The AAPS Journal (2025) 27:138

[Link]

REVIEW ARTICLE-THEME

Antibody-Drug Conjugates: Innovations and Therapeutic Frontiers

Recent Advances in Bioanalytical Methods for Quantification

and Pharmacokinetic Analyses of Antibody–Drug Conjugates
R. M. Naseer Khan1 · Yi Zeng1 · Abdul‑Azeez A. Lanihun1 · Oluwatobi T. Arisa1 · Jessica L. Horner2 ·
William D. Figg1,2

Received: 25 April 2025 / Accepted: 24 July 2025

This is a U.S. Government work and not under copyright protection in the US; foreign copyright protection may apply 2025

Abstract
Antibody–drug conjugates (ADCs) represent a rapidly expanding class of therapeutics, uniquely combining the specificity
of monoclonal antibodies with the potency of cytotoxic small-molecule payloads. Due to their inherent structural complexity
and heterogeneous composition, accurate characterization and quantification of ADCs pose significant bioanalytical chal-
lenges. This review discusses recent advancements in bioanalytical methodologies, including ligand binding assays (LBAs),
liquid chromatography-tandem mass spectrometry (LC–MS/MS)-based approaches, and emerging hybrid LBA-LC–MS/MS
platforms. In addition, this review will discuss pharmacokinetic (PK) modeling approaches essential to ADC development,
ranging from population PK models to mechanistic frameworks, including physiologically based pharmacokinetic (PBPK)
and quantitative systems pharmacology (QSP) models. These modeling strategies allow detailed characterization of ADC
absorption, distribution, metabolism, and elimination processes while also accounting for complexities introduced by payload
deconjugation and drug-to-antibody ratio variability. By integrating robust bioanalytical methods with advanced modeling
techniques, this review provides researchers with essential insights to enhance ADC characterization, inform experimental
design, and ultimately facilitate the development of safer, more effective therapeutic candidates.

Keywords analytical chemistry · antibody–drug conjugate · bioanalytical · cancer · clinical Pharmacology · ECLIA ·
ELISA · FT-ICR · LC–MS/MS · pharmacokinetic

Introduction In 2000, gemtuzumab ozogamicin became the first ADC

to receive accelerated FDA approval for the treatment of
Antibody–drug conjugates (ADCs) are a fast-emerging acute myeloid leukemia (AML) (2). It was initially with-
class of chemotherapeutics, with over a dozen approved by drawn due to safety concerns, but later reapproved following
the U.S. Food and Drug Administration (FDA) to date (see adjustments to its dosing regimen that improved its safety
Table I). While some, such as Lumoxiti and Blenrep, have and efficacy profile. Since then, multiple ADCs have been
since been withdrawn from the market, more than 100 ADCs approved and have become essential for the treatment of
remain in clinical trials for various cancer treatments (1). various cancers, including hematologic malignancies, as
well as urothelial, lung, and breast cancers. A recent exam-
ple is datopotamab deruxtecan, approved for the treatment
Guest Editor: Baolin Zhang of unresectable or metastatic, hormone receptor-positive,
* William D. Figg
human epidermal growth factor receptor 2 (HER2)-negative
wdfigg@[Link] breast cancer (3).
ADCs are designed to selectively deliver potent cytotoxic
1
Clinical Pharmacology Laboratory, Clinical Center, National agents to cancer cells and consist of three key components:
Institutes of Health, 9000 Rockville Pike, Building 10, Room
5A03, Bethesda, Maryland 20892, USA
a cytotoxic payload, an antibody, and a linker that cova-
2
lently attaches the payload to amino acids on the antibody
Clinical Pharmacology Program, Office of the Clinical
Director, Center for Cancer Research, National Cancer
(4). ADCs can incorporate multiple linkers, increasing the
Institute, Bethesda, Maryland, USA concentration of payloads delivered. These are typically

Vol.:(0123456789)
138 Page 2 of 15 The AAPS Journal (2025) 27:138

Table I  FDA-approved ADCs Brand Generic Approval Year Target Bioanalytical Assay
and the Bioanalytical Assays
used for the Quantification of Mylotarg* gemtuzumab ozogamicin 2000; 2017 CD33 ELISA
ADCs or their Components LC–MS/MS
in Serum or Plasma for
Adcetris brentuximab vedotin 2011 CD30 ELISA
Submission
LC–MS/MS
Kadcyla trastuzumab emtansine 2013 HER2 ELISA
LC–MS/MS
Besponsa inotuzumab ozogamicin 2017 CD22 ELISA
LC–MS/MS
Polivy polatuzumab vedotin 2019 CD79b Immunoaffinity LC–MS/MS
Padcev enfortumab vedotin 2019 Nectin-4 ELISA
LC–MS/MS
Enhertu trastuzumab deruxtecan 2019 HER2 ECLIA
LC–MS/MS
Trodelvy sacituzumab govitecan 2020 TROP2 ECLIA
LC–MS/MS
Blenrep* Belantamab mafodotin 2020 BCMA ECLIA
LC–MS/MS
Zynlonta loncastuximab tesirine 2021 CD19 ECLIA
LC–MS/MS
Tivdak tisotumab vedotin 2021 TF ELISA
LC–MS/MS
Elahere mirvetuximab soravtansine 2022 FRα ELISA
LC–MS/MS
Datroway datopotamab deruxtecan 2025 TROP2 Gyrolab LBA
LC–MS/MS
*
Mylotarg was withdrawn in 2010 and reapproved in 2017. Blenrep withdrawn in 2022 but has been reap-
proved in the UK in 2025

conjugated to the antibody via lysine residues or reduced in the target cancer cell (12, 13). The released payload may
interchain disulfide bonds, resulting in a heterogeneous diffuse beyond the initial target cell, leading to the destruc-
mixture of ADCs with variable drug–antibody ratios (DAR tion of neighboring cells, also known as the bystander effect
0–8) (see Fig. 1) (5–7). Notably, deconjugation of the cyto- (see Fig. 2) (14). For ADCs with non-cleavable linkers, the
toxic payload, occurring either enzymatically or chemi- payload remains conjugated to the antibody until proteo-
cally can take place in systemic circulation, resulting in lytic degradation occurs within the lysosome, resulting in
a gradual decrease in the average DAR over time. While a degraded monoclonal antibody (mAb) moiety and a pay-
linker stability primarily impacts ADC safety and efficacy load with a positively charged lysine, that is trapped in the
by increasing the systemic levels of the cytotoxic payload, target cell (15) Alternately, ADCs with cleavable linkers
systemic deconjugation can complicate bioanalytical assays may undergo enzymatic cleavage within the endosome or
by increasing analyte heterogeneity and can pose additional lysosome, leading to payload release. The bystander effect
challenges for accurate pharmacokinetic (PK) analysis (8, 9). is commonly observed in ADCs with cleavable linkers non-
Thus, during the design stage of an ADC, selecting a linker polar payloads. Payloads with non-polar properties are more
with substantial in vivo stability is a critical parameter (10). membrane permeable, enabling their diffusion into neigh-
Both cleavable and non-cleavable linkers have been used in boring cells (15). In the context of a heterogeneous tumor
approved ADCs, and there has been continuous progress in microenvironment, where both antigen-positive and antigen-
linker technology to prevent premature release of cytotoxic negative cells contribute to cancer progression, the bystander
drugs in non-target cells and systemic circulation (10, 11). effect theoretically has the potential to improve both thera-
Mechanistically, once the ADC reaches the target cell, its peutic efficacy and patient outcomes. However, conversely,
antibody component recognizes and binds to the correspond- it can also affect nearby healthy cells (16).
ing cell surface antigen, initiating internalization via endocy- ADCs exhibit a complex PK profile, driven by the
tosis. The ADC is first enclosed within an early endosome, distinct properties and interactions of their components,
which subsequently matures into a late endosome. This late which influence absorption, distribution, metabolism,
endosome then combines with a lysosome, facilitating pay- and elimination (ADME). After the infusion of an ADC,
load release. Once released, the payload induces apoptosis systemic circulation may contain the fully intact ADC or
The AAPS Journal (2025) 27:138 Page 3 of 15 138

A
OH

O O
O O O
H H HN
S N N N Enzyme Cleavage O
N N NH
H H --
O O O -- N
- - Payload Release
O -
O
Maleimide Cysteine Cleavable peptide O N
Conjugation Linker
O OH F
HN
O
Payload DXd
N
O
N
O OH F

B
Hu 276 mAb
O
O Payload Release
HN S
O
N --
-- N
- --
O H
Dibromo maleimide N

O O
Phosphine Azide

O N
Ph P O
Ph O O OH
HN
O O
O N
O
-- N
-- H
H
Payload - - - NH
Release
MMAF
N NH2

N
N

IMQ

Fig. 1  Schematic depiction of various conjugation strategies utilized load ADC, CD276 monoclonal antibody (mAb) conjugated to mono-
during the design phase of single and dual-payload antibody–drug methyl auristatin F (MMAF) and imiquimod (IMQ), linked through
conjugates (ADCs). a The structure of the recently FDA-approved either cysteine or lysine residues using two distinct linkers. Payloads
datopotomab deruxtecan, which comprises a topoisomerase I inhibi- are shown in color red; cleavable peptide linkers are shown in color
tor (DXd payload), covalently attached to the antibody via a tetra- blue and conjugation site is shown in color black. Adapted from refer-
peptide-based cleavable linker. The cytotoxic payload is conjugated ences 36 and 59 and created using ChemDraw
through maleimide-cysteine linkage. b The structure of the dual-pay-

its individual components, including the unconjugated and possibly nonlinear distribution and elimination (17).
antibody and unconjugated cytotoxic payload. As the However, unlike unconjugated antibodies, ADCs, particu-
large-molecule antibody component displays different PK larly those with higher DAR, tend to exhibit faster clear-
profiles from the small-molecule payload, PK analysis of ance due to the addition of the cytotoxic payload (17).
an ADC requires a comprehensive understanding of each In addition to the inherent complexity of assessing the PK
component. ADC PK is primarily driven by the antibody profile of ADCs, accurately characterizing these molecules
component, and can display both the benefits and limita- through bioanalytical analysis presents unique challenges. In
tions of unconjugated monoclonal antibodies, such as, low contrast to small molecule compounds, which can be reli-
volume of distribution, slow clearance, a long half-life, ably quantified using established techniques such as liquid
 138 Page 4 of 15 The AAPS Journal (2025) 27:138

Fig. 2  Mechanism of action of an ADC and subsequent quantifica- plete enzymatic payload release, resulting in apoptosis. 3 Samples are
tion by different bioanalytical platforms. 1 mAb is conjugated to collected from either preclinical or clinical studies at different time
cytotoxic payload via a linker to yield an ADC. 2 The ADC binds its intervals that include various ADC components, 4 that can be quanti-
target antigen, triggering internalization into the cancer cell. It first fied either by LC–MS/MS 5 or through ligand binding assays such as
enters an early endosome, which then progresses to a late endosome, ELISA and/or 6 through hybrid LBA – LC – MS/MS quantification.
and if the linker is cleavable, payload cleavage begins here. The late Adapted in parts from references 4 and recreated using [Link]
endosome then fuses with the acidic lysosomal compartment to com-

chromatography-mass spectrometry (LC–MS), large bio- Ligand Binding Assays

molecules like ADCs require more sophisticated analytical
approaches. This complexity arises from the heterogeneous Ligand binding assays (LBAs) have been widely used for
composition of ADCs, which can exist as fully conjugated, the quantification of large molecules, including ADCs (20).
partially conjugated, or completely unconjugated species, In ADC analysis, LBA’s are primarily used to quantify the
each with distinct structural and functional properties that total antibody and conjugated antibody components (21),
influence safety and efficacy profiles. Therefore, a thorough and very rarely, to measure the unconjugated payload (22).
understanding of ADC PK and pharmacodynamics (PD) LBAs leverage specific antigen–antibody interactions to
requires bioanalytical methods capable of differentiating measure analytes, offering high sensitivity, cost-effective-
between fully intact and partially unconjugated forms to ness, suitability for high-throughput analyses, and ease of
accurately characterize their distribution, metabolism, and method implementation (Table II) (23, 24). However, LBAs
clearance in biological systems. have limitations for ADC characterization, including the ina-
There are multiple FDA guidance available on approaches bility to holistically differentiate DAR species due to ADC
for ADC drug development including bioanalytical assay heterogeneity, and challenges in detecting unconjugated
development and validation and PK considerations (18, 19), payload or partially degraded ADC species. Additionally,
as presented below. These methods are extensively used in the development of specific reagents used in LBAs, such as
the pre-clinical phase of ADC drug development as well anti-idiotype antibodies and anti-payload antibodies, can be
as in various phases of clinical trials to assess safety and resource-intensive and time-consuming (24).
efficacy of the drug.
The AAPS Journal (2025) 27:138 Page 5 of 15 138

Table II  Comparison of Various Analytical Methods Employed for the Quantification of ADCs
Analytical Method Principle Advantages Limitations Typical Analytes Quantification Range
Measured

Ligand-Binding Assay Specific antibody- - High throughput - Cross-reactivity - Total antibody Medium—High
(LBA) (21–23) antigen interaction - Cost-effective issues - Conjugated antibody
- High sensitivity for - Limited DAR speci- - Payload (rarely)
antibodies ficity
- Matrix interference
LC–MS/MS (bottom- - High specificity
Proteolytic digestion - Time-consuming - Payload Medium—High
up) (25) followed by mass - High sensitivity for sample preparation - Site-specific conju-
spectrometry payloads - Loss of intact struc- gation
- Site-specific infor- tural context
mation
LC–MS/MS (middle- Partial digestion/ - Good balance - Intermediate - Payload Medium
down) (26, 27) reduction followed by between structural complexity of data - DAR
mass spectrometry detail and simplicity interpretation - Partially reduced
- Localization of con- ADC components
jugation sites
LC–MS/MS (top- Direct analysis of - Retains intact struc- - Lower sensitivity - Intact ADC Low—Medium
down, intact) (28) intact ADCs via tural context than bottom-up - DAR distribution
high-resolution mass - Rapid workflow - Complex spectral
spectrometry - Direct DAR meas- interpretation
urement
Hybrid LBA–LC–MS/ Immunocapture - Combines specificity - Method development - Total antibody, Medium—High
MS (29) enrichment com- of LBA and accu- can be complex - Conjugated antibody
bined with LC–MS/ racy of MS - Requires specialized - Payload
MS quantification - Overcomes instrumentation
traditional LBA
limitations (DAR
resolution)

ADC Quantification on LBA platforms Monomethyl auristatin E (MMAE) ADCs and total antibod-
ies in cynomolgus monkey sera. For MMAE-ADC detec-
Various types of LBA platforms are currently utilized for tion, a mouse anti-MMAE antibody served as the capture
ADC quantification, each offering distinct benefits in sensi- antibody, while an horseradish peroxidase (HRP)-conjugated
tivity, throughput, and multiplexing potential, with selection mouse anti-human immunoglobulin G – fragment crystalliz-
of the platform based on the specific requirements of the able (IgG-Fc) antibody was used for detection. To measure
assay. Commonly used LBA platforms for biotherapeutic total antibody (hRS7), a recombinant human trophoblast
analysis include chromogenic, luminescence, time-resolved cell surface antigen two extracellular domain (hTROP-2-
fluorescence, electrochemiluminescence, label-free, fluores- ECD-His) was employed as the capture reagent, with the
cence polarization, and imaging in combination with exist- same HRP-conjugated mouse anti-human IgG-Fc antibody
ing platforms (30). for detection. The assay demonstrated a dynamic range of
For ADC quantification, Enzyme-linked immunosorbent 0.3–35.0 ng/mL for MMAE-ADCs and 0.2–22.0 ng/mL for
assay (ELISA), Electrochemiluminescence immunoassay total antibodies (32). This underscores the highly sensitive
(ECLIA), and Gyrolab® have been commonly reported in nature of ELISA, particularly for conjugated antibody and
the literature. ELISA is the most established of the vari- total antibody quantifications. Despite the several benefits of
ous LBA platforms for PK profiling and immunogenicity ELISA, its use is limited by the potential for matrix interfer-
testing of large molecules in complex matrices. This is due ence and limited dynamic range.
to its cost-effectiveness and simplicity of procedure (no ECLIA, like ELISA, is based on the specific interaction
sample extraction) (24). ELISA is a heterogeneous enzyme between an antibody and its corresponding antigen. It uti-
immunoassay (EIA) technique in which one of the reaction lizes specialized reagents, including a capture antibody that
components is either nonspecifically adsorbed or covalently targets the antigen, and a labeled antibody for detection.
bound to a solid-phase surface, such as a microtiter well, To enable detection, the capture antibody is immobilized
magnetic particle, or plastic bead, and operates based on the on a solid matrix, such as a microplate or magnetic bead,
principle of enzymatic signal detection (31). Pie et al. (2022) while the labeled antibody is tagged with a luminescent
developed an ELISA-based bioanalytical method to quantify marker and an electrochemically active molecule, allowing
138 Page 6 of 15 The AAPS Journal (2025) 27:138

for signal generation and detection (31). The assay formats from Singulex Inc., Cytometric Bead Array (CBA, Becton,
used to detect analytes are similar to ELISA: direct, indirect, Dickinson and Company), Ella (ProteinSimple/Bio-Techne),
competitive, and sandwich (33–35). The ECLIA platform is and Immuno-PCR (Imperacer®) from Chimera Biotec
increasingly used to replace traditional ELISA due to higher GmbH. These instruments facilitate ultrasensitive biothera-
sensitivity in the femtomolar/106 nM range, wider dynamic peutic quantification and expedite analytical turnaround
range, and minimal matrix effects, as well as limited cross- times. To the best of our knowledge, these products have
reactivity with reagents and analytes (36). Skidmore et al. yet to be applied specifically to ADC quantification, but their
(2020) employed an ECLIA-based bioanalytical method to use for antibody quantification has been reported (42, 43),
quantify ARX788 (conjugated ADC) and total antibodies indicating their potential application for ADC bioanalysis.
in a preclinical study. ARX788 is a site-specific Anti-HER2
ADC generated by conjugating Amberstatin antibody with LC‑MS/MS Developments in the Quantification
monomethyl auristatin F (MMAF) payload via a noncleav- of ADCs
able hydroxylamine- polyethylene glycol linker. To quantify
the conjugated ADC, the anti-Amberstatin antibody (AMB- PK analyses of ADCs have traditionally relied on LBAs, but
20) was used as the capture antibody, while biotinylated due to past limitations with LBAs, such as cross-reactivity,
AMB-20 and streptavidin SULFO-TAG served as detection limited ability to quantify unconjugated payload, and chal-
reagents. For total antibody quantification, a recombinant lenges in differentiating various ADC species, LC–MS-
human ErbB2/HER2-Fc Chimera Protein served as the cap- based methods have increasingly been used to improve
ture antigen, with detection achieved using a biotinylated accuracy and specificity (44, 45). LC–MS methods enable
goat anti-human kappa light chain antibody and streptavi- the quantification of ADCs at various levels of structural
din SULFO-TAG (37). Nevertheless, a major limitation of resolution, including intact analysis (top-down, under native
ECLIA is the requirement for specialized equipment, rea- or denaturing conditions) (46), partially reduced analysis
gents, and solvents, making it a more expensive option com- (middle-down) (47), and peptide-level characterization (bot-
pared to ELISA (37, 38). tom-up) (see Fig. 3) (48). This flexibility allows for a more
The Gyrolab® system is an advanced automated immuno- precise assessment of ADC heterogeneity, metabolism, and
assay platform that operates within a compact microfluidic clearance compared to traditional LBAs. In addition, this
disc with preloaded streptavidin columns to facilitate affinity approach allows for sensitive and accurate quantification of
interactions. With the aid of capillary and centrifugal forces, the small molecule payload, a critical parameter for assess-
this nanoliter-scale flow-through technology controls liquid ing both therapeutic efficacy and potential off-target toxicity.
flow through the affinity column. By integrating operations Recently, the LC–MS approach has undergone signifi-
within the microfluidic device, the instrument minimizes cant technological advancements. Notably, there have been
hands-on time for sample and reagent preparation while improvements in ionization techniques, particularly with
maintaining flexibility to accommodate various reagents and the development of nano-electrospray ionization, which
analytes. The system is designed to accommodate a wide employs ultra-fine capillary needles to generate highly
concentration range and various sample matrices, making charged droplets from minimal sample volumes, thereby
it suitable for applications such as biomarker analysis, PK, significantly enhancing analytical sensitivity and resolution
PD studies, and bioprocess development (39). In a Gyrolab (49, 50). Furthermore, there has been an increase in the use
system, the capture and detection reagents may consist of of novel mass analyzers, such as the Orbitrap mass spec-
antibodies, proteins, peptides, or nucleic acid-based binders, trometer and the Fourier transform ion cyclotron resonance
which can be commercially sourced or developed in-house. (FT-ICR) mass spectrometer, which offers exceptionally
Essentially, capture reagents must be biotin-labeled, while high mass resolution and accuracy (51, 52). Building on
detection reagents require a fluorescence tag, such as Alexa these advances, recent studies have demonstrated the grow-
Fluor® 647, compatible with the system’s 635-nm red laser ing use of LC–MS/MS approaches for the quantification of
detection. Notably, the sample volume used in a Gyrolab® ADCs in biological matrices. A few illustrative examples
is minimal compared to conventional immunoassays, rang- are highlighted below.
ing from 0.02 to 4 𝜇 L (40). Okamoto et al. (2020) utilized a Top-down proteomics involves the direct analysis of
Gyrolab immunoassay to determine concentrations of trastu- intact proteins without digestion, enabling the characteriza-
zumab deruxtecan (T-DXD) and total antibody in the plasma tion of sequence variants, post-translational modifications
of HER2-positive tumor-bearing mice and achieved a lower (PTMs), and DARs (Table II). Li et al. recently reported
quantification limit of 200 ng/ml (41). an intact quantification workflow for cysteine-conjugated
Other advanced LBA immunoassay systems include the ADCs using native mass spectrometry (nMS) (28), con-
Single Molecule Array (SIMOA™), a digital ELISA tech- necting capillary size exclusion chromatography (SEC)
nology by Quanterix, Single Molecule Counting (SMC™) directly to the mass spectrometer. Their workflow involved
The AAPS Journal (2025) 27:138 Page 7 of 15 138

Fig. 3  Overview of general sample preparation and LC–MS/MS separate the light and heavy chain fragments for subsequent LC–MS/
analysis of ADC and unconjugated payload. ADC from biological MS quantification. The top-down proteomics approach enables direct
matrix is typically enriched using immunoaffinity capture. In bottom- quantification of the intact ADC. For released payload quantification,
up proteomics, the enriched ADC is digested using trypsin, generat- the biological matrix is spiked with an internal standard, and plasma
ing peptide fragments that are then quantified by LC–MS/MS using a proteins are precipitated using organic solvents prior to LC–MS/MS
surrogate peptide. Middle-down proteomics involves either reducing quantification. Adapted in parts from reference 48 and recreated using
cysteine-conjugated ADCs with DTT or partially digesting them to [Link]

automated affinity purification followed by nMS analysis, DAR as 8, consistent with homogenous cysteine conjuga-
allowing monitoring of average DAR and in vivo stability tion. The ADC was further digested with IdeS to identify
of new linker payloads with minimal method development. conjugation sites, revealing three major species: a drug-con-
Mouse PK study samples with two ADCs were analyzed jugated light chain, a heavy chain (Fd subunit) conjugated
with this method and showed an excellent linear dynamic to three drugs, and unconjugated Fc/2 regions. Subsequent
range of 5–100 µg/mL, and strong linear correlations simi- peptide mapping of T-DXd with trypsin or pepsin digestion
lar to conventional targeted bottom-up proteomics approach provided extensive sequence coverage (96% for heavy chain,
using a surrogate peptide method that had sensitivity from 100% for light chain), confirming cysteine conjugations at
0.026–100 µg/mL (28). positions C223, C229, and C232 (heavy chain) and C214
Deslignière et al. also investigated the intact quantifica- (light chain) (53). Despite growing interest in top-down pro-
tion of T-DXd as a model compound to illustrate the poten- teomics, lower sensitivity still remains a challenge, particu-
tials of state-of-the-art LC methods coupled to nMS under larly for proteins with very high molecular weight.
denaturing conditions using LC–MS and included peptide Middle-down proteomics bridges the gap between top-
mapping and middle-down analysis for comparison (53). down and bottom-up proteomics, providing complementary
The samples in this study were either dissolved in water or structural insights. This approach involves partial digestion
phosphate buffer with no additional purification employed of mAbs into larger peptides (> 3 kDa), enabling detailed
prior to the analysis. Using nMS, they characterized the analysis of the primary sequence and accurate localization
138 Page 8 of 15 The AAPS Journal (2025) 27:138

of drug conjugation sites with enhanced sensitivity. A recent and thiol-adduct formations. Notably, EAD generated diagnos-
study developed an online LC–MS/MS method coupled with tic ions enabling clear differentiation of constitutional isomers
high-resolution Fourier transform ion cyclotron resonance and provided structural details of biotransformation species
(FTICR) for rapid analysis of reduced ADCs at the subu- by selectively cleaving disulfide bonds without disrupting
nit level. A concurrent reduction strategy was applied to a the peptide backbone. Thus, demonstrating that bottom-up
model cysteine-linked ADC to generate six distinct subu- approaches can effectively quantify and characterize ADC
nits: unconjugated light chain (Lc0), drug-conjugated light species and DAR variants in complex in vivo biological sam-
chain (Lc1), unconjugated heavy chain (Hc0), and heavy ples (56).
chains conjugated with one to three drugs (Hc1–3). This In addition, to assess the drug safety and efficacy of an
approach enabled simultaneous assessment of partially and ADC, it is imperative to fully characterize the released cyto-
fully reduced ADC forms, facilitating rapid and compre- toxic payload in samples after intravenous injection of an
hensive characterization of critical ADC attributes, includ- ADC. Due to its high sensitivity, LC–MS/MS enables the
ing DAR quantification, primary sequence characterization, quantification of released payloads even at very low con-
and localization of drug conjugation regions (Table II) (26). centrations. A sensitive and rapid LC–MS/MS method was
Yuan et al. recently reported a digestion-free middle-down recently developed to quantify the unconjugated payload,
MS (DF-MDMS) approach for quantifying conjugated Aur0101, from the investigational ADC PYX-201 in patients
payloads from four ADCs with varying DARs and linker with advanced solid tumors (57, 58). Following a simple
cleavage properties, using the Zeno TOF 7600 (SCIEX) extraction from 25 µL of human plasma, Aur0101 was quan-
with collision-induced dissociation (CID) (27). The method tified by LC–MS/MS with a linear range of 0.036 to 16.823
employed immunocapture of the ADC using magnetic beads nM (57). Similarly, Xu et al. published a fully validated
and optional reduction and deglycosylation, enabling sensi- method to analyze exatecan, its prodrug, and ARV-825 in
tive quantification of the conjugated payload (as low as 50 rat plasma, demonstrating its use in a proof-of-principle PK
ng/mL) in ADCs with homogenous conjugation. However, it study of exatecan following intravenous administration to
is pertinent to note that multiple factors could contribute to Sprague–Dawley rats (59).
DF-MDMS assay sensitivity. In this study both ADCs with With continued advancements, current research has
homogeneous cysteine conjugation and with DAR8 showed expanded into incorporating multiple payloads, either two
excellent sensitivity but ADC2 DAR4 and trastuzumab cytotoxic agents or a combination of one cytotoxic payload
emtansine had much higher LLOQ, even though ADC2 has and one immunoregulating agent into a single ADC. Lev-
the same payload as ADC1. Hence, ADCs with stochastic engo et al. developed a method to construct homogenous
heterogeneous conjugation inherently have a distribution of dual-drug ADCs (60), while McKertish et al. demonstrated
DAR species, thus affecting the sensitivity of each species synergistic cytotoxic efficacy from dual-drug ADCs (61).
(27). In addition, Linker-payload intrinsic ionization and dis- Additionally, a recent study has introduced a novel approach
sociation properties are also key contributing factors towards that incorporates a cytotoxic payload and an immunoregu-
assay sensitivity of ADCs (27). lating agent into a single mAb for treating triple-negative
Finally, perhaps the most commonly used method, bottom- breast cancer (62). LC–MS/MS remains crucial for quan-
up proteomics, involves breaking down antibodies into small tifying these multi-payload ADCs. Mak et al. developed a
peptides using proteases, enabling precise site-specific analysis simple and sensitive workflow enabling simultaneous quan-
of drug conjugation (Table II) (25, 54). While this approach tification of six ADC payloads—7-ethyl-10-hydroxycampt-
provides detailed information on conjugation sites, it inher- othecin (SN-38), Methotrexate (MTX), Deruxtecan (DXd),
ently disrupts the overall protein structure, limiting insights MMAE (47), MMAF and Calicheamicin (CM) by LC–MS/
into higher-order architecture. Despite these limitations, MS with validated quantification range of 0.4–100 nM
bottom-up proteomics remains a valuable tool for character- for SN-38, MTX, and DXd, 0.04–100 nM for MMAE and
izing drug-loading patterns or even detailed structural char- MMAF, and 0.4 −1000 nM for CM (63). Using this work-
acterization of ADCs (55, 56). Recent studies have further flow, the PK profile of unconjugated MMAE was assessed
illustrated the utility of bottom-up proteomics for analyzing in mouse serum post intravenous ADC administration (5 mg/
in vivo ADC samples. For instance, a study employing tryp- kg), demonstrating minimal levels of unconjugated MMAE
tic digestion combined with ligand-binding affinity and LC- (Cmax = 2.1 ± 0.5 nM) in circulation, thus confirming strong
MRMHR (multiple reaction monitoring with high-resolution linker stability (64).
MS) successfully characterized biotransformation products Moreover, LC–MS/MS approaches have emerged as a
of AZD8205. Using CID (collision-induced dissociation) and highly efficient and versatile tool for the analysis of ADCs
EAD (electron-activated dissociation), the study accurately in recent years. Its ability to analyze multiple components
identified payload deconjugation sites, distinguished linker- of ADCs from a single sample makes the technique highly
payload species, and characterized disulfide rearrangements sought after compared to other available techniques.
The AAPS Journal (2025) 27:138 Page 9 of 15 138

Hybrid LBA‑LC‑MS/MS Assays for the Quantification the intact conjugate and its individual components – the
and PK Analysis of ADCs large molecule antibody and the small-molecule cytotoxic
payload – each of which contributes uniquely to the over-
Hybrid immunocapture-liquid chromatography coupled with all disposition and therapeutic effect. Mathematical models
tandem mass spectrometry (LBA-LC–MS/MS) has emerged are created to characterize the PK characteristics of a drug,
as a powerful analytical technique for the quantification of i.e., its absorption, distribution, metabolism, and elimination
ADCs and total antibodies in biological matrices. These (ADME), and its PD, which includes safety and efficacy end-
hybrid approaches combine the specificity and selectiv- points. These models use concentration data obtained from
ity of ligand-binding assays with the accuracy, sensitivity, the discussed assays to provide a comprehensive understand-
and structural resolution of mass spectrometry, effectively ing of drug behavior. They are then used to guide decision
addressing common limitations associated with traditional making, optimize dosing regimens, and predict efficacy or
LBAs (Table II). adverse events. Additionally, mechanistic models may be
Huang et al. (29) recently developed a hybrid immuno- used to simulate drug behavior in on-target and off-target
capture-liquid chromatography coupled with multiple reac- tissues, as well as the deconjugation rate. There are multiple
tion monitoring (LBA-LC-MRM) method for quantifying types of models, and their relevance to ADCs is covered
unconjugated mAbs, site-specific cysteine conjugated ADCs further in this section. An overview of modeling approaches
(DAR 2), and site-non-specific cysteine-conjugated ADCs across the drug development process can be found in Fig. 4.
(DAR 8), achieving a linear dynamic range between 1–10 Modeling approaches of ADCs must capture the PK/
µg/mL. Intact ADC quantification was compared with a PD of the overall ADC with additional consideration of the
surrogate peptide analysis approach targeting peptides from variability of the deconjugation process, and the PK of the
both the light and heavy chains, as well as conjugated pay- released cytotoxic payload. Typically, the PK of the ADC
loads. When applied to PK analysis in rat plasma samples, is mainly driven by the antibody component. However, the
the intact quantification method revealed biotransformation small-molecule cytotoxic payload can be released prior to
products that were not detected using the surrogate peptide internalization, diffuse across cell membranes, or affected
method, demonstrating the additional insights attainable by efflux transporters, allowing these drugs to escape the
through intact mass spectrometric analyses (29) (Table II). target tumor cells and affect nearby cancerous cells or off-
Similarly, Suh et al. evaluated a hybrid micro-flow liquid target healthy tissues, a phenomenon known as the bystander
chromatography-tandem mass spectrometry (µLC-MS/MS) effect. Although preclinical evidence suggests that hetero-
approach for preclinical PK analysis of ADCs using minimal geneous tumors may benefit from the bystander effect, it
plasma volumes. Their approach utilized semi-automated remains challenging to measure this effect in clinical set-
solid-phase extraction (SPE) to streamline sample prepa- tings (15). Although it is difficult to directly measure, the
ration and significantly reduce reagent and plasma sample bystander effect and higher DAR of trastuzmab deruxtecan
consumption. Compared to traditional LBAs, the µLC-MS/ (T-DXd) is theorized to play a role in the differences found
MS platform provided reliable quantification of total anti- in clinical activity in heterogeneous cancers and cancers
body, intact antibody, and total ADC in mouse plasma, dem- with lower target antigen compared to another trastuzumab-
onstrating its reliability for robust ADC PK analyses (65). based ADC (15, 67). However, recent ADCs with improved
Further highlighting the versatility of hybrid methods, linker stability, which would theoretically prevent the induc-
Yin et al. introduced an immunoaffinity LC–MS/MS assay tion of the bystander effect, have been shown to be effective
using a signature peptide from the complementarity-deter- even in the absence of a bystander effect, suggesting the
mining regions (CDRs) as a surrogate for quantifying TAb bystander effect should not be the sole focus in the develop-
levels from PYX-201 in human plasma (66). Collectively, ment of novel ADCs. (15) Additionally, once the ADC is
these studies show the effectiveness of hybrid LBA-LC–MS/ degraded, the payload can accumulate in certain off-target
MS platforms in addressing traditional LBA limitations, organs, such as the liver (17) Thus, the appropriate PK
such as matrix interference and lack of DAR differentia- modeling approach depends on the specific questions being
tion, while enhancing analytical rigor and depth in ADC investigated and the stage of drug development.
quantification. Building on these advances in quantification, Due to the complex nature of ADCs, multiple analytes
recent efforts have turned toward characterizing the pharma- are collected throughout preclinical and clinical develop-
cokinetic behavior of ADCs in greater detail. ment. During the early clinical phase, the population PK/
PD approach is categorized into single-analyte, two-analyte,
Pharmacokinetic Considerations and Approaches and semi-mechanistic multiple-analyte models. Single-
analyte population PK models focus on the specific analyte
Characterizing the pharmacokinetics of ADCs can be incred- determined to be the most clinically relevant to efficacy and
ibly complex, as it requires evaluating the PK profile of both safety, primarily the conjugate form, measured as conjugated
138 Page 10 of 15 The AAPS Journal (2025) 27:138

Fig. 4  Impact of pharmacokinetic modeling approaches throughout the drug development process. Adapted from reference 66

antibody or conjugated payload. Conjugated antibody con- further quantifies the relationship between large-molecule
centration is the most commonly used analyte for ADC PK antibody components and the rate of payload deconjuga-
analyses; however, the interpretation of these measurements tion. It can also model the dynamics of conjugate stability
is complicated by ADC heterogeneity. ADC species with and the subsequent formation and clearance of the unconju-
different DARs can exhibit distinct pharmacologic prop- gated payload. These population PK approaches also allow
erties, meaning that measured conjugated antibody levels for the exploration of covariates or the impact of patient
might not reliably predict therapeutic response. On the other demographics such as body weight, age, sex, ethnicity, and
hand, assays that quantify the conjugated payload provide hepatic or renal impairment on the PK. The subsequent
limited insight into overall antibody exposure, as low con- population PK-derived exposure metrics, considered better
centrations of ADCs with high DAR may appear similar to than observed exposure metrics, which are affected by vari-
high concentrations of ADCs with lower DAR (17). This ability across trials and sampling time, are then used for PK/
can complicate the comparison of the PK of an ADC to the PD modeling or exposure–response analysis and are critical
expected pharmacologic effect. Improvements in both the for phase 2/3 dose selection (68). Both further explore the
analytical methods to better describe the DAR distribution relationship between ADC concentration and efficacy and
and chemistry techniques to create homogenous ADCs have safety endpoints.
helped address this issue. Comparatively, semi-mechanistic models often include
Two-analyte population PK models are far more varied, multiple-analyte analysis to characterize the mechanism
involving combinations such as total antibody/conjugated related to proteolytic degradation, deconjugation, and
payload, total antibody/conjugated antibody, conjugated unconjugated payload formation, often built on preclini-
payload/unconjugated payload, and conjugated antibody/ cal data (68). These established models can then be used
unconjugated payload (68). The two-analyte approach to predict one analyte from another, decreasing the amount
The AAPS Journal (2025) 27:138 Page 11 of 15 138

of sampling and bioanalytical assays needed. This semi- seen in Fig. 5. Internalization of the ADC in vitro can be
mechanistic approach is a blend of the traditional top-down measured by labeling the ADC with a pH-sensitive fluores-
empirical PK/PD approach and the bottom-up mechanistic cent dye (73, 74) and tracking its uptake into an appropri-
systems biology approach, resulting in the formation of a ate preclinical model, such as a 2D cell line, 3D spheroid,
new discipline. In recent years, advancements in mechanis- or organoid (75), using the Incucyte Live-Cell Analysis
tic approaches, such as physiologically based pharmacoki- System or fluorescence microscopy (73, 74). Internaliza-
netic (PBPK) and quantitative systems pharmacology (QSP) tion assays are used to confirm that the ADC enters its tar-
modeling, have enabled further characterization of ADCs. get cell enabling payload release. Additionally, by utiliz-
Although QSP is a relatively new discipline compared to ing non-cancerous cell lines, these experiments can reveal
PBPK modeling and lacks a universally agreed-upon defi- whether the ADC enters non-tumor cells, potentially leading
nition, it has been suggested that PBPK modeling may be to increased toxicity in vivo. Once inside the target cancer
considered a subset within the broader scope of QSP. QSP cell, the preclinical efficacy of the ADC may be assessed
was first defined in a National Institutes of Health (NIH) using in vitro assays that measure apoptosis or changes in
working group white paper, which recognized the need for cell viability (76, 77). Furthermore, these assays may also
a new approach to translating preclinical discoveries into be used to evaluate the bystander effect of ADCs in vitro.
clinical progress by integrating computational modeling of By combining antigen-positive and antigen-negative cells in
biological and pharmacologic systems (69, 70). co-culture systems, and comparing ADC efficacy in mono-
Both approaches require extensive data from multiple cultures versus co-cultures, one can determine whether the
analytes obtained through robust assays and often incorpo- ADC facilitates killing of antigen-negative cells through
rate information obtained from both preclinical and clini- bystander effect mechanisms (78).
cal studies. As the name implies, PBPK further integrates Preclinical in vivo animal studies also provide valuable
physiological information, including different organs and data for mechanistic models, as cell line-derived or patient-
tissues, allowing the estimation of drug concentrations in derived xenografts often offer a more accurate representation
tissue that would be otherwise difficult to measure in the of patient drug responses and toxicity compared to in vitro
real world (71). Additionally, PBPK models can be further systems (79). Prior to treatment, xenograft tumors can be
used to explore possible drug-drug interactions (DDI) of the evaluated for antigen expression using immunohistochem-
unconjugated payload and possibly justify approval without istry, with the quantification of antigen levels helping to pre-
a dedicated DDI clinical trial (68). While the complexity dict the ADC’s ability to bind target cells (80). ADC efficacy
of PBPK models have been a barrier, Simcyp Simulator, a in vivo is typically communicated by measuring changes in
PBPK modeling and simulation software, has also released tumor volume following treatment (73). Additionally, track-
an ADC module that could make further PBPK approaches ing the overall survival of the study animals provides impor-
more accessible. This module was tested using a model for tant information on preliminary toxicity, which is crucial for
enfortumab vedotin and unconjugated MMAE to predict ADCs with a narrow therapeutic index (81). QSP models
drug-drug interactions (DDI), demonstrating the applicabil- can then integrate these PD metrics to estimate ADC activity
ity of the ADC module during drug development. in clinical settings and identify potential off-target effects.
There are similarities between QSP and previously dis- Overall, as our understanding of ADCs and bioanalytical
cussed methods, as QSP evolved from multiple comple- methods improves, further advancements in QSP and PBPK
mentary modeling approaches, including systems biology, modeling can streamline the ADC development pipeline.
PKPD, and PBPK. Although both QSP and PBPK are heav- With further developments in ADC technology, such as
ily mechanistic approaches, PBPK models predominantly improvements in cytotoxic payload conjugation, allowing
focus on answering ADME and PK questions, while QSP homogenous high DAR molecules and even multiple pay-
models focus on PD efficacy and safety outcomes and dis- load ADCs, new considerations regarding the PK must be
ease pathophysiology (69). Recent advancements have taken into account.
improved our understanding of intracellular mechanisms, Currently, most ADCs have a single unique cytotoxic pay-
tumor heterogeneity, the bystander effect, and the mecha- load and exhibit highly variable DAR, which can change
nisms behind ADC toxicity (72). These mechanistic or semi- over time. The pharmacological activity of ADCs is primar-
mechanistic approaches are often favored from preclinical ily driven by the cytotoxic payload. However, ADCs with a
development to the early stages of clinical development, high DAR tend to have faster clearance than those with a
particularly during the transition from preclinical to clinical lower DAR, possibly leading to decreased exposure. As a
phases. They are used to leverage preclinical data to predict result, most clinical stage ADCs have an average DAR of
clinical responses when clinical data is lacking. 3.5–4 (17, 82). Efforts to homogenize ADCs provide more
Examples of the preclinical data used depend on the spe- robust and reliable pharmacokinetics, with better plasma sta-
cific aspect of the ADC’s PD being investigated and can be bility and fewer off-target effects (83). Currently, Enhertu
138 Page 12 of 15 The AAPS Journal (2025) 27:138

Fig. 5  An overview of preclinical techniques to analyze ADC PD in cancer models. In vivo, immunohistochemistry staining of tumor tis-
vitro and in vivo. In vitro, pH-sensitive dyes assist in the visualiza- sue confirms the presence of the cell surface antigen that the ADC
tion of the ADC internalization process. ADC-mediated apoptosis or targets. In CDXs or PDXs, effective ADCs trigger a measurable
subsequent decreases in cell viability may be assessed in cell-based decrease in tumor volume

(fam-trastuzumab deruxtecan-nxki) and Trodelvy (sacitu- about unforeseen toxicity, as there is a current lack of in vivo
zumab govitecan-hziy), are two FDA-approved ADCs con- data (85). The pharmacokinetics, safety, efficacy, and pos-
sidered to be homogeneous as they approach the theoretical sible drug-drug interactions of the individual payloads must
maximum DAR, 8, both of which include cleavable linkers also be further considered, as the complex modifications
allowing for the bystander effect. required for multi-payload ADCs may lead to new issues
Additionally, there is research into conjugating multiple (86). As multi-payload ADCs are still in the early stages
distinct payloads onto the same antibody. Given the het- of research with limited in vivo data, further research and a
erogeneous nature of tumors, resistance can develop with more comprehensive quantitative approach using the meth-
many current treatment regimens involving a combination ods mentioned in this review, such as QSP, will be neces-
of drugs with different mechanisms of action (84). Theo- sary to characterize the safety and efficacy of multi-payload
retically, conjugating multiple payloads onto the same ADC ADCs (86).
could offer the same benefit and possible synergistic effects
as discussed previously. Conversely, there may be concerns
The AAPS Journal (2025) 27:138 Page 13 of 15 138

Conclusions need to obtain permission directly from the copyright holder. To view a
copy of this licence, visit [Link]

The successful development and clinical implementation of

ADC therapeutics are heavily dependent on accurate bio-
analytical characterization and subsequent robust pharma-
cokinetic modeling. This review highlights the importance References
of analytical methods, including LBAs, LC–MS/MS, and
1. Liu K, Li M, Li Y, Li Y, Chen Z, Tang Y, et al. A review of the
innovative hybrid LCA-MS/MS platforms, in overcoming clinical efficacy of FDA-approved antibody-drug conjugates in
traditional quantification challenges such as ADC heteroge- human cancers. Mol Cancer. 2024;23(1):62.
neity, interference effects, sensitivity, and limited differen- 2. Bross PF, Beitz J, Chen G, Chen XH, Duffy E, Kieffer L, et al.
Approval summary: gemtuzumab ozogamicin in relapsed acute
tiation of analytes. Furthermore, we emphasized the critical
myeloid leukemia. Clin Cancer Res. 2001;7(6):1490–6.
role of PK modeling approaches, ranging from population 3. Mullard A. FDA approves TROP2-targeted antibody–drug conju-
PK methods to mechanistic frameworks like PBPK and QSP, gate for breast cancer. Nat Rev Drug Discov. 2025;24(159).
in elucidating ADC disposition, payload dynamics, and drug 4. Fu Z, Li S, Han S, Shi C, Zhang Y. Antibody drug conjugate: the
“biological missile” for targeted cancer therapy. Signal Transduct
tissue interactions. The integration of these models enables Target Ther. 2022;7(1):93.
researchers to better understand and predict ADC behavior, 5. Tang Y, Tang F, Yang Y, Zhao L, Zhou H, Dong J, et al. Real-
which is crucial for informing dosing strategies and thera- time analysis on drug-antibody ratio of antibody-drug conjugates
peutic decisions. It is evident that future advances in ADC for synthesis, process optimization, and quality control. Sci Rep.
2017;7(1):7763.
design, including multi-payload integration and optimiza- 6. McCombs JR, Owen SC. Antibody drug conjugates: design and
tion of DAR homogeneity, will necessitate further refine- selection of linker, payload and conjugation chemistry. Aaps Jour-
ment of analytical methods and PK modeling strategies. By nal. 2015;17(2):339–51.
combining state-of-the-art bioanalytical quantification and 7. Panowski S, Bhakta S, Raab H, Polakis P, Junutula JR. Site-
specific antibody drug conjugates for cancer therapy. mAbs.
sophisticated PK modeling methods, scientists will be able 2014;6(1):34–45.
to gain deeper insights into ADC pharmacology, streamlin- 8. Zuo P. Capturing the magic bullet: pharmacokinetic principles and
ing the development process and enabling more effective, modeling of antibody-drug conjugates. AAPS J. 2020;22(5):105.
safer therapeutic options for patients. 9. Kopp A, Hofsess S, Cardillo TM, Govindan SV, Donnell J,
Thurber GM. Antibody-drug conjugate sacituzumab govitecan
drives efficient tissue penetration and rapid intracellular drug
release. Mol Cancer Ther. 2023;22(1):102–11.
Author Contributions All authors have reviewed the article, and wrote,
10. Su Z, Xiao D, Xie F, Liu L, Wang Y, Fan S, et al. Antibody-drug
edited, and proofread the manuscript. All authors approved the final
conjugates: Recent advances in linker chemistry. Acta Pharm Sin
text.
B. 2021;11(12):3889–907.
11. Bargh JD, Isidro-Llobet A, Parker JS, Spring DR. Cleav-
Funding Open access funding provided by the National Institutes of
able linkers in antibody–drug conjugates. Chem Soc Rev.
Health. This research was supported by the Intramural Research Pro-
2019;48(16):4361–74.
gram of the National Institutes of Health (NIH). The contributions of
12. Chau CH, Steeg PS, Figg WD. Antibody-drug conjugates for can-
the NIH author(s) were made as part of their official duties as NIH
cer. Lancet. 2019;394(10200):793–804.
federal employees, are in compliance with agency policy requirements,
13. Kalim M, Chen J, Wang S, Lin C, Ullah S, Liang K, et al. Intra-
and are considered Works of the United States Government. How-
cellular trafficking of new anticancer therapeutics: antibody-drug
ever, the findings and conclusions presented in this paper are those of
conjugates. Drug Des Devel Ther. 2017;11:2265–76.
the author(s) and do not necessarily reflect the views of the NIH or
14. Staudacher AH, Brown MP. Antibody drug conjugates and
the U.S. Department of Health and Human Services.
bystander killing: is antigen-dependent internalisation required?
Br J Cancer. 2017;117(12):1736–42.
Declarations 15. Giugliano F, Corti C, Tarantino P, Michelini F, Curigliano G.
Bystander effect of antibody-drug conjugates: fact or fiction? Curr
Conflict of interest WDF receives research funding from Propella Oncol Rep. 2022;24(7):809–17.
Therapeutics Inc, and has active licenses with Aevis Bio and Novartis. 16. Lee HK, Kim B, Ko YG, Chung SW, Shim WS, Choi S-Y, et al.
All remaining authors declare no competing interests. Enhancing the bystander effect of antibody-drug conjugate by
using a novel caspase-3 cleavable peptide linker to overcome
Open Access This article is licensed under a Creative Commons Attri- tumor heterogeneity. J Control Release. 2025;382:113738.
bution 4.0 International License, which permits use, sharing, adapta- 17. Lin K, Tibbitts J. Pharmacokinetic considerations for antibody
tion, distribution and reproduction in any medium or format, as long drug conjugates. Pharm Res. 2012;29(9):2354–66.
as you give appropriate credit to the original author(s) and the source, 18. Centre for Drug Evaluation and Research FDA. M10 bioanalyti-
provide a link to the Creative Commons licence, and indicate if changes cal method validation and study sample analysis: guidance for
were made. The images or other third party material in this article are industry. 2022. Contract No.: 03/28/2025.
included in the article’s Creative Commons licence, unless indicated 19. Centre for Drug Evaluation and Research [Link]. Clinical phar-
otherwise in a credit line to the material. If material is not included in macology conisderations for antibody-drug conjugates: Guidance
the article’s Creative Commons licence and your intended use is not for industry. 2024.
permitted by statutory regulation or exceeds the permitted use, you will
 138 Page 14 of 15 The AAPS Journal (2025) 27:138

20. Thway TM. Fundamentals of large-molecule protein thera- technology for the qualitative and quantitative assessment of sero-
peutic bioanalysis using ligand-binding assays. Bioanalysis. logical responses to vaccination. Malar J. 2020;19(1):159.
2016;8(1):11–7. 39. Engström J. Gyrolab Immunoassays: Miniaturization, Automa-
21. Qin Q, Gong L. Current analytical strategies for antibody-drug tion, and Integration into a Rapid Workflow. In: Matson RS, edi-
conjugates in biomatrices. Molecules. 2022;27(19):6299. tor. ELISA: Methods and Protocols. Springer, US: New York, NY;
22. Sanderson RJ, Hering MA, James SF, Sun MM, Doronina SO, 2023. p. 109–27.
Siadak AW, et al. In vivo drug-linker stability of an anti-CD30 40. Engström J. Gyrolab immunoassays: miniaturization, automa-
dipeptide-linked auristatin immunoconjugate. Clin Cancer Res. tion, and integration into a rapid workflow. Methods Mol Biol.
2005;11(2 Pt 1):843–52. 2023;2612:109–27.
23. Qin Q, Gong L. Current analytical strategies for antibody-drug 41. Okamoto H, Oitate M, Hagihara K, Shiozawa H, Furuta Y, Ogitani
conjugates in biomatrices. Molecules. 2022;27(19). Y, et al. Pharmacokinetics of trastuzumab deruxtecan (T-DXd),
24. Cahuzac H, Devel L. Analytical methods for the detection and a novel anti-HER2 antibody-drug conjugate, in HER2-positive
quantification of ADCs in biological matrices. Pharmaceuticals. tumour-bearing mice. Xenobiotica. 2020;50(10):1242–50.
2020;13(12):462. 42. Costa OR, Verhaeghen K, Roels S, Stangé G, Ling Z, Pipeleers D,
25. Jiang Y, Rex DAB, Schuster D, Neely BA, Rosano GL, Volkmar et al. An analytical comparison of three immunoassay platforms
N, et al. Comprehensive overview of bottom-up proteomics using for subpicomolar detection of protein biomarker GAD65. PLoS
mass spectrometry. ACS Meas Sci Au. 2024;4(4):338–417. One. 2018;13(3):e0193670.
26. Larson EJ, Zhu Y, Wu Z, Chen B, Zhang Z, Zhou S, et al. Rapid 43. Wessels U, Mackeben K, Stubenrauch K. Use of Ella® to facilitate
analysis of reduced antibody drug conjugate by online LC-MS/MS drug quantification and antidrug antibody detection in preclinical
with fourier transform ion cyclotron resonance mass spectrometry. studies. Bioanalysis. 2019;11(3):153–64.
Anal Chem. 2020;92(22):15096–103. 44. Zheng J, Mehl J, Zhu Y, Xin B, Olah T. Application and chal-
27. Yuan J, Tan HY, Huang Y, Rosenbaum AI. Digestion-free middle- lenges in using LC–MS assays for absolute quantitative anal-
down mass spectrometry method for absolute quantification of ysis of therapeutic proteins in drug discovery. Bioanalysis.
conjugated payload from antibody-drug conjugates. Anal Chem. 2014;6(6):859–79.
2024;96(42):16475–80. 45. Lee JW, Kelley M, King LE, Yang J, Salimi-Moosavi H, Tang
28. Li Y, Wang Y, Shenoy VM, Niu S, Jenkins GJ, Sarvaiya H. Intact MT, et al. Bioanalytical approaches to quantify “total” and
quantitation of cysteine-conjugated antibody-drug conjugates “free” therapeutic antibodies and their targets: technical chal-
using native mass spectrometry. Rapid Commun Mass Spectrom. lenges and PK/PD applications over the course of drug develop-
2024;38(15):e9774. ment. Aaps J. 2011;13(1):99–110.
29. Huang Y, Mou S, Wang Y, Mu R, Liang M, Rosenbaum AI. Char- 46. Zhou M, Lantz C, Brown KA, Ge Y, Paša-Tolić L, Loo JA,
acterization of antibody-drug conjugate pharmacokinetics and et al. Higher-order structural characterisation of native proteins
in vivo biotransformation using quantitative intact LC-HRMS and and complexes by top-down mass spectrometry. Chem Sci.
surrogate analyte LC-MRM. Anal Chem. 2021;93(15):6135–44. 2020;11(48):12918–36.
30. Spriggs FP, Zhong ZD, Safavi A, Jani D, Dontha N, Kant A, et al. 47. Pandeswari PB, Sabareesh V. Middle-down approach: a choice to
Ligand binding assays in the 21st Century laboratory: platforms. sequence and characterize proteins/proteomes by mass spectrom-
Aaps j. 2012;14(1):113–8. etry. RSC Adv. 2019;9(1):313–44.
31. Deshpande N, Suryawanshi PV, Tripathy S. Unveiling the 48. Zhang Y, Fonslow BR, Shan B, Baek M-C, Yates JR III. Pro-
quest: crafting an enzyme-linked immunosorbent assay tein analysis by shotgun/bottom-up proteomics. Chem Rev.
(ELISA) technique to uncover COVID-19 antibodies. Cureus. 2013;113(4):2343–94.
2024;16(8):e66659. 49. Chua ZQ, Prabhu GRD, Wang Y-W, Raju CM, Buchowiecki K,
32. Pei M, Liu T, Ouyang L, Sun J, Deng X, Sun X, et al. Enzyme- Ochirov O, et al. Moderate signal enhancement in electrospray
linked immunosorbent assays for quantification of MMAE-con- ionization mass spectrometry by focusing electrospray plume with
jugated ADCs and total antibodies in cynomolgus monkey sera. J a dielectric layer around the mass spectrometer’s orifice. Mol-
Pharm Analysis. 2022;12(4):645–52. ecules. 2024;29(2):316.
33. Mrosewski I, Dannheim V, Klett R, Urbank M, Stobbe S, Ittner 50. Hsu P-C, Urban PL. Electric field-modulated electrospray ioniza-
JR, et al. Rare coincidence: Macro-thyroid-stimulating hormone tion mass spectrometry for quantity calibration and mass tracking.
and multiple manufacturer-specific interferences in thyroid hor- J Am Soc Mass Spectrom. 2024;35(9):2064–72.
mone immunoassays. Ann Clin Biochem. 2024;61(6):484–9. 51. Xu T, Wang Q, Wang Q, Sun L. Mass spectrometry-intensive
34. Sankiewicz A, Zelazowska-Rutkowska B, Lukaszewski Z, Her- top-down proteomics: an update on technology advancements and
manowicz A, Gorodkiewicz E. An array SPRi biosensor for the biomedical applications. Anal Methods. 2024;16(28):4664–82.
determination of follicle-stimulating hormone in blood plasma. 52. Cochran D, Powers R. Fourier transform ion cyclotron resonance
Sensors (Basel). 2023;23(24):9686. mass spectrometry applications for metabolomics. Biomedicines.
35. Stelmachowska-Banaś M, Ostrowska M, Goszczyński TM, 2024;12(8):1786.
Kowalski K, Korbonits M, Kapuścińska R, et al. Macro-GH – A 53. Deslignière E, Diemer H, Erb S, Coliat P, Pivot X, Detappe A,
clinical entity causing a diagnostic challenge – A case report. Clin et al. A combination of native LC-MS approaches for the com-
Chim Acta. 2023;546:117392. prehensive characterization of the antibody-drug conjugate tras-
36. Soderstrom CI, Spriggs FP, Song W, Burrell S. Comparison of tuzumab deruxtecan. FBL. 2022;27(10).
four distinct detection platforms using multiple ligand binding 54. Robotham AC, Kelly JF. Chapter 1 - LC-MS characterization of
assay formats. J Immunol Methods. 2011;371(1):106–13. antibody-based therapeutics: recent highlights and future pros-
37. Skidmore L, Sakamuri S, Knudsen NA, Hewet AG, Milutinovic pects. In: Matte A, editor. Approaches to the Purification. Analysis
S, Barkho W, et al. ARX788, a site-specific anti-HER2 antibody- and Characterization of Antibody-Based Therapeutics: Elsevier;
drug conjugate, demonstrates potent and selective activity in 2020. p. 1–33.
HER2-low and T-DM1-resistant breast and gastric cancers. Mol 55. Wang T, Huang Z-A, Zhou M, Wang R, Li Y, Guo L, et al. Drug
Cancer Ther. 2020;19(9):1833–43. deconjugation-assisted peptide mapping by LC–MS/MS to iden-
38. Bolton JS, Chaudhury S, Dutta S, Gregory S, Locke E, Pierson tify conjugation sites and quantify site occupancy for antibody-
T, et al. Comparison of ELISA with electro-chemiluminescence drug conjugates. J Pharm Biomed Anal. 2024;243:116098.
The AAPS Journal (2025) 27:138 Page 15 of 15 138

56. Yang J, Tan HY, Yuan J, Huang Y, Rosenbaum AI. Detailed struc- Vicini P. Quantitative and systems pharmacology in the post-
tural elucidation of antibody-drug conjugate biotransformation genomic era: new approaches to discovering drugs and under-
species using high resolution multiple reaction monitoring mass standing therapeutic mechanisms - an NIH white paper by the
spectrometry with orthogonal dissociation methods. ACS Phar- QSP workshop group. 2011. Accessed on: 04 March 2025.
macol Transl Sci. 2025;8(1):113–23. 71. Kuepfer L, Niederalt C, Wendl T, Schlender J-F, Willmann S,
57. Yin F, Adhikari D, Li Y, Turner D, Shane Woolf M, Lebarbenchon Lippert J, et al. Applied concepts in PBPK modeling: how to
D, et al. A sensitive and rapid LC-MS/MS assay for quantitation build a PBPK/PD model. CPT Pharmacometrics Syst Pharmacol.
of free payload Aur0101 from antibody drug conjugate (ADC) 2016;5(10):516–31.
PYX-201 in human plasma. J Chromatogr B. 2023;1226:123786. 72. Lam I, Pilla Reddy V, Ball K, Arends RH, Mac GF. Develop-
58. Meesters R, Voswinkel S. Bioanalytical method development and ment of and insights from systems pharmacology models of anti-
validation: from the USFDA 2001 to the USFDA 2018 guidance body-drug conjugates. CPT Pharmacometrics Syst Pharmacol.
for industry. J Appl Bioanal. 2018;4(3):67–73. 2022;11(8):967–90.
59. Xu X, Liu Y, Yan Q, Bai C, Ran X, Li J, et al. Development 73. Gupta A, Michelini F, Shao H, Yeh C, Drago JZ, Liu D, et al.
and optimization of a high-throughput LC-MS/MS method for EGFR-directed antibodies promote HER2 ADC internalization
the simultaneous determination of Exatecan and its Cathepsin and efficacy. Cell Rep Med. 2024;5(11):101792.
B-sensitive prodrug, and ARV-825 in rat plasma: Application to 74. Lin S, Zhang Y, Yao J, Yang J, Qiu Y, Zhu Z, et al. DB-1314, a
pharmacokinetic study. J Pharm Biomed Anal. 2025;258:116746. novel DLL3-targeting ADC with DNA topoisomerase I inhibi-
60. Levengood MR, Zhang X, Hunter JH, Emmerton KK, Miyamoto tor, exhibits promising safety profile and therapeutic efficacy
JB, Lewis TS, et al. Orthogonal cysteine protection enables homo- in preclinical small cell lung cancer models. J Transl Med.
geneous multi-drug antibody-drug conjugates. Angew Chem Int 2024;22(1):766.
Ed. 2017;56(3):733–7. 75. Porter RJ, Murray GI, McLean MH. Current concepts in tumour-
61. Mckertish CM, Kayser V. A novel dual-payload ADC for the derived organoids. Br J Cancer. 2020;123(8):1209–18.
treatment of HER2+ breast and colon cancer. Pharmaceutics. 76. Lewis GD, Li G, Guo J, Yu S-F, Fields CT, Lee G, et al. The
2023;15(8):2020. HER2-directed antibody-drug conjugate DHES0815A in
62. Zhou ZZ, Si Y, Zhang J, Chen K, George A, Kim S, et al. A dual- advanced and/or metastatic breast cancer: preclinical characteri-
payload antibody-drug conjugate targeting CD276/B7-H3 elicits zation and phase 1 trial results. Nat Commun. 2024;15(1):466.
cytotoxicity and immune activation in triple-negative breast can- 77. Perez Bay AE, Faulkner D, DaSilva JO, Young TM, Yang K,
cer. Can Res. 2024;84(22):3848–63. Giurleo JT, et al. A bispecific METxMET antibody-drug con-
63. Mak SY, Chen S, Fong WJ, Choo A, Ho YS. A simple and highly jugate with cleavable linker is processed in recycling and late
sensitive LC–MS workflow for characterization and quantification endosomes. Mol Cancer Ther. 2023;22(3):357–70.
of ADC cleavable payloads. Sci Rep. 2024;14(1):11018. 78. Singh AP, Sharma S, Shah DK. Quantitative characterization of
64. Hu X-Y, Wang R, Jin J, Liu X-J, Cui A-L, Sun L-Q, et al. An in vitro bystander effect of antibody-drug conjugates. J Pharma-
EGFR-targeting antibody–drug conjugate LR004-VC-MMAE: cokinet Pharmacodyn. 2016;43(6):567–82.
potential in esophageal squamous cell carcinoma and other malig- 79. Cekanova M, Rathore K. Animal models and therapeutic molecu-
nancies. Mol Oncol. 2019;13(2):246–63. lar targets of cancer: utility and limitations. Drug Des Devel Ther.
65. Suh M-J, Powers JB, Daniels CM, Wu Y. Enhanced pharma- 2014;8:1911–21.
cokinetic bioanalysis of antibody–drug conjugates using hybrid 80. Ascione L, Crimini E, Trapani D, Marra A, Criscitiello C, Curigli-
immunoaffinity capture and microflow LC-MS/MS. AAPS J. ano G. Predicting response to antibody drug conjugates: a focus
2023;25(4):68. on antigens’ targetability. Oncologist. 2023;28(11):944–60.
66. Yin F, Adhikari D, Peay M, Cortes D, Garada M, Shane Woolf 81. Williams M, Spreafico A, Vashisht K, Hinrichs MJ. Patient selec-
M, et al. Development and validation of a hybrid immunoaffinity tion strategies to maximize therapeutic index of antibody-drug
LC–MS/MS assay for quantitation of total antibody (TAb) from conjugates: prior approaches and future directions. Mol Cancer
an antibody drug conjugate (ADC) PYX-201 in human plasma. J Ther. 2020;19(9):1770–83.
Chromatogr B. 2023;1228:123844. 82. Beck A, Goetsch L, Dumontet C, Corvaïa N. Strategies and chal-
67. Ascione L, Guidi L, Prakash A, Trapani D, LoRusso P, Lou lenges for the next generation of antibody–drug conjugates. Nat
E, et al. Unlocking the potential: biomarkers of response to Rev Drug Discovery. 2017;16(5):315–37.
antibody-drug conjugates. Am Soc Clin Oncol Educ Book. 83. Walsh SJ, Iegre J, Seki H, Bargh JD, Sore HF, Parker JS, et al. Gen-
2024;44(3):e431766. eral dual functionalisation of biomacromolecules via a cysteine
68. Li C, Chen SC, Chen Y, Girish S, Kaagedal M, Lu D, et al. bridging strategy. Org Biomol Chem. 2020;18(22):4224–30.
Impact of physiologically based pharmacokinetics, population 84. Dagogo-Jack I, Shaw AT. Tumour heterogeneity and resistance to
pharmacokinetics and pharmacokinetics/pharmacodynamics in cancer therapies. Nat Rev Clin Oncol. 2018;15(2):81–94.
the development of antibody-drug conjugates. J Clin Pharmacol. 85. Journeaux T, Bernardes GJL. Homogeneous multi-payload anti-
2020;60(Suppl 1):S105-s19. body–drug conjugates. Nat Chem. 2024;16(6):854–70.
69. Bradshaw EL, Spilker ME, Zang R, Bansal L, He H, Jones 86. Tao J, Gu Y, Zhou W, Wang Y. Dual-payload antibody–drug con-
RDO, et al. Applications of quantitative systems pharmacol- jugates: Taking a dual shot. Eur J Med Chem. 2025;281:116995.
ogy in model-informed drug discovery: perspective on impact
and opportunities. CPT Pharmacometrics Syst Pharmacol. Publisher's Note Springer Nature remains neutral with regard to
2019;8(11):777–91. jurisdictional claims in published maps and institutional affiliations.
70. Peter K. Sorger SRBA, Darrell R. Abernethy, Russ B. Altman,
Kim L. R. Brouwer, Andrea Califano, David Z., D'Argenio RI,
William J. Jusko, Richard Lalonde, Douglas A. Lauffenburger,,
Brian Shoichet JLS, Shankar Subramaniam, Piet Van der Graaf,