With R&D returns for large cap biopharma companies falling to the lowest level in nine years (Deloitte, 1), the escalating cost of bringing medicines to market means that the biopharma R&D process must be examined in detail. One important factor is the development and validation of immunoassays that can deliver high quality data efficiently and in a timely manner. Speed and quality are intertwined, and while manual ELISA has been a major workhorse in biopharma R&D, automation is a key factor in speeding up the generation of high quality data with minimum hands-on time, minimum need for re-runs, and using readily validated assays that meet regulatory requirements.
Addressing the challenges of immunoassay development in a high pressure environment
A recent survey on the use of Ligand Binding Assays (LBA) that we carried out highlighted the interest in running immunoassays, mainly for pharmacokinetics (PK)/pharmacodynamics (PD) studies of biotherapeutics (73% of respondents) or biomarkers/targets (69%), and safety assessment (47%). Matrix interference (72%), inconsistent method and reagent performance (52%), data quality (sensitivity, 46%; specificity, 37%; precision, 26%), together with time-consuming method development (63%) were named as major challenges or limitations, and it is these aspects we will be looking at here.
Ensuring your immunoassay is fit-for-purpose
A successful bioanalytical strategy is essential in driving efficient drug development by helping to establish drug target engagement, determining drug dosage, providing early indications of drug safety, and may also help in preparing the clinical phase of trials e.g. patient selection. To achieve this means developing fit-for-purpose assays with performance that meets the minimal acceptance criteria to support the intended purpose of the data (2).
High quality data builds confidence in decision-making
The quality needed to achieve fit-for-purpose can be multi-facetted. Measurements of analytes need to be close to the actual level (accuracy) in a consistent manner (precision). Data points also need to represent the level of the correct analyte in a background of other analytes (selectivity), and the assay needs to be able to detect the analyte at a low level (sensitivity). Not only that, the assay needs to deliver high quality data, time and time again (robustness), even with small variations in method parameters, such as the presence of varying amounts of interfering substances from complex matrices such as serum. Being able to use a low minimum required dilution (MRD) to avoid matrix effects, and a broad analytical range will minimize the need to dilute samples, will increase assay performance and also save time.
Data quality starts with the antibody reagents
The FDA guidance for Bioanalytical Method Validation indicates that the accuracy and precision of ligand binding assays used within- and between runs should be ± 20% of nominal concentrations, except ±25% at lower limit of quantitation (LLOQ) and upper limit of quantitation (ULOQ) (3). Corresponding targets according to EMA guidelines are 15% and 20% (4). Generating this level of data quality starts with choosing high quality assay components.
Perhaps the most important factor is ensuring that high performance antibodies are used to capture and detect the analyte. These antibodies determine the specificity and selectivity of the assay, which are measures of whether an antibody binds solely to a unique epitope (high specificity) from a single antigen in a single species (high selectivity), or binds to similar epitopes on multiple molecules from different species. Cross-reactivity is the opposite of specificity. High affinity antibodies are the best choice for immunoassays due to their ability to quickly generate stable immune complexes that provide the most sensitive detection.
Ideally, the antibody reagents are incorporated into a ready-to-use kit that has already been quality-controlled by the manufacturer. Once high performance antibody reagents or kits have been identified it is also important to monitor lot-to-lot variation based on acceptance criteria determined in your own lab (5).
The need to minimize processing errors
Next comes the ability to minimize errors in the workflow, from avoiding pipetting errors to placement of samples and reagents in the correct wells of a multi-well plate. This means simplifying the workflow as much as possible, and minimizing manual operations that can cause error. Automation is a natural step in this process.
High quality data supports the emergence of immunoassay singlicates
Singlicate analysis is becoming more of a standard procedure for automated immunoassay analyzers used in clinical chemistry. Immunoassays, however, have historically been performed in duplicate to reduce the risk of error, but improvements in assay reagents as well as automation mean than singlicate analysis is now possible. For example, a group at Bristol-Myers Squibb used a singlicate analysis run on an automated immunoassay platform, backed up by tools to detect aberrant results, to study a biotherapeutic through three development stages: discovery (PK/TK), GLP-toxicology, both in cynomolgus monkey, and the clinical phase. They concluded that, “In the fast-paced world of drug development where ’speed to patient’ is key, we believe that this approach bears negligible risk and can significantly increase benefit to the physicians and patients”, (6).
Singlicate analysis in LBA has been supported by published data and presentations from conferences and workshops in the last few years (7), and many pharmaceutical companies are now submitting singlicate data for regulatory approval. There is clearly a paradigm shift towards the acceptance of singlicate analysis using robust analytical platforms, providing it is backed by appropriate validation.
Meeting deadlines relies on timely delivery of reliable data
The fit-for-purpose approach also means developing methods that support efficient and therefore cost-effective decision-making in drug development. This search for efficiency involves a number of factors involved in method development, validation and running assays routinely.
The most direct route to shortening assay development time is to use a ready-to-use kit that a manufacturer has verified, validated and run through quality control to guarantee high quality results. An alternative is a detailed description of a robust assay provided by the immunoassay vendor that reduces the need for method development. If this is not available, then a carefully chosen immunoassay platform that minimizes errors will enable rapid method development based on as few variables as possible, and deliver high quality data that readily supports Design of Experiments (DOE) to speed up assay development.
Once the assay has been developed and validated, a high throughput automated platform should give you the turnaround time you need and also free up a scientist’s time for other tasks such as data analysis. The high precision, accuracy, robustness and broad analytical range that is possible with automated systems will also minimize re-runs and smoothen data-driven decision-making. Added to that, a platform that can also run different assays in parallel will enable you to get all the data you need at once.
Clearing regulatory hurdles is key to getting to the market quickly
The 2013 FDA Draft Guidance for Bioanalytical Method Validation (3) was the first guidance to formally discuss fit-for-purpose bioanalytical methods for biomarkers. The draft guidance emphasized that laboratory should determine the level of method validation required unless the biomarker data are intended to support label claims. Similar information can be found from the EMA, which has released a guideline that defines ‘key elements necessary for the validation of bioanalytical methods. The guideline focuses on the validation of the bioanalytical methods generating quantitative concentration data used for pharmacokinetic and toxicokinetic parameter determinations. Guidance and criteria are given on the application of these validated methods in the routine analysis of study samples from animal and human studies’. (4).
Validation builds on insights during method development
Validation can be speeded up with a reliable immunoassay platform, based on data acquired during method optimization for a number of key parameters:
- Range of quantitation – Lower Limit Of Quantitation (LLOQ) and Upper Limit Of Quantitation (ULOQ)
- Dilutional Linearity – multiple dilutions of spiked samples
- Robustness – consistency when assay conditions are changed
- Ruggedness – consistency when routines are changed
There is a wealth of information on validation of immunoassays (8–11), but with so many different factors to consider, off-the-shelf kits are attractive alternatives to taking on the design and validation of immunoassays in-house.
Streamlined automated workflows pull it all together
Transferring a manual ELISA to an automated platform frequently results in considerable improvements in data quality and speed. For example, employing the streamlined workflow achieved by Gyrolab® systems can reduce the time needed to run an assay from 10h on a manual ELISA with 9 key steps to about 1h and only 3 key steps. Automation is also critical in improving robustness and making singlicates reliable, and smoothens the transfer of assays between pharma and CRO, which is a key factor in modern biopharma R&D.
The automation achieved in the flow-through structures of Gyrolab CD also enables the routine generation of data with high accuracy and precision. The system consistently delivers data with CVs of less than 20%, and often below 10% for a range of applications (6, 12, 13).
Choosing the right immunoassay platform is an important step in improving R&D productivity to bring safe and efficacious biotherapeutics onto the market as quickly as possible. Automation of streamlined workflows based on high performance assays will deliver the data needed to support pre-clinical and clinical studies, as well as support process development and manufacturing of both biotherapeutics and agents used in cell and gene therapy.
To find out more about robust immunoassays that can save you time and deliver high quality listen to the webinar on Biopharmaceutical drug development:from CMC to clinical PK
- Deloitte report: https://www2.deloitte.com/content/dam/Deloitte/uk/Documents/life-sciences-health-care/deloitte-uk-measuring-return-on-pharma-innovation-report-2018.pdf
- Editorial: Implementing fit-for-purpose biomarker assay approaches: a bioanalytical perspective. Cowan KJ, Bioanalysis 2016 8(12), 1221–1223
- Bioanalytical Method Validation Guidance for Industry (FDA)
- Guideline on bioanalytical method validation
- Analysis of Reagent Lot-to-Lot Comparability Tests in Five Immunoassay Items. Hyun Soo Kim, et al. Annals of Clinical & Laboratory Science, 2012 42 (2), 165. Pubmed
- Singlicate Ligand Binding Assay Using an Automated Microfluidic System: a Clinical Case Study. Jiang H, et al. AAPS J. 2017 Sep;19(5):1461-1468.
- Singlicate analysis: should this be the default for biomarker measurements using ligand-binding assays? Ye Z, et al. Bioanalysis. 2018 Jun 1;10(12):909-912.
- Recommendations for the bioanalytical method validation of ligand-binding assays to support pharmacokinetic assessments of macromolecules. DeSilva B et al. Pharm Res. 2003 Nov;20(11):1885-900.: Pubmed link
- Fit-for-purpose method development and validation for successful biomarker measurement. Lee JW et al Pharm Res. 2006 Feb;23(2):312-28. Epub 2006 Jan 12.: Pubmed link
- Poster: Validation of a ligand binding assay. Eckersten A. et al : Download
- A practical guide to immunoassay method validation. Andreasson U et al. Front Neurol. 2015; 6: 179. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4541289/pdf/fneur-06-00179.pdf
- A microfluidic approach to high throughput quantification of host cell protein impurities for bioprocess development. Heo, JH et al, Pharm. Bioprocess. (2014) 2(2), 129–139.
- Evaluation of Multiple Immunoassay Technology Platforms to Select the Anti-Drug Antibody Assay Exhibiting the Most Appropriate Drug and Target Tolerance. Collet-Brose, J et al., J. Immunology Research, 2016, Article ID 5069678, 15 pages