Applying Pattern Recognition as a Robust Approach for Silicone Oil Droplet Identification in Flow-Microscopy Images of Protein Formulations

Sub-Visible Particle Classification and Label Consistency Analysis for Flow-Imaging Microscopy Via Machine Learning Methods

Sub-visible particles can be a quality concern in pharmaceutical products, especially parenteral preparations. To quantify and characterize these particles, liquid samples may be passed through a flow-imaging microscopy instrument that also generates images of each detected particle. Machine learning techniques have increasingly been applied to this kind of data to detect changes in experimental conditions or classify specific types of particles, primarily focusing on silicone oil. That technique generally requires manual labeling of particle images by subject matter experts, a time-consuming and complex task. In this study, we created artificial datasets of silicone oil, protein particles, and glass particles that mimicked complex datasets of particles found in biopharmaceutical products. We used unsupervised learning techniques to effectively describe particle composition by sample. We then trained independent one-class classifiers to detect specific particle populations: silicone oil and glass particles. We also studied the consistency of the particle labels used to evaluate these models. Our results show that one-class classifiers are a reasonable choice for handling heterogeneous flow-imaging microscopy data and that unsupervised learning can aid in the labeling process. However, we found agreement among experts to be rather low, especially for smaller particles (< 8 µm for our Micro-Flow Imaging data). Given the fact that particle label confidence is not usually reported in the literature, we recommend more careful assessment of this topic in the future.

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.

Particles in Biopharmaceutical Formulations, Part 2: An Update on Analytical Techniques and Applications for Therapeutic Proteins, Viruses, Vaccines and Cells

Particles in biopharmaceutical formulations remain a hot topic in drug product development. With new product classes emerging it is crucial to discriminate particulate active pharmaceutical ingredients from particulate impurities. Technical improvements, new analytical developments and emerging tools (e.g., machine learning tools) increase the amount of information generated for particles. For a proper interpretation and judgment of the generated data a thorough understanding of the measurement principle, suitable application fields and potential limitations and pitfalls is required. Our review provides a comprehensive overview of novel particle analysis techniques emerging in the last decade for particulate impurities in therapeutic protein formulations (protein-related, excipient-related and primary packaging material-related), as well as particulate biopharmaceutical formulations (virus particles, virus-like particles, lipid nanoparticles and cell-based medicinal products). In addition, we review the literature on applications, describe specific analytical approaches and illustrate advantages and drawbacks of currently available techniques for particulate biopharmaceutical formulations.

Towards quantification and differentiation of protein aggregates and silicone oil droplets in the low micrometer and submicrometer size range by using oil-immersion flow imaging microscopy and convolutional neural networks

Biopharmaceutical product characterization benefits from the quantification and differentiation of unwanted protein aggregates and silicone oil droplets to support risk assessment and control strategies as part of the development. Flow imaging microscopy is successfully applied to differentiate the two impurities in the size range larger than about 5 µm based on their morphological appearance. In our study we applied the combination of oil-immersion flow imaging microscopy and convolutional neural networks to extend the size range below 5 µm. It allowed to differentiate and quantify heat stressed therapeutic monoclonal antibody aggregates from artificially generated silicone oil droplets with misclassification rates of about 10% in the size range between 0.3 and 5 µm. By comparing the misclassifications across the tested size range, particles in the low submicron size range were particularly difficult to differentiate as their morphological appearance becomes very similar.