05 — Bioanalysis: Clinical Stage
Overview
Clinical bioanalysis of ADCs supports first-in-human (FIH) through registrational trials (Phase 1–3). Methods must be fully validated per ICH M10/FDA BioA guidance. The clinical bioanalytical package is substantially more complex than non-clinical: it covers PK across multiple analytes, immunogenicity (ADA), pharmacodynamic (PD) biomarkers, and increasingly, exposure–response (E-R) and exposure–safety relationships.
1. Clinical PK Analyte Strategy
Regulatory expectation is that at minimum three analytes are characterized in clinical trials:
| Analyte | Measurement Purpose | Regulatory Driver |
|---|---|---|
| Total Antibody (TAb) | Antibody PK, half-life, FcRn recycling, deconjugation rate | Always required |
| Conjugated Antibody (cAb) | Active ADC exposure; correlates with efficacy and payload-related tox | Always required |
| Free Payload | Off-target toxicity driver; linker stability confirmation in humans | Required; scope depends on payload toxicity profile |
Optional (case-by-case):
- Catabolites: Measured if a catabolite is pharmacologically active (e.g., Lys-SMCC-DM1 from Kadcyla)
- Conjugated payload by hybrid LBA-MS: More precise cAb alternative; increasingly preferred in newer programs
- Total payload: Sum of conjugated + free; rarely measured directly
Relationship Between Analytes Over Time
In circulation, ADCs undergo deconjugation (payload loss without antibody clearance):
t=0 post-dose: [TAb] ≈ [cAb] (all antibody is conjugated)
t=24h: [TAb] > [cAb] (some payload lost; DAR declining)
t=days: [TAb] >> [cAb] (significant deconjugation; naked Ab accumulates)
The ratio cAb/TAb over time is a linker stability surrogate in humans. Non-cleavable thioether ADCs (e.g., T-DM1) show slower cAb decline vs. TAb compared to cleavable ADCs.
2. Method Validation Requirements (ICH M10 / FDA)
All clinical bioanalytical methods must be fully validated (GCP-compliant) before use in registrational studies. For FIH/Phase 1, validated methods are expected at study start.
Key Validation Parameters
| Parameter | LBA (ELISA/MSD) | Chromatographic (LC-MS/MS) |
|---|---|---|
| Selectivity | 10 individual lots of matrix (incl. 5% hemolyzed, lipemic) | Same |
| Calibration range | Min. 6 non-zero standards; quadratic or 4PL fit | Min. 6 standards; linear or 1/x² weighting |
| LLOQ | ≤20% CV and ≤20% bias; anchored by signal:noise ≥10 | ≤20% CV and ≤20% bias |
| Accuracy/precision | ≥3 QC levels; ≤15% CV; ±15% bias at each QC (±20% at LLOQ) | Same |
| Dilutional linearity | Hook effect assessment; parallelism with clinical samples | Dilutional integrity |
| Matrix effects | Post-column infusion or matrix-matched IS | Post-extraction matrix effect (≤25% CV) |
| Stability | Bench-top, freeze-thaw (≥3 cycles), long-term (covers study duration) | + processed sample stability |
| Reagent QC | Critical reagents (anti-drug Ab, antigen) lot-to-lot comparison; bridging if changed | Not applicable |
Critical Reagents for ADC LBA
Critical reagents are a defining feature of LBA validation for ADCs:
| Reagent | Function | Risk |
|---|---|---|
| Capture antibody (anti-Fc or anti-antigen) | Pulls ADC from matrix | Lot variability in binding affinity → assay drift |
| Detection antibody (anti-drug or anti-linker) | Generates signal proportional to conjugated payload | Anti-drug Ab supply, affinity, and specificity critical |
| Reference standard (characterized ADC) | Calibrator; defines concentration scale | DAR of reference standard must match study material; stability over program lifetime |
| Positive control (PC) | Demonstrates assay performance | Must be prepared from same lot as standard if possible |
Regulatory expectation: Critical reagent acceptance criteria defined before use; lot-to-lot bridging studies performed; reagent stability in storage documented.
3. Immunogenicity — ADA Testing
ADAs (Anti-Drug Antibodies) can profoundly affect ADC PK, efficacy, and safety. ADC immunogenicity is complex because the ADA response can be directed against:
- The antibody (anti-idiotype, anti-Fc)
- The linker
- The payload
- Linker-payload conjugate (neoepitope created by conjugation)
Testing Algorithm
Screen (all samples) → Confirmatory (positives) → Titer (confirmed+) → Characterization
| Step | Purpose | Cutoff |
|---|---|---|
| Screening assay | Detect potential ADA (high sensitivity) | 1% false-positive rate (statistical cutoff) |
| Confirmatory assay | Confirm specificity via drug competition (≥20–30% inhibition) | Drug at 25–100× endogenous ADC concentration |
| Titer assay | Quantify ADA level in positive samples | Reciprocal dilution at cutoff signal |
| Characterization | Isotyping (IgG, IgM, IgE), neutralizing Ab (NAb), anti-drug specificity | Case-by-case |
Assay Formats
| Format | Principle | Advantage | Limitation |
|---|---|---|---|
| Electrochemiluminescence (ECL) bridging assay | ADC-biotin + ADC-ruthenium; ADA bridges to generate signal (MSD platform) | Low matrix interference; widely accepted | Drug tolerance often low |
| Direct ELISA | Plate-bound ADC captures ADA | Simple; high throughput | High non-specific binding; low drug tolerance |
| Acid dissociation (ADA) | Acidification disrupts ADC-ADA complex; re-neutralization allows ADA binding | Improves drug-tolerant ADA sensitivity | May denature ADC epitopes |
Drug tolerance is a key concern: circulating ADC competes with plate-bound ADC for ADA binding, suppressing signal. Strategies to improve drug tolerance:
- Acid dissociation step (pH 3.5)
- Use of affinity-purified anti-idiotype to deplete circulating ADC (not widely used)
- Bead-based extraction + re-neutralization
ADA Impact on ADC PK
| ADA Type | PK Effect | Clinical Consequence |
|---|---|---|
| Non-neutralizing (binding Ab) | Accelerated clearance (immune complex formation) | Reduced exposure → reduced efficacy |
| Neutralizing (NAb) | Block target binding or FcRn recycling | Loss of PK and efficacy |
| Anti-drug (payload-directed) | Accelerated payload clearance if circulating | May reduce free payload toxicity — or have no effect |
| Anti-PEG | Accelerated clearance for PEGylated ADCs | Relevant for PEGylated linker designs |
Clinical examples:
- T-DM1 (Kadcyla): ADA incidence ~5%; minimal clinical impact on PK — thioether linker neoepitopes relatively non-immunogenic
- Brentuximab vedotin (Adcetris): ADA incidence ~30% in HL; some patients show accelerated clearance
4. Biomarkers in Clinical ADC Trials
4a. Soluble Target Antigen (sTAA)
Shed antigen competes with cell-bound antigen for ADC binding → reduces effective tumor exposure:
| ADC | Soluble Antigen | Clinical Significance |
|---|---|---|
| Mylotarg | Soluble CD33 (sCD33) | High sCD33 → reduced ADC delivery to AML blasts |
| Besylomab | Soluble CD22 | Correlated with treatment response in B-ALL |
| T-DM1 | Soluble HER2 (sHER2/HER2 ECD) | Elevated sHER2 → reduced T-DM1 PK exposure (accelerated antigen-mediated clearance) |
Assay: ELISA with anti-antigen antibody; validated in serum or plasma.
4b. Target Expression (Tissue Biomarkers)
| Marker | Method | Purpose |
|---|---|---|
| HER2 IHC/ISH | IHC (0/1+/2+/3+) or FISH (amplification) | Patient selection; eligibility criterion |
| TROP-2 IHC | IHC H-score | Predictive biomarker for sacituzumab govitecan response |
| FRα IHC | Ventana FOLR1 (SP368) | Eligibility for mirvetuximab soravtansine |
| CD30 IHC | Dako CD30 (clone Ber-H2) | Eligibility for brentuximab vedotin |
4c. Pharmacodynamic Biomarkers
| Biomarker | Assay | ADC/Payload Class |
|---|---|---|
| γH2AX (phospho-H2AX) | IHC or flow cytometry | DNA-damaging payloads (calicheamicin, PBDs) — marker of DSBs |
| Mitotic index | IHC (phospho-histone H3, PHH3) | Tubulin inhibitors (MMAE, DM1) — G2/M arrest |
| Caspase-3 (cleaved) | IHC | Apoptosis across payload classes |
| ctDNA | Digital PCR, NGS panel | Tumor response and resistance mutation tracking |
5. Exposure–Response (E-R) Analysis
Regulatory guidance increasingly expects E-R modeling to support dose selection and labeling. For ADCs:
| Exposure Metric | Typical E-R Endpoint |
|---|---|
| cAb AUC₀–τ (cycle 1) | Objective response rate (ORR), PFS |
| Free payload Cmax | Specific toxicities (e.g., peripheral neuropathy for MMAE) |
| TAb trough (Ctrough) | ADA incidence correlation |
| sTAA (baseline) | PK exposure (antigen-mediated clearance) |
E-R models are submitted in population PK (popPK) analyses to FDA/EMA and used to:
- Justify flat dosing (mg/kg vs. flat mg dose)
- Support dose modifications in special populations (renal/hepatic impairment)
- Select pediatric doses (if applicable)
6. Special Analytical Topics
6a. Incurred Sample Reanalysis (ISR)
Required for GCP-validated methods. ISR confirms method reproducibility on actual study samples:
- Minimum 10% of total study samples (or 67 samples, whichever is greater)
- ≥66.7% of ISR samples must meet ±20% of original (LBA) or ±15% (LC-MS/MS)
- ADC-specific challenge: ISR window must be within biomarker stability window; frozen sample stability must cover ISR timing
6b. Calibration in Human Matrix
Human matrix variability (ADA presence, endogenous antigen, bilirubin, lipids) can affect LBA assays:
- Use pooled human plasma/serum for calibrators and QCs (screened for ADAs, low endogenous antigen)
- Or use stripped matrix (acid-stripped or charcoal-stripped) if interferences cannot be avoided
- Document selectivity against lipemia, hemolysis, bilirubinemia
6c. Multi-Analyte Phase Considerations
| Trial Phase | TAb | cAb | Free Payload | ADA | Biomarker |
|---|---|---|---|---|---|
| FIH (Phase 1, dose escalation) | Validated | Validated | Validated | Validated screen/confirm | Exploratory |
| Phase 1 expansion | Validated | Validated | Validated | Full ADA panel | Semi-validated |
| Phase 2 | Validated | Validated | Validated (if informative) | Full, incl. titer | Validated (if primary endpoint) |
| Phase 3 (pivotal) | Validated | Validated | Case-by-case | Full | Validated if used for selection/endpoint |
7. Regulatory Submissions
Key Documents Referencing Bioanalysis
- IND: Preliminary method descriptions; validated methods for GLP toxicology
- CTA: European equivalent; same BMV standard (ICH M10)
- Integrated Summary of Clinical Pharmacology (ISCP): PK across analytes; popPK model; E-R relationships
- NDA/BLA Module 2.7.2: Summary of clinical PK studies including bioanalytical methods
- Module 5.3.4: Analytical reports supporting clinical studies
Current Regulatory Guidances
| Document | Applicability |
|---|---|
| FDA Guidance: Bioanalytical Method Validation (2018) | US; LBA and chromatographic methods |
| ICH M10 (Step 4, 2022) | Global harmonized BMV; includes LBA-specific chapter |
| EMA Guideline on BMV (2012, under revision) | EU; will be superseded by ICH M10 implementation |
| FDA Guidance: Immunogenicity Assessment for Therapeutic Protein Products (2019) | ADA testing framework |
| AAPS Recommendations for the Bioanalytical Method Validation of ADCs (2020 white paper) | Industry consensus; referenced in submissions |
Key Papers
- Kaur et al. (2013) AAPS J — ADC bioanalysis recommendations
- Gorovits et al. (2013) Bioanalysis — hybrid LBA-MS for ADC
- Stevenson et al. (2021) Bioanalysis — ADA challenges for ADCs
- ICH M10 Bioanalytical Method Validation (2022)
- FDA BMV Guidance (2018)
- Cowan et al. (2023) Clin Pharmacol Ther — exposure-response analysis of T-DXd in DESTINY-Breast trials