Skip to main content

Joint Disposition Properties and Comprehensive Pharmacokinetic Characterization of Antibody–Drug Conjugates

Abstract

Antibody–drug conjugates (ADCs) comprise 3 distinct parts: a specific antibody carrier (mAb), a linker, and a cytotoxic payload. Typical pharmacokinetic (PK) characterization of ADCs remains fragmented using separate noncompartmental analyses (NCA) of individual analytes, offering little insight into the dynamic relationships among the ADC components, and the safety and efficacy implications. As a result, it is exceedingly difficult to compare ADCs in terms of favorable PK characteristics. Therefore, there is a need for characterizing ADCs using the joint disposition properties critical for understanding the fate of an ADC complex and clinical implications. In this communication, we describe 3 joint disposition metrics (JDMs) for integrated NCA of ADCs based on a combination of common analytes of ADC, payload, conjugated payload, and total mAb. These JDMs were derived, each in a simple form of a ratio between appropriate PK parameters of two analytes, from the presumed drug delivery scheme behind typical ADC designs, in terms of (1) linker stability, (2) therapeutic exposure ratio, and (3) effective drug-to-antibody ratio in vivo. The validity of the JDM-based PK characterization was examined against model-based analyses via their applications to 3 clinical candidates: PF-06650808, PF-06647020, and PF-06664178. For instance, the linker stability estimates for PF-06650808, PF-06647020, and PF-06664178 were 0.31, 0.14, and 0.096, respectively, from the JDM-based analyses vs. 0.23, 0.11, and 0.086 by the model-based approach. Additionally, the JDMs were estimated for a number of FDA-approved or otherwise well-documented ADCs, showing their utilities in comparing ADCs in terms of favorable PK characteristics.

Graphical Abstract

INTRODUCTION

ADCs are one of the fast-growing classes of anticancer agents which comprise mAbs conjugated to cytotoxic payloads via synthetic linkers. The central idea behind ADC designs is to improve the therapeutic index (the ratio of the maximum tolerated to the minimally efficacious doses) which is generally low for cytotoxic drugs used in conventional chemotherapy. As such, the mAb is used as a specific carrier for optimal delivery of the cytotoxic payload to targeted cancer cells, while leaving normal cells relatively unscathed, which typically involves the following steps: ADC binding to the antigen on targeted cancer cells, internalization, and deconjugation/degradation in the endo-lysosomal system releasing the payload into the cytoplasm. The linker plays an important role in the optimal payload delivery. It allows an efficient release of the payload following internalization, while ensuring the ADC stability in off-target tissues such that the ADC is maintained in a nontoxic state. The majority of payloads utilized in ADCs are highly potent, often cytotoxic in the picomolar range, as only a small fraction (< 1–2%) of the dose localizes to the cancer cells (1,2,3). However, because of their potencies, payloads by far drive the toxicities of ADCs (4). Although the concept of ADCs is relatively straightforward, there are significant challenges at various aspects of ADC development compared to single-component drug entities, such as small molecular drugs and monoclonal antibody biologics. One of these areas is the PK characterization of ADCs, especially when a comparative PK analysis is performed to determine the impact of a specific design change, such as adoption of a new mAb, linker, or conjugation process, relative to the desired disposition characteristics.

Typical PK characterization of ADCs remains fragmented using separate NCA of individual analytes related to its conjugated or unconjugated components, as if these analytes were not related (5,6,7,8,9), providing no or little insight into the dynamic relationships between the ADC components. These relationships are critical for understanding of the PK of a given ADC as well as its safety and efficacy implications. For instance, when a single-analyte-based PK analysis reveals that the ADC under consideration has a short half-life, it offers no further information on what extent to which this instability finding is due to the linker (deconjugation) or mAb carrier (degradation), nor does it inform about its implication for the therapeutic index. In recent years, there have been increased efforts to analyze more than one analytes together using a model-based approach. These analyses were performed in a population PK analysis framework for the purpose of summarizing the PK exposure data across studies and identifying the covariate effects (10,11,12,13), or in a more mechanistic framework for a comprehensive description of ADC disposition (e.g., Shah et al.(14)). In the latter case, the number of model parameters tends to be large relative to the amount of PK information from typical clinical studies, especially during the early clinical development, such that it can be quite challenging to obtain reliable model parameter estimates. Therefore, there is a need for integrated NCA of ADCs in terms of the joint disposition properties that are critical for understanding the fate of an ADC complex as a whole and its clinical implications.

In this communication, we present 3 JDMs for integrated NCA of ADCs in terms of the linker stability, therapeutic exposure ratio, and effective drug–antibody ratio (DAR) in vivo. As a result, these JDMs define what constitute a PK profile favorable for an ADC. The JDMs were derived, in a form of a ratio between appropriate PK parameters of two analytes, from the presumed drug delivery scheme behind typical ADC designs based on a combination of commonly used analytes. The JDMs, as well as the presumed framework, were further examined for verification using model-based approaches via their applications to 3 clinical ADC candidates, with the PK data from their respective first-in-patient (FIP) studies. In addition, utilities of the JDMs were explored for comparative assessments among ADCs in terms of favorable PK characteristics, using FDA-approved or otherwise well-documented ADC products and candidates.

THEORY

Framework Description

Figure 1 illustrates the presumed drug delivery scheme or disposition framework behind typical ADC designs based on commonly used bioanalytical measurements for concentrations of payload, ADC (or conjugated mAb), and total mAb (TAb, ADC plus unconjugated mAb) in the blood. The mathematical representation of a full version of the framework, including the target mediated degradation process, is provided in Table S1.

Fig. 1
figure 1

Following entering the system, the ADC undergoes distribution between the assigned compartments and elimination, via a first-order process. Two elimination pathways operate in parallel, one postulated as a deconjugation (CLdc) and the other as a degradation (CLdg), both resulting in the release of payload, with only the deconjugation producing unconjugated mAb. In certain cases, a third elimination pathway involving the ADC-target complex (ADC-R) or target-mediated degradation (CLAR) may be added. In addition, the deconjugation pathway may be approached by using sequential steps for further elucidating the payload release from the deconjugation process

The framework is centered around the basic structure that captures the joint disposition properties critical for understanding of the drug delivery scheme. The basic structure delineates the payload release from ADC elimination through two parallel processes, deconjugation and degradation, by leveraging the information on unconjugated mAb resulting from the deconjugation process. Additionally, it assumes that the ADC and unconjugated mAb exhibit one-compartment disposition characteristics sharing the same distribution space (VA) and the payload has two-compartment (V1 and V2) disposition characteristics, with ADC administration via intravenous (iv) infusion over a duration of T (Table S1). The concentrations of ADC (\({C}_{ADC}\)), unconjugated mAb (\({C}_{mAb}\)), and payload (\({C}_{PL,1}\)) may be expressed as follows:

$${C}_{ADC}=\frac{-{k}_{0}(1-{e}^{\frac{{CL}_{ADC}}{{V}_{A}}\cdot IT})}{{CL}_{ADC}}{e}^{-\frac{{CL}_{ADC}}{{V}_{A}}\cdot t}$$
(1.1)
$${C}_{mAb}=\frac{-{CL}_{dc}\cdot {k}_{0}(1-{e}^{\frac{{CL}_{ADC}}{{V}_{A}}\cdot IT})}{{CL}_{ADC}({CL}_{mAb}-{CL}_{ADC})}{e}^{-\frac{{CL}_{ADC}}{{V}_{A}}\cdot t}+\frac{{-CL}_{dc}\cdot {k}_{0}(1-{e}^{\frac{{CL}_{mAb}}{{V}_{A}}\cdot IT})}{{CL}_{mAb}({CL}_{ADC}-{CL}_{mAb})}{e}^{-\frac{{CL}_{mAb}}{{V}_{A}}\cdot t}$$
(1.2)
$$C_{PL\mathit,\mathit1}=\frac{DAR\cdot f\cdot k_0}{V_1}\frac{{CL}_{ADC}}{V_A}\left\{\frac{-(\frac{Q_{PL}}{V_2}-\alpha)(1-e^{\alpha\cdot IT})}{\alpha(\beta-\alpha)(\frac{{CL}_{ADC}}{V_A}-\alpha)}e^{-\alpha\cdot t}+\frac{-(\frac{Q_{PL}}{V_2}-\beta)(1-e^{\beta\cdot IT})}{\beta(\alpha-\beta)(\frac{{CL}_{ADC}}{V_A}-\beta)}e^{-\beta\cdot t}\right.\left.+\frac{-(\frac{Q_{PL}}{V_2}-\frac{{CL}_{ADC}}{V_A})(1-e^{\frac{{CL}_{ADC}}{V_A}\cdot IT})}{\frac{{CL}_{ADC}}{V_A}(\alpha-\frac{{CL}_{ADC}}{V_A})(\beta-\frac{{CL}_{ADC}}{V_A})}e^{-\frac{{CL}_{ADC}}{V_A}\cdot t}\right\}$$
(1.3)

Here \({k}_{0}\) is the infusion rate and IT is the infusion time (IT = t during infusion and IT = T after the end of infusion); DAR is the drug (payload)–antibody ratio; \({CL}_{ADC}\) and \({CL}_{mAb}\) are the ADC and mAb clearance, and \({CL}_{ADC}\) has the deconjugation and degradation components, i.e., \({CL}_{ADC}={CL}_{dc}+{CL}_{dg}\); and \({Q}_{PL}/{V}_{1}\) and \({Q}_{PL}/{V}_{2}\) are the rate constants of payload intercompartmental distribution and they have the following relationship to other rate constants: α + β = \(\frac{{Cl}_{PL}}{{V}_{1}}+\frac{{Q}_{PL}}{{V}_{1}}+\frac{{Q}_{PL}}{{V}_{2}}\) and αβ = \(\frac{{Cl}_{PL}}{{V}_{1}}+\frac{{Q}_{PL}}{{V}_{2}}\). Furthermore, it is assumed for simplicity that all payload drug (f = 100%) appears in \({V}_{1}\) following its release from the ADC degradation and deconjugation processes. In cases where \({CL}_{mAb}\) can be approximated by \({CL}_{dg}\), Eq. (1.2) may be simplified with \({CL}_{dc}={CL}_{ADC}-{CL}_{mAb}\).

Additional insight may be gained when the conjugated payload (CPL) information is also available and certain assumptions regarding the deconjugation process can be made. For further elucidating the linker stability property, under the assumption that the deconjugation process consists of DAR sequential steps starting with conversion of the ADC with a payload substitution level of DAR, \({X}_{ADC}^{DAR}\), to the one with DAR-1 substitution level, \({X}_{ADC}^{DAR-1}\), and ending with conversion of the ADC with 1 substitution level, \({X}_{ADC}^{1}\), to \({X}_{mAb}\), one may obtain an alternative version of the basic structure, with a more detailed description of deconjugation pathways for the mAb formation (also see Table S2):

$$\frac{dX_{ADC}}{dt}=-\frac{{CL}_{dc}^1}{V_A}X_{ADC}^1-\frac{{CL}_{dg}}{V_A}X_{ADC}+In(k_0,T)$$
(1.4)
$$\frac{d{X}_{mAb}}{dt}=-\frac{{CL}_{mAb}}{{V}_{A}}{X}_{mAb}+\frac{{CL}_{dc}^{1}}{{V}_{A}}{X}_{ADC}^{1}$$
(1.5)

and for the payload release:

$$\frac{dX_{CPL}}{dt}=-\sum\nolimits_{i=1}^{DAR}\frac{{CL}_{dc}^i}{V_A}X_{ADC}^i-\frac{{CL}_{dg}}{V_A}X_{CPL}+DAR\cdot In(k_0,T)$$
(1.6)
$$\frac{d{X}_{PL,1}}{dt}=\sum\nolimits_{i=1}^{DAR}\frac{{CL}_{dc}^{i}}{{V}_{A}}{X}_{ADC}^{i}+\frac{{CL}_{dg}}{{V}_{A}}{X}_{CPL}-\frac{{CL}_{PL}}{{V}_{1}}{X}_{PL,1}-\frac{{Q}_{PL}}{{V}_{1}}{X}_{PL,1}+\frac{{Q}_{PL}}{{V}_{2}}{X}_{PL,2}$$
(1.7)

where In(k0,T) is the drug input at a rate of \({k}_{0}\) over an interval of T, and \({CL}_{dc}^{i}\) is the clearance for deconjugation step i (i = 1, 2,..., DAR). The concentration of ADC with i substitution level can be expressed as (Table S2)

$${C}_{ADC}^{i}=\frac{{X}_{ADC}^{i}}{{V}_{A}}={k}_{0}\prod_{j=i+1}^{DAR}{CL}_{dc}^{j} {\sum }_{l=i}^{DAR}\frac{(1-{e}^{\frac{{CL}_{ADC}^{l}}{{V}_{A}}IT})}{{CL}_{ADC}^{l}{\prod }_{\begin{array}{c}j=0\\ j\ne l\end{array}}^{i}({CL}_{ADC}^{j}-{CL}_{ADC}^{l})}{e}^{-\frac{{CL}_{ADC}^{l}}{{V}_{A}}\cdot t}$$
(1.8)

where \({CL}_{ADC}^{i}=\) \({CL}_{dc}^{i}+{CL}_{dg}\). Note that \({C}_{ADC}={\sum }_{i=1}^{DAR}{C}_{ADC}^{i}\) and \({C}_{CPL}={\sum }_{i=1}^{DAR}{i\cdot C}_{ADC}^{i}\). It is clear from Eqs. (1.4) and (1.6) that the only last deconjugation step producing mAb is accounted for in the ADC elimination, whereas every deconjugation step is accounted for in the CPL elimination.

ADC Joint Disposition Properties

Within the framework described by Eqs. (1.1)–(1.8), the ADC joint disposition properties representing the linker stability, therapeutic exposure ratio, and effective DAR in vivo can be defined in a form of combining appropriate exposure or (clearance) parameters from two individual components/analytes, as summarized in Table I, based on the following exposure relationship to the dose \({D}_{j}\) for individual analyte j, after single-dose administration:

Table I Joint Disposition Metrics and Annotations
$$D_j={CL}_j\int_0^\infty{C(t)}_jdt={{CL}_j\cdot AUC}_j$$
(2.1)

or, at steady state (over τ after the nth dose), \({D}_{j}={CL}_{j}{\int }_{n\tau }^{(n+1)\tau }{C(t)}_{j} dt=\) \({CL}_j\cdot{AUC}_j^\tau\).

The ratio of \({AUC}_{mAb}/{AUC}_{ADC}\), or alternatively \({AUC}_{mAb}/{AUC}_{TAb}\), measures the linker stability in terms of the amount of the ADC dose eliminated via deconjugation. Note that \({{AUC}_{mAb}=AUC}_{TAb}-{AUC}_{ADC}\). The ratio of \({AUC}_{mAb}/{AUC}_{ADC}\) can be expressed as

$$\frac{{AUC}_{mAb}}{{AUC}_{ADC}}=\frac{{{CL}_{mAb}AUC}_{mAb}}{{{CL}_{mAb}AUC}_{ADC}}$$
(2.2)

The numerator of the right side represents the amounts of the ADC dose eliminated via deconjugation. For the basic structure where \({{CL}_{mAb}AUC}_{mAb}={{CL}_{dc}AUC}_{ADC}\), Eq. (2.2) may be written as

$$\frac{{AUC}_{mAb}}{{AUC}_{ADC}}=\frac{{CL}_{dc}}{{CL}_{mAb}}$$
(2.3)

i.e., the ratio of \({AUC}_{mAb}/{AUC}_{ADC}\) represents the ADC deconjugation clearance relative to the mAb degradation clearance. For the alternative structure (Eqs. (1.4)–(1.7)) in which \({{CL}_{mAb}AUC}_{mAb}={CL}_{dc}^{1}{AUC}_{ADC}^{1}\), Eq. (2.2) may be written as

$$\frac{{AUC}_{mAb}}{{AUC}_{ADC}}=\frac{{CL}_{dc}^{^{\prime}}}{{CL}_{mAb}}$$
(2.4)

where \({CL}_{dc}^{^{\prime}}={CL}_{dc}^{1}\frac{{AUC}_{ADC}^{1}}{{AUC}_{ADC}}\), the deconjugation clearance adjusted for the fraction of \({AUC}_{ADC}\) contributed from \({AUC}_{ADC}^{1}\) (also see Table S2). In cases where the ADC degradation clearance (\({CL}_{dg}\)) is comparable to the mAb clearance, the ratio of \({AUC}_{mAb}/{AUC}_{ADC}\) is a measure of the amount of ADC eliminated via deconjugation relative to that via degradation. In these cases, similar to Eq. (2.3) or Eq. (2.4), it can be shown that the ratio of \({AUC}_{mAb}/{AUC}_{TAb}\) is an alternative measure of the linker stability in terms of the amount of ADC eliminated via deconjugation relative to the total ADC dose, i.e.,

$$\frac{{AUC}_{mAb}}{{AUC}_{TAb}}=\frac{{{CL}_{mAb}AUC}_{mAb}}{{{CL}_{mAB}({AUC}_{mAb}+AUC}_{ADC})}=\frac{{{CL}_{dc}AUC}_{ADC}}{{{CL}_{dc}{AUC}_{ADC}+{CL}_{dg}AUC}_{ADC}}=\frac{{CL}_{dc}}{{Cl}_{ADC}} ,\mathrm{ or }\frac{{AUC}_{mAb}}{{AUC}_{TAb}}=\frac{{CL}_{dc}^{^{\prime}}}{{CL}_{ADC}}$$
(2.5)

When the linker is highly stable or \({CL}_{dc}\) is negligible, no or little unconjugated mAb is produced and both ratios of \({AUC}_{mAb}/{AUC}_{TAb}\) and \({AUC}_{mAb}/{AUC}_{ADC}\) approach 0. When the linker is highly unstable, the ratio of \({AUC}_{mAb}/{AUC}_{TAb}\) may take a value close to 1, while the ratio of \({AUC}_{mAb}/{AUC}_{ADC}\) can far exceed 1.

The ratio of \({CL}_{PL}/{CL}_{CPL}\), or alternatively \({CL}_{PL}/{CL}_{ADC}\), provides a measure of the therapeutic index for the drug delivery scheme behind typical ADC designs. For this purpose, the term, “therapeutic exposure ratio” is used which determines what level of \({AUC}_{CPL}\) (or \({AUC}_{ADC}\)) can be achieved when the \({AUC}_{PL}\) is at its maximum tolerated level, and it is defined by the ratio of \({CL}_{PL}/{CL}_{CPL}\) (or \({CL}_{PL}/{CL}_{ADC}\)) as shown below. An ADC with improved therapeutic index should have a greater exposure ratio allowing a higher \({AUC}_{CPL}\) (or \({AUC}_{ADC}\)) to ensure delivery of the payload doses to cancer cells required for the therapeutic effect, while maintaining \({AUC}_{PL}\) below its maximum tolerated level. With Eq. (2.1), recognizing that \({D}_{CPL}={D}_{PL}\) and the assumption that f = 100%, it follows that

$${AUC}_{CPL}/{AUC}_{PL}= {CL}_{PL}/{CL}_{CPL}$$
(2.6)

Similarly, recognizing that in a molar unit \({D}_{PL}=DAR\cdot {D}_{ADC}\), it follows that

$${AUC}_{ADC}/{AUC}_{PL}=(1/DAR)\cdot{CL}_{PL}/{CL}_{ADC},\mathrm{or}\;{DAR\cdot AUC}_{ADC}/{AUC}_{PL}={CL}_{PL}/{CL}_{ADC}$$
(2.7)

It should be apparent that, for the basic structure (Eqs. (1.1)–(1.4)), \({CL}_{ADC}={CL}_{CPL}\) as both are defined by the same \({CL}_{dg}\) and \({CL}_{dc}\). For the alternative structure (Eqs. (1.4)–(1.7)), while both \({CL}_{ADC}\) and \({CL}_{CPL}\) share the same \({CL}_{dg}\) component, their deconjugation components are different (Table S2). Note that the resulting difference between \({CL}_{ADC}\) and \({CL}_{CPL}\) should be minor, when the deconjugation contribution is small relative to the degradation counterpart.

The ratio of \({AUC}_{CPL}/{AUC}_{ADC}\) is the mean DAR (\(\overline{DAR }\)), a measure of the effective DAR in vivo relative to the nominal value of DAR. The number of payload molecules (DAR(t)) released per one ADC molecule eliminated at time t following ADC administration may be expressed as the ratio of molar concentration of CPL over that of ADC:

$$DAR\left(t\right)={C(t)}_{CPL}/{C(t)}_{ADC}$$
(2.8)

From the moment analysis, the \(\overline{DAR }\) is given by

$$\overline{DAR }= {\int }_{0}^{\infty }DAR(t)\left\{\frac{{C(t)}_{ADC}}{{\int }_{0}^{\infty }{C(t)}_{ADC}dt}\right\}dt$$
(2.9)

Incorporating Eq. 2.8 into Eq. 2.9 leads

$$\overline{DAR }= {\int }_{0}^{\infty }{C(t)}_{CPL}dt/{\int }_{0}^{\infty }{C(t)}_{ADC}dt$$
(2.10)

i.e., the ratio of \({AUC}_{CPL}/{AUC}_{ADC}\) is the mean DAR(t) in vivo. Conceptually, it is an average number of payload molecules that could be released per one ADC molecule internalized, taking a value between 1 and DAR, largely dependent on the specifics of deconjugation steps. Therefore, relative to the nominal DAR value, it is a measure of the payload delivery effectiveness on a 1 to DAR scale.

MATERIALS AND METHODS

The JDMs, as well as the ADC disposition framework from which the JDMs were derived, were further evaluated through the following case studies: (1) PK characterization of 3 clinical ADC candidates using the JDM- and model-based approaches, (2) the impact of linker stability on ADC disposition via simulations using Eqs. (1.4)–(1.7), and (3) utilities of the JDMs for assessing among ADCs in terms of favorable PK characteristics.

Model-based PK Characterization of ADC Candidates

The presumed ADC disposition framework is based on the basic structure (Eqs. (1.1)–(1.3)) which, in a concise form, captures all dynamic relationships among analytes of ADC, payload, and TAb in the blood. The model-based PK characterization was carried out in case study 1 to verify the presumed ADC disposition framework against the corresponding PK observations with 3 clinical ADC candidates: PF-06650808, PF-06647020, and PF-06664178. The main interest of the study was to determine the individual contributions to the total ADC clearance by deconjugation and degradation using the JDM- and model-based approaches. PF-06650808 is an anti-Notch3 ADC composed of the humanized IgG1 mAb linked to an auristatin-based cytotoxic payload (Aur0101) utilizing a valine-citrulline (vc)-based enzymatically cleavable linker with a DAR of 4 (15, 16). PF-06647020 is an ADC comprising a humanized anti-PTK7 IgG1 mAb joined to Aur0101 by a cleavable vc-based linker with a DAR of 4 (17, 18). PF-06664178 is an ADC composed of a humanized anti-Trop-2 IgG1 mAb conjugated using a cleavable AcLys-vcAur0101 linker-payload with a DAR of 2 (19, 20).

Cycle 1 ADC, payload, and TAb concentration–time data for PF-06650808, PF-06647020, and PF-06664178 were obtained from their FIP studies (ClinicalTrials.gov ID: NCT02129205, NCT02222922, and NCT02122146, respectively). The PF-06647020 FIP study consisted of both dose escalation and expansion phases while the other two were limited to only the dose escalation. The design and procedures of each study have been previously described (16, 18, 20). The PK samples for ADC, payload, and TAb were collected at pre-dose and various times post-dose (Table S5) and quantified using validated bioanalytical procedures (Table S6).

The basic structure of the framework, Eqs. (1.1)–(1.3), was used to characterize the ADC, payload, and TAb concentration–time data for PF-06650808 at doses of 0.2 to 6.4 mg/kg every 3 weeks, and those for PF-06664178 at doses of 0.15 to 4.8 mg/kg every 3 weeks. The model-based characterizations assumed the degradation clearance for mAb is the same as that for ADC (\({CL}_{dg}\)). For PF-06647020, a target-mediated process was added to the basic structure, as the PK of PF-06647020 appeared to behave in a dose-dependent manner over a dose range of 0.2 to 3.7 mg/kg (18). In this expanded model, it is assumed that the distribution volumes for target and ADC-target complex are the same as that for ADC. The mathematical model description is provided in Table S3. Additional information on stochastic model components for both the basic and expanded structures is provided in Table S3. While estimations of the disposition parameters were the primary focus of the modeling exercise, typical model performance evaluations were also conducted.

Nonlinear mixed-effect model parameter estimations were performed using NONMEM version 7.4.3 or higher (Ellicott City, MD, USA). The ADVAN6 and first-order conditional estimation method with interaction were used for the analysis. Bootstrap runs were performed using the software package Perl-speaks-NONMEM (Version 4.9.0 or higher, Uppsala Pharmacometrics, Uppsala, Sweden). R (version 3.6.2) was used to visualize the NONMEM outputs.

The Effect of Linker Stability on ADC Disposition

The PK profiles of two versions of an ADC candidate with a DAR of 4 were simulated in case study 2 using the alternative version of the basic structure of which the deconjugation process consists of DAR sequential steps as described (Eqs. (1.4)–(1.7)), with version 2 having a more stable linker slowing down the deconjugation process compared to version 1. The goal of this case study was to show the effects of a reduced deconjugation on disposition of the ADC, as determined by the integrated NCA vs. the separate NCA. The concentration–time profiles of ADC, payload, TAb, and CPL were simulated using the parameters given in Table S4. As a reference, the PK profile of payload delivered as a bolus at the same dose as carried by the ADC was also generated. The AUC for each analyte was estimated using the standard NCA method, and the JDMs,\({AUC}_{mAb}/{AUC}_{TAb}\),\({AUC}_{CPL}/{AUC}_{PL}\), and\({AUC}_{CPL}/{AUC}_{ADC}\), were calculated as defined by Eqs. (2.5), (2.6), and (2.10), respectively.

Utilities of JDMs for Assessing Among ADCs in Terms of Favorable PK Characteristics

Utilities of the JDMs were explored for comparing among a group of ADCs that share one or more common ADC design element(s), in terms of whether or to what extent they have a favorable PK profile relative to others in the group, in case study 3. For this purpose, a number of ADCs were selected based on the criteria that their safety and anticancer activity profiles have been well documented previously in the literature and that the PK information is available for ADC, payload, and TAb. All current FDA-approved ADCs were examined, and three did not have the same set of PK information as the rest in terms of analytes used and were not included for the utility assessments. The included ADCs were loosely divided into groups based on their commonality. In addition, the JDMs were calculated for PF-06650808, PF-06647020, and PF-06664178 for comparisons with those obtained from the model-based approach.

The JDMs, \({AUC}_{mAb}/{AUC}_{TAb}\), \({CL}_{PL}\)/\({CL}_{ADC}\), and \({AUC}_{CPL}/{AUC}_{ADC}\), were estimated using an appropriate ratio between the reported means of \({AUC}_{ADC}\), \({AUC}_{TAb}\), \({AUC}_{mAb}\)(\({AUC}_{TAb}\)- \({AUC}_{ADC}\)), \({AUC}_{CPL}\), \({AUC}_{PL}\), \({CL}_{ADC}\), and \({CL}_{PL}\) at the approved label dose or maximum tolerated dose (MTD). For 3 ADCs where more than one means were available from multiple studies/groups, these means were averaged with each weighed by its sample size.

RESULTS

Totals of 320, 792, and 237 PK blood samples (for concentrations of ADC and payload and TAb) from 40, 101, 31 patients for PF-06650808, PF-06647020, and PF-06664178, respectively, were used for the model parameter estimations in case study 1. The estimates of clearance and volume parameters of interest, including the 90% confidence intervals from bootstrap procedures, are shown in Table II. Individual predictions for concentrations of ADC, payload, and TAb are shown in comparison with the respective observed concentrations, for each ADC, in Figs. S1S3. The results from model performance evaluations are provided in Figs. S4S6. The visual predictive check (VPC) plots for PK profiles of PF-06647020 ADC, TAb, and payload for the 2.8 mg/kg expansion group are shown in Fig. 2.

Table II Disposition Parameter Estimates (90% CIs§) Based on the Basic Structure for PF-06664178 and PF-06650808, and Expanded Structure for PF-06647020
Fig. 2
figure 2

VPC plots for PK profiles of PF-06647020 ADC (a), TAb (b), and payload (c), where the dashed lines represent the 5th percentile, median, and 95th percentile for the observed concentrations; the solid lines represent the medians of the simulated concentrations and the shaded areas indicate the 5th and 95th percentiles of the simulated concentrations; and the dotted lines represent the levels of low limit of quantitation (LLOQ) of the assays, and all observations below the LLOQ were placed at 0.5 times the respective LLOQ level

Figure 3 shows the results from the case study 2 in which the concentration–time profiles for ADC, TAb, payload, and CPL were simulated based on Eqs. (1.4)–(1.7) for 2 versions of ADCs with only a difference in the linker stability. The estimates of relevant single-analyte-based PK parameters and those of JDMs are provided in Table S4.

Fig. 3
figure 3

The PK profiles of two versions of an ADC candidate were simulated using Eqs. (1.4)–(1.7), with version 2 (ADC (V2)) having a more stable linker slowing down the deconjugation process compared to version 1 (ADC (V1)), to show the effects of a reduced deconjugation on disposition of ADC, TAb, payload and CPL, and on DAR. In addition, the PK profile of payload after iv bolus at the same dose carried by the ADC was simulated as a reference (see Table S4 for the case study details)

A total of 10 ADC products and candidates were used in case study 3 in exploring utilities of the JDMs for the assessment among ADCs in terms of favorable PK characteristics (5,6,7,8,9, 16, 18, 20,21,22). The calculated JDMs are shown in Table III together with other related information, in groups of anti-HER2 mAb, acid labile linker, and MMAE or Aur0101 payload.

Table III Evaluation of ADCs in Terms of Favorable PK Characteristics Using the Joint Disposition Metrics

DISCUSSION

PK characterization of ADCs is typically carried out using separate NCA of individual analytes, offering no or little insight into the dynamic relationships among the ADC components, and the safety and efficacy implications. The integrated NCA proposed here takes full advantage of commonly used ADC analytes, i.e., concentrations of ADC, payload, conjugated payload, and total mAb in the blood, by sorting out how these ADC analytes are related to one another in the drug delivery scheme behind typical ADC designs and what clinical implication each relationship is associated with. As a result, it characterizes the whole ADC complex in terms of the joint disposition properties representing the linker stability, therapeutic exposure ratio, and effective DAR in vivo. To the best of our knowledge, for the first time, the concept of the ADC joint disposition properties and the utility for integrated NCA are described. It has been exceedingly difficult, if not impossible, to compare ADCs in terms of whether or to what extent they have a favorable PK profile due to the lack of an integrated NCA framework for characterizing ADCs in terms of the joint disposition properties critical for understanding the fate of an ADC complex and its clinical implications.

ADC Joint Disposition Metrics

The ratio of \({AUC}_{mAb}/{AUC}_{TAb}\) (or \({AUC}_{mAb}/{AUC}_{ADC}\)) offers a measure of the linker stability in terms of ADC loss from deconjugation. A ratio of \({AUC}_{mAb}/{AUC}_{TAb}\) close to zero indicates the ADC has a highly stable linker, whereas a ratio close to the opposite end of 1 corresponds to an unstable linker. The ratio of \({CL}_{PL}\)/\({CL}_{CPL}(\mathrm{or}\;{CL}_{PL}\)/\({CL}_{ADC}\)) provides a measure of the therapeutic exposure ratio of an ADC in terms of the ability to attain the exposure to CPL (or ADC) at a sufficiently high level to ensure the delivery of required payload doses to cancer cells, while maintaining exposures to the payload below the maximum tolerated level. For achieving a wider margin between exposures to CPL (or ADC) and payload as a more favorable PK characteristic, this ratio calls for the use of a mAb carrier with a lower \({CL}_{CPL}\) (or \({CL}_{ADC}\)) and a payload with a higher \({CL}_{PL}\) (Table III).

The ratio of \({AUC}_{CPL}/{AUC}_{ADC}\) is a measure of the effective DAR in vivo as an average number (\(\overline{DAR }\)) of payload molecules that could be released per one ADC molecule internalized. It is conceivable that the \(\overline{DAR }\) takes a value close to 1 at one end where the \({C}_{ADC}^{1}\) species quickly predominates in the entire ADC population and DAR at the other end where the \({C}_{ADC}^{DAR}\) species is present predominantly (Eq. (1.8)). An example of the \(\overline{DAR }\) having a low value would be with the alternative structure (Eqs. (1.4)–(1.7)) when the rates for all deconjugation steps i > 1 are extremely fast and that for step i = 1 is slow. An example of the other extreme would be a case where the deconjugation is negligible or it takes place with all payloads being simultaneously released at the same rate for each ADC molecule, such that the number of payload molecules [\(DAR(t)\)] released per one ADC molecule eliminated equals to DAR. Clearly, an ADC with a lower value of \(\overline{DAR }\) is less favorable pharmacokinetically than that with the same nominal DAR value but a higher \(\overline{DAR }\) in terms of their payload delivery effectiveness.

While the calculation of each JDM is straightforward using a ratio between appropriate PK parameters from the separate NCA of two analytes, independent of structure related assumptions, other assumptions (or limitations) related to the interpretations of some JDMs or used in their derivations should be mentioned. The ratio of \({CL}_{PL}\)/\({CL}_{CPL}(\mathrm{or}\;{CL}_{PL}\)/\({CL}_{ADC}\)) as a measure of the therapeutic exposure ratio would be less useful if the given ADC does not fall in the drug delivery scheme behind typical ADC designs, whose ADC- and target interaction–related toxicities, for example, are significant relative to those driven by the payload. When the CPL information is available, more detailed interpretations of the linker stability estimate may be made, requiring the assumptions regarding specifics of the deconjugation process, such as those used in the alternative structure (Eqs. (1.4)–(1.7)), in conjunction with the ratio of\({AUC}_{CPL}/{AUC}_{ADC}\). For convenient, the assumption of linearity was used in the JDM derivations (Eq. (2.1)). In cases where the clearance of one or more components is concentration dependent, some JDMs (e.g., \({CL}_{PL}\)/\({CL}_{ADC}\)) should be interpreted in the context of specific doses, such as the therapeutic dose or maximum tolerated dose, as generally done in the NCA practice.

The Impact of Linker Stability on ADC Disposition

The PK profiles of two versions of an ADC candidate with a DAR of 4 were simulated using Eqs. (1.4)–(1.7) to show the effects of the reduced deconjugation clearance on disposition of the ADC, TAb, payload, and CPL (Fig. 3). The case study was intended to help understand how these effects are captured by the integrated NCA vs. the separate NCA. Version 2 of the ADC was eliminated at a slower rate than version 1, due to a decreased contribution from deconjugation to the total ADC clearance, while the TAb of both versions is cleared at the same degradation rate. Consequently, version 2 had a smaller ratio of \({AUC}_{mAb}/{AUC}_{TAb}\) (0.063) than version 1 (0.33). This in turn resulted in changes in payload PK profile with a reduced peak concentration and less steep slope in the terminal disposition phase for version 2 compared to version 1, though no change in \({AUC}_{PL}\). Note that the payload terminal phase corresponded to the slow payload release from the ADC elimination, rather than the payload elimination which was fast as shown in the payload PK profile after iv bolus administration (Fig. 3). This, as a result of typical ADC designs, is also apparent from Eq. (1.3) as the term with the slowest rate constant (\({CL}_{ADC}\)/\({V}_{A}\)) becomes dominating at later times. Additionally, the use of a more stable linker in version 2 improved the margin between exposures to the CPL and payload, with the ratio of \({AUC}_{CPL}/{AUC}_{PL}\) increased to 100 for version 2 from 62 for version 1. Moreover, version 2 had a higher ratio of \({AUC}_{CPL}/{AUC}_{ADC}\) (3.22) than version 1 (2.79) in favor of more efficient payload delivery to the target cells. Clearly, version 2 with a more stable linker has more favorable PK characteristics than version 1.

Utilities of JDMs for Assessing Among ADCs in Terms of Favorable PK Characteristics

The proposed JDMs allow one to define and compare ADCs, in terms of the linker stability, therapeutic exposure ratio, and effective DAR in vivo. To this end, the JDMs were calculated for a number of FDA-approved ADC products or otherwise well-documented ADC candidates to explore their utilities in assessing among ADCs in terms of favorable PK characteristics (Table III).

For the anti-HER2 ADCs, T-DXd had a smaller ratio of \({AUC}_{mAb}/{AUC}_{TAb}\) (0.12 vs. 0.57 for T-DM1), indicating a minimal deconjugation involved in the ADC elimination due to the use of a more stable linker. The ratio of \({CL}_{PL}/{CL}_{ADC}\), a metric of the therapeutic exposure ratio, for T-DXd was 545, corresponding to a wide margin between the exposures to the ADC and payload. Although there was no CPL information available for estimating the \({AUC}_{CPL}/{AUC}_{ADC}\), this ratio is expected to be close to the nominal DAR value since the deconjugation was minimal. Clearly, the PK characteristics of T-DXd are highly desirable compared to T-DM1, consistent with the recent results from a head-to-head phase 3 trial where T-DXd demonstrated superior efficacy versus T-DM1 in patients with HER2-positive breast cancer (23).

All 3 MMAE-based ADCs had a similar value for the ratio of \({AUC}_{mAb}/{AUC}_{TAb}\) between 0.45 and 0.60, indicating the deconjugation contributed considerably to the respective ADC elimination. CDX-011 had a lower ratio of \({CL}_{PL}/{CL}_{ADC}\), likely due to its higher ADC clearance. Note that the clearance for MMAE was similar for the 3 ADCs, as expected. This may explain in part why CDX-011 failed at its maximum tolerated dose to show advantage in response rate and survival in patients with metastatic breast cancer (24). Furthermore, this ratio estimate for ASG-22ME favors use of the fractionated dosing regimen (3 × 1.25 mg/kg/week per a 4-week cycle) as opposed to a larger dose once every 3 weeks.

Among the Aur0101-based ADCs, the ratio of \({AUC}_{mAb}/{AUC}_{TAb}\) was small, revealing a minimal deconjugation, for PF-06664178 (0.096) and PF-06647020 (0.14), while that for PF-06650808 (0.31) was appreciably greater. These \({AUC}_{mAb}/{AUC}_{TAb}\) ratios were essentially the same as the ratios of \({CL}_{dc}/{CL}_{ADC}\) obtained from the model-based approaches (Table II), as expected based on Eq. (2.5). The ratio of \({CL}_{PL}/{CL}_{ADC}\) for PF-06647020 was approximately 2.5 times greater than the other two, likely because the ADC had a lower clearance from degradation as well. This advantage for PF-06647020 was consistent with the safety and early efficacy data from these Aur0101-based ADCs (16, 18, 20), and supported further investigations of PF-06647020. Also note that the clearance for Aur0101 was somewhat higher than the closely related MMAE, in favor of a higher ratio of \({CL}_{PL}/{CL}_{ADC}\).

For IMMU-130 and IMMU-132, there was PK information available for 4 analytes: ADC, TAb, payload, and CPL. With their mAbs against different targets, both ADCs used the SN-38 payload and an acid labile linker thought to be favorable for the “bystander effect” towards neighboring cancer cells (25, 26). The ratio of \({AUC}_{mAb}/{AUC}_{TAb}\) was comparable between the two, with a value consistent with a greater contribution by deconjugation to the ADC elimination. However, they had a ratio of \({CL}_{PL}/{CL}_{ADC}\) similar to those for MMAE-based ADCs. Moreover, for both ADCs, the ratio of \({AUC}_{CPL}/{AUC}_{ADC}\) (\(\overline{DAR }\)) was close to their nominal DAR values, suggesting that the \({C}_{ADC}^{DAR}\) species predominated relative to other ADC species.

Model-Based PK Characterization of ADC Candidates

The basic structure (Eqs. (1.1)–(1.3)) was based on the payload release from ADC elimination through two parallel processes, deconjugation and degradation, by taking advantage of the information on unconjugated mAb formation only from the deconjugation process. This simple structure adequately described the time courses of ADC, payload, and TAb concentrations over ranges of doses studied for PF-06650808 and PF-06664178 (Figs. S1 and S3). It is quite common for a mAb drug, like PF-06647020, to exhibit a dose-dependent PK behavior. The basic structure expanded with a target mediated degradation described well the PK profile for PF-06647020, as well as those for the payload and TAb, over a wide range of doses (Fig. S2), with the target-mediated clearance plateaued toward its capacity at approximately 2.1 mg/kg (Fig. 4). It should be noted that the sample sizes at lower doses, over which the target-mediated degradation became saturated, were small, and therefore the contributions from these dose groups to the overall parameter estimation were relatively limited.

Fig. 4
figure 4

Individual predictions of cycle 1 concentrations of PF-06647020 ADC (red solid line), TAb (black dotted line), and payload (blue dashed line), identified with dose and dummy patient ID, over the dose range of 0.2 to 3.7 mg/kg. For the 2.1, 2.8, and 3.7 mg/kg groups where n ≥ 4, the middle two in terms of AUCADC estimates were selected as representatives of the respective group (Fig. S2)

The parameter estimations with the basic and expanded structures allowed simultaneous determinations of all clearance and volume terms involved in the ADC elimination pathways as well as the payload disposition (Table II). The total clearance was similar for PF-06664178 and PF-06650808, ranging from 1.94 to 2.40 L/day, and lower for PF-06647020 at 1.21 L/day. Between deconjugation and degradation, the contribution by deconjugation to the total ADC clearance was minimal for both PF-06664178 and PF-06647020, while appreciable for PF-06650808 (also see the comparison above with those by the ratios of \({AUC}_{mAb}/{AUC}_{TAb}\)). The half-life for payload elimination from the blood was relatively short, between 5.2 and 9.2 h.

The model evaluations showed, with the PF-06647020 expansion group at 2.8 mg/kg, that the model tracked also well at a population level (Fig. 2). The goodness-of-fit plots (Figs. S4S6) showed that the population and individual model predictions adequately described the observed data with no apparent trend in the individual weighted residuals over the concentration range and time points evaluated.

CONCLUSION

We presented 3 JDMs for integrated NCA of ADCs in terms of the linker stability, therapeutic exposure ratio, and effective DAR in vivo based on a combination of analytes of ADC, payload, conjugated payload, and total antibody. These JDMs allow to define what constitutes a PK profile favorable for an ADC with clinical implications. The validity of the JDMs, as well as the presumed ADC disposition framework, was verified via their applications to 3 clinical ADC candidates, including that the same linker stability assessments were obtained between the JDM- and model-based approaches. Moreover, utilities of the JDMs were demonstrated in assessing among the selected ADC candidates and products in terms of favorable PK characteristics.

References

  1. Diamantis N, Banerji U. Antibody-drug conjugates–an emerging class of cancer treatment. Br J Cancer. 2016;114:362–7.

    CAS  Article  Google Scholar 

  2. Panowski S, Bhakta S, Raab H, Polakis P, Junutula JR. Site-specific antibody drug conjugates for cancer therapy. MAbs. 2014;6:34–45.

    Article  Google Scholar 

  3. Teicher BA, Chari RV. Antibody conjugate therapeutics: challenges and potential. Clin Cancer Res. 2011;17:6389–97.

    CAS  Article  Google Scholar 

  4. Donaghy H. Effects of antibody, drug and linker on the preclinical and clinical toxicities of antibody-drug conjugates. MAbs. 2016;8:659–71.

    CAS  Article  Google Scholar 

  5. Girish S, Gupta M, Wang B, Lu D, Krop IE, Vogel CL, et al. Clinical pharmacology of trastuzumab emtansine (T-DM1): an antibody-drug conjugate in development for the treatment of HER2-positive cancer. Cancer Chemother Pharmacol. 2012;69:1229–40.

    CAS  Article  Google Scholar 

  6. Yamashita T, Shimomura A, Takano T, Sagara Y, Watanabe J, Tokunaga E, et al. Abstract P1–18–12: A phase 1, multicenter, open-label study to assess the effect of [fam-] trastuzumab deruxtecan (T-DXd; DS-8201a) on QTc and pharmacokinetics in subjects with HER2-expressing metastatic and/or unresectable breast cancer. Cancer Res. 2020;80(4 Suppl):P1–18–12.

  7. U.S. Food and Drug Administration. Drugs@FDA: FDA:Approved Drugs, Brentuximab Vedotin. 2011.

  8. Rosenberg J, Sridhar SS, Zhang J, Smith D, Ruether D, Flaig TW, et al. EV-101: a phase I study of single-agent enfortumab vedotin in patients with nectin-4-positive solid tumors, including metastatic urothelial carcinoma. J Clin Oncol. 2020;38:1041–9.

    CAS  Article  Google Scholar 

  9. Ocean AJ, Starodub AN, Bardia A, Vahdat LT, Isakoff SJ, Guarino M, et al. Sacituzumab govitecan (IMMU-132), an anti-Trop-2-SN-38 antibody-drug conjugate for the treatment of diverse epithelial cancers: Safety and pharmacokinetics. Cancer. 2017;123:3843–54.

    CAS  Article  Google Scholar 

  10. Yin O, Xiong Y, Endo S, Yoshihara K, Garimella T, AbuTarif M, et al. Population Pharmacokinetics of trastuzumab deruxtecan in patients with HER2-positive breast cancer and other solid tumors. Clin Pharmacol Ther. 2021;109:1314–25.

    CAS  Article  Google Scholar 

  11. Li H, Han TH, Hunder NN, Jang G, Zhao B. Population pharmacokinetics of brentuximab vedotin in patients with CD30-expressing hematologic malignancies. J Clin Pharmacol. 2017;57:1148–58.

    CAS  Article  Google Scholar 

  12. Hibma J, Knight B. Population pharmacokinetic modeling of gemtuzumab ozogamicin in adult patients with acute myeloid leukemia. Clin Pharmacokinet. 2019;58:335–47.

    CAS  Article  Google Scholar 

  13. Lu D, Lu T, Gibiansky L, Li X, Li C, Agarwal P, et al. Integrated two-analyte population pharmacokinetic model of polatuzumab vedotin in patients with non-Hodgkin lymphoma. CPT Pharmacometrics Syst Pharmacol. 2020;9:48–59.

    CAS  Article  Google Scholar 

  14. Shah DK, Haddish-Berhane N, Betts A. Bench to bedside translation of antibody drug conjugates using a multiscale mechanistic PK/PD model: a case study with brentuximab-vedotin. J Pharmacokinet Pharmacodyn. 2012;39:643–59.

    Article  Google Scholar 

  15. Geles KG, Gao Y, Sridharan L, Giannakou A, Yamin T-T, Golas J, et al. Abstract 1697: therapeutic targeting the NOTCH3 receptor with antibody drug conjugates. Can Res. 2015;75(15 Supplement):1697–1697.

    Article  Google Scholar 

  16. Rosen LS, Wesolowski R, Baffa R, Liao KH, Hua SY, Gibson BL, et al. A phase I, dose-escalation study of PF-06650808, an anti-Notch3 antibody-drug conjugate, in patients with breast cancer and other advanced solid tumors. Invest New Drugs. 2020;38:120–30.

    CAS  Article  Google Scholar 

  17. Damelin M, Bankovich A, Bernstein J, Lucas J, Chen L, Williams S, et al. A PTK7-targeted antibody-drug conjugate reduces tumor-initiating cells and induces sustained tumor regressions. Sci Transl Med. 2017;9(372).

  18. Maitland ML, Sachdev JC, Sharma MR, Moreno V, Boni V, Kummar S, et al. First-in-human study of PF-06647020 (Cofetuzumab Pelidotin), an antibody-drug conjugate targeting protein tyrosine kinase 7, in advanced solid tumors. Clin Cancer Res. 2021;27:4511–20.

    CAS  Article  Google Scholar 

  19. Strop P, Tran TT, Dorywalska M, Delaria K, Dushin R, Wong OK, et al. RN927C, a site-specific trop-2 antibody-drug conjugate (ADC) with enhanced stability, is highly efficacious in preclinical solid tumor models. Mol Cancer Ther. 2016;15:2698–708.

    CAS  Article  Google Scholar 

  20. King GT, Eaton KD, Beagle BR, Zopf CJ, Wong GY, Krupka HI, et al. A phase 1, dose-escalation study of PF-06664178, an anti-Trop-2/Aur0101 antibody-drug conjugate in patients with advanced or metastatic solid tumors. Invest New Drugs. 2018;36:836–47.

    CAS  Article  Google Scholar 

  21. Ott PA, Hamid O, Pavlick AC, Kluger H, Kim KB, Boasberg PD, et al. Phase I/II study of the antibody-drug conjugate glembatumumab vedotin in patients with advanced melanoma. J Clin Oncol. 2014;32:3659–66.

    CAS  Article  Google Scholar 

  22. Dotan E, Cohen SJ, Starodub AN, Lieu CH, Messersmith WA, Simpson PS, et al. Phase I/II trial of labetuzumab govitecan (anti-CEACAM5/SN-38 antibody-drug conjugate) in patients with refractory or relapsing metastatic colorectal cancer. J Clin Oncol. 2017;35:3338–46.

    CAS  Article  Google Scholar 

  23. Cortés J, Kim S, Chung W, Im S, Park YH, Hegg R, et al. Trastuzumab deruxtecan (T-DXd) vs trastuzumab emtansine (T-DM1) in patients (Pts) with HER2+ metastatic breast cancer (mBC): results of the randomized phase III DESTINY-Breast03 study. Ann Oncol. 2021;32 (suppl_5):S1283-S1346. https://doi.org/10.1016/annonc/annonc741.

  24. Nagayama A, Vidula N, Ellisen L, Bardia L. Novel antibody–drug conjugates for triple negative breast cancer. Ther Adv Med Oncol. 2020;12:1–12.

    Article  Google Scholar 

  25. Goldenberg DM, Cardillo TM, Govindan SV, Rossi EA, Sharkey RM. Trop-2 is a novel target for solid cancer therapy with sacituzumab govitecan (IMMU-132), an antibody-drug conjugate (ADC). Oncotarget. 2015;6:22496–512.

    Article  Google Scholar 

  26. Govindan SV, Cardillo TM, Rossi EA, Trisal P, McBride WJ, Sharkey RM, et al. Improving the therapeutic index in cancer therapy by using antibody-drug conjugates designed with a moderately cytotoxic drug. Mol Pharm. 2015;12:1836–47.

    CAS  Article  Google Scholar 

Download references

Acknowledgements

We would like to acknowledge previous contributions from Christopher J. Zopf, PhD to the ADC PK characterization framework and its application to PF-06664178.

Funding

This work was sponsored by Pfizer.

Author information

Authors and Affiliations

Authors

Contributions

K.H.L., J.H.W., S.P., D.Y., and X.M. wrote the manuscript. K.H.L., J.H.W., S.P., and X.M. designed the research, performed the research, and analyzed the data.

Corresponding author

Correspondence to Jason H. Williams.

Ethics declarations

Conflict of Interest

J.H.W. and D.Y. are employees of Pfizer and hold Pfizer stock. K.H.L., S.P., and X.M. were employees of Pfizer at the time of this analysis and hold Pfizer stock.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Below is the link to the electronic supplementary material.

Supplementary file1 (PDF 2455 KB)

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Liao, K.H., Williams, J.H., Palani, S. et al. Joint Disposition Properties and Comprehensive Pharmacokinetic Characterization of Antibody–Drug Conjugates. AAPS J 24, 73 (2022). https://doi.org/10.1208/s12248-022-00717-x

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1208/s12248-022-00717-x

KEY WORDS

  • antibody–drug conjugate
  • pharmacometrics
  • drug-to-antibody ratio
  • linker stability
  • therapeutic index