veloci Biopharma analyzer brings this innovation to pharmaceutical quality control, offering:
A-TEEM Spectroscopy
A-TEEM Spectroscopy
Research Papers, Posters and Talks on A-TEEM
Nathan Fuenffinger – Merck & Co., Inc., Jennifer Pai Wang – Merck & Co., Inc., Louis Obando – Merck & Co., Inc.
Post-packaging identification of all manufactured lots is a general provision set forth by the Food and Drug Administration for biological products to be distributed within the United States. With this being the last test prior to the manufacturer being able to release a batch, it is desirable to have an analytical technique which is fast, simple-to-use, and logistically efficient. Ideally, such a technique would enable identity testing in the same general location that the packaging occurs to avoid the significant amount of lead time associated with transferring material to an offsite testing facility.
Simultaneous absorbance, transmission, and fluorescence excitation-emission matrix (A-TEEM) spectroscopy is a technique which has the potential to provide rapid identification of vaccine products. In this study, we examined nearly 1,100 spectra from 92 lots of seven similarly formulated attenuated live-virus vaccine products using EEM spectroscopy. The dataset was visualized by examining the resulting 3D fluorescence images and by performing exploratory analysis with principal component analysis. Each vaccine product appeared to give a unique spectral fingerprint, of which the specificity was derived from different intensities of the same peak, or the absence or presence of additional peaks. Finally, classification models were developed using soft independent modelling class analogy (SIMCA) for the purpose of identifying each vaccine product. External test and validation sets consisting of 84 vials from 42 lots not used during calibration were used to establish the accuracy of the models. In all, the SIMCA models achieved 100% accuracy in predicting the identity of the vaccines in roughly 1,300 total instances during calibration, testing, and validation, thus proving the models suitable for their intended use.
Nicole Ralbovsky, PhD – Merck
Process Analytical Technology (PAT) is increasingly explored within large molecule pharmaceuticals, with a particular emphasis in the vaccine space. PAT aims to provide increased process understanding and control through real-time monitoring of critical quality attributes and key process parameters, as well as detection of process deviations. Downstream purification in vaccine manufacturing processes can be complex and require copious analytical characterization. Here, we showcase the development of a PAT method for use as an In Process Control during clinical manufacturing of a vaccine drug substance. In addition, we illustrate how PAT can also be used for rapid characterization of a vaccine drug product. In the first case, SoloVPE UV-Vis spectroscopy is leveraged as a rapid and accurate total protein concentration method; in the second case, Raman hyperspectral imaging with machine learning are employed for analytical characterization and understanding of a drug product formulation. Both efforts illustrate the advantages of PAT in vaccine process development and clinical manufacturing spaces, and indicate the utility of PAT for process optimization and improvement.
Rita C. Barral – Sanofi, Process Analytical Technology – Mammalian Platform, Framingham, Massachusetts, USA, Marina Hincapie – Sanofi, Process Analytical Technology – Mammalian Platform, Framingham, Massachusetts, USA, John Bobiak – Sanofi, Manufacturing, Science & Technologies – Mammalian Platform, Framingham, Massachusetts, USA, Esra Kiran – Sanofi, Manufacturing, Science & Technologies – Mammalian Platform, Framingham, Massachusetts, USA, Lia Tescione – Sanofi, Manufacturing, Science & Technologies – Mammalian Platform, Framingham, Massachusetts, USA
In bioprocessing, one of the fastest-growing areas is Process Analytical Technology (PAT), which seeks to bring innovation to biologics process development and manufacturing. The term PAT encompasses a broad range of advanced analytical technologies, including analyzers, soft sensors, automated sampling systems, data management platforms, and automation frameworks. Overall, PAT is a data-rich approach that leads to a better understanding of processes because results are obtained within minutes or even seconds, as opposed to hours or days.
Cell culture media used in mammalian upstream processes are water-based solutions containing many different components, hence their complexity. These components include, but are not limited to, amino acids, vitamins, and metals, among others. It is difficult to analyze such mixtures due to their complex nature. For routine screening purposes, rapid spectroscopy-based techniques, specifically Raman spectroscopy, have been employed; however, this method falls short in differentiating types of cell culture media when the variations are only in the micromolar range.
In this study, we investigated an alternative multi-attribute method to analyze cell culture media that offers more sensitivity. Our investigation focused on the use of Absorbance-Transmission Excitation-Emission Matrix (A-TEEM) in conjunction with chemometric methods, such as Parallel Factor Analysis (PARAFAC) and Nonlinear Partial Least Squares (NPLS). With the method created, we were able to observe compositional changes between different cell culture media and assess the quality of the media, differentiating between expired and non-expired samples.
In general, A-TEEM could become a more suitable alternative to Raman spectroscopy-based methods through this study, which would greatly improve our understanding of variability in cell culture media.
Lyufei Chen, PhD – HORIBA Instruments, Jeffrey Julien – HORIBA Instruments
The variability in cell culture media significantly impacts cellular growth, productivity, and the quality profile of expressed therapeutic proteins. With the biopharmaceutical industry expanding, there is a growing need to monitor the consistency of cell culture media, both between batches and over time. In this presentation, we will showcase a novel qualification and quantification method applied to commercially available cell media. This method is based on A-TEEM, a multimodal spectroscopic technique that integrates UV/Vis spectroscopy (A-T for Absorbance-Transmission) and 3D fluorescence (EEM for Excitation Emission Matrix). We will also introduce the use of PARAFAC (Parallel factor analysis), which decomposes fluorescence excitation emission matrices into their underlying chemical components. By combining A-TEEM with PARAFAC, we can qualify cell media by clustering and classifying batches, as well as quantify cell media to provide concentration information on major components such as tryptophan, tyrosine, pyridoxine, folic acid, and riboflavin. This technique will provide a sensitive alternative tool for cell media profiling that is easy to use and requires minimal sample preparation.
Alan G. Ryder, PhD (he/him/his) – Nanoscale Biophotonics Laboratory, University of Galway, Matheus A. de Castro – Nanoscale Biophotonics Laboratory, University of Galway
Poly-N-isopropylacrylamide (PNIPAm), a thermos-responsive polymer, highly soluble in water below its Lower Critical Solution Temperature(LCST), is widely used in biomedical applications like drug delivery. Nipam is however, very prone to aggregation both in solid and solution states which can be induced relatively innocuous sources such as prolonged storage in humid environments and air-water interface stresses. Thus, being able to quickly, inexpensively, and accurately measure size the size and aggregation state of PNIPAm in solution is critical for applications where stoichiometric molecular ratios are required.
Dynamic light scattering (DLS) is probably the most widely available, and inexpensive nanoparticle sizing technique, but there are limitations in respect to speed of measurement, concentration and polydispersity that need to be carefully considered. Fluorescence Correlation Spectroscopy (FCS) using covalent or non-covalent fluorophores as probes can also be employed, however, there are disadvantages in terms of measurement complexity, sample manipulation, and usable sample concentrations.
However, we have found that commercially available PNIPAm were intrinsically fluorescent, with two distinct types of emission, presumably from the initiator molecules used during synthesis. By measuring the intrinsic emission from PNIPAm using polarized Excitation-Emission Matrix (pEEM) spectroscopy, one can get a more detailed and informative measurement of size changes in solution. pEEM measurements can be implemented as parallel or perpendicular polarization modes which when coupled with the Rayleigh scatter signals provide useful information about particle size changes. The big advantages of pEEM polymer analysis in an industrial context is that itis label free, needs minimal sample handling, and can be implemented using conventional benchtop fluorometers. Here we show that PNIPAm particle size (reference data from DLS measurements) were correlated (R2>0.9) with the Rayleigh scattering band from the parallel polarized measurements, and that the change in emission properties during aggregation provides a convenient method for monitoring polymer aggregation induced by air-water interfaces.
Brendon M. Lyons, M.S. Chemistry – Bristol Myers Squibb, Jeffrey Julien – HORIBA Instruments, Lyufei Chen, PhD – HORIBA Instruments
Absorbance - Transmission, Excitation Emission Matrix (A-TEEM) spectroscopy was evaluated for a novel application of screening falsified biologic drug products. Some falsified biologics drug products are challenging to screen with fast, inexpensive analytical tools. For example, authentic BMS biologics which are falsified by market diversion (i.e. smuggling) require highly specific analytical methods to verify authenticity to ensure patient safety. However, traditional peptide mapping techniques are too slow, expensive, and labor intensive for screening suspect products in a high-throughput forensics laboratory. A proof-of-principle study was performed in collaboration with the analytical instrument vendor, HORIBA, to evaluate the specificity of A-TEEM for similar immunoglobulin G (IgG) protein structures. A-TEEM is a three-dimensional spectroscopic technique, which can characterize the composition and local environment of the aromatic amino acid residues tyrosine and tryptophan. Two groups of commercial drug substance IgG proteins were selected based on primary structure similarity: 1. Six (6) proteins with the same number of tyrosine and tryptophan residues per molecule and 2. Five (5) proteins with high sequence similarity (93-95% homologous). The A-TEEM spectra were measured in the same buffer on a HORIBA Aqualog spectrometer. The A-TEEM profiles were analyzed using a multivariate statistical technique named Parallel Factor Analysis (PARAFAC) to extract the reproducible features of the fluorescence profiles. Although the A-TEEM plots appear visually broad and featureless, the PARAFAC model was able to distinguish these IgG samples. This talk will present results of the application of A-TEEM spectroscopy and chemometrics to the discrimination of similar biologic drug substances for the purpose of suspect biologic drug screening.
Prabuddha Mukherjee – Sartorius Stedim North America, Inc.
With the onset of COVID that has left a huge impact in our society, research in the Bio-tech industry has seen a paradigm shift in designing assays, monitoring cell cultures, implementing novel imaging modalities and subsequent analytical algorithms for better understanding of the processes and improve its efficacy. While the current standard for detection still relies heavily on fluorescence-based assays, the chemical impact of these fluorescent tags cannot be overlooked. Thus, there is a significant emphasis on designing label free detection strategies, which rely solely on the molecular response of the target species. Certain molecular responses like intrinsic fluorescence of cofactors NADPH, FAD tracks the metabolic state of the cells, while vibrational methods are able to track the chemical composition and distribution of proteins, lipids and nucleic acids in cells. While these strategies are extremely promising, they often suffer from low signal from the samples posing significant challenges especially as process analytical tools. Here, we would be presenting multi-photon label free detection strategies (fluorescence and Raman) that is potentially promising for bioreactor applications. Lastly, we will outline how these technologies when coupled effectively with the present offline tools can offer more reliable analytical tools
Ashley Love – University of Nottingham
Excitation emission matrix fluorescence spectroscopy is a rapid technique, which is not only information “rich” but also very sensitive. This work explores the potential that this approach offers to provide key insights into vital parameters, such as molecular environments, configurations and dynamics. This is achieved by analyzing the information contained within an EEM using a variety of multivariate methods, allowing us gain insights into these parameters which are often obscured in traditional fluorescence spectroscopy.
Another avenue where EEM fluorescence spectroscopy is invaluable is in process monitoring, as the ability to monitor and optimise reactions in real-time is of upmost importance. The ability to do this is critical in the manufacturing processes of active pharmaceutical agents and essential medicines. By combining in-line optical spectroscopies with flow chemistry, the ability to rapidly monitor and autonomously control reactions becomes possible, allowing efficient, algorithmically driven process control and optimization to take place. Here we utilize the advantages of EEM fluorescence spectroscopy alongside an array of other spectroscopic techniques to monitor reactions in real-time and autonomously optimize, model and probe key reaction metrics.
John Bobiak - Bristol-Myers Squibb
Chemically defined media is utilized in biopharmaceutical processes to suppl cells with nutrients for growth and protein production. Spectroscopic tools can plat an important role in demonstrating control of prepared medium properties, notable due to measurement speed and simplicity. Excitation-Emission Fluorescess (EEM) is one spectroscopic tool that may be employed to detect prepared medium variations imposed during batch preparation and.pr by medium hold time.s Semi-quantitative and quantitative models (Partial Least Squares and Principal Component Analysis) were developed using the reshaped two-dimensional data matrix provided by EEM. Such models were able to readily detect differences in medium preparation procedures, and score trends observed at medium hold times correlated wit process performance. Furthermore a semi-quantitative PLS model for two amino acides was developed through gravimetric spiking, and applied to illustrate sensitivity for both formulation errors and medium stability, Stability-indicting EEM methods like the one shown here, are broadly applicable to trend prepared media critical attributes in development and commercial settings.
Michael George – University of Nottingham
Photochemistry and electrochemistry are potentially very powerful tools for manufacturing not least because energy is delivered to reacting molecules far more selectively than by bulk heating in an atom efficient manner. Indeed, more than a century ago, Ciamician, presented a very powerful vision of where photochemistry could lead us [Science 1912, 36, 385-394]. By comparison, its penetration into chemical manufacture remains comparatively modest because of a whole series of issues, mostly centered on the problems of carrying out large-scale photochemical reactions both efficiently and safely. In recent years we have been addressing some of the challenges of making photochemistry and electrochemical synthesis greener, more energy efficient and more widely accessible. This presentation will cover our activity particularly aiming for generic approaches for linking and scaling up multi-step processes in the context of photo-, electro- and thermal- chemistry on the kg/day scale particularly focused on continuous photo-redox and oxidative and reductive electrochemical processes enabling 1-10 kg/day productivity in Taylor Vortex reactors with a very small footprint. Existing and new PAT approaches (Raman, IR and EEMs) are used in combination with autonomous self-optimization for process development. The main aim focusses on improving sensitivity, specificity, dynamic range and the speed of data acquisition.
Nicole Ralbovsky, PhD – Merck
Process Analytical Technology (PAT) is increasingly explored within large molecule pharmaceuticals, with a particular emphasis in the vaccine space. PAT aims to provide increased process understanding and control through real-time monitoring of critical quality attributes and key process parameters, as well as detection of process deviations. Downstream purification in vaccine manufacturing processes can be complex and require copious analytical characterization. Here, we showcase the development of a PAT method for use as an In Process Control during clinical manufacturing of a vaccine drug substance. In addition, we illustrate how PAT can also be used for rapid characterization of a vaccine drug product. In the first case, SoloVPE UV-Vis spectroscopy is leveraged as a rapid and accurate total protein concentration method; in the second case, Raman hyperspectral imaging with machine learning are employed for analytical characterization and understanding of a drug product formulation. Both efforts illustrate the advantages of PAT in vaccine process development and clinical manufacturing spaces, and indicate the utility of PAT for process optimization and improvement.
Do you have any questions or requests? Use this form to contact our specialists.
