Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2020 Sep 4.
Published in final edited form as: J Clin Pharmacol. 2019 Dec 2;60(2):149–163. doi: 10.1002/jcph.1545

Concept of Pharmacologic Target-Mediated Drug Disposition (TMDD) in Large-Molecule and Small-Molecule Compounds

Guohua An 1
PMCID: PMC7472685  NIHMSID: NIHMS1622559  PMID: 31793004

Abstract

Target-mediated drug disposition (TMDD) is a term to describe a nonlinear pharmacokinetics (PK) phenomenon that is caused by high-affinity binding of a compound to its pharmacologic targets. As the interaction between drug and its pharmacologic target belongs to the process of pharmacodynamics (PD), TMDD can be viewed as a consequence of “PD affecting PK”. Although there are numerous TMDD related papers in the literature, most of them focus on characterizing TMDD using various mathematical models, which may not be suitable for those readers who have little interest on mathematical modeling and only want to have an understanding of the basic concept. The goal of this review is to serve as a “primer” on TMDD. This review explains 1) how TMDD happens; 2) why large-molecule and small-molecule compounds exhibiting TMDD demonstrate substantially different nonlinear PK behaviors; 3) what nonlinear PK profiles look like in large-molecule and small-molecule compounds exhibiting TMDD, using pegfilgrastim, erythropoietin, ABT-384, and linagliptin as case examples; 4) how to identify whether the nonlinear PK of a compound is due to TMDD or not.

Keywords: ABT-384, erythropoietin, linagliptin, nonlinear pharmacokinetics, pegfilgrastim, Target-mediated drug disposition

Scope of This Review

Target-mediated drug disposition (TMDD) is a term to describe a nonlinear pharmacokinetics (PK) phenomenon that is caused by high-affinity binding of a compound to its pharmacologic targets 1. As the interaction between drug and its pharmacologic target belongs to the process of pharmacodynamics (PD), TMDD can be viewed as a consequence of “PD affecting PK”, which is truly an interesting phenomenon and to certain aspect even counterintuitive as the default knowledge we always have is “PK affecting PD”. Although the concept of TMDD is not new (generated 25 years ago), there are a number of misunderstandings among pharmaceutical scientists on this concept. One misunderstanding is that “TMDD only occurs in large-molecule compounds”. Actually the concept of TMDD was originally generated by Levy in 1994 based on the unusual nonlinear PK of a number of small-molecule drugs 1. However, this concept did not receive immediate attention when Levy raised it for the first time due to the relatively low prevalence of TMDD in small-molecule drugs. Since early 2000s, the shape of new drug discovery has been greatly changed following biotechnology revolution, with more and more protein drugs being developed and launched in the market. Along with the blossoming of large-molecule drug development, the concept of TMDD has gained broad attention and has become well-known because numerous protein drugs demonstrate nonlinear PK imparted by TMDD due to specific binding to their pharmacological targets 2-10. Considering the high prevalence of TMDD in large-molecule drugs, it is not surprising that the message from Levy was only half-taken, with the concept of TMDD being widely accepted but the examples of small-molecule drugs used to generate the concept being ignored/forgotten, at least by many pharmaceutical scientists. In the past 10 years, increasingly potent small molecule drugs acting on highly specific targets have been developed. As a result, the prevalence of TMDD in small molecule compounds has increased significantly 11-17. It is really about time to increase the awareness that TMDD can occur in small molecule compounds. Another misunderstanding is that “compounds with high affinity for their pharmacological targets will have TMDD”. High affinity binding of a compound to its pharmacological targets is a necessary but not a sufficient condition for TMDD. A compound with high affinity binding to its target could demonstrate linear PK simply because it also has substantial non-specific binding to tissue and/or plasma, which masks the saturable target binding process. So how does TMDD happen? Does TMDD in large-molecule compounds demonstrate similar nonlinear PK behavior as that in small-molecule compounds? How to quickly identify whether the nonlinear PK of a compound is due to TMDD or not? The aim of this review is to clarify our misunderstandings mentioned earlier and address the questions listed above. Although there are numerous TMDD related papers in the literature, most of them focus on characterizing TMDD using various mathematical models, which may not be suitable for readers with limited math background (or little interest on mathematical modeling) and only want to have an understanding of the basic concept. This review targets this type of reader. The goal of this review is to serve as a “primer” on TMDD. I hope the reader will have a fundamental understanding of this concept after reading this review.

How Does TMDD Happen?

We can consider a compound exhibiting TMDD when this compound 1) has nonlinear PK across the dose range evaluated, and 2) the nonlinearity is caused by the binding of drug molecules to its pharmacological target (i.e., PD) affecting the disposition of drug in the body (i.e., PK). There are a number of prerequisites for the occurrence of TMDD. First, the binding affinity of the compound to its pharmacological target(s) needs to be high. Without high affinity binding, the drug molecules won’t be able to strongly engage with its target. Second, the compound needs to specifically bind to this target. The reason is fairly obvious - if a compound not only binds to its target but also has substantial non-specific tissue binding at the same time, then the target binding, even with high-affinity, can easily be masked by the binding to those nonspecific sites whose capacity is usually larger. Large-molecule compounds usually have high-affinity binding to their pharmacological targets with minimal nonspecific tissue binding. That is why the prevalence of TMDD is high in large-molecule compounds. Lastly, the capacity of the pharmacological target needs to be relatively low. If the capacity of the target is very high and none of the doses evaluated is close to saturation, then the drug PK will be essentially linear.

When all these prerequisite conditions are met, how does TMDD happen? For a compound that specifically binds to a high-affinity-low-capacity target, a significant fraction of the dose will be sequestered by the target when a low dose is given, such that only a small fraction of the dose is available in the systemic circulation. With increase in doses, the portion of the dose that is “trapped” by the target is smaller and smaller, and consequently higher and higher fraction of the population of drug molecules is available for systemic circulation. At high doses, the target is saturated due to its low capacity, and the fraction of dose that binds to the target is minimal compared to the total dose. As a result, the kinetics of the system is essentially linear (i.e. dose proportional) at high doses. Therefore, for a compound exhibiting TMDD, no matter if it is a small molecule or large molecule, and no matter whether it is given orally or via i.v. bolus, its non-linearity usually occurs (or is more pronounced) at low, single doses. TMDD behavior following multiple dose regimens is less intuitive and will be explained in detail in later sections. Although both large-molecule and small-molecule compounds can have TMDD and both have nonlinear PK at low doses when TMDD happens, they demonstrate substantially different nonlinear PK behaviors. Why is there such a difference if both types of compound undergo target-mediated drug disposition? The object of the next section is to answer this question.

Why Large-Molecule and Small-Molecule Compounds Exhibiting TMDD Demonstrate Substantially Different Nonlinear PK Behaviors?

Although TMDD can occur in both large-molecule compounds and small-molecule compounds, the observed nonlinear PK profiles are substantially different between these two types of compounds. There are three reasons for their different PK behaviors:

Reason #1. The fate of their drug-target complex is different.

After a large-molecule binds to its target, the drug-target complex has two fates: a) some drug-target complex molecules will dissociate back to free drug molecules and free target, and b) at the same time other drug-target complex molecules will undergo subsequent internalization, endocytosis, and degradation into amino acids which will be recycled by our body for other biosynthetic processes. The second fate of the drug-target complex can be viewed as “drug and target destroy each other”, a target-mediated elimination process that represents a major elimination pathway for most protein drugs. In contrast, small-molecule compounds usually have no such mechanism. The elimination of small-molecule compounds is mainly accomplished by metabolism via drug metabolizing enzymes and/or excretion via urine/bile as unchanged compounds. Consequently, when a small-molecule compound binds to its targets, the drug-target complex usually will not be degraded, and all of the drug-target complex molecules will eventually dissociate back to free drug molecules and free target. Because of the different fates of their drug-target complexes, the overall pattern of clearance differs between these two types of compounds.

Reason #2. Large-molecule compounds exhibiting TMDD often have much stronger interplay between their PK and PD compared with that of small-molecule compounds exhibiting TMDD.

The PD effect of many protein drugs is to boost a certain type of cells (except for antibodies which usually deplete their target cells), and interestingly the pharmacologic targets of these protein drugs are often located in the same type of cells. In this case, when protein drug molecules bind to their pharmacologic targets, on one hand the drug-target complex stimulates cell synthesis and consequently more target cells are generated, while on the other hand the same drug-target complex undergoes degradation and consequently determines the elimination of protein drugs. The whole process is essentially an intensive interplay between drug PK and PD – the higher the drug concentration, the more the target cells generated; the more the target cells generated, the more pharmacologic targets present (i.e. increase the synthesis rate of the targets and correspondingly expand the target pool); the more pharmacologic targets present, the faster/more the drug molecules cleared. Typical examples of protein drugs with this type of “PK-PD loop” include Pegfilgrastim, a recombinant granulocyte colony-stimulating factor (G-CSF), and Epoetin Alfa, a recombinant human erythropoietin (detailed information is provided in the next section). It should be noted that antibodies do not have such strong PKPD interplay as antibody-drug complex has no impact on the synthesis of their targets. In contrast, for small molecule compounds exhibiting TMDD, the PD part that affects their PK is only limited to the association and dissociation processes between drug and target, and there is no involvement of the process of “target destroys the drug” or dynamic change of target expression as seen in large molecule compounds.

Reason #3. For small-molecule compounds exhibiting TMDD, in general the dissociation of drug-target complex is a much slower process than that in large-molecule compounds exhibiting TMDD.

In general, both types of compounds have a fast association process (i.e., quick binding process to form drug-target complex), as reflected by a large associate rate constant kon value. The difference lies in the dissociation process. For large-molecule compounds, the drug-target complex dissociates back to free drug and free target fairly quickly (may take minutes or hours), as reflected by a large dissociation rate constant koff value. In contrast, small-molecule compounds exhibiting TMDD often have a firm and long lasting target binding property, resulting in a slow dissociation process which may take days or even weeks (i.e. small dissociation rate constant koff value) 12, 15. The slow dissociation of drug-target complex, together with the low capacity of the target, is the root cause of several unusual PK behaviors commonly seen in small-molecule compounds exhibiting TMDD, such as nonlinear PK observed after first dose disappearing with repeated doses, unusual drug accumulation pattern, and long terminal elimination phase which tends to converge to similar concentration values regardless of the doses given 12, 18. These unusual PK behaviors observed in small-molecule compounds exhibiting TMDD are distinct as these features neither exist in large-molecule compounds exhibiting TMDD nor in any small-molecule compounds whose nonlinear PK is caused by other reasons (e.g. saturation of drug metabolizing enzymes, drug transporters, etc.). Typical examples of small-molecule compounds with these features include ABT-384, a potent 11β-Hydroxysteroid Dehydrogenase Type 1 (HSD-1) inhibitor, and linagliptin, a potent Dipeptidyl Peptidase (DPP)-4 inhibitor (detailed information is provided in the next section).

When keeping the above different mechanisms in mind, the answer to the question “why both types of compounds undergoing TMDD have substantially different nonlinear PK behaviors?” is self-evident. A natural follow-up question arises - what do these nonlinear PK profiles look like in large-molecule and small-molecule compounds exhibiting TMDD? This question is addressed in the next section.

What Do the Nonlinear PK Profiles Look Like in Large-Molecule and Small-Molecule Compounds exhibiting TMDD?

Various large-molecules, including recombinant G-CSF products (filgrastim and pegfligrastim), erythropoietin, thrombopoietin, macrophage colony-stimulating factor (M-CSF), interferon-β 1a, and various monoclonal antibodies, bispecific antibodies, have been reported to have nonlinear PK imparted by TMDD 2-4, 6-8, 10, 19-21. In this section, recombinant G-CSF products, erythropoietin, and antibodies are used as case examples to demonstrate what nonlinear PK profiles look like in large molecules exhibiting TMDD. Similarly, many small molecules have been reported to have nonlinear PK due to TMDD, which I have summarized them in detail in a mini-review 12. The small-molecule compounds covered in that mini-review include warfarin 22, imirestat 23, enalaprilat 24, perindoprilat25, cilazaprilat 26, selegiline 27, bosentan 28, ABT-384 11, 15, linagliptin 13, 29, and valdagliptin30. After the publication of the mini-review, additional small-molecule compounds exhibiting TMDD have been reported; this includes ASP366231, soluble epoxide hydrolase (sEH) inhibitors (TPPU and TCPU)32, and two investigational compounds from Roche (names were not provided) 16. In this section, ABT-384 and linagliptin are used as case examples to demonstrate how nonlinear PK profiles look in small molecules exhibiting TMDD.

Case example 1. Large-molecule compound – Filgrastim and Pegfilgrastim

G-CSF is an endogenous hematopoietic growth factor which acts at all stages of neutrophil development and is a very potent late-acting growth factor. G-CSF increases the proliferation and differentiation of neutrophils from committed progenitor cells. G-CSF receptors are present on both myeloid progenitor cells in the bone marrow and peripheral neutrophils. Both filgrastim and pegfligrastim are recombinant human G-CSF. The only difference is that pegfligrastim contains polyethylene glycol (PEG) moieties in its structure and correspondingly has longer half-life due to the shielding effect of the PEG moieties. Both filgrastim and pegfligrastim are effective in reducing the severity and duration of neutropenia (i.e. low neutrophil count) and its complications. Both agents have been widely used in patients with cancer who receive myelosuppressive chemotherapy or bone marrow transplantation, and patients with severe chronic neutropenia 33.

Filgrastim demonstrates clear nonlinear PK, with its clearance decreasing with increasing dose. Several studies have shown that the serum concentrations of filgrastim are inversely correlated with the number of circulating neutrophils 33, suggesting that the saturable clearance pathway of filgrastim is neutrophil-mediated. In addition, the clearance of recombinant human G-CSF was found to be closely related not only to the number of circulating neutrophils but also to the percentage of G-CSF receptor-positive neutrophils 34, indicating that the neutrophil-mediated clearance is mediated by G-CSF receptors on neutrophils and neutrophil precursors.

Similar to filgrastim, pegfilgrastim also has nonlinear PK with saturable clearance pathway. In a study conducted in rats, the plasma concentrations of pegfilgrastim and the absolute neutrophil count (ANC) were measured following pegfilgrastim every other day for two weeks 33. As shown in Figure 1, the plasma concentrations of pegfilgrastim after the last dose were lower than that after the first dose, while the ANC after the last dose was higher than after the first. The parallel change in pegfilgrastim exposure and ANC indicated that the increased clearance of pegfilgrastim after repeated doses is due to the expansion of neutrophil and neutrophil precursor mass33. In addition, as shown in Figure 1, this time-dependent nonlinearity was more pronounced with lower dose groups (50 and100 ug/kg) than with higher dose groups (500 and 1000 ug/kg), which is expected since neutrophil-mediated clearance of pegfilgrastim that predominating at low doses may get saturated at higher doses. A further mechanism study was conducted in G-CSF receptor-knockout mice 35. Following a single i.v. dose of filgrastim or pegfilgrastim at 10 or 100 ug/kg, both drugs had significantly higher exposure in the G-CSF receptor-knockout mice than in wild-type mice. The PK of filgrastim or pegfilgrastim, which was nonlinear in wild-type mice, turned into linear in G-CSF receptor-knockout mice 35. In addition, the clearance of pegfilgrastim in G-CSF receptor-knockout mice is only about 30% of that in wild-type mice (1.5 vs 4.4 mL/hr/kg) following 10 ug/kg dose and about 50% of that in wild-type mice (1.8 vs 3.5 mL/hr/kg) 35. This confirmed that, the neutrophil-mediated clearance of pegfilgrastim is mediated by G-CSF receptors on neutrophils and neutrophil precursors, and this elimination pathway represents the major elimination pathway when pegfilgrastim was given at ≤ 100 ug/kg.

Figure 1.

Figure 1.

Open in a new tab

Mean pegfilgrastim plasma concentration-time profile and absolute neutrophil count (ANC)-time profile in rats following subcutaneous administration of pegfilgrastim every other day for two weeks at four dose levels (50, 100, 500, and 1000 ug/kg). This figure was adapted from references 33.

When pegfilgrastim was administered in humans, neutrophil-mediated clearance of pegfilgrastim was also clearly observed. Johnston et al evaluated the concentration-time profiles of pegfilgrastim and ANC-time profiles following different doses of pegfilgrastim (30, 100, and 300 ug/kg) in patients with non-small cell lung cancer before and after chemotherapy36. Pegfilgrastim PK showed a similar trend in all evaluated doses. To highlight the interesting interplay between PK and PD, Figure 2 only shows the result from the 100 ug/kg dose group. As shown in the lower panel of Figure 2, ANC increased after pegfligrastim was given at Day 0, greatly decreased after chemotherapy was given at Day 15, and then gradually recovered from Day 22 after dosing again at Day 1736. An interesting PK profile of pegfligrastim was observed (the upper panel of Figure 2) – although the maximum concentrations of pegfilgrastim following the 2nd dose (i.e. after chemotherapy) was similar to the 1st dose (i.e. before chemotherapy), there was a prolonged plateau in the concentrations after chemotherapy, which began to decline only at the onset of neutrophil recovery at around Day 2236. The pegfligrastim PK observed in this study clearly showed how its PD affected its PK, and it represents an excellent example of large-molecule compound exhibiting TMDD.

Figure 2.

Figure 2.

Open in a new tab

Pegfilgrastim serum concentration-time profile and absolute neutrophil count (ANC)-time profile in cancer patients single pegfilgrastim administration at 100 ug/kg before and after chemotherapy. This figure was adapted from references 33 and 36.

Case example 2. Large-molecule compound – rhEPO

Recombinant human erythropoietin (rhEPO) stimulates erythropoiesis by binding to EPO receptors located on the surface of erythroid progenitor cells, and it has been widely used as a major therapeutic agent for treatment of anemia caused by chronic kidney disease and chemotherapy. Similar to those recombinant human G-CSF products, the clearance of rhEPO also decreases with increasing doses (i.e. dose-dependent) and increases with repeated doses (i.e. time-dependent). The dose- and time-dependent change in rhEPO clearance is believed to be caused by EPO receptor-mediated elimination. Following a single dose regimen, rhEPO clearance deceases at high doses due to saturation of the EPO receptors. Following a multiple dose regimen, the pool of erythroid progenitors is expanded and correspondingly more EPO receptors are generated. As a result, rhEPO clearance increases after repeated doses because of the increased capacity of the EPO receptor. There are a number of lines of evidence indicate that rhEPO is mainly degraded by its pharmacological target, the EPO receptor, which is in greatest abundance on erythroid progenitors located primarily in bone marrow. Widness’ group evaluated the role of bone marrow in rhEPO elimination by determining rhEPO PK in sheep both before and after chemotherapy-induced bone marrow ablation 37. As shown in Figure 3, after busulfan-induced bone marrow ablation, rhEPO plasma concentrations in sheep were much higher than that obtained before busulfan administration, with substantial reduction in rhEPO clearance (up to 80% decrease). The same group also evaluated rhEPO elimination in patients undergoing hematopoietic transplantation and observed a similar result 38. Another interesting experiment that Widness’ group conducted was to evaluate phlebotomy-induced changes in rhEPO PK in sheep 39. In this experiment, anemia to hemoglobin levels of 3-4 g/dL was induced by removing a large amount of blood and replacing it with the same amount of saline. The PK of rhEPO was evaluated prior to phlebotomy, Day 1 post-phlebotomy, and Day 7 post-phlebotomy 39. As shown in Figure 4, rhEPO exposure on Day 1 post-phlebotomy was much higher than pre-phlebotomy condition, while rhEPO exposure on Day 7 post-phlebotomy was significantly lower than pre-phlebotomy condition. This indicated decreased rhEPO elimination on Day 1 post-phlebotomy and increased elimination on Day 7 post-phlebotomy. The initial reduction in rhEPO clearance is likely due to a transient saturation of EPO receptor resulting from the phlebotomy induced high rhEPO concentration. The later increase in rhEPO clearance is due to the expanded pool of progenitor cells (and consequently more EPO receptors) in response to the need of restoring hemoglobin levels.

Figure 3.

Figure 3.

Open in a new tab

A representative plasma 125I-rhEPO concentration-time profiles in a sheep before and after busulfan administration. This figure was adapted from references 37.

Figure 4.

Figure 4.

Open in a new tab

A representative plasma 125I-rhEPO concentration-time profiles in a sheep during three study phases (i.e. before phlebotomy; Day 1 post phlebotomy; and Day 7 post phlebotomy). This figure was adapted from references 39.

Case example 3. Large-molecule compounds – monoclonal antibodies (mAbs)

Numerous mAbs have been reported to have nonlinear PK which is believed to be related to clearance of the mAb via saturable target-mediated mechanisms (e.g. receptor-mediated endocytosis and subsequent lysosomal degradation of the mAb-receptor complex). At low mAb concentrations, total clearance (CL) is relatively unaffected, and target-mediated elimination represents the major elimination pathway. With increase in mAb concentrations, target-mediated elimination pathway starts to become saturated. As a result, total CL decreases substantially. Due to the concentration-dependent biphasic clearance, mAbs concentration-time profiles often demonstrate S-shaped curves at high therapeutic doses if mAbs samples are collected over a prolonged time period (Figure 5). Peletier and Gabrielsson did comprehensive model simulation and described some of the key features of antibody TMDD 40. For IgG antibodies, in addition to target-mediated elimination, they also undergo Fc-receptor-mediated elimination. It has been reported that IgG can be salvaged from lysosomal catabolism in cells expressing the FcRn, which helps to explain the much longer half-life of IgG compared with other immunoglobulin classes 41. FcRn-mediated salvage pathway is capacity limited and is expected to become saturated at very high doses (e.g. >100 mg/kg). As the doses of therapeutic mAbs that are commonly administered are less than 10 mg/kg, saturation of FcRn and subsequent increase in mAb clearance is less common 41. Therefore, at therapeutic doses, the nonlinear PK of a mAb is mainly due to its pharmacologic target-mediated elimination.

Figure 5.

Figure 5.

Open in a new tab

Simulated anti-CD4 mAb TRX1 concentration time profiles following doses of 1 mg/kg, 5 mg/kg and 10 mg/kg. This figure was adapted from reference 46 46

Case example 4. Small-molecule compound – ABT-384

11β-HSD1 is an enzyme that converts inactive cortisone to active cortisol and participates in the regulation of glucocorticoid levels in tissues. The 11β-HSD1enzyme is primarily expressed in liver, fat, and brain. ABT-384 is a selective and potent HSD-1 inhibitor, with the apparent Ki of 5-10 nM across all species tested 11. An in vitro experiment showed that ABT-384 quickly forms a tight drug-HSD-1 complex but has a very slow dissociation rate. The dissociation rate constant (koff) of drug-HSD-1 complex could not be estimated because of the minimal dissociation during the whole time period of the in vitro experiment 11. ABT-384 demonstrated clear nonlinear PK in animals. In addition, a preclinical study carried out in monkeys showed that ABT-384 was readily distributed in liver, the organ with highest HSD-1 expression, and was retained for months. ABT-384 was found to be measurable in monkey liver after more than 2 months washout and the liver concentration of ABT-384 on day 77 was a constant value, regardless of the dose administered 11.

The clinical PK of ABT-384 was evaluated following once daily dose for 7 days at six dose levels (1, 2, 4, 8, 30, and 100 mg) in healthy adults 15. A typical way to evaluate the PK linearity of a compound is to plot dose-normalized concentration-time profiles and see if those profiles superimpose. Nonlinear PK does not obey the rule of superposition and consequently concentration/dose versus time profiles do not superimpose. Figure 6 shows the dose-normalized concentration-time profiles of ABT-384 on Day 1 and Day 7. On Day 1, ABT-384 concentration/dose versus time profiles superimposed at high dose groups (≥8 mg doses) but not at low dose groups, indicating the nonlinear PK of ABT-384 at low doses when it was given for the first time. In addition, there was a clear trend such that the lower the dose, the greater the extent of the nonlinearity of ABT-384 with the first dose. Interestingly, on Day 7 the dose-normalized concentration-time profiles of ABT-384 superimposed at all doses evaluated in this study, indicating that the nonlinear PK observed on Day 1 disappeared with repeated doses 15.

Figure 6.

Figure 6.

Open in a new tab

Dose-normalized plasma concentration –time profiles of ABT-384 on study Day 1 and Day 7. ABT-384 was administered once-daily doses for 7 days at six dose levels (1, 2, 4, 8, 30, and 100 mg). ABT-384 demonstrated substantial nonlinear PK, especially in low dose groups, on Day 1, but linear PK on Day 7. This figure was adapted from reference 11.

Based on the in vitro data, in vivo monkey PK data obtained in plasma and liver, as well as human PK data, the unusual PK of ABT-384 is believed to be caused by the specific binding of ABT-384 to its high-affinity-low-capacity pharmacological target HSD-1. At low doses, ABT-384 binds to the target site (i.e. HSD-1) with high affinity to a significant extent relative to the dose, such that ABT-384 PK is greatly affected. The lower the dose, the larger is the fraction of the dose being “trapped” to the target site and, consequently, the lower the fraction of the ABT-384 molecules (relative to dose) being available in the systemic circulation. This explains why the lower the dose, the greater the extent of ABT-384 nonlinearity with the first dose (Figure 6, Day 1 plot). Because of the low capacity of the target site, at relatively high doses (≥ 8 mg), the binding site is saturated rapidly and the kinetics of the system is essentially linear (i.e., dose proportional). This explains why ABT-384 demonstrated linear PK with the first dose in higher dose groups (Figure 6, Day 1 plot). Because of the low capacity nature of the target site, together with the very slow dissociation of drug-target complex, the target can also be saturated rapidly with repeated low doses. As a result, in low dose groups, the substantial nonlinear PK of ABT-384 observed following the first dose disappeared after repeated administrations of low dose of ABT-384 (Figure 6, Day 7 plot).

Case example 5. Small-molecule compound – Linagliptin

Linagliptin is a highly selective and long acting DPP-4 inhibitor which is used for glycaemic control in patients with type 2 diabetes mellitus. DPP-4 is an exopeptidase that exists in both soluble form in plasma and membrane-bound form in several tissues. Although DPP-4 is present in both plasma and tissues, it seems that plasma DPP-4 plays a bigger role than tissue DPP-4 in the nonlinear PK of linagliptin. Linagliptin demonstrates concentration-dependent plasma protein binding in vitro, with the fraction of linagliptin bound to plasma protein approximately 99% at low concentrations (< 1nmol/L) and dropping to 70-80% at high concentrations (> 100 nmol/L) 42. The concentration-dependent binding of linagliption to plasma protein, which was associated with very high affinity constants, is caused by specific, high affinity and saturable binding of linagliption to plasma DPP-4. Direct supporting evidence is that plasma protein binding of linagliptin was concentration-independent (fraction of bound is about 70% at all drug concentrations) in DDP-4 knockout mice and DDP-4-deficient rats 42. Linagliptin demonstrated substantial nonlinear PK in preclinical species and there is strong evidence that DPP-4 is the root cause of the observed unusual PK behavior. Retlich et al evaluated the role of DPP-4 in linagliptin PK by measuring linagliptin plasma concentrations following single i.v. doses of 0.01, 0.1, 0.3, 1, 3, 10, and 50 mg/kg of [14C]linagliptin in DPP-4 deficient or wild-type rats in a parallel group design 13. As shown in Figure 7, in wild-type rats, the exposure of linagliptin increased less than dose proportional with both CL and Vss increasing with the increase of dose, and the nonlinear PK was most pronounced in the low dose groups (up to 1 mg/kg). In contrast, linagliptin displayed linear PK in DPP-4 deficient rats, as reflected by the superimposed concentration/dose time profiles across all doses evaluated 13. In addition, compared with wild-type rats, the terminal half-life of linagliptin in DPP-4 deficient rats was significantly shorter. Collectively, these results indicate that the high affinity and saturable binding of linagliptin to plasma DPP-4 should be the main or only contributor to the nonlinear PK of linagliptin observed in wild-type rats.

Figure 7.

Figure 7.

Open in a new tab

Dose-normalized linagliptin mean plasma concentration –time profiles after single i.v. bolus administration of linagliptin to A) wild type rats and B) DPP-4 deficient rats. Linagliptin demonstrated substantial nonlinear PK in wild type rats, but linear PK in DPP-4 deficient rats. This figure was adapted from reference 13.

Consistent with preclinical observations, linagliptin exhibited similar nonlinear PK behavior in human using the therapeutic dose range of1-10 mg of linagliptin once daily 29: a) less than dose-proportional increase in linagliptin exposure, with both apparent volume of distribution and clearance of linagliptin being increased with increasing dose; and b) uncommon accumulation characteristics of linagliptin after repeated doses. Although linagliptin has long terminal half-life (~110 to 130 hours), the accumulation ratio of linagliptin is relatively small and dose-dependent, ranging from 2.0-fold in the 1 mg dose group to 1.2-fold in the 10 mg dose group.

The nonlinear PK behavior of linagliptin in human can also be explained by specific, firm and saturable binding of linagliptin to plasma DPP-4. At very low doses, a substantial fraction of linagliptin dose is bound to DPP-4 in plasma, and consequently only limited unbound linagliptin molecules (i.e. free drug molecules) are available for tissue distribution and elimination. With increasing dose, the pool of DPP-4 in plasma is saturated, and subsequently more linagliptin molecules are present as free form and available for tissue distribution and elimination. As a result, both volume of distribution and clearance of linagliptin are increased with the increase in doses. In addition, because of the tight and long lasting binding of linagliptin to DPP-4, the dissociation of the linagliptin/DPP-4 complex is slow, leading to a very long terminal half-life of linagliptin.

Both ABT-384 in case study 3 and linagliptin in case study 4 display unusual and substantial nonlinear PK due to their binding to high-affinity-low-capacity pharmacological targets as well as slow dissociation of the drug-target complex. However, the nonlinear PK behaviors of ABT-384 and linagliptin are substantially different, even demonstrating opposite trends for some PK features. The reason for this disagreement is the different location of their pharmacologic targets. HSD-1, the target of ABT-384, is located in tissues, while DPP-4, the target of linagliptin, is located in plasma and tissues with plasma DPP-4 playing a bigger role.

The four case examples provided in this section clearly show how complicated drug PK can be when PD greatly affects PK. Although compounds exhibiting TMDD all demonstrate unusual and complex nonlinear PK, it is difficult to generalize their PK behaviors as large-molecule compounds exhibiting TMDD have substantially different nonlinear PK behaviors compared with small molecule compounds exhibiting TMDD, and small molecule compounds with targets located in tissues have substantially different nonlinear PK behaviors compared with small molecules whose targets are located in plasma. Then how to identify whether the nonlinear PK of a compound is due to TMDD or not? That is the topic we will discuss in the next section.

How to Identify Whether the Nonlinear PK of a Compound is Due to TMDD or Not?

A. Nonlinear PK features shared among compounds exhibiting TMDD

As noted earlier, there are substantially different TMDD behaviors among a) large molecules, b) small molecules with targets located in tissues, and c) small molecules with targets located in plasma. Except for nonlinear PK being more pronounced at low doses, it is difficult to find other common behaviors among all three drug categories. My group has reported with great detail on the general nonlinear PK features in small molecule compounds exhibiting TMDD with different locations of their targets 12, 18. To facilitate the identification of TMDD, general nonlinear PK features imparted by TMDD within each drug category are briefly summarized again in this section.

  1. For large-molecule compounds exhibiting TMDD, the common features include: 1) nonlinear PK is most pronounced at low doses; 2) clearance is high at low dose, then decreases at high doses (due to saturation of the target-mediated elimination); 3) volume of distribution decreases with increase in dose; 4) time-dependent clearance will be observed if the pharmacological target is located in the same type of cells that the compound boosts. Clearance will be higher after repeated doses due to the expanded target pool; 5) intensive interplay between drug PK and PD.

  2. For small-molecule compounds exhibiting TMDD whose targets are located in tissues, common features include: 1) apparent volume of distribution is very large at low dose, decreases quickly as dose increases, and reaches a limit value at high doses; 2) real clearance is dose-independent since drug-target complex does not degrade. However apparent clearance (Dose/AUC) may be dose-dependent due to assay limitations (i.e. unreliable AUC estimate at low doses); 3) nonlinear PK with the first dose, but linear PK with repeated doses; 4) unusually high accumulation of drug following repeated low doses, which can’t be explained by the compound’s known half-life; 5) following a single dose regimen, there is a long terminal elimination phase which tends to converge to similar concentration values, regardless of the doses given. The prerequisite to capture this feature is to have an analytical assay sensitive enough to measure the very low concentrations in the terminal phase and sample collection period long enough to cover the time needed for the slow dissociation of drug-target complex (i.e. slow release of drug molecules from the binding site to the blood stream).

  3. For small-molecule compounds exhibiting TMDD whose targets are located in plasma, the common features include: 1) apparent volume of distribution increases as dose increases; 2) fraction of unbound drug in plasma is low at low doses and increases with increase in dose; 3) following single dose regimen, clearance increases with the increase in dose. Following multiple dose regimen, clearance increases following repeated doses. 4) unusually low accumulation of drug following repeated low doses, which can’t be explained by the compound’s known long half-life; 5) nonlinear PK observed following the first dose persists after repeated doses; and 6) following single escalating doses, a relatively long terminal elimination phase is observed which tends to converge to similar concentration values.

B. Experiments that can be used to determine if nonlinear PK is due to TMDD

  1. Pharmacokinetic experiment using pharmacological target knock-out animals. The best method for TMDD identification is to compare drug PK following different doses between wild-type animals and pharmacological target knock-out animals. If the nonlinear PK of a drug observed in wild-type animals is truly due to TMDD, the drug will have linear PK in the pharmacological target knock-out animals (i.e. all dose-normalized concentration- time profiles superimpose across all doses evaluated). Such experiments have been done for linagliptin (Figure 7), fligrastim and pegfligrastim; all displayed linear PK in the pharmacological target knock-out animals 13, 35.

  2. In vivo displacement experiments with co-administration of pharmacological target binding displacer. An alternative method for TMDD identification is the in vivo displacement experiment. In this method, a low dose of the tested compound will be administered first. Then a large dose of a displacer (which binds to the same target with high affinity) will be administered after the PK of the tested compound reaches its terminal phase. For a compound with TMDD, following a low dose, a considerable portion of the dose will still bind to its pharmacological target even if its PK reaches the terminal phase in plasma. When a second compound (i.e. displacer) competing for the same target with high affinity is given later with a large dose, the first compound will be displaced by the second compound and free drug molecules will be available for systemic circulation. As a result, the tested compound (i.e. first compound) will have a second peak right after the displacer administration. A typical example is an in vivo displacement experiment conducted by Lee et al on TPPU and TCPU 32, both of which are potent sEH inhibitors. In that study, a low dose of TPPU(0.3 mg/kg) was first administered subcutaneously, then a high dose of TCPU (3 mg/kg) was administered 168 hours later 32. Figure 8 shows the plasma concentration time profile of TPPU during the time period of 0-216 hours following TPPU administration at 0 hour and TCPU at 168 hours. As shown in Figure 8, there is a clear second peak of TPPU right after TCPU administration at 168 hr, indicating that considerable amount of TPPU molecules were still bound to its target even 168 hours after the dose, and these bound TPPU molecules were displaced by TCPU32.

  3. Tracer interaction methods with a trace amount of radio labeled drug followed by a large dose of non-radio labeled (i.e. cold) drug. Another alternative method for TMDD identification is to give same drug, but first a trace amount of radio labeled drug is administered, then a large dose of non-radio labeled (i.e. cold) drug is given after the radio-labeled drug reaches elimination phase. A typical example is a tracer interaction experiment conducted by Veng-Pedersen’s group on rhEPO 43. In that study 43, an iv bolus injection of rhEPO tracer (125I-rhEPO) was injected at time zero. At 4.0 hr a high dose of non-radio labeled rhEPO was given via iv bolus injection, which resulted in a peak of the radio labeled rhEPO tracer (Figure 9). The rationale behind the tracer interaction experiment is the same as the in vivo displacement experiment.

Figure 8.

Figure 8.

Open in a new tab

In vivo displacement assay of potent sEH inhibitors TPPU and TCPU conducted in mice. TPPU, the tested compound, was given at 0.3 mg/kg at time 0 (subcutaneous injection). TCPU, the displacer, was given at 3 mg/kg at time 168 hr (subcutaneous injection). The concentrations of TPPU were measured during the time period of 0-216 hr. The 2nd peak of TPPU occurred right after TCPU administration, indicating the displacement of TPPU by TCPU at the sEH binding site. This figure was adapted from reference 32.

Figure 9.

Figure 9.

Open in a new tab

Tracer interaction method demonstrated in newborn lamb. Low dose of tracer 125I-rhEPO was injected at time zero. Large dose of nontracer rhEPO was injected at 4 hr, and resulted in a significant perturbation in the tracer level that diminished as the nontracer was eliminated. This figure was adapted from reference43.

Conclusions and Future Considerations

For compounds exhibiting TMDD in their therapeutic concentration ranges, it can be quite challenging in terms of study design, rational dose regimen selection, as well as establishing relationships between drug exposure and response. For compounds exhibiting TMDD, extra caution is needed when carrying out microdosing studies as TMDD may confound study results, leading to incorrect expectations for the PK at therapeutic doses. In addition, the substantial difference between the PK following the first dose and the subsequent dose(s), may lead to order/sequence effect and potentially could significantly influence the results in cross-over clinical studies, such as bioequivalence and bioavailability studies. To ensure the quality of those clinical studies, utilizing TMDD principles to select the appropriate dose(s) as well as sufficient wash out time will be important. As compounds exhibiting TMDD often demonstrate complicated nonlinear PK, the relationship between drug exposure and response is no longer intuitive and consequently the dose regimen design can be quite challenging. To optimize the dose regimen, it is highly valuable to utilize pharmacometric modeling approach to elucidate the quantitative relationship between drug exposure and response. The first TMDD mathematical model was proposed by Mager and Jusko in 200144. Since then, numerous papers on TMDD models have been published21, 40, 45-48. Dua et al published a tutorial on TMDD models and summarized the various TMDD models that have been developed since 2001 45. Mathematical modeling is beyond the scope of this review. Readers who are interested in TMDD models are encouraged to start with Mager and Jusko’s TMDD model paper, or the tutorial paper published by Dua et al 44, 45.

ACKNOWLEDGMENTS

The author gratefully acknowledges valuable comments from Dr. Peter Schotland, Dr. Xiaoyu Yan, and Dr. Brett Fleisher. Partially supported by National Institutes of Health (NIH) US Public Health Service Program Project Grant P01 HL046925.

References

  • 1.Levy G Pharmacologic target-mediated drug disposition. Clin Pharmacol Ther. 1994;56(3): 248–252. [DOI] [PubMed] [Google Scholar]
  • 2.Abraham AK, Kagan L, Kumar S, Mager DE. Type I interferon receptor is a primary regulator of target-mediated drug disposition of interferon-beta in mice. J Pharmacol Exp Ther.334(1): 327–332. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Buchwalder PA, Buclin T, Trinchard I, Munafo A, Biollaz J. Pharmacokinetics and pharmacodynamics of IFN-beta 1a in healthy volunteers. J Interferon Cytokine Res. 2000;20(10): 857–866. [DOI] [PubMed] [Google Scholar]
  • 4.Yan X, Lowe PJ, Fink M, Berghout A, Balser S, Krzyzanski W. Population Pharmacokinetic and Pharmacodynamic Model-Based Comparability Assessment of a Recombinant Human Epoetin Alfa and the Biosimilar HX575. J Clin Pharmacol. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.El-Komy MH, Widness JA, Veng-Pedersen P. Pharmacokinetic analysis of continuous erythropoietin receptor activator disposition in adult sheep using a target-mediated, physiologic recirculation model and a tracer interaction methodology. Drug metabolism and disposition: the biological fate of chemicals. 2011;39(4): 603–609. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Jin F, Krzyzanski W. Pharmacokinetic model of target-mediated disposition of thrombopoietin. The AAPS journal. 2004;6(1): 86–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Kakkar T, Sung C, Gibiansky L, et al. Population PK and IgE pharmacodynamic analysis of a fully human monoclonal antibody against IL4 receptor. Pharm Res. 2011;28(10): 2530–2542. [DOI] [PubMed] [Google Scholar]
  • 8.Meijer RT, Koopmans RP, ten Berge IJ, Schellekens PT. Pharmacokinetics of murine anti-human CD3 antibodies in man are determined by the disappearance of target antigen. J Pharmacol Exp Ther. 2002;300(1): 346–353. [DOI] [PubMed] [Google Scholar]
  • 9.Bauer RJ, Gibbons JA, Bell DP, Luo ZP, Young JD. Nonlinear pharmacokinetics of recombinant human macrophage colony-stimulating factor (M-CSF) in rats. J Pharmacol Exp Ther. 1994;268(1): 152–158. [PubMed] [Google Scholar]
  • 10.Schropp J, Khot A, Shah DK, Koch G. Target-Mediated Drug Disposition Model for Bispecific Antibodies: Properties, Approximation, and Optimal Dosing Strategy. CPT: pharmacometrics & systems pharmacology. 2019;8(3): 177–187. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.An G, Liu W, Dutta S. Small-molecule compounds exhibiting target-mediated drug disposition - A case example of ABT-384. J Clin Pharmacol. 2015;55(10): 1079–1085. [DOI] [PubMed] [Google Scholar]
  • 12.An G Small-Molecule Compounds Exhibiting Target-Mediated Drug Disposition (TMDD): A Minireview. J Clin Pharmacol. 2017;57(2): 137–150. [DOI] [PubMed] [Google Scholar]
  • 13.Retlich S, Withopf B, Greischel A, Staab A, Jaehde U, Fuchs H. Binding to dipeptidyl peptidase-4 determines the disposition of linagliptin (BI 1356)--investigations in DPP-4 deficient and wildtype rats. Biopharm Drug Dispos. 2009;30(8): 422–436. [DOI] [PubMed] [Google Scholar]
  • 14.Yamazaki S, Shen Z, Jiang Y, Smith BJ, Vicini P. Application of target-mediated drug disposition model to small molecule heat shock protein 90 inhibitors. Drug metabolism and disposition: the biological fate of chemicals. 2013;41(6): 1285–1294. [DOI] [PubMed] [Google Scholar]
  • 15.An G, Liu W, Katz DA, Marek GJ, Awni W, Dutta S. Population pharmacokinetics of the 11beta-hydroxysteroid dehydrogenase type 1 inhibitor ABT-384 in healthy volunteers following single and multiple dose regimens. Biopharm Drug Dispos. 2014;35(7): 417–429. [DOI] [PubMed] [Google Scholar]
  • 16.Smith DA, van Waterschoot RAB, Parrott NJ, Olivares-Morales A, Lave T, Rowland M. Importance of target-mediated drug disposition for small molecules. Drug Discov Today. 2018;23(12): 2023–2030. [DOI] [PubMed] [Google Scholar]
  • 17.van Waterschoot RAB, Parrott NJ, Olivares-Morales A, Lave T, Rowland M, Smith DA. Impact of target interactions on small-molecule drug disposition: an overlooked area. Nature reviews Drug discovery. 2018;17(4): 299. [DOI] [PubMed] [Google Scholar]
  • 18.Bach T, Jiang Y, Zhang X, An G. General Pharmacokinetic Features of Small-Molecule Compounds Exhibiting Target-Mediated Drug Disposition (TMDD): A Simulation-Based Study. J Clin Pharmacol. 2019;59(3): 394–405. [DOI] [PubMed] [Google Scholar]
  • 19.D’Cunha R, Schmidt R, Widness JA, et al. Target-mediated disposition population pharmacokinetics model of erythropoietin in premature neonates following multiple intravenous and subcutaneous dosing regimens. European journal of pharmaceutical sciences : official journal of the European Federation for Pharmaceutical Sciences. 2019: 105013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.D’Cunha R, Widness JA, Yan X, Schmidt RL, Veng-Pedersen P, An G. A Mechanism-Based Population Pharmacokinetics Model of Erythropoietin in Premature Infants and Healthy Adults Following Multiple Intravenous Doses. J Clin Pharmacol. 2019;59(6): 835–846. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Ng CM, Stefanich E, Anand BS, Fielder PJ, Vaickus L. Pharmacokinetics/pharmacodynamics of nondepleting anti-CD4 monoclonal antibody (TRX1) in healthy human volunteers. Pharm Res. 2006;23(1): 95–103. [DOI] [PubMed] [Google Scholar]
  • 22.Levy G, Mager DE, Cheung WK, Jusko WJ. Comparative pharmacokinetics of coumarin anticoagulants L: Physiologic modeling of S-warfarin in rats and pharmacologic target-mediated warfarin disposition in man. J Pharm Sci. 2003;92(5): 985–994. [DOI] [PubMed] [Google Scholar]
  • 23.Brazzell RK, Mayer PR, Dobbs R, McNamara PJ, Teng RL, Slattery JT. Dose-dependent pharmacokinetics of the aldose reductase inhibitor imirestat in man. Pharm Res. 1991;8(1): 112–118. [DOI] [PubMed] [Google Scholar]
  • 24.Till AE, Gomez HJ, Hichens M, et al. Pharmacokinetics of repeated single oral doses of enalapril maleate (MK-421) in normal volunteers. Biopharm Drug Dispos. 1984;5(3): 273–280. [DOI] [PubMed] [Google Scholar]
  • 25.Lees KR, Kelman AW, Reid JL, Whiting B. Pharmacokinetics of an ACE inhibitor, S-9780, in man: evidence of tissue binding. Journal of pharmacokinetics and biopharmaceutics. 1989;17(5): 529–550. [DOI] [PubMed] [Google Scholar]
  • 26.Francis RJ, Brown AN, Kler L, et al. Pharmacokinetics of the converting enzyme inhibitor cilazapril in normal volunteers and the relationship to enzyme inhibition: development of a mathematical model. J Cardiovasc Pharmacol. 1987;9(1): 32–38. [PubMed] [Google Scholar]
  • 27.Laine K, Anttila M, Huupponen R, Maki-Ikola O, Heinonen E. Multiple-dose pharmacokinetics of selegiline and desmethylselegiline suggest saturable tissue binding. Clin Neuropharmacol. 2000;23(1): 22–27. [DOI] [PubMed] [Google Scholar]
  • 28.Weber C, Schmitt R, Birnboeck H, et al. Pharmacokinetics and pharmacodynamics of the endothelin-receptor antagonist bosentan in healthy human subjects. Clin Pharmacol Ther. 1996;60(2): 124–137. [DOI] [PubMed] [Google Scholar]
  • 29.Retlich S, Duval V, Graefe-Mody U, Jaehde U, Staab A. Impact of target-mediated drug disposition on Linagliptin pharmacokinetics and DPP-4 inhibition in type 2 diabetic patients. J Clin Pharmacol. 2010;50(8): 873–885. [DOI] [PubMed] [Google Scholar]
  • 30.Landersdorfer CB, He YL, Jusko WJ. Mechanism-based population pharmacokinetic modelling in diabetes: vildagliptin as a tight binding inhibitor and substrate of dipeptidyl peptidase IV. British journal of clinical pharmacology. 2012;73(3): 391–401. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Bellaire S, Walzer M, Wang T, Krauwinkel W, Yuan N, Marek GJ. Safety, Pharmacokinetics, and Pharmacodynamics of ASP3662, a Novel 11beta-Hydroxysteroid Dehydrogenase Type 1 Inhibitor, in Healthy Young and Elderly Subjects. Clinical and translational science. 2019;12(3): 291–301. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Lee KSS, Yang J, Niu J, et al. Drug-Target Residence Time Affects in Vivo Target Occupancy through Multiple Pathways. ACS Cent Sci. 2019;5(9): 1614–1624. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Yang BB. Integration of Pharmacokinetics and Pharmacodynamics into the Drug Development of Pegfilgrastim, a Pegylated Protein. Chapter 15 of book “Pharmcokinetics and Pharmacodynamics of Biotech Drugs”. 2006: 373–393. [Google Scholar]
  • 34.Terashi K, Oka M, Ohdo S, et al. Close association between clearance of recombinant human granulocyte colony-stimulating factor (G-CSF) and G-CSF receptor on neutrophils in cancer patients. Antimicrobial agents and chemotherapy. 1999;43(1): 21–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Kotto-Kome AC, Fox SE, Lu W, Yang BB, Christensen RD, Calhoun DA. Evidence that the granulocyte colony-stimulating factor (G-CSF) receptor plays a role in the pharmacokinetics of G-CSF and PegG-CSF using a G-CSF-R KO model. Pharmacological research. 2004;50(1): 55–58. [DOI] [PubMed] [Google Scholar]
  • 36.Johnston E, Crawford J, Blackwell S, et al. Randomized, dose-escalation study of SD/01 compared with daily filgrastim in patients receiving chemotherapy. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2000;18(13): 2522–2528. [DOI] [PubMed] [Google Scholar]
  • 37.Chapel S, Veng-Pedersen P, Hohl RJ, Schmidt RL, McGuire EM, Widness JA. Changes in erythropoietin pharmacokinetics following busulfan-induced bone marrow ablation in sheep: evidence for bone marrow as a major erythropoietin elimination pathway. J Pharmacol Exp Ther. 2001;298(2): 820–824. [PubMed] [Google Scholar]
  • 38.Widness JA, Schmidt RL, Hohl RJ, et al. Change in erythropoietin pharmacokinetics following hematopoietic transplantation. Clin Pharmacol Ther. 2007;81(6): 873–879. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Chapel SH, Veng-Pedersen P, Schmidt RL, Widness JA. Receptor-based model accounts for phlebotomy-induced changes in erythropoietin pharmacokinetics. Experimental hematology. 2001;29(4): 425–431. [DOI] [PubMed] [Google Scholar]
  • 40.Peletier LA, Gabrielsson J. Dynamics of target-mediated drug disposition: characteristic profiles and parameter identification. J Pharmacokinet Pharmacodyn. 2012;39(5): 429–451. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Dirks NL, Meibohm B. Population pharmacokinetics of therapeutic monoclonal antibodies. Clinical pharmacokinetics. 2010;49(10): 633–659. [DOI] [PubMed] [Google Scholar]
  • 42.Fuchs H, Tillement JP, Urien S, Greischel A, Roth W. Concentration-dependent plasma protein binding of the novel dipeptidyl peptidase 4 inhibitor BI 1356 due to saturable binding to its target in plasma of mice, rats and humans. The Journal of pharmacy and pharmacology. 2009;61(1): 55–62. [DOI] [PubMed] [Google Scholar]
  • 43.Veng-Pedersen P, Widness JA, Wang J, Schmidt RL. A tracer interaction method for nonlinear pharmacokinetics analysis: application to evaluation of nonlinear elimination. Journal of pharmacokinetics and biopharmaceutics. 1997;25(5): 569–593. [DOI] [PubMed] [Google Scholar]
  • 44.Mager DE, Jusko WJ. General pharmacokinetic model for drugs exhibiting target-mediated drug disposition. J Pharmacokinet Pharmacodyn. 2001;28(6): 507–532. [DOI] [PubMed] [Google Scholar]
  • 45.Dua P, Hawkins E, van der Graaf PH. A Tutorial on Target-Mediated Drug Disposition (TMDD) Models. CPT: pharmacometrics & systems pharmacology. 2015;4(6): 324–337. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Yan X, Mager DE, Krzyzanski W. Selection between Michaelis-Menten and target-mediated drug disposition pharmacokinetic models. J Pharmacokinet Pharmacodyn. 2010;37(1): 25–47. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Mager DE, Krzyzanski W. Quasi-equilibrium pharmacokinetic model for drugs exhibiting target-mediated drug disposition. Pharm Res. 2005;22(10): 1589–1596. [DOI] [PubMed] [Google Scholar]
  • 48.Gibiansky L, Gibiansky E, Kakkar T, Ma P. Approximations of the target-mediated drug disposition model and identifiability of model parameters. J Pharmacokinet Pharmacodyn. 2008;35(5): 573–591. [DOI] [PubMed] [Google Scholar]

RESOURCES