Biologics: The Unique Bioanalytical Program Requirements
Paula M. Jardieu, Ph.D l Senior Vice President, ICON Development Solutions
The biopharmaceutical industry has dramatically expanded the number of recombinant proteins marketed as therapeutics, and biologics comprise one of the fastest growing therapeutic categories.
The drug development paradigm for small molecules cannot be directly applied to large molecules. All too often, attempts are made to develop biotherapeutics as âbig drugsâ without fully shifting the development approach to account for their intricate physicochemical properties. This shift in paradigm and corresponding methodology is also required for the analytical assays utilized to characterize these macromolecules.
The majority of new biotechnology derived therapeutics are recombinant proteins which include monoclonal antibodies, cytokines and growth factors. Biologics differ significantly from traditional drugs since they target specific components of a disease state instead of broadly affecting biological pathways. For example, biotherapeutics for the treatment of rheumatoid arthritis target very precise disease producing molecules such as the pro-inflammatory cytokine Tumor Necrosis Factor (TNF) as opposed to traditional drugs such as steroids, which affect multiple biological pathways.
The biopharmaceutical industry has dramatically expanded the number of recombinant proteins marketed as therapeutics, and biologics comprise one of the fastest growing therapeutic categories. In 2009, sales are estimated to reach $90 billion, up from $80.5 billon in 2008. In addition, an estimated $10 billion worth of biologics are expected to come off patent by 2010. The increased number of biological agents used as therapeutics and the promise of biosimilars has resulted in an increase in the number of CROs offering analytical support for the quantification and characterization of these therapeutics. Well designed pre-clinical and clinical bioanalytical programs for biologics not only include the typical assays for characterization of the pharmacokinetics and pharmacodynamics (biomarkers) of the molecules, but also include assays to detect and characterize their immunogenicity profile.
Immunogenicity is the immune response against foreign objects which can include recombinant human proteins. Even with identical or nearly identical sequences to native human proteins, the majority of therapeutic biologics, especially after chronic administration, elicit an immune response in preclinical animal species and often in humans. This is an inherent property of recombinant human proteins which are manufactured in non-human host cell lines, since these non-human hosts can produce a protein with altered primary or secondary structure compared to the protein produced in human cells. Antibodies produced by this immune response can result in alterations in the pharmacokinetic profile of the recombinant human protein due to its enhanced clearance from the body. This alteration may affect the interpretation of pre-clinical toxicity data. In human studies, patients may develop antibodies that neutralize the biological activity of the therapeutic or increase its clearance, which may lead to decreased therapeutic levels and a loss of efficacy. In addition to a decrease in therapeutic levels, relationships have been noted between antibody formation and adverse effects, such as infusion reactions and serum sickness- like clinical reactions. Typically, immunogenicity is characterized by using assays to detect Anti-Drug Antibodies (ADA) against the protein therapeutic. For these reasons, ADA assays go hand-in-hand with pharmacokinetic and biomarker assays for the characterization of biologics.
Pharmacokinetic Assays
Pharmacokinetics impacts all stages of drug development since appropriate pharmacokinetic properties of a biologic can lead to improved efficacy and safety. Pre-clinical pharmacokinetic studies are critical for planning and interpreting toxicological studies and are therefore essential for human risk assessment. Additionally, preclinical PK parameters are used for extrapolation to safe and effective human doses via interspecies scaling.
Adequate bioanalytical methods are crucial to producing valid pharmacokinetic data. The methods that are primarily used in these evaluations are Ligand Binding Assays (LBAs). Unlike small-molecule drugs, there is no physicochemical property of a macromolecule that can be used for detection. Instead, the specificity and selectivity of the assay depends on the interactions of antibodies against the therapeutic candidates. The response observed in these methods (color change, fluorescence or chemiluminescence) is proportional to the concentration of the therapeutic.
This is the basis for the assay format known as Enzyme-Linked ImmunoSorbent Assay (ELISA), the cornerstone for detection of biologics. In a âsandwichâ ELISA, a solid support (usually a polystyrene microtiter plate) is coated via adsorption with an antibody to the biologic. A patientâs sample containing an unknown amount of the biotherapeutic is added and is specifically captured by the antibody. The plate is washed with a mild detergent solution to remove any proteins that are not specifically bound (i.e. separation of bound from free). A second, so called, detection antibody is added which is covalently linked to an enzyme through bioconjugation. After the final wash step, the plate is developed by adding the enzymatic substrate to produce the signal. This signal is proportional to the amount of biologic present in the sample. In the case of a fluorescence ELISA, light of the appropriate âexcitationâ wavelength is applied to the sample, the antibody complexes fluorescence and the amount of biologic in the sample can be inferred from the intensity of the fluorescence. In the case of an absorbance assay, chromogenic substrates are used and for chemiluminescent assays, chemiluminescent substrates are employed. An illustration of a chromogenic ELISA is shown in Figure 1.
To insure that an LBA used in support of pharmacokinetic studies can withstand regulatory scrutiny, the assay is validated and continually monitored throughout its use by random reanalysis of incurred samples. Typical parameters assessed during validation are included in Table 1. These parameters are used throughout the industry to insure assay robustness.
In addition to method validation, incurred sample reanalysis (ISR) has become a requirement for bioanalysis to demonstrate method reproducibility. Typically 5-10% of in vivo derived study samples are reanalyzed at random from peak and trough levels of the PK curve. Agreement between the initial and repeat analysis must be within defined acceptance criteria. If reanalysis fails, a failure investigation is undertaken to identify the source of the failure. Reproducibility of incurred sample reanalysis is important to prove assay validity.
Anti-Drug Assays (ADA)
There are many factors that play a role in the development of an immune response to a therapeutic protein. There are factors like the route and dose, with subcutaneous administration eliciting a higher incidence of antibodies than intravenous administration. Factors contributing to immunogenicity can also be related to manufacturing and formulation (e.g. aggregates) of the molecule which can make it more immunogenic than soluble proteins. If the molecule is used as replacement therapy for an endogenous molecule, these antibodies to the recombinant protein can cross-react with the endogenous molecule. For instance, rare cases of unexplained severe anemia and resistance to recombinant human erythropoietin (rHuEPO) in patients treated with (rHuEPO) have been attributed to the development of anti-rHuEPO antibodies which cross-react with the endogenous erythropoietin molecule.
More and more antibodies are used as therapeutic agents. Examples
of therapeutic antibodies are TNF-inhibitors like infliximab
(Remicade®) and adalimumab (Humira®) used for rheumatoid
arthritis (RA), and antibodies directed against the CD20-antigen,
rituximab (Mabthera®) or the EGFâR, cetuximab (Erbitux®) used for
cancer. The first generation of therapeutic antibodies was of murine
origin and was associated with high incidence of anti-drug
responses. The next generation of chimeric antibodies (i.e. mouse/
human) had lower incidence of antibody formation. Technology
now exists to generate fully human antibodies, but these still
maintain the potential to be immunogenic. Immunogenicity of
leading biotherapeutics for the treatment of RA is shown in Table 2.
An immune response to a product does not mean the drug cannot be developed. The consequences of an immune reaction to a therapeutic protein range from transient appearance of antibodies without any clinical consequences to severe life threatening conditions. For these reasons, detection and management of immunogenicity is an essential part of the development and post marketing of biologics to minimize risk to the patient.
Assay formats used to measure anti-drug antibodies include ELISA and more recently ECLA (ElectroChemiLuminescence Assay). Rather than metabolism of substrates to generate a signal as in ELISA, the ECLA platform relies on signals from electrochemiluminescent labels on the detecting antibody, which emits light when electrochemically stimulated.
Development of assays to measure and characterize antibodies formed against a therapeutic protein is not without technical challenges. First, there are rarely appropriate positive controls. In most cases, it is impossible to find human immune serum specific for a particular human protein to use to evaluate assay performance. Therefore, development and validation of anti-drug assays are based on monoclonal or polyclonal antibodies prepared in nonhuman species which serve as the positive control. Unlike PK assays, which use the biologic as the reference standard, anti-drug assays use these surrogate antibodies. For this reason, no quantitation of the response is possible, since the immune response of a patient to the biologic will be very different than the response of a rabbit or monkey. Instead, ADA assays measure the signal above a negative response derived statistically from samples from patients with no known exposure to the biologic. This signal is termed the cutpoint of the assay and is used to distinguish samples that are positive for anti-drug antibodies from those that are negative. A confirmatory test (based on the same assay format) is then performed to provide evidence that the signal above the cutpoint is due to a specific anti-drug antibody. These assays involve testing a sample in the presence or absence of excess drug. A decrease in assay signal in the presence of drug is taken to indicate the presence of anti-drug antibodies and the sample is confirmed positive. Because no quantification is possible, the means for expressing the concentration of a specific antibody varies substantially. Most often, the amount of ADA is expressed as a titer value. However, a titer value for any one sample can vary based on the assay format and reagents. The stratified approach used to characterize anti-drug responses is described in Figure 2.
As is the case with PK assays, ADA assays require validation to insure reproducibility. Validation parameters for ADA assays are outlined in Table 3. These parameters are based on the industry best practice recommendations.
PD/Biomarkers
There is a growing interest in biomarkers, previously referred to as pharmacodynamic markers. Biomarkers of disease play an important role in medicine and have begun to assume a greater role in drug discovery and development. This increasing interest has been largely driven by the focus on reaching early go/no-go decisions for compounds entering early clinical development. Pharmacodynamic or biomarker assays which measure pharmacologic responses to therapeutic intervention are an effective strategy for improving the selection of suitable drug candidates.
There are two basic approaches to the identification and selection of biomarkers. The traditional method is based on knowledge of the mechanism of the disease state. However, newer omics-based approaches such as genomics, transcriptomics, proteomics, and metabolomics, combined with pattern recognition statistics, are also being used to identify novel biomarkers. Optimum biomarker development and application will require a team approach because of the nature of biomarker selection, validation and application.
Biomarkers run the gamut from medical imaging to physiological responses to in vitro assays for plasma proteins putatively involved in disease mechanisms. Protein biomarker assays often employ ELISA or ECLA, the same platforms used for PK and ADA assays. Validation of these assays is less straightforward since they measure the response of an endogenous molecule to treatment with the biologic and reliable reference standards often do not exist. Well known protein biomarkers used in various disease states include hs-CRP (cardiac disease), PSA (prostate cancer), Abeta (Alzheimerâs) and neopterin (infectious disease). These analyses are used in conjunction with clinical outcomes to validate the putative marker as an indicator of safety or efficacy of the biologic. The ultimate goal is to use panels of biomarkers as validated surrogate end points which can substitute for conventional physiological measurements often requiring frequent hospital visits and increased expenditures.
Successful applications of biomarkers will need to be effectively communicated with all of the stakeholders including regulatory agencies to promote the acceptance of biomarkers in drug development. In the future, pharmaceutical companies will not only develop drugs, they also will develop the biomarker kits that will be part of using the drug. This is already true for Herceptin for the treatment of breast cancer. Before patients can be dosed with the biologic, a diagnostic test for Her-2, the target of the therapeutic is required. This may increase the complexity of drug development but it will improve patient outcome.
Conclusion
Biologics are not just âbig drugsâ. Based on their intricate physicochemical properties compared to small molecules, they demonstrate complex pharmacokinetic and immunogenicity profiles which strongly influence pre-clinical and clinical testing strategy. The development of biologics requires a well designed pre-clinical and clinical bioanalytical program to help address these issues. The high attrition rates in the development process can be significantly reduced by employment of suitable PK, biomarker and ADA assays to guide decision making and appropriately balance risks and benefits.