Essentials in Bioassay Development

Caleb Foster/Stock.Adobe.com

A bioassay measures biological activity or potency of a drug product (DP) or drug substance (DS), for which the most important critical quality attribute (CQA) is potency or relative potency if it is compared to a standard. Concentration measures how much of the molecule is in solution, and the bioassay measures how biologically active the protein is. Every biological product such as vaccines, hormones, antibodies, cell therapies, complex therapeutic proteins, and gene therapies all need a bioassay or multiple bioassays to measure biological activity.

A bioassay is an interesting combination of science and statistical methods to determine potency or relative potency.  The International Council for Harmonization (ICH) does not discuss bioassays to any great degree; however, United States Pharmacopeia (USP) does. USP<1032> covers biological development, USP<1033> covers validation, and USP<1034> covers analysis of bioassay data and methods of calculation.

Failure to develop the bioassay correctly will cause problems with drug approval/licensure, excessive invalids, and excessive out of specification (OOS).  Most of the problems with a bioassay can be avoided with correct development, qualification, validation, and control of the reference standard. 

Essential steps in developing a bioassay

Developing and validating a bioassay requires 12 essential steps, including determining the science that creates the signal that indicates biological activity and the dose response as well as number of doses used in the bioassay. In addition, one must develop a reference for relative potency as well as determine the method of calculation and design the bioassay calculator. Another step involves evaluating the robustness of the bioassay method while optimizing set points and limits. The other steps include standardizing and automating the bioassay, evaluating specificity and stability indicating, qualifying the bioassay, validating it, and monitoring reference potency as well as conducting stability checks. 

Determine signal indicating biological activity
Every bioassay generates a signal based on the activity of the molecule. The signal measures the change in protein concentration/dilution and biological activity. Each bioassay uses a specific technique to measure the change in activity or potency. It may react with the presence of another analyte in solution, modify cellular DNA, stimulate the immune response, cause florescence, stimulate hormonal activity, generate antibodies, measure toxicity, etc. The action of the drug generates a signal that can be measured with light, weight, or concentration.  

There are two different types of signals that can be generated by a bioassay, a linear dose response and a sigmoidal (s-shaped) dose response. Either approach may be used. Sigmoidal bioassay designs are preferred if there is a clear saturation in the dose response and there is no compelling reason to limit the number of doses. One solution does not fit all bioassays, however, and each bioassay needs to be evaluated based on the data at hand to ensure the right approach is used. This is particularly true for gene therapy and cell therapy bioassays.

Determine doses used in bioassay
If a sigmoidal design is used, a four- or five-parameter logistics (4PL or 5PL) curve is fit to the data. Nine doses are recommended to fit a 4PL model, three doses in the lower asymptote, three doses in the upper asymptote, and three doses in the linear range. If a parallel line analysis (PLA) fit is used, a minimum of four doses is recommended within the linear range. Dose masking allows for the identification and removal of a dose at the low or high end of the dose curve. Three consecutive doses are required by guidance. Dose and curve fitting selection are among the most critical aspects of bioassay development.

Develop reference for relative potency
A reference is selected to evaluate potency relative to the reference. This helps to control for cell-line variation, sample handing, animal biological activity, reagent effects, food sources, media, etc. References are often selected from early clinical lots so that there is a clinical connection and clinical relevance associated with the reference. Secondary references are measured and created to provide an understanding of quality should the primary reference be retired or depleted. Calculator design should control/correct for difference in standards.

In defining bioassays and relative potency, the USP states (1):

“Because of the inherent variability in biological test systems (including that from animals, cells, instruments, reagents, and day-to-day and between-lab), an absolute measure of potency is more variable than a measure of activity relative to a Standard. This has led to the adoption of the relative potency methodology. Assuming that the Standard and Test materials are biologically similar, statistical similarity (a consequence of the Test and Standard similarity) should be present, and the Test sample can be expected to behave like a concentration or dilution of the Standard. Relative potency is a unitless measure obtained from a comparison of the dose-response relationships of Test and Standard drug preparations.”

CLICK FIGURE TO ENLARGE Figure 1. Four parameter logistics (4PL) versus parallel line analysis (PLA) method of potency calculation.

Determine calculation method
There are two primary methods by which a bioassay potency and relative potency are calculated; 4PL with nine plus doses and PLA with four doses. In cases where animal life is a consideration, PLA is preferred. In cases where there is no clear upper asymptote, PLA is required. This is often the case with infectious viral vectors. For PLA, the doses must be carefully selected to stay in the linear range. Figure 1 shows the two methods of calculation.  Potency is the difference in the intercepts for the PLA, and Potency is the inflection point (50% change in activity) from the 4PL curve. Relative potency is calculated using the following equation, potency reference/potency test article * 100.  The ratio depends on the bioassay and how the activity is measured.

Characterize robustness, optimize set points and limitsUSP <1033> states that a bioassay needs to be characterized for robustness during development. Normally this is accomplished with a risk assessment (what can influence accuracy and repeatability) followed by a designed experiment with critical factors that may influence the bioassay. Normally three determinations are performed at the 100% target dilution for each experimental run. Accuracy and repeatability are evaluated for each of the run conditions, and the method is optimized to find the most stable conditions of the bioassay. A design space is created where the assay is accurate and repeatable and the edge of failure analysis indicates limits for critical reagent concentration, temperature, pH, etc.

Design bioassay calculator
Generally, SAS or SAS/JMP are recommended as software for bioassay development and validation because they are the only fully programable/customizable solutions that allow for all of the below items. Other canned bioassay software have serious restrictions that make them undesirable for bioassay computation and reporting. Elements of the calculator design for a bioassay include the following: 

• Complete all data entry and quality control data for integrity

• Log₁₀ transform the data as appropriate (dose and or response) for model fitting

• Identify and remove within dose outliers (Jackknife z)

• Fit the model (4PL or PLA)

• Identify the subset of doses to be used (PLA only, dose masking)

• Remove between dose (model) outliers (Jackknife or Studentized Residuals)

• Check systems suitability/positive control

• Check test article and reference validity

• Calculate potency and relative potency and 95% confidence intervals

• If out of specification or the confidence interval is too wide run additional n where indicated in the standard operating procedures (run rules for the bioassay)

• Combine n=1 or more to determine the reportable result and 95% confidence interval or declare the bioassay run invalid.

CLICK TABLE TO ENLARGE Table I. Evaluation of parallelism and systems suitability. 

There are several techniques for the evaluation of parallelism. Table I provides a summary of methods for the evaluation of parallelism. Other systems suitability criteria include the measurement of a known control, relative variation (%CV) of repeated measures, and curve depth for signal strength. Systems suitability most often is the EC50 of the reference standard (positive control).

Standardize and automate the bioassay
Develop the code/script for the bioassay. Make sure to generate all reports as PDFs. Extract all critical tracking and trending data from the bioassay and store all reportable information and results in a table form for easy analysis. Update the LIMS system and, where possible, automate all calculations and data transfer.

Evaluate specificity

Specificity is normally evaluated as a part of bioassay development or validation. There are two different ways of showing assay specificity or selectivity. The first method indicates only the specific protein generates a signal and blanks/placebos or other proteins generate no signal. The second is interference of other materials such as media, formulation buffer, forced degradation materials. Any interference is demonstrated by a shift in the intercept when spiked in. It can also be viewed as the change in accuracy in the presence of interfering compounds.

Evaluate stability indicating
Normally as part of development there is a desire to show the bioassay is stability indicating. This is done by thermal, light, pH or chemical degradation of the protein and measuring the change in potency. Normally 4 to 5 time points is desired and a statistically significant linear or none linear change in the signal is sufficient to prove the bioassay is stability indicating.

Qualify the bioassay

• To qualify a bioassay the following is recommended:

• Accuracy, five concentrations, n=6 or more at each concentration

• Repeatability, five concentrations, n=6 or more at each concentrations

• Dilutional linearity, five concentrations exceeding the upper specification limit (USL) and lower specification limit (LSL)

• Range where the bioassay is accurate, repeatable, and linear

• Intermediate precision, design of experiment (DOE) design to isolate the analyst, day, and instrument variance components.

Validate the bioassay
To validate a bioassay, the following is recommended:

• Accuracy, five concentrations, n=6 or more at each concentration

• Repeatability, five concentration, n=6 or more at each concentrations

• Dilutional linearity, five concentrations exceeding the USL and LSL

• Range where the bioassay is accurate, repeatable, and linear

• Intermediate precision, DOE design to isolate the analyst, day, and instrument variance components

• Robustness, set point, and operational range

• Stability indicating, significant change in signal under degraded conditions.

Monitor/control reference potency and check stability
It is critical to monitor the performance of the bioassay over time. The unconstrained EC50 standard is generally considered the best measure of the stability and consistency of the assay. To detect if assay drift is a method issue or standard issue a comparison to the unconstrained EC50 should be made. If both the reference and the test are drifting, it is change in the assay over time but no impact to relative potency. If the test is stable and the reference is trending, it will impact relative potency and should be corrected or a new reference selected.

Conclusion

Bioassays are powerful tests of relative potency and biological activity but need care in the design of the assay and the analysis to make sure they are a reliable indication of the change in potency. Validation of the assay will demonstrate the accuracy, repeatability, and linearity of the assay and its fit for use in the measurement of relative potency. Care in the selection of the dose response, outlier detection and removal, masking, transformation, and weighting will make the assay more stable and repeatable. 

References

1. USP, <1032>, “Design and Development of Biological Assay,” USP–NF 38(US Pharmacopeial Convention, Rockville, MD, December 2015).
2. USP, <1034>, “Analysis of Biological Assays,” USP–NF 38(US Pharmacopeial Convention, Rockville, MD, December 2015).
3. USP, <1033>, “Biological Assay Validation,” USP–NF 38(US Pharmacopeial Convention, Rockville, MD, December 2015).
4. T. Little, BioPharm International29 (1) 50–55 (2016). 

About the author

Thomas A. Little^*, PhD, drlittle@bioassaysciences.com, is president and CEO of BioAssay Sciences. 

^*To whom correspondence should be addressed.

Article Details

BioPharm International
Vol. 32, No. 11
November 2019
Pages: 42–45

Citation

When referring to this article, please cite it as T.A. Little, “Essentials in Bioassay Development,” BioPharm International 32 (11) 2019.