PK/PD, ADA, and biomarker assays explained for biotech founders

Pharmacokinetic (PK) assays measure how much drug is in the blood over time. Pharmacodynamic (PD) assays measure what the drug does to its target. Anti-drug antibody (ADA) assays detect the immune response a patient mounts against the drug. Biomarker assays measure biology that signals disease or response. Together they tell a Phase 1 team whether exposure, activity, and immunogenicity match the plan, and all must exist before first dose.

What do PK and PD assays actually measure?

A PK assay quantifies drug concentration in serum or plasma at timed points after dosing. Those concentrations become exposure parameters: peak concentration, half-life, and area under the curve. Small molecules are usually measured by liquid chromatography with mass spectrometry. Biologics are usually measured by ligand-binding assays, in which a capture reagent binds the drug and a labeled detector reports the amount.

FDA's adoption of ICH M10 covers chromatographic and ligand-binding assays that measure the parent drug and active metabolites in nonclinical and clinical subjects [1]. The same document states that biomarker bioanalysis and immunogenicity methods are not within its scope [1]. That boundary tells you which assays have fixed criteria and which do not.

A PD assay measures the drug's effect on its intended biology: receptor occupancy, depletion of a cell population, or a shift in a downstream marker. In Phase 1, PK and PD together show whether the predicted dose reaches target-saturating concentrations. The honest limit is statistical. Cohorts are small, PK varies between patients, and PD markers often do not move at low doses.

What is an anti-drug antibody assay, and why is it tiered?

FDA's 2019 guidance recommends a multi-tiered approach. A sensitive screening assay flags samples that may contain binding antibodies. Positives move to a confirmatory assay, usually a competition format in which excess drug erases the signal if the antibodies are specific. Confirmed positives are then characterized by titration and, where warranted, neutralization assays [2]. The design traces to industry recommendations published in 2008 that still shape cut-point setting [4].

FDA considers a false-positive rate of approximately 5 percent desirable for screening and recommends 1 percent for the confirmatory cut point [2]. Screening and confirmatory IgG and IgM assays should reach a sensitivity of at least 100 nanograms per milliliter, tighter than the 250 to 500 nanograms per milliliter FDA traditionally accepted [2]. FDA also recommends testing early whether circulating drug masks the antibodies [2].

The reason regulators care is clinical. FDA's 2014 immunogenicity guidance describes immune responses that alter PK, neutralize efficacy, cause adverse events, or cross-react with an endogenous protein, and asks for a risk-based approach [5]. Harmonized terms such as treatment-emergent ADA come from a 2014 consensus paper [6]. Neutralizing antibody assays are the hardest tier, and dedicated validation recommendations exist for them [7].

What is a biomarker assay, and how is it different from a PK assay?

FDA and NIH define a biomarker as "a defined characteristic that is measured as an indicator of normal biological processes, pathogenic processes, or responses to an exposure or intervention" [8]. Their joint BEST resource, begun after the agencies prioritized shared terminology in 2015, names seven categories, including pharmacodynamic or response, predictive, and safety biomarkers [8], [9].

The technical difference from a PK assay is the analyte. A PK assay measures a foreign molecule against a clean blank matrix. A biomarker assay measures something the body already makes, so there is no true blank and the meaningful result is a change from baseline. That is why M10 leaves biomarkers out [1].

The accepted standard for biomarker methods is fit-for-purpose validation, in which rigor is scaled to how the data will be used [10]. Formal qualification is separate: FDA qualifies a biomarker for a defined context of use, not the method, through the three-stage process the 21st Century Cures Act requires [8]. Almost no Phase 1 program needs it.

What does validated mean, and when is fit-for-purpose enough?

For a PK ligand-binding assay, M10 defines full validation precisely. Accuracy must be within 20 percent of nominal and precision at or below 20 percent CV, both relaxed to 25 percent at the limits of quantitation. Total error may not exceed 30 percent, or 40 percent at those limits, and the calibration curve needs at least six concentration levels [1].

M10 also requires incurred sample reanalysis, in which real study samples are re-assayed to prove reproducibility: 10 percent of samples for studies of 1,000 or fewer, and 10 percent of the first 1,000 plus 5 percent of the remainder above that [1]. Bioanalysis supporting GLP toxicology must also conform to 21 CFR Part 58, which exists to assure the integrity of safety data [11].

ADA assays follow a staged standard. FDA states that Phase 1 and Phase 2 samples may be tested with suitable screening, confirmatory, and sometimes neutralization assays, while pivotal samples require fully validated assays [2]. When risk is high, FDA may require assays before the trial starts and real-time testing; otherwise sponsors may bank samples until suitable assays exist [2]. The tradeoff is deferred cost, because any later method change forces bridging across stored samples.

When does each assay need to be ready for a Phase 1 trial?

The 9 to 12 month lead time I plan for comes from dependencies, not bench work. ADA cut points need treatment-naive serum from the target population, which is scarce in a rare pediatric disease. The positive control antibody usually has to be raised in animals. PK assays need the clinical drug lot.

ICH S6(R1) shapes the nonclinical side. It calls for antibody measurement in repeat-dose toxicology studies, characterized by titer, responding animals, and neutralizing status, and correlated with PK and PD changes [3]. It also states that antibody formation in animals is not predictive of antibody formation in humans [3]. Nonclinical ADA data explain the tox study; they do not forecast clinical immunogenicity.

What did I learn building the assay suite for a Phase 1 biologic?

For a first-in-class biologic in a rare pediatric leukemia, I oversaw development of the PK, PD, ADA, biomarker, potency, and safety assays for the first-in-human trial through contract laboratories in a virtual model. The hardest constraint was blood volume. Pediatric sampling limits forced us to rank every assay and time point, and some exploratory biomarkers were cut.

The second lesson was that the protocol and the assay plan must be written together. Sampling windows, tube types, and storage temperatures belong in the protocol because the assays depend on them. In my experience the most common Phase 1 bioanalytical failure is not a bad assay. It is a good assay applied to samples handled in a way the method was never validated for.

This article is part of the Life Sciences Consulting practice at Hamilton Atlas Group — see the full capability list there.

References

1. U.S. Food and Drug Administration. M10 Bioanalytical Method Validation and Study Sample Analysis. FDA, 2022.
2. U.S. Food and Drug Administration. Immunogenicity Testing of Therapeutic Protein Products - Developing and Validating Assays for Anti-Drug Antibody Detection. FDA, 2019.
3. International Council for Harmonisation. Preclinical Safety Evaluation of Biotechnology-Derived Pharmaceuticals S6(R1). ICH, 2011.
4. Shankar G, Devanarayan V, Amaravadi L, et al. Recommendations for the validation of immunoassays used for detection of host antibodies against biotechnology products. Journal of Pharmaceutical and Biomedical Analysis. 2008;48(5):1267-1281.
5. U.S. Food and Drug Administration. Immunogenicity Assessment for Therapeutic Protein Products. FDA, 2014.
6. Shankar G, Arkin S, Cocea L, et al. Assessment and Reporting of the Clinical Immunogenicity of Therapeutic Proteins and Peptides - Harmonized Terminology and Tactical Recommendations. The AAPS Journal. 2014;16(4):658-673.
7. Gupta S, Devanarayan V, Finco D, et al. Recommendations for the validation of cell-based assays used for the detection of neutralizing antibody immune responses elicited against biological therapeutics. Journal of Pharmaceutical and Biomedical Analysis. 2011;55(5):878-888.
8. U.S. Food and Drug Administration. About Biomarkers and Qualification. FDA, 2024.
9. U.S. Food and Drug Administration. BEST Resource Taxonomy. FDA, 2024.
10. Lee JW, Devanarayan V, Barrett YC, et al. Fit-for-Purpose Method Development and Validation for Successful Biomarker Measurement. Pharmaceutical Research. 2006;23(2):312-328.
11. Legal Information Institute, Cornell Law School. 21 CFR § 58.1 - Scope. Code of Federal Regulations, 2025.