Quantification and Characterization of Subvisible Proteinaceous Particles in Opalescent mAb Formulations Using Micro-Flow Imaging

Micro-flow imaging (MFItm) has been shown to be more sensitive than light obscuration (LO) methods for measuring subvisible proteinaceous particles in protein formulations. Given the potential challenges in detecting particulates in opalescent mAb formulations, the accuracy of MFI to size and count particles in opalescent solutions was investigated and compared to LO and membrane microscopy methods. Proteinaceous monoclonal antibody (mAb) particles, generated either by chemical denaturation or agitation stress, polystyrene and glass particles were used as model systems for measurements in opalescent mAb solutions. The sizing and counting accuracies of MFI were unaffected by the opalescence of the medium. Using glass particles as a model system for proteinaceous particles, MFI was able to detect relatively low particle concentrations (~ 10/mL) in opalescent solutions. MFI showed excellent linearity (R2 = 0.9969) for quantifying proteinaceous particles in opalescent solutions over a wide range of particle concentrations (~ 20-160,000/mL). Analyses of MFI particle image intensities revealed significant differences in the transparency of proteinaceous particles as a function of their size and mode of generation. LO method significantly underestimated proteinaceous particles, particularly those in the 2–10 μm size range. The less opaque proteinaceous particles were relatively more underestimated by the LO method in opalescent solutions.     

The Coulter Principle for Analysis of Subvisible Particles in Protein Formulations

The Coulter principle can be used for analysis of subvisible particles in protein formulations. The approach has several advantages including: an orthogonal operating principle, high sensitivity, ability to detect very small particles, excellent reproducibility, and high-resolution size information. This minireview discusses some of the important considerations that must be taken into account when utilizing the Coulter principle for subvisible particle analysis in protein formulations.     

Deep Convolutional Neural Network Analysis of Flow Imaging Microscopy Data to Classify Subvisible Particles in Protein Formulations

Flow-imaging microscopy (FIM) is commonly used to characterize subvisible particles in therapeutic protein formulations. Although pharmaceutical companies often collect large repositories of FIM images of protein therapeutic products, current state-of-the-art methods for analyzing these images rely on low-dimensional lists of “morphological features” to characterize particles that ignore much of the information encoded in the existing image databases. Deep convolutional neural networks (sometimes referred to as “CNNs or ConvNets”) have demonstrated the ability to extract predictive information from raw macroscopic image data without requiring the selection or specification of “morphological features” in a variety of tasks. However, the inherent heterogeneity of protein therapeutics and optical phenomena associated with subvisible FIM particle measurements introduces new challenges regarding the application of ConvNets to FIM image analysis. We demonstrate a supervised learning technique leveraging ConvNets to extract information from raw images in order to predict the process conditions or stress states (freeze-thawing, mechanical shaking, etc.) that produced a variety of different protein particles. We demonstrate that our new classifier, in combination with a “data pooling” strategy, can nearly perfectly differentiate between protein formulations in a variety of scenarios of relevance to protein therapeutics quality control and process monitoring using as few as 20 particles imaged via FIM.     

折光率对微流成像法检测蛋白制剂不溶性微粒的影响

目的：蛋白制剂中不溶性微粒的含量是衡量样品质量的重要指标之一,为了更为准确地检测不溶性微粒的含量和粒径,本研究探讨了溶液折光率对于微流成像系统检测不溶性微粒的影响。方法：本研究以牛血清白蛋白（BSA）为例,通过常见的外界刺激条件（冷冻-解冻）制备高浓度的蛋白质不溶性微粒,并将此微粒稀释至不同折光率的溶液（由PEG1000、海藻糖制备）中,利用微流成像系统检测不溶性微粒的含量。结果：当溶液的折光率接近蛋白质不溶性微粒折光率时,利用微流成像技术检测的不溶性微粒含量低于实际的微粒含量。此外,随着溶液折光率的增加,采用微流成像技术检测出的不溶性微粒的粒径也随之减小。结论：蛋白质溶液的折光率发生改变,会影响利用微流成像技术检测不溶性微粒的准确性。因此,利用微流成像技术检测蛋白制剂中不溶性微粒时,需要考虑到制剂处方的折光率对于检测不溶性微粒的影响,必要时可以采用稀释的方法降低折光率对蛋白质颗粒的屏蔽作用。     

Flow Microscopy Imaging Is Sensitive to Characteristics of Subvisible Particles in Peginesatide Formulations Associated With Severe Adverse Reactions

The presence of subvisible particles in formulations of therapeutic proteins is a risk factor for adverse immune responses. Although the immunogenic potential of particulate contaminants likely depends on particle structural characteristics (e.g., composition, size, and shape), exact structure-immunogenicity relationships are unknown. Images recorded by flow imaging microscopy reflect information about particle morphology, but flow microscopy is typically used to determine only particle size distributions, neglecting information on particle morphological features that may be immunologically relevant. We recently developed computational techniques that utilize the Kullback-Leibler divergence and multidimensional scaling to compare the morphological properties of particles in sets of flow microscopy images. In the current work, we combined these techniques with expectation maximization cluster analyses and used them to compare flow imaging microscopy data sets that had been collected by the U.S. Food and Drug Administration after severe adverse drug reactions (including 7 fatalities) were observed in patients who had been administered some lots of peginesatide formulations. Flow microscopy images of particle populations found in the peginesatide lots associated with severe adverse reactions in patients were readily distinguishable from images of particles in lots where severe adverse reactions did not occur.     

Collaborative Study for Analysis of Subvisible Particles Using Flow Imaging and Light Obscuration: Experiences in Japanese Biopharmaceutical Consortium

The evaluation of subvisible particles, including protein aggregates, in therapeutic protein products has been of great interest for both pharmaceutical manufacturers and regulatory agencies. To date, the flow imaging (FI) method has emerged as a powerful tool instead of light obscuration (LO) due to the fact that (1) protein aggregates contain highly transparent particles and thereby escape detection by LO and (2) FI provides detailed morphological characteristics of subvisible particles. However, the FI method has not yet been standardized nor listed in any compendium. In an attempt to assess the applicability of the standardization of the FI method, we conducted a collaborative study using FI and LO instruments in a Japanese biopharmaceutical consortium. Three types of subvisible particle preparations were shared across 12 laboratories and analyzed for their sizes and counts. The results were compared between the methods (FI and LO), inter-laboratories, and inter-instruments (Micro Flow Imaging and FlowCam). We clarified the marked difference between the detectability of FI and LO when counting highly transparent protein aggregates in the preparations. Although FlowCam provided a relatively higher number of particles compared with MFI, consistent results were obtained using the instrument from the same manufacturer in all 3 samples.     

Imaging Flow Cytometry for Sizing and Counting of Subvisible Particles in Biotherapeutics

Imaging flow cytometry (IFC), a technique originally designed for cellular imaging, is featured by the parallel acquisition in brightfield (BF), fluorescence (FL), and side scattering channels. Introduced to the field of subvisible particle analysis in biopharmaceuticals roughly ten years ago, it has the potential to yield additional information, e.g., on particle origin. Here, we present an extensive, systematic development of masks for IFC image analysis to optimize the accuracy of size determination of polystyrene beads and pharmaceutically relevant particles (protein, silicone oil) in BF and FL channels. Based on the developed masks, particle sizing and counting by IFC are compared to flow imaging microscopy (FIM). Mask verification based on fluorescent polystyrene particles revealed good agreement between sizes obtained from IFC and FIM. In the evaluation of counting accuracy, IFC reported lower concentrations for polystyrene particle standards than FIM. For the analysis of fluorescently stained silicone oil and protein particles however, IFC FL imaging reported higher particle concentrations in the low micrometer size range. Overall, we identified IFC as suitable tool to generate supportive data for particle characterization purposes or trouble shooting activities, but not as routine quantitative technique, e.g., for subvisible particle analysis during drug product development or quality control.     

Combining Machine Learning and Backgrounded Membrane Imaging: A Case Study in Comparing and Classifying Different Types of Biopharmaceutically Relevant Particles

This study investigates how backgrounded membrane imaging (BMI) can be used in combination with convolutional neural networks (CNNs) in order to quantitatively and qualitatively study subvisible particles in both protein biopharmaceuticals and samples containing synthetic model particles. BMI requires low sample volumes and avoids many technical complications associated with imaging particles in solution, e.g., air bubble interference, low refractive index contrast between solution and particles of interest, etc. Hence, BMI is an attractive technique for characterizing particles at various stages of drug product development. However, to date, the morphological information encoded in brightfield BMI images has scarcely been utilized. Here we show that CNN based methods can be useful in extracting morphological information from (label-free) brightfield BMI particle images. Images of particles from biopharmaceutical products and from laboratory prepared samples were analyzed with two types of CNN based approaches: traditional supervised classifiers and a recently proposed fingerprinting analysis method. We demonstrate that the CNN based methods are able to efficiently leverage BMI data to distinguish between particles comprised of different proteins, various fatty acids (representing polysorbate degradation related particles), and protein surrogates (NIST ETFE reference material) only based on BMI images. The utility of using the fingerprinting method for comparing morphological differences and similarities of particles formed in distinct drug products and/or laboratory prepared samples is further demonstrated and discussed through three case studies.     

Optimization of Infrared Microscopy to Assess Secondary Structure of Insulin Molecules Within Individual Subvisible Particles in Aqueous Formulations

The analysis of subvisible particles is currently challenging but pivotal to the understanding and control of the quality of protein therapeutics. While a range of characterization methods is available for subvisible particles, information on the protein conformation in a particle—considered a possible parameter in eliciting unwanted immunogenicity of protein therapeutics—is especially challenging in the lower micrometer range using existing analytical technologies. Using 6 different protein particle populations, we show that transmission Fourier transform infrared (FTIR) microscopy can determine protein secondary structure in single particles down to 10 μm. The analytical setup presented here is able to immobilize protein particles and obtain transmission FTIR spectra on individual protein particles in their intact aqueous environment. Spectra of dried particles, on the other hand, were found to occasionally differ from spectra of particles in aqueous environment. In summary, using the analytical setup described in this study, transmission FTIR microscopy uniquely provides information on single protein particles in particle populations in their aqueous environment without interference from the background protein solution.     

Subvisible Particle Analysis of 17 Monoclonal Antibodies Approved in China Using Flow Imaging and Light Obscuration

In the study, subvisible particles in 205 samples from 17 commercial mAb drug products approved in China were analyzed using light obscuration (LO) and flow imaging microscopy (FIM) methods. For each method, a total 633 tests (runs) were performed. In the tests, samples in state of lyophilized powder or syringe package had significantly higher particle concentrations. It was confirmed by analyzing the 205 drug product samples that FIM particle counts are generally higher than LO counts. The cause of the higher counts of FIM method than LO counts was examined by looking into the contribution of proteinaceous, translucent particles in the samples. The data of the study showed that the number of proteinaceous, translucent particles was a factor in the elevated counts of FIM method compared to LO method.     

Differentiation of Subvisible Silicone Oil Droplets from Irregular Standard Dust Particles

New methods are being applied to distinguish silicone oil droplets from subvisible particles in therapeutic protein formulations. The need to standardize these methods and compare them to established methods is critical to increase the understanding of the risks from protein aggregation and other subvisible particulate matter. We present the use of medium test dust (MTD) as a stable subvisible particle standard that can be distinguished from silicone oil by flow imaging methodology and the combination of light obscuration and microscopic methods. We further present the use of binary classification techniques to characterize the ability of flow imaging to distinguish between particle types (i.e., silicone vs. protein, silicone vs. standard, etc.) as a function of particle size. For the differentiation of silicone oil and MTD, the aspect ratio attribute was as good as or better than any other characteristic or combination of two characteristics applied to distinguish this particle population. However, the value of the discrimination by flow imaging was limited to particles larger than 5 μm. © 2013 Wiley Periodicals, Inc. and the American Pharmacists Association J Pharm Sci 102:1696-1700, 2013     

Aerosol Dispersion of Respirable Particles in Narrow Size Distributions Using Drug-Alone and Lactose-Blend Formulations

PURPOSE: To examine the effect of formulation type on the aerosolization of respirable particles in narrow size distributions. METHODS: Aerosol dispersion of two formulation types (drug alone and 2% w/w drug-lactose blends) containing micronized or spray-dried fluticasone propionate (FP) particles (d50% = 1.3 to 9.6 microm, GSD = 1.8 to 2.2) were examined using cascade impaction at 60 l/min with low and high resistance inhaler devices: Rotahaler and Inhalator, respectively. RESULTS: The aerosol dispersion of FP particles was significantly affected by the particle size, particle type, inhaler device, and formulation type. Interactions were observed between all factors. Generally, greater powder entrainment was obtained with smaller d50%. Higher emitted doses were obtained from drug-alone formulations of spray-dried FP particles and lactose blends of micronized FP particles. Greater aerosol dispersion of spray-dried FP particles was obtained using lactose-blend formulations with d50% around 4 microm. Greater aerosol dispersion of micronized FP particles was obtained using formulations of drug alone. Larger d50% produced larger mass median aerodynamic diameters. CONCLUSIONS: Small changes in the particle size within the 1-10-microm range exerted a major influence on aerosol dispersion of jet-milled and spray-dried FP particles using drug-alone and lactose-blend formulations.     

Utility of Three Flow Imaging Microscopy Instruments for Image Analysis in Evaluating four Types of Subvisible Particle in Biopharmaceuticals

Subvisible particles (SVPs) are a critical quality attribute of parenteral and ophthalmic products. United States Pharmacopeia recommends the characterizations of SVPs which are classified into intrinsic, extrinsic, and inherent particles. Flow imaging microscopy (FIM) is useful as an orthogonal method in both the quantification and classification of SVPs because FIM instruments provide particle images. In addition to the conventionally used FlowCam (Yokogawa Fluid Imaging Technologies) and Micro-Flow Imaging (Bio-Techne) instruments, the iSpect DIA-10 (Shimadzu) instrument has recently been released. The three instruments have similar detection principles but different optical settings and image processing, which may lead to different results of the quantification and classification of SVPs based on the information from particle images. The present study compares four types of SVP (protein aggregates, silicone oil droplets, and surrogates for solid free-fatty-acid particles, milled-lipid particles, and sprayed-lipid particles) to compare the results of size distributions and classification abilities obtained using morphological features and a deep-learning approach. Although the three FIM instruments were effective in classifying the four types of SVP through convolutional neural network analysis, there was no agreement on the size distribution for the same protein aggregate solution, suggesting that using the classifiers of the FIM instruments could result in different evaluations of SVPs in the field of biopharmaceuticals.     

Rapid Quantification of Protein Particles in High-Concentration Antibody Formulations

Current technologies for monitoring the subvisible particles that may be generated during fill-finish operations for protein formulations are cumbersome. Measurement times are generally too long for real-time analysis, and the high protein concentrations that are characteristic of many antibody products interfere with common optical techniques for particle analysis. To rapidly monitor protein particle levels in high-concentration protein solutions, we developed a fluorescence-based method that uses extrinsic fluorescent dyes such as 4,4'-dianilino-1,1'-binaphthyl-5,5'-disulfonic acid that are sensitive to the presence of aggregated protein. To test the method, antibody formulations containing various concentrations of protein particles were generated by application of various mechanical and freeze-thaw stresses. After addition of fluorescent dyes, fluorescence intensities were measured and compared to fluorescence intensities in particle-free formulations. The differences in fluorescence intensities were linearly proportional to protein particle levels, which for calibration purposes were measured offline by fluid imaging microscopy and protein assays. Protein particle levels could be measured without requiring sample dilution, even in high-concentration (e.g., 40 mg/mL) antibody formulations.     

Contribution of Intravenous Administration Components to Subvisible and Submicron Particles Present in Administered Drug Product

Particulate matter present in drug products intended for parenteral administration to patients is typically monitored and controlled in the finished drug product to minimize potential risks to patients. In contrast to particulates found in drug products, the current study evaluated particulates representative of materials and operations typically used in the dose preparation and administration of drug products. A comprehensive assessment of intrinsic and extrinsic sources of subvisible and submicron particulates arising from materials associated with subcutaneous and intravenous dose preparation and administration was conducted. In particular, particles arising from disposable syringes, commercial sterile diluents, and intravenous supplies were quantitated using established methods for subvisible (light obscuration, flow imaging) and submicron particles (resistive pulse sensing). Each of these sources contributed varying amounts of particulates; therefore, owing to sources from materials required for administration, it is inadequate to assume that the total particulate load delivered to patients arises solely from the drug product. Careful consideration of the administration method and supplies used can improve the predictability of particulate levels present in dose preparations or administration volumes.     

Flow Imaging: Moving Toward Best Practices for Subvisible Particle Quantitation in Protein Products

The potential impact of subvisible particles (SVPs) in protein therapeutic products has received a great deal of attention recently. As a result, new analytical methods have emerged to characterize and quantify SVPs. Among these, flow imaging (also called flow microscopy) has been widely employed. As we have used both FlowCAM® and Micro-Flow Imaging? in a variety of development projects, we wished to share our experiences and observations, with the intention of fostering a discussion that will lead to best practices for the use of flow imaging to quantify SVP levels in biopharmaceuticals. © 2013 Wiley Periodicals, Inc. and the American Pharmacists Association J Pharm Sci 102:1133–1134, 2013     

Micro-Flow Imaging: Flow Microscopy Applied to Sub-visible Particulate Analysis in Protein Formulations

The need to monitor, measure, and control sub-visible proteinaceous particulates in biopharmaceutical formulations has been emphasized in recent publications and commentaries. Some of these particulates can be highly transparent, fragile, and unstable. In addition, for much of the size range of concern, no practical measurement method with adequate sensitivity and repeatability has been available. A complication in measuring protein particulates in many formulations is the simultaneous presence of other particle types such as silicone micro-droplets, air bubbles, and extrinsic contaminants. The need has therefore been identified for new analytical methods which can accurately measure and characterize sub-visible particulates in formulations. Micro-flow imaging has been shown to provide high sensitivity in detecting and imaging transparent protein particles and a unique capability to independently analyze such populations even when other particle types are present.Key words: light obscuration, micro-flow imaging, particle sizing, protein aggregation, protein formulation     

Subvisible Particles in Solutions of Remicade in Intravenous Saline Activate Immune System Pathways in In Vitro Human Cell Systems

Among patients that receive Remicade® therapy, more than 20% have adverse infusion related reactions and approximately 50% have immunogenic responses.1, 2, 3 Upon characterization of initial Remicade®-IV solution we observed a high concentration of subvisible particles that could inadvertently be delivered to patients. This solution was processed through the IV infusion system, mimicking the typical clinical administration setup - either with or without an in-line filter connected to the IV line. The samples generated thereafter were tested using various in vitro assays for activation of the innate immune system via cytokine release in whole blood and in peripheral blood mononuclear cell (PBMC) cultures, and activation of the Toll like receptors (TLRs). Activation of the adaptive immune system was evaluated by monitoring upregulation of surface receptors on dendritic cells (DCs) and CD4+ T cell proliferation in response to IV solution of Remicade®. Our results indicate that subvisible particles in Remicade®-saline solution have a significant role in activation of the immune system but there are extrinsic factors potentially contributed by the in-line filters or other process parameters that also contribute to immune system activation.     

A Random Forest Approach for Counting Silicone Oil Droplets and Protein Particles in Antibody Formulations Using Flow Microscopy

Purpose

To evaluate a random forest model that counts silicone oil droplets and non-silicone oil particles in protein formulations with large class imbalance.

Methods

In this work, we present a novel approach for automated image analysis of flow microscopy data based on random forest classification enabling rapid analysis of large data sets. The random forest approach overcomes many of the limitations of traditional classification schemes derived from simple filters or regression models. In particular, the approach does not require a priori selection of important morphology parameters.

Results

We analyzed silicone oil droplets and non-silicone oil particles observed in four model systems with protein concentrations of 20, 50 and 125 mg/mL. Filters based on random forests achieve higher classification accuracies when compared to regression based filters. Additionally, we showcase a procedure that allows for accurate counting of particles ≥1 μm.

Conclusions

Our method is generally applicable for classification and counting of different classes of particles as long as class morphologies are differentially expressed.

     

Screening of Monoclonal Antibody Formulations Based on High-Throughput Thermostability and Viscosity Measurements: Design of Experiment and Statistical Analysis

The purpose of this study was to demonstrate the utility of combining a design of experiment (DOE) approach with high-throughput formulation screening to identify the main factors affecting protein thermostability and solution viscosity. The optimization of buffer compositions was guided by statistical analysis of the data to obtain the targeted combination of low viscosity and high thermostability. Different monoclonal antibody (mAb) formulation variables were evaluated in the study to achieve optimization of two parameters: (i) thermostability characterized by temperature of hydrophobic exposure and (ii) viscosity. High-throughput measurements were employed to characterize both parameters. The significance of each factor and the two-way interactions between them was studied by multivariable regression analysis. An experimental design was used to estimate the significance of all factors, including interaction effects. The range of optimal buffer compositions that maximized thermostability and minimized viscosity of a mAb formulation was determined. The described high-throughput methods are well suited for characterization of multiple protein formulation compositions with minimized resources such as time and material. The DOE approach can be successfully applied to the screening of mAb formulations early in the development lifecycle.