2017 Host Cell Protein Conference
4th Annual | Sam Francisco, CA | May 10-12, 2017
2017 HCP Speaker Abstracts
Svetlana Bergelson, Biogen
Roundtable Discussion Panel
Xuezhi Bi, Bioprocessing Technology Institute, A*STAR
Title: An Effective SWATH-MS Workflow for the Analysis of CHO HCPs
Abstract: Residual host cell proteins (HCPs) impurity is one of the critical quality attributes (CQAs) in CHO-produced biotherapeutic monoclonal antibodies (mAbs). In order to monitor individual CHO HCP clearance, a sensitive, robust and effective SWATH-MS workflow was developed to identify and quantify these impurities. Using this workflow on premixed control samples comprising known ratios of pure mAb to residual HCPs, a total of 2413 proteins were identified with up to 60% of the proteins accurately quantified within ±20% of theoretical values. We next applied the workflow on downstream purification intermediates of a mAb product; we were able to identify 1300 and 1465 HCPs across the process intermediates in two independent runs, and to show significant reduction of 600 HCPs after Protein A affinity purification and two polishing steps. Subsequent spiking of the samples with known quantities of standard proteins enabled the absolute quantification of individual HCPs and resulted in improved detection sensitivity down to 1 fmol with a 1 µg total protein load – a five-fold higher sensitivity than the previously reported detection limit. To date, we have generated a comprehensive in-house CHO-K1 spectral ion library covering more than 8200 proteins (identified by ≥ 2 unique peptides at 1% FDR). Using a high performance streamlined software pipeline, we will demonstrate that the SWATH-MS workflow can be used for high-throughput and robust monitoring and quantification of HCPs during downstream purification processes of mAb production; the workflow also provides improved detection sensitivity and quantification accuracy of low-level HCPs.
Jonas Borch-Jensen, Novo Nordisk
Title: Performance of HCP LC-MS/MS scaled from nanoflow to standard LC flow using a Thermo Q Exactive Plus MS
Abstract: Subject areas
LC methods
Scaling of LC methods
Thermo MS platform
Tobey Gooding, Covance
Title: Anti-HCP Antibody Reagents and Assay Development From a CRO Perspective: Industry Wide Data, Trends and Recommendations
Abstract: As a CRO, Covance serves clients in every facet of the scientific community. This experience gives us broad exposure to the practices and approaches of the Biopharmaceutical Industry to meet regulatory requirements to consistently remove Host Cell Proteins (HCP) during production of bioproduct drug substances. Ongoing client requests for work suggest a continued migration in the industry toward process-specific/platform assays and away from commercially available kits. When developing immunoassays, the quality of the finished product begins with the ability to generate antibodies with high titer, sufficient quantities to span development through samples testing and broad immunoreactivty. A number of considerations must be made to ensure the overall quality and suitability of the antibodies for their intended purpose, ranging from species selection and prescreening activities through carefully tailored immunization schedules and purification strategies. Other considerations involve the development of a robust ELISA. Assay development activities should include thorough evaluation of the performance of the antibodies across the range of sample conditions expected to be observed throughout the purification process of the product material. This presentation will discuss strategies employed within Covance Laboratories toward producing a de novo HCP ELISA, focusing on factors that can produce a final assay format that is suitable and appropriate for its intended use.
Contributing Authors:
Tobey Gooding, James Mellon, Covance Inc.
Feny Gunawan, Genentech
Title: Overcoming the Challenges of Updating an HCP Control Strategy in a Marketed Product
Abstract: In the lifecycle of a commercial product, updates to the manufacturing processes and control systems are to be expected. In the case of process related impurities, such as Host Cell Proteins (HCP), several unique challenges need to be considered when updating their control strategy. For any process change, the HCP should remain within the currently approved levels. However, as analytical technologies improved we are able to detect impurities at much lower levels and/or identify those that are previously unknown. This presentation will focus on a case study in which we updated the HCP control system of a marketed product, 20 years post approval, with modern analytical technologies and current regulatory expectations. In the midst of this update, a new co-purifying HCP — that was not detectable in the approved process — was discovered. The strategies and considerations for overcoming these challenges will be discussed.
Contributing Authors:
Feny Gunawan1, Denise Krawitz3, Julie Nishihara2, Marty Vanderlaan3, and Heidi Zhang1
1 Analytical Operations, Genentech
2 Protein Analytical Chemistry, Genentech
3 Independent
Lea Hagigi, Biogen
Title: CHO Strain vs. Culture Process: Comparing HCP Reagent Made From Different CHO Lines On Measurement of HCPs in the Same Products
Abstract: (Pending)
Markus Haindl, Biogen
Title: A holistic phase appropriate strategy for HCP-limits and HCP
Abstract: (Pending)
Xian Huang, Bristol-Myers Squibb
Title: An LC-MS/MS Workflow for Identification and Quantification of E. Coli Residual Host Cell Proteins to Support Biotherapeutic Purification Process Development
Abstract: Introduction: In recent years, LC-MS/MS has been increasingly employed as a complementary approach to ELISA for tracking residual host cell proteins (HCPs) since it can provide both identities and quantities for individual HCPs detected. However, when 2-D LC and/or nanoflow/capillary column as well as isotopic-labeling sample preparation are incorporated, the MS-based approach has been limited by the method complexity and reliability. In this study, a simple and robust label-free quantitative 1-D LC-MS/MS workflow was developed and applied to a purification process for a 12 kDa protein produced by periplasmic expression in E. coli. Because the periplasmic expression utilizes whole cell homogenization, high levels of HCPs must be cleared by the downstream process. In particular, several HCPs were visualized by SDS-PAGE that were not recognized by ELISA antibodies.
Methods: In-process pools and final drug substance with known drug concentrations from different downstream processes were digested with trypsin and spiked with an internal standard (several equal-molar tryptic peptides of a non-microbial protein). LC-MS/MS was carried out using a Waters Acquity UPLC interfaced with a Thermo Q Exactive™ Plus Hybrid Quadrupole-Orbitrap™ mass spectrometer through an electrospray ionization source. The LC separation of peptides was performed on an Acquity UPLC C18 column with gradient elution followed by ESI-MS/MS data-dependent acquisition. The raw data were analyzed using Proteome Discoverer with filtering settings for E. Coli protein database search with high confidence. An identified protein quantity is determined by protein peak area on the basis of its top 3 unique peptides and the use of an internal standard to enable quantification.
Preliminary Results: A simple and reliable 1-D LC-MS/MS workflow was developed for label-free quantification of HCPs. The analyses gave identities and quantities of individual HCPs detected in samples. Residual E. Coli HCPs at a level of as low as 30 ng/mg can be identified and quantified by this workflow. The method was applied in tracking residual E. Coli HCPs in the in-process pools from sequential purification steps and final drug substance. This allowed monitoring of clearance of the high abundance HCPs not detected by ELISA. It furthermore allowed identification of HCPs present in drug substance at levels suggesting lack of detection by ELISA, and will enable future process development aimed at their removal and ELISA development for better routine monitoring of residual HCPs.
Novel Aspects: Application of a simple and reliable LC-MS/MS workflow for HCP identification and quantification in support of bioprocess development and troubleshooting.)
Contributing Authors:
Xian Huang, Jeffrey Meyer, Derek Hackathorn, and Karen De Jongh
Bristol-Myers Squibb, Seattle, WA 98102, USA
Eric B. Johansen, AbbVie
Title: Quantitative Mass Spectrometry Strategies for Detection, Relative and Absolute Quantitation of Host Cell Proteins. Data dependent, Independent and targeted analyses
Abstract: Quantitative mass spectrometry strategies are rapidly developing greater sensitivity and selectivity. Here we will discuss workflows and best practices from detection and relative abundance to absolute quantitation by LCMS as they relate to host cell proteins and peptides.
Thomas Kofoed, Alphalyse
Title: Identification and Quantification of HCP’s by SWATH LC MS for Process Development of a Novel Biologic
Abstract: HCP analysis by ELISA for novel biologics under development is difficult because no platform or product specific ELISA has been developed yet. For biologics expressed in E.coli inclusion bodies it may not even be possible to develop a product specific ELISA because immunization reagents also contain the drug substance.
Here we present a HCP analysis strategy and analysis results for a novel biologic, where a platform ELISA is not available, using generic ELISA methods in combination with SWATH LC MS/MS for identification and quantification of individual HCP’s.
The generic ELISA assays are based on off-the-shelf kits from different vendors. The SWATH LC MS/MS method is based on a 1D microflow LC method in combination with SWATH data independent acquisition on a Sciex trippleTOF 6600 mass spectrometer. The microflow 1-dimensional chromatography enables very fast and robust chromatography compared to multi-dimensional nanoflow chromatography. The SWATH independent data acquisition enables very reproducible HCP identification and quantification compared to data dependent workflows.
Contributing Authors:
Thomas Kofoed, Marie Grimstrup, Janne Crawford, Jakob Bunkeborg, Ejvind Mørtz, Alphalyse A/S, Denmark.
Lars Skriver, René Egebro, Savara Pharmaceuticals Austin, TX, USA.
Denise Krawitz, Consultant
Workshop 2 Instructor
Denise Krawitz, Consultant
Roundtable Discussion Panel
Kelvin Lee, University of Delaware
Title: The 2017 Chinese Hamster Ovary Reference Genome
Abstract: (Pending)
Ingo Lindner, Roche
Title: Case Study for a Validation of a MS Based HCP Quantification Method – Requirements and Limitation
Abstract: (Pending)
John Matthews, Genentech
Title: Case Study: Clinical perspectives on Phospholipase B-Like 2 Protein, a Host Cell Impurity in Lebrikizumab Clinical Material
Abstract: (Pending)
Laura McIntosh, Caprion Biosciences
Title: Lessons Learned: HCP Analysis of a Variety of Biologics and Biosimilars using LC-MS/MS
Abstract: Gel-free, label-free mass spectrometry (MS) is an orthogonal method to ELISA, and has an invaluable role in characterization of HCPs. We have established robust workflows for the characterization, monitoring and quantitation of low level HCPs by LC-MS/MS.
For identification and relative quantification of HCPs, samples are subjected to trypsin digestion, fractionation to reduce the complexity of the tryptic peptide mix due to the presence of the DS in very high amounts, and high mass accuracy shotgun RPLC-MS/MS. PQCs (Process Quality Controls) are analyzed alongside the samples as a measure of the instrument and analytical procedure performance. HCPs are identified by using an unbiased bioinformatics workflow with multiple search engines to match detected peptides against a custom process-specific HCP database (i.e., CHO, human, mouse, yeast and bacteria), and spectral counts provide a qualitative assessment of relative abundance of each HCP. The sensitivity of the method depends on the particular characteristics of the DS, but has been shown to reproducibly identify spiked proteins in the presence of drug substance between 1-10 ppm.
For absolute quantitation of HCPs, a multiplexed targeted LC-MRM/MS assay is developed using stable-isotope labelled peptides representing the specific HCPs to be monitored. QC samples and calibration curves are run alongside study samples, and stable-isotope labelled peptides are spiked into all samples as internal standards. The sensitivity of detection is routinely achieved in the 1 ppm range.
Caprion now has experience with the analysis of numerous different novel biologics and biosimilar products (monoclonal antibodies, recombinant proteins, vaccines, peptide products). The goals and results from each study are always slightly different which has enhanced our overall understanding of HCP analysis by mass spectrometry, particularly with regards to analytical design (i.e., appropriate controls to include) and bioinformatics / data interpretation (database challenges and solutions). These learnings will be highlighted along with results with regards to sensitivity of HCP detection, types of HCPs identified, and overlap in different products.
Contributing Authors:
Laura McIntosh, John Babetas, Laetitia Cortes, Lorella Di Donato, Rudolf Guilbaud, Stéphane Parent and Michael Schirm, Caprion Biosciences, Montréal, Canada
Emily Menesale, Biogen
Title: Case Study: HCP Antigen Stability
Abstract: (Pending)
Ned Mozier, Pfizer
Roundtable Discussion Panel
John Rolf, Amgen
Title: HCP Control Strategy Reassessment for a CHO-derived Protein Therapeutic
Abstract: (Pending)
Oliver Schirm, Caprion Biosciences Inc.
Title: Data Analysis, Quantitation and Reporting for Mass Spectrometry-Based HCP Studies
Abstract: The high specificity and sensitivity of mass spectrometry enables high HCP coverage, providing identification and quantitative information on individual HCPs. Non-targeted LC-MS/MS based workflows and targeted LC-MRM/MS workflows have been developed which routinely achieve sensitivity in the 1-10ppm range for both HCP identification and absolute quantitation. Case studies will be shown which illustrate how data analysis, quantitation and reporting can be performed for various HCP applications, including, for example: characterization of in-process samples and DS, comparability studies of Biosimilars to Innovators and absolute quantification of HCPs. For non-targeted data analysis, creation of appropriate databases for CHO, E.coli and human production systems will be covered, in addition to selection of false discovery rates, usage of different search engines and spectral counting vs. intensity based measurements. For quantitative data analysis, an automated adaptation of Skyline will be presented.
Contributing Authors:
Oliver Gingras, Laetitia Cortes, Laura McIntosh, Michael Schirm
Lars Skriver, Savara Pharmaceuticals
Title: Identification and Quantification of HCP’s by SWATH LC MS for Process Development of a Novel Biologic
Abstract: HCP analysis by ELISA for novel biologics under development is difficult because no platform or product specific ELISA has been developed yet. For biologics expressed in E.coli inclusion bodies it may not even be possible to develop a product specific ELISA because immunization reagents also contain the drug substance.
Here we present a HCP analysis strategy and analysis results for a novel biologic, where a platform ELISA is not available, using generic ELISA methods in combination with SWATH LC MS/MS for identification and quantification of individual HCP’s.
The generic ELISA assays are based on off-the-shelf kits from different vendors. The SWATH LC MS/MS method is based on a 1D microflow LC method in combination with SWATH data independent acquisition on a Sciex trippleTOF 6600 mass spectrometer. The microflow 1-dimensional chromatography enables very fast and robust chromatography compared to multi-dimensional nanoflow chromatography. The SWATH independent data acquisition enables very reproducible HCP identification and quantification compared to data dependent workflows.
Contributing Authors:
Thomas Kofoed, Marie Grimstrup, Janne Crawford, Jakob Bunkeborg, Ejvind Mørtz, Alphalyse A/S, Denmark.
Lars Skriver, René Egebro, Savara Pharmaceuticals Austin, TX, USA.
Mark Smales, University of Kent
Title: Monitoring & Identifying Critical HCPs During CHO Cell Bioprocessing: Do They Exist & Why We Shouldn’t Have to Reclibrate HCP Assays Between Projects
Abstract: (Pending)
Martha Stapels, Sanofi Genzyme
Title: Increased Throughput and Accuracy in Host Cell Protein Quantitation using Spectral Library Searches
Abstract: (Pending)
Kevin Van Cott, University of Nebraska
Workshop 1 Instructor
Thomas Waerner, Boehringer Ingelheim
Host Cell Proteins Seen as Critical Quality Attributes
Abstract: (Pending)
Thomas Waerner, Boehringer Ingelheim
Roundtable Discussion Panel
Fengqiang Wang, Merck
Title: Knowledge Based HCP Risk Control Strategy During Downstream Process Development with Recent Case Studies
Abstract: (Pending)
Michael Wiedman, Boehringer Ingelheim
Roundtable Discussion Panela
Boyan Zhang, Mab-Works
Title: Development of a Fast LC-MS/SWATH Method for Real-Time Identification and Quantitation of the Residual HCPs During the Manufacturing Process of a Monoclonal Antibody Therapeutic
Abstract: Subject areas
DIA – SWATH
Quantitation
Case-study, real system
Ying Zhang, Pfizer
Title: Quantitative Investigation of HCP Impurities: Bridging the Gap Between ELISA & Orthogonal LC-MS/MS Analysis
Abstract: (Pending)
2017 HCP Posters
David Chimento, Rockland Immunochemicals
Title: Enhanced HCP Coverage Analysis utilizing Multiplexed 2D Electrophoresis
Abstract: A crucial prerequisite in the drug manufacturing process is an efficient analysis of Host Cell Protein (HCP) impurities that result from process specific expression conditions as well as downstream purification procedures. Present guidelines call for minimum levels of HCP contaminants that may be left behind during the purification process from the expression hosts. To investigate the presence of residual contamination of the final biopharmaceutical product, the development of polyclonal antibodies with maximum coverage against naive HCP lysate provides a valuable tool to demonstrate product purity.
Quality of the HCP specific antibody is most commonly analyzed via ELISA and orthogonal methods like 1D and 2D-PAGE assays where the coverage of the antibody is confirmed. Determining anti-HCP antibody coverage is essential to ascertaining the robustness of the antibody in comprehensive reactivity toward HCPs during process points.
Here we evaluate the coverage of a generic anti-CHO-HCP antibody by 2D-PAGE separation using two experimental methods for the analysis: (A) chemiluminescent based detection method; and (B) two-dimensional differential in blot electrophoresis (2D-DIBE). For (A): the proteins detected by the antibody on a membrane (WB) via a chemiluminescent detection system, are equated to the proteins separated on a separate SDS-gel and visualized by an in-gel protein stain. For (B): 2D-DIBE utilizes a fluorescent multiplexing methodology, where the CyDye™ pre-labeled and 2D separated proteins transferred to the membrane can directly be visualized and compared to the proteins detected by CyDye™ pre-labeled HCP-antibodies on the same membrane.
Sensitivity and antibody reactivity of the two methods is compared and analyzed.
Contributing Authors:
M. Sayeed1, P. Beckett2, S. Voordijk3, J. Hirano4, Y. Zang1, D. Chimento1, and K. Abarca Heidemannen1
1Rockland Immunochemicals Inc., Limerick, PA 19646; 2GE Healthcare Life Sciences, Marlborough, MA 01752; 3Geneva Bioinformatics (GeneBio) SA, Geneva, Switzerland; 4GE Healthcare Life Sciences, Uppsala, Sweden
Catalin Doneanu, Waters Corporation
Title: A Novel Data Acquisition Mode for Identification and Quantification of Low-Abundance Host Cell Proteins in Biopharmaceuticals
Abstract: Residual host cell proteins (HCPs) are a major component of biopharmaceutical process related impurities. They are trace level contaminants (1-100 ppm) sometimes present in highly purified biopharmaceutical proteins, but they can potentially elicit an unpredictable immune response in patients or they can affect the drug activity or stability. For these reasons, there are regulatory requirements for HCP identification and quantification.
Here we describe an LC/MS assay able to identify and quantify HCPs from high-purity monoclonal antibodies (mAbs). The assay relies on a novel data independent acquisition mode recently implemented on a quadrupole/time-of-flight (QTOF) mass spectrometer, namely SONAR data acquisition. Instead of transmitting all peptide ions produced by the electrospray source, in SONAR mode the quadrupole slides over the mass range of interest during the time required for recording a single MS spectrum by the TOF analyzer. The precursor ions produced by co-eluting peptides are separated by the quadrupole and their MS/MS fragmentation spectra are recorded individually. In this way SONAR offers additional selectivity, by producing clean, interference-free fragmentation spectra.
The utility of the LC/MS assay was demonstrated using a high-purity mAb produced in a murine cell line (NIST mAb, 100 mg/mL). Following denaturation, reduction, alkylation and trypsin digestion, the mAb sample was spiked with four protein digest standards in the range of 1-250 ppm. The peptide mixture was separated on a 2.1 mm ID column specifically designed for chromatography with high sample loads (50-500 µg mAb digest loaded on-column). Highresolution (> 30,000) data-independent acquisition of peptide precursors and fragments in SONAR mode was performed on a QTOF instrument. Spiked proteins as well as five HCPs were identified and quantified in a single injection with an LLOQ of the assay of 10 ppm.
Contributing Authors:
Catalin Doneanu1, Sarah Lennon2, Malcolm Anderson2, Ying Qing Yu1, Laetitia Denbigh2, and Weibin Chen1
1 Waters Corporation, Milford, MA, USA
2 Waters Corporation, Wilmslow, UK
Jeffrey L. Turner, Millipore Sigma
Title: Characterization and Quantification of Residual Host Cell Proteins in CHO Using a Stable Isotope Labeled Standard
Abstract: Recent advances in mass spectrometry instrumentation have allowed for tremendous improvements in the ability to detect large numbers of proteins and to quantify proteins across broad dynamic ranges. However, the adaptation and extension of Host Cell Protein (HCP) analysis onto LCMS platforms has been hindered by 1) complexities in the large-scale quantification of contaminant proteins and 2) the lack of historical reference sets and data for assessment of product purity. To help overcome these limitations, we have developed and characterized a stable isotope labeled (SIL) protein extract from Chinese Hamster Ovary (CHO) for use as an internal standard during LCMS analyses. The protein standard was produced from CHO-K1 cells cultured in serum-free medium which had been enriched with 13C615N4-Arg and 13C615N2-Lys amino acids. Standard proteomic analysis of the secreted protein fraction from this sample on an Orbitrap QE+ demonstrated over 2300 proteins from the CHO cell proteome (cricetulus griseus). The breadth of proteins obtained and high levels of isotopic incorporation (>98%) were critical in being able to quantifiably assess a broad dynamic range of host cell proteins within a variety of sample inputs. The resultant mixture has demonstrated utility for the development and optimization of downstream process development, as a reference set for historical comparison of HCP variability across production runs, and the LC-MS characterization and quantification of residual proteins within commercial CHO-derived samples.
Contributing Authors:
Jeffrey L. Turner, Pegah R. Jalili, Zhiyun Cao, and Kevin Ray; Millipore Sigma, St. Louis, MO
Sonja Voordijk, Geneva Bioinformatics (GeneBio)
Title: Robust HCP Coverage Analysis with Dedicated Melanie Software
Abstract: HCP impurities must be carefully identified, minimized and monitored throughout development and manufacturing of biopharmaceutical products to guarantee patient safety and drug efficacy. While HCP ELISA is a critical component of HCP contaminant detection, regulatory agencies require demonstration that the polyclonal antibody mixture used in the ELISA is broadly reactive against a wide range of potential HCPs.
During the development and validation of the antibodies, 2D gel electrophoresis followed by Western blotting is a standard approach to determine coverage of the HCP specific antibody, i.e. the percentage of immunodetection the antibody offers for the total population of HCP. Yet, vastly different spot patterns often seen between independent gels and blots, as well as subjective and time-consuming image analysis with unsuitable tools have represented serious challenges to coverage assessment.
Here we show how the application of 2D Differential in Blot Electrophoresis (2D-DIBE) for the characterization of an anti-CHO cell antibody removes the need for subjective and laborious blot-to-gel matching. Image analysis with the dedicated Melanie Coverage software further reduces subjectivity and analysis time. This is especially achieved by providing interactive 3D views at all steps of the simple workflow, providing valuable context wherever user decisions are required. A three-level spot filter introduces the notion of uncertain spots, in addition to spots considered absent or present on a given image. Remaining low number of uncertain spots can be reviewed easily and quickly, while the coverage is instantaneously updated.
Analysis of 2D-DIBE data with Melanie therefore provides reliable, reproducible, unbiased and fast coverage results to validate antibody reagents for their intended use.
Contributing Authors:
S. Voordijk1, M. Sayeed2, G. Bouchet3, S. Catherinet3, D. Walther3, P. Beckett4, I. Xenarios3, J. Hirano4, K. Abarca Heidemannen2
1Geneva Bioinformatics (GeneBio) SA, Geneva, Switzerland; 2Rockland Immunochemicals Inc., Limerick, PA 19646, USA; 3SIB Swiss Institute of Bioinformatics, Geneva, Switzerland; 4GE Healthcare Life Sciences, Uppsala, Sweden