Cambridge Healthtech Institute’s 2nd Annual

Characterizing Aggregates & Impurities

Detect. Quantify. Monitor.

March 4-5, 2019 * The Westin Alexandria * Alexandria, VA


Product and process-related impurities impact biopharmaceutical development at every stage from discovery to development to manufacturing. Detecting, identifying and characterizing these variants and impurities are of high regulatory concern, due to the potential correlation between aggregation and increased immunogenicity of the biotherapeutics.

Among the impurities, aggregates such as subvisible and submicron particles, and host cell proteins are of particular challenge. The 2nd Annual Characterizing Aggregates & Impurities conference brings together best practices and new insights in the identification, monitoring and characterization of these impurities and contaminants.

Final Agenda


7:30 am Registration and Morning Coffee

8:30 Chairperson’s Opening Remarks

Christopher Yu, PhD, Principal Scientist, Protein Analytical Chemistry, Genentech, Inc.


8:40 Analytical Strategies for Characterizing Residuals in Autologous Chimeric Antigen Receptor (CAR) T-Cell Therapy

Yue_HaiHai Yue, PhD, Senior Scientist, Analytical Development, Juno Therapeutics, A Celgene Company

The manufacturing of autologous CAR T-cell products starts with highly heterogeneous leukapheresis materials from patients and involves the use of various raw materials. This complex process may result in a formulation containing process- and product-related residuals. A phase-appropriate, risk-based analytical development strategy for possible residual characterization will be presented.

9:10 Analytical Method for In-Process Control and Characterization for Bi-specifics

Tao_YuanqiYuanqi Tao, PhD, Investigator II, Integrated Biologics Profiling, Novartis Biologics, NIBR

9:40 Improving Subvisible Particle Characterization with Deep Learning

Neelima Chavali, PhD, Engineer, Development Supply Chain, Amgen, Inc.

Deep learning represents specialized neural networks that can discover intricate patterns in digital images. In this work, we describe how we apply deep learning to overcome the limitation with flow microscopy particle classification by going beyond Aspect Ratio. We demonstrate that this leads to higher accuracy and precision in sub-visible particle classification.

10:10 Networking Coffee Break


10:40 Evaluation of Particle Characterization Methods and Their Applications for Therapeutic Protein Development

Yoen Joo Kim, MSc, Scientist II, Analytical Sciences, MedImmune

Various particle characterization methods are employed for therapeutic protein development. The methods include light obscuration, flow microscopy, IR microscopy and visible particles detection. The suitability of the methods for determining the number of particles, the size of particles, morphology, or chemical identification will be discussed. Case studies will be presented to demonstrate the applications of the methods to support the development of therapeutic protein.

11:10 Aggregates Quantification in Wide Range of Size Using Orthogonal Methods

Uchiyama_SusumuSusumu Uchiyama, PhD, Professor, Biotechnology, Osaka University

Proteinous particles are generated by various stresses during manufacturing, transportation, storage and administration of therapeutic proteins. Sizes of such particles range from tens of nanometers to hundreds of micrometers and should be characterized according to the size ranges. In this talk, characterization methods of proteinous particles especially for the estimation of number concentration will be introduced with strengths and weaknesses of each method.

11:40 Non-Arrhenius Behavior of Peptide Fibril Formation in the Presence of Stabilizing Excipients and Its Impact on Formulation Development

Smith_KatelynKatelyn Smith, PhD, Senior Scientist, Pharmaceutical Sciences, Merck & Co., Inc.

Peptide fibrillation is a significant challenge in the successful development of a peptide therapeutic. The peptide discussed herein exhibits Arrhenius-based fibrillation kinetics where faster aggregation occurs with increasing temperature. Interestingly, the addition of stabilizing excipients reverses this trend: fibril formation is observed at lower temperatures but is inhibited at higher temperatures. This phenomenon has great implications on our understanding of formulation shelf-life stability where samples are stored at low temperatures.

12:10 pm Luncheon Presentation (Sponsorship Opportunity Available) or Enjoy Lunch on Your Own

12:40 Session Break

1:40 Chairperson’s Remarks

Katelyn Smith, PhD, Senior Scientist, Pharmaceutical Sciences, Merck & Co., Inc.


1:45 Semi-Quantitative Method for Measuring Inherent Visible Particles in Protein Products

Mazaheri_MaryamMaryam Mazaheri, MS, Assoc CMC Lead, Viela Bio

2:15 New Reference Materials for Monitoring Proteinaceous Particles in Biotherapeutics

Telikepalli_SrivalliSrivalli Telikepalli, PhD, Research Chemist, Biomolecular Measurement Division, NIST

We are developing subvisible and visible particle standards using a polymer made of ethylene tetrafluoroethylene to better mimic the morphology and optical properties of proteinaceous particles than currently available particles standards. Applications of the ETFE particles to standardize subvisible particle measurements and to standardize the visual inspection process will be discussed.

2:45 Impact of Chemical Modifications on the Immunogenicity of IgG Subvisible Particles

Bjoern_BollBjörn Boll, PhD, Head of Particle Lab and HOS Analytics, Physical Chemical Analytics, Novartis

The theoretical concerns regarding the potential immunogenicity of proteinaceous aggregates and subvisible particles in protein therapeutics have being widely debated. This talk will present the detailed mechanistic studies on the biological impact of aggregates and sub-visible proteinaceous particles in a hIgG1 transgenic mouse model. The results are discussed within the current status of related literature of in vivo and in vitro studies.

3:15 Networking Refreshment Break

3:45 Breakout Discussion Session

Join your peers and colleagues in facilitated, small-group discussions to exchange ideas, share best practices, discuss challenges and make new contacts.

Protein Footprinting - HDX, FPOP and Other Approaches

Moderator: Michael L. Gross, PhD, Professor, Chemistry, Immunology and Medicine, Washington University in St. Louis

  • Advantages and disadvantages of HDX
  • Advantages and disadvantages of FPOP
  • FPOP and other approaches vs. Synchrotron footprinting
  • Role of other footprinting approaches (e.g., NEM, GEE, BHD)

Challenges in Establishing Bioassay Acceptance Criteria


Liming Shi, MS, MA, Executive Director, Quality Control, CMAB Biopharma Inc.

Marla Abodeely, PhD, Head, Bioassay/Immunoassay Method Development, Shire Pharmaceuticals

  • Stage appropriate acceptance criteria set up along with method development
  • Accumulation of so-called "enough" historical bioassay data
  • Both scientific and statistically sound bioassay acceptance criteria
  • Challenges in bioassay method validation and transfer

Analytical Tools and Techniques for Host Cell Protein Characterization

Moderator: Christopher Yu, PhD, Principal Scientist, Protein Analytical Chemistry, Genentech, Inc.

Computational Re-design of Protein Solubility

Moderator: Salvador Ventura, PhD, Director, Institute of Biotechnology and Biomedicine, Autonomous University of Barcelona

  • The role of protein sequence
  • The role of protein structure
  • The role of protein stability
  • The role of exrinsic factors


4:45 Keynote Introduction

Katelyn Smith, PhD, Senior Scientist, Pharmaceutical Sciences, Merck & Co., Inc.


4:50 KEYNOTE PRESENTATION I: Mass Spectrometry for Higher Order Structure and Properties of Therapeutic Proteins

Michael L. Gross, PhD, Professor, Chemistry, Immunology, and Medicine, Washington University in St. Louis

We are developing methods for protein therapeutics to determine higher order structure and to assure the structure has not changed. Our methods bring understanding to hidden conformations that characterize mutant proteins from bacterial and viral sources and to amyloid-forming proteins. Of additional interest are methods for determining protein affinity with small molecules (drug candidates). The measurements strategies are mass-spectrometry-based protein footprinting (HDX, FPOP, and other labeling strategies) and native MS with ion mobility.

5:25 KEYNOTE PRESENTATION II: Combining Structural Aggregation Propensity and Stability Predictions to Re-Design Protein Solubility

Salvador Ventura, PhD, Director, Institute of Biotechnology and Biomedicine, Autonomous University of Barcelona

The aggregation propensity of each particular protein seems to be shaped by evolution according to its natural abundance in the cell. The production and downstream processing of recombinant polypeptides implies attaining concentrations that are orders of magnitude above their natural levels, often resulting in their aggregation. We demonstrate that our AGGRESCAN 3D structural aggregation predictor allows designing mutations at specific positions in the structure that improve the solubility of proteins without compromising their conformation. Our data indicate that the solubility of unrelated proteins can be easily tuned by in silico-designed non-destabilizing amino acid changes at their surfaces.

6:00 Welcome Reception in the Exhibit Hall with Poster Viewing

7:00 End of Day


8:00 am Morning Coffee

8:30 Chairperson’s Remarks

Salvador Ventura, PhD, Director, Institute of Biotechnology and Biomedicine, Universitat Autonoma de Barcelona


8:35 Characterization of Aggregation Propensities of Protein Therapeutics by Hydrogen/Deuterium Exchange Mass Spectrometry

Huang_RichardRichard Huang, PhD, Senior Research Investigator, Pharmaceutical Candidate Optimization, R&D, Bristol-Myers Squibb

Characterization of therapeutic antibody aggregate is a critical task in pharmaceutical industry to optimize product efficacy and prevent adverse effects. This presentation illustrates the utility of HDX-MS for the characterization of higher order structures of antibody aggregates. HDX information combined with molecular modeling, and other orthogonal analytical techniques, revealed the antibody aggregate structures and the mechanism of aggregate removal using Protein A chromatography. It also provided a general strategy for in-depth understanding in antibody and other biologics development.

9:05 Discovering and Quantifying Polysorbate Degrading HCP Activities by Mass Spectrometry

Yu_ChristopherChristopher Yu, PhD, Principal Scientist, Protein Analytical Chemistry, Genentech, Inc.

Residual levels of host cell proteins (HCPs) are hypothesized to contain lipase-like activities that result in hydrolytic degradation of polysorbate molecules used as a surfactant in monoclonal antibody therapeutics. We present an investigation that highlights the use of mass spectrometry in the identification of HCPs with polysorbate degrading activities. We also describe a mass spectrometric method that quantitatively measures free fatty acids and its use in bioprocess optimization towards reducing polysorbate degradation.

9:35 Practical Approaches Using Liquid Chromatography-Mass Spectrometry (LC-MS) to Characterize Protein Aggregates

Mugabe_SheilaSheila Mugabe, MSc, Senior Development Assoc, Macrogenics Inc.

Monoclonal antibodies and bispecific DART® molecules are being developed for a variety of indications including immune-oncology. This presentation will discuss practical approaches using case studies of LC/MS based characterization of the above molecules during process development and late-stage analytical characterization stages.

10:05 Coffee Break in the Exhibit Hall with Poster Viewing

10:45 Challenges in Analyzing Highly Viscous Proteins: Identification of SVPs with Micro Flow Imaging (MFI)

Zahir Akhunzada, PhD, DPST, Bristol Myers Squibb

Subvisible particles (SVPs) and protein aggregates formation in protein drug products is a great cause for concern, because of their potential for immunogenicity, and loss of therapeutic efficacy. The tendency to form aggregates is even greater in protein drug products with high concentrations due to molecular crowding and dipole-dipole interactions that leads to self-association and increase in viscosity. Analyzing such highly viscous products pose a number of challenges because of the fluid dynamics. The first part of this presentation discusses strategy to analyze highly viscous proteins by Microflow Imaging (MFI). The second part will discuss how to distinguish, both potentially proteinaceous and non-proteinaceous SVPs in protein formulations by using MFI in conjunction with the MVAS (MFI View Analysis Suite) software (ProteinSimple-Bio-techne) and Lumetics software (Lumetics Inc.) for report generation.

11:15 Improving the Accuracy of Subvisible Particle Analysis Using the Results from Measurement of Orthogonal Properties

Cavicchi_RichardRichard Cavicchi, Research Physicist, Biomolecular Measurement Division, Material Measurement Laboratory, NIST

A number of orthogonal methods have been used for the analysis of subvisible particles in biopharmaceuticals. It is not unusual for these methods, each with their own calibration and operational protocols, to produce seemingly conflicting results. To reconcile results obtained through different measurement methods we have performed simultaneous measurements using various pairs of orthogonal methods, for example Flow Imaging and Brownian Motion Analysis. These results indicate correction factors which can improve accuracy and provide better correlation between methods.

11:45 Close of Conference