ADC Bioanalysis Knowledge Base

FDA_Immunogenicity

7 min read

FDA Immunogenicity Guidances for Therapeutic Protein Products (2014 + 2019)

概述 Overview

Two complementary FDA guidances govern ADC immunogenicity programs:

DocumentYearFocusSource
Immunogenicity Assessment for Therapeutic Protein Products2014 (Final)Overall strategy, risk assessment, clinical impacthttps://www.fda.gov/media/85017/download
Immunogenicity Testing of Therapeutic Protein Products — Developing and Validating Assays for Anti-Drug Antibody Detection2019 (Final)ADA assay development and validation methodologyhttps://www.fda.gov/media/119788/download

These guidances apply to all biologic therapeutics including ADCs. The 2019 assay guidance is more directly actionable for bioanalysis teams; the 2014 strategy guidance governs study design and clinical impact reporting.

核心要点 Key Points

1. Immunogenicity Risk Assessment Framework (2014 Guidance)

The 2014 guidance introduces a risk-based tiered approach — not all programs require the same immunogenicity testing depth:

Factors determining immunogenicity risk tier:

  • Product-related: molecular type (chimeric > humanized > fully human), Fc modifications, glycosylation pattern, aggregation propensity
  • Patient-related: immune status, concomitant immunosuppressive medications, indication
  • Treatment-related: route of administration, dosing frequency, dose level, treatment duration

For ADCs specifically: Additional risk factors include:

  • Linker-payload neoepitopes: The drug-linker moiety on the antibody creates novel chemical structures not present in the unconjugated antibody → potential anti-linker, anti-payload, or anti-linker-payload ADA
  • High DAR ADCs (e.g., T-DXd DAR ~8): More hydrophobic surface → higher aggregation risk → higher immunogenicity risk
  • Payload cytotoxicity: Paradoxically, immunosuppressive effects of cytotoxic payloads may reduce ADA incidence in heavily pretreated patients

2. Tiered Testing Algorithm (2019 Guidance)

The standard 3-tier (or 4-tier) testing cascade:

All Study Samples
       ↓
TIER 1: Screening Assay (high sensitivity)
       ↓ (if signal > cutpoint)
TIER 2: Confirmatory Assay (specificity confirmation by competition)
       ↓ (if ≥20-30% inhibition with free drug)
TIER 3: Titer Assay
       ↓ (selected positives)
TIER 4: Characterization (isotype, NAb, target of ADA)

Tier 1 — Screening:

  • Statistical cutpoint: 1% false-positive rate from drug-naïve matrix lots (≥50 individual donors)
  • Run in duplicate; signal threshold set during validation
  • Must include control (positive control = ADA reference standard; negative control = blank matrix)

Tier 2 — Confirmatory:

  • Confirm specificity by adding excess drug (25–100× nominal concentration) to the sample
  • Criterion: ≥20–30% inhibition of signal by competing drug confirms ADA specificity
  • Samples with inhibition below threshold are reported as screening-positive/confirmatory-negative

Tier 3 — Titer:

  • Serial dilution of confirmed ADA-positive samples until signal drops below cutpoint
  • Titer = reciprocal of last dilution still positive
  • Report as geometric mean titer for population summaries

Tier 4 — Characterization:

  • Isotype (IgG, IgM, IgE): IgE can drive hypersensitivity reactions
  • Neutralizing antibody (NAb): Cell-based or competitive binding assay to test if ADA blocks drug’s pharmacological activity
  • Target of ADA: Anti-idiotype (Fab-targeting), anti-Fc, anti-linker, anti-payload — use biotinylated component-specific antigens in competition studies

3. Assay Format Recommendation (2019 Guidance)

Preferred format: ECL bridging assay (MSD platform)

In a bridging assay:

  • Drug is labeled with biotin (capture) and with ruthenium/SULFO-TAG (detection)
  • ADA bridges the two drug molecules → generates signal
  • Requires sufficient ADA avidity to form stable bridge

Advantages of ECL bridging:

  • Low non-specific binding vs. plate ELISA
  • High sensitivity; S/N ratio typically 5–10×
  • Commercially validated MSD instruments and reagents
  • Species-independent (no secondary antibody needed)
  • Isotype-independent

Alternative: Direct ELISA (drug-coated plate + secondary anti-human-IgG) — simpler but lower drug tolerance and higher non-specific binding.

4. Drug Tolerance — Critical ADC Challenge

Why drug tolerance matters: Circulating ADC (cAb) in patient plasma competes with plate-bound or labeled drug for ADA binding sites. If cAb concentration is high (e.g., within 1–3 half-lives of last dose), drug suppresses signal → false-negative ADA result.

Drug tolerance of ECL bridging assay: Typically 10–50 µg/mL range (depends on format and drug concentrations used). At trough ADC concentrations in patients receiving q3W dosing, cAb may still be 20–100 µg/mL → drug tolerance inadequate.

Strategies to improve drug tolerance (endorsed by 2019 guidance):

StrategyMechanismDrug Tolerance Improvement
Acid dissociation (pH 3.5)Dissociates ADA-drug complex → free ADA re-captured after neutralization3–10× improvement (to ~100–1000 µg/mL drug)
Bead-based extractionParamagnetic beads coated with anti-human Fc capture ADA; wash away circulating drug2–5× improvement; complex
High concentration drug competition (confirmatory only)At high drug concentrations, excess drug out-competes circulating drug for ADA → not applicable to screening, but relevant to confirmatory cutpoint designN/A for screening
Anti-idiotype depletionRemove circulating drug using solid-phase anti-idiotype; releases ADATechnically challenging; rare

Recommendation: Acid dissociation + neutralization is the most widely implemented and accepted strategy. Implement for all ADC programs with expected trough concentrations >50 µg/mL.

5. ADA Sampling Strategy

Minimum sampling requirements (from 2019 guidance):

  • Pre-dose (baseline, Cycle 1 Day 1)
  • At least one on-treatment time point (typically end of Cycle 2 or 3)
  • At end of treatment / last dose
  • Follow-up post-treatment (30- and 90-day post-last-dose)
  • More frequent sampling in FIH trials

Sample timing guidance for ADC programs:

  • Draw ADA sample immediately before dosing (trough = lowest drug concentration = highest drug tolerance)
  • Pre-dose samples have highest drug tolerance — ideally >1 half-life from last dose
  • Use trough samples to maximize ADA detection sensitivity

常见问题和挑战，具体案例和解决方案

Challenge 1: Anti-Linker ADA vs. Anti-Idiotype ADA — How to Distinguish?

Problem: Patient develops confirmed ADA. Is it against the unique antibody idiotype (anti-Id), against the linker-payload neoepitope, or against the drug-linker complex? This matters because anti-linker ADA might affect multiple drugs using the same linker (e.g., Val-Cit linker class effect).

Solution:

  1. Use biotinylated naked antibody (unconjugated parent mAb) in a competition assay in confirmatory tier → if naked mAb inhibits ≥20%, ADA target is antibody idiotype or Fc
  2. Use biotinylated linker-payload moiety (synthesized independently) as competing antigen → if linker-payload inhibits ≥20%, ADA is anti-linker-payload
  3. Report both results; document in NDA/BLA BioA section

Case: In early brentuximab vedotin (Adcetris) trials, ~30% of patients developed ADA. Characterization revealed both anti-idiotype and anti-MMAE-linker components. FDA reviewed this carefully and required separate monitoring of anti-drug neoepitope ADA. This became a model for subsequent Val-Cit ADC programs.

Challenge 2: Confirmatory False-Negatives at High Payload-Drug Concentrations

Problem: The confirmatory assay uses a fixed drug concentration (e.g., 100 µg/mL drug) to compete with ADA for binding. If circulating drug concentration in the sample is already >100 µg/mL (e.g., peak sample collected at Cmax of 200 µg/mL), the added competitor cannot out-compete background → confirmatory assay gives false-negative.

Solution:

  1. Specify in the confirmatory assay SOP that samples with expected drug concentrations >X µg/mL (where X is the drug competition concentration) should be diluted before confirmatory testing
  2. Report the dilution factor used in confirmatory assay; acknowledge that this increases the apparent ADA titer threshold
  3. Use acid dissociation step to pre-treat all samples before confirmatory assay to partially mitigate drug interference

Challenge 3: Low ADA Incidence — Is the Assay Sensitive Enough?

Problem: A program reports 3% ADA incidence (3/100 patients). The sponsor argues the drug is non-immunogenic. Reviewer asks: could the assay be insensitive and missing ADA?

FDA expectation: Demonstrate assay sensitivity with a minimum detectable ADA concentration in the presence of the worst-case drug concentration expected in study samples. Report drug tolerance level as a key assay performance parameter.

Solution:

  1. Validate and report drug tolerance (the drug concentration in sample at which positive control ADA is still detectable at ≥3× S/N)
  2. Report % of samples collected at concentrations below drug tolerance threshold
  3. If >20% of samples are above drug tolerance: implement enhanced methods (acid dissociation) or increase sampling at trough time points

Graph View

Backlinks